Curriculum Vitae Rebecca L

Home , Intrauterine device

Drug Policy Implications of Inhaled Cannabis: Driving Skills and Subjective Effects, Vaporized Cannabinoid Pharmacokinetics, and Interactions with Alcohol

Item Type dissertation

Authors Hartman, Rebecca Lynn

Publication Date 2015

Abstract Increasing medical and recreational cannabis legalization and shifting public attitudes are accompanied by increased driving under the influence of cannabis (DUIC) cases and changing administration routes, including vaporization. Cannabis' driving ef...

Keywords oral fluid; vaporization; Alcohol; Blood; Cannabis; Driving Under the Influence; Marijuana Smoking; Volatilization

Download date 11/10/2021 06:37:59

Link to Item http://hdl.handle.net/10713/4585 CURRICULUM VITAE

REBECCA L. HARTMAN

CONTACT

Email: [email protected] [email protected]

EDUCATION

Doctor of Philosophy, Toxicology August 2010 – May 2015 Graduate Program in Life Sciences, School of Medicine University of Maryland Baltimore, Baltimore, MD Advisor: Marilyn A. Huestis, PhD Dissertation: “Drug Policy Implications of Inhaled Cannabis: Driving Skills and Psychoactive Effects, Vaporized Pharmacokinetic Disposition, and Interactions with Alcohol”

Bachelor of Arts, cum laude, Chemistry August 2004 – January 2008 College of Arts and Sciences Cornell University, Ithaca, NY

PROFESSIONAL EXPERIENCE

Pre-doctoral Fellow August 2010 – present Chemistry and Drug Metabolism, Intramural Research Program, National Institute on Drug Abuse, NIH, Baltimore, MD

Toxicologist I March 2010 – August 2010 Forensic Toxicology Laboratory Monroe County Office of the Medical Examiner, Rochester, NY

Toxicology Intern January 2008 – March 2010 Forensic Toxicology Laboratory Monroe County Office of the Medical Examiner, Rochester, NY

Student Research Assistant September 2006 – December 2007 Advisor: J. Tom Brenna, PhD Undergraduate research on anabolic steroids in urine, in support of anti-doping Cornell University, Ithaca, NY

Teaching Assistant August 2006 – December 2006 Department of Chemistry, College of Arts and Sciences Cornell University, Ithaca, NY

Summer Laboratory Aide May 2007 – August 2007 Monroe County Environmental Laboratory, Rochester, NY May 2006 – August 2006

PROFESSIONAL AFFILIATIONS

Society of Forensic Toxicologists Student Member 2010 – present The International Association of Forensic Toxicologists Member 2010 – present Editor, Social Media Committee 2013 – present American Academy of Forensic Sciences, Toxicology Associate Member 2015 – present Student Member 2013 – 2015

HONORS

June K. Jones Scholarship, American Academy of Forensic Sciences 2015 Forensic Sciences Foundation Student Scholarship, 2015 American Academy of Forensic Sciences Educational Research Award, Society of Forensic Toxicologists 2013 Annual Merit Scholarships, Cornell Alumni Association of Rochester 2005, 2006, 2007

EDITORIAL TASKS

Invited Reviewer for Journal of Analytical Toxicology

INVITED PRESENTATIONS

R.L. Hartman* and M.A. Huestis. Cannabis Effects on Driving Skills. The International Association of Forensic Toxicologists Young Forensic Toxicologists’ Young Scientists Symposium, Funchal, Madeira, Portugal, September 2013.

R.L. Hartman* and M.A. Huestis. Cannabis Effects on Driving Skills. Centre of Forensic Sciences annual Drugs and Driving Symposium, Toronto, ON, June 2013.

PUBLICATIONS

R.L. Hartman, T. Brown, G. Milavetz, A. Spurgin, DA Gorelick, G. Gaffney, and M.A. Huestis. Controlled Cannabis Vaporizer Administration: Blood and Plasma Cannabinoids With and Without Alcohol. Clin Chem 2015. In press.

M.S. Castaneto, A. Wohlfarth, N.A. Desrosiers, R.L. Hartman, D.A. Gorelick, and M.A. Huestis. Synthetic Cannabinoids: pharmacokinetics and detection methods in biological matrices. Drug Metab Rev 2015. doi: 10.3109/03602532.2015.1029635

R.L. Hartman, S. Anizan, M. Jang, T.L. Brown, K. Yun, D.A. Gorelick, G. Milavetz, A. Spurgin, G. Gaffney, and M.A. Huestis. Cannabinoid disposition in oral fluid after controlled vaporizer administration with and without alcohol. Forensic Toxicol 2015. Epub 2015 doi: 10.1007/s11419-015-0269-6

M.S. Castaneto, D.A. Gorelick, N.A. Desrosiers, R.L. Hartman, S. Pirard, and M.A. Huestis. Synthetic Cannabinoids: Epidemiology, Pharmacodynamics, and Clinical Implications. Drug Alcohol Depend 2014. 144:12-41. Epub 2014 doi: 10.1016/j.drugalcdep.2014.08.005

R.L. Hartman, N.A. Desrosiers, A.J. Barnes, Keming Yun, K.B. Scheidweiler, E.A. Kolbrich-Spargo, D.A. Gorelick, R.S. Goodwin, and M.A. Huestis. 3,4- Methylenedioxymethamphetamine (MDMA) and Metabolite Disposition in Blood and Plasma Following Controlled Oral Administration. Anal Bioanal Chem 2014. 406(2): 587-99. Epub 2013 doi: 10.1007/s00216-013-7468-y

N.A. Desrosiers, A.J. Barnes, R.L. Hartman, K.B. Scheidweiler, E.A. Kolbrich-Spargo, D.A. Gorelick, R.S. Goodwin, and M.A. Huestis. Oral fluid and plasma 3,4- methylenedioxymethamphetamine (MDMA) and metabolite correlation after controlled oral MDMA administration. Anal Bioanal Chem 2013. 405(12): 4067-76. Epub 2013 doi: 10.1007/s00216-013-6848-7

R.L. Hartman and M.A. Huestis. Cannabis Effects on Driving Skills. Clin Chem 2013. 59(3): 478-92. Epub 2012 Dec 7 doi: 10.1373/clinchem.2012.194381

J.M. Beno, R. Hartman, C.Wallace, D.Nemeth, and S. LaPoint. Homicidal Methanol Poisoning in a Child. Journal of Analytical Toxicology. J Anal Toxicol 2011. 35 (7): 524-528.

CONTINUING EDUCATION / TRAINING

Introduction to Uncertainty in Forensic Chemistry and Toxicology 9 June 2014 1, 2, & 3 (online), Research Triangle Institute International American College of Medical Toxicology Seminars in 9 May 2011 Forensic Toxicology: Stimulants, Arlington, VA Expert Witness Testimony, Albany, NY 14-15 September 2009 Vehicle and Traffic Law for Toxicologists, Albany, NY 8-9 September 2008

ABSTRACTS

R.L. Hartman*, M. Jang, A. Spurgin, K. Yun, D.A. Gorelick, G. Milavetz, T.L. Brown, G. Gaffney and M.A. Huestis. Cannabinoid Disposition in Oral Fluid after Controlled Cannabis Vaporizer Administration. American Academy of Forensic Sciences annual conference, Orlando, FL, February 2015.

R.L. Hartman*, T.L. Brown, A. Spurgin, D.A. Gorelick, G. Milavetz, G. Gaffney, and M.A. Huestis. Cannabis and Low-Dose Alcohol Effects on Standard Deviation of Lateral Position (Lane Weaving) in Simulated Driving After Controlled Administration. The International Association of Forensic Toxicologists annual conference, Buenos Aires, Argentina, November 2014.

R.L. Hartman*, S. Anizan, M. Jang, T.L. Brown, K. Yun, D.A. Gorelick, G. Milavetz, G. Gaffney, and M.A. Huestis. Draeger® DrugTest 5000 On-Site Oral Fluid Cannabinoid Screening Performance after Cannabis Vaporization. Society of Forensic Toxicologists annual conference, Grand Rapids, MI, October 2014.

R.L. Hartman*, D.A. Gorelick, T.L. Brown, R. Compton, D. Smither, G. Gaffney, and M.A. Huestis. Cannabinoids Disposition in Blood Following Controlled Cannabis Administration by Volcano® Vaporizer. Society of Forensic Toxicologists annual conference, Orlando, FL, October/November 2013.

R.L. Hartman*, D. Nemeth, D. Nemeth, M. Salamone, P. Schantz, J. Fellows, K. Mahoney, L. Kadlec, M.A. Huestis, and J.M. Beno. Drug Recognition Expert (DRE) Success Rates in Western New York. The International Association of Forensic Toxicologists annual conference, Funchal, Madeira, Portugal, September 2013.

M.A. Huestis*, A. Wohlfarth*, R.L. Hartman*, and N.A. Desrosiers*. Current Research Initiatives in Toxicology at the National Institute on Drug Abuse. [Presentation entitled Driving Under the Influence of Cannabis in Western New York.] American Academy of Forensic Sciences annual conference, Washington, DC, February 2013.

R.L. Hartman* and M.A. Huestis. The Effects of Cannabis on Driving Skills. Society of Forensic Toxicologists annual conference, Boston, MA, July 2012.

R.L. Hartman*, N.A. Desrosiers, A.J. Barnes, E.A. Kolbrich-Spargo, K.B. Scheidweiler, D.A. Gorelick, R.S. Goodwin, E.A. Stein, and M.A. Huestis. Acute Oral 3,4- Methylenedioxymethamphetamine (MDMA) Effects on Immediate, Short Term and Delayed Memory. Society of Forensic Toxicologists-The International Association of Forensic Toxicologists joint annual conference, San Francisco, CA, September 2011.

N.A. Desrosiers*, A.J. Barnes, R.L. Hartman, K.B. Scheidweiler, E.A. Kolbrich-Spargo, D.A. Gorelick, R.S. Goodwin, and M.A. Huestis. Oral Fluid and Plasma 3,4- Methylenedioxymethamphetamine (MDMA) and Metabolite Correlation after Controlled Oral MDMA Administration. Society of Forensic Toxicologists-The International Association of Forensic Toxicologists joint annual conference, San Francisco, CA, September 2011.

J. Beno*, E. Baker, D. Nemeth, and R. Hartman. Detection of driving under the influence of cocaine in a hit and run accident based on the analysis of blood smears on the deployed airbag using LC/MS/MS technology. Society of Forensic Toxicologists annual conference, Oklahoma City, OK, October 2009.

EDITORIALS

R.L. Hartman and M.A. Huestis. Letter to the Editor regarding “Trends in Alcohol and Other Drugs Detected in Fatally Injured Drivers in the United States, 1999-2010”. Am J Epidemiology 2014. 180(8):862-3. doi: 10.1093/aje/kwu251

Abstract

Title: Drug Policy Implications of Inhaled Cannabis: Driving Skills and Subjective

Effects, Vaporized Cannabinoid Pharmacokinetics, and Interactions with Alcohol

Rebecca L. Hartman, Doctor of Philosophy, 2015

Dissertation Directed by: Professor Dr. Dr. (h.c.) Marilyn A. Huestis, Chief, Chemistry and Drug Metabolism, National Institute on Drug Abuse, National Institutes of Health

Increasing medical and recreational cannabis legalization and shifting public attitudes are accompanied by increased driving under the influence of cannabis (DUIC) cases and changing administration routes, including vaporization. Cannabis’ driving effects are poorly understood, and cannabis driving per se laws heavily debated— especially when considering blood collection delays. Cannabis and alcohol are frequently encountered together in drugged driving cases, but interactive effects are not fully elucidated. Oral fluid (OF) is an advantageous alternative matrix for documenting cannabis exposure, with interest in correlating cannabis’ effects with OF cannabinoid concentrations.

This research aimed to 1) evaluate cannabis’ effect on driving lateral control relative to blood ∆9-tetrahydrocannabinol (THC) concentrations, with and without alcohol; 2) assess vaporized cannabis’ subjective effects and pharmacokinetic disposition in blood, plasma, and OF, with and without alcohol. Current cannabis smokers drank placebo or low-dose alcohol and inhaled vaporized placebo, low (2.9%), or high (6.7%)-

THC vaporized bulk cannabis (6 conditions, within-subject). Participants drove 45min

simulated drives 0.5h post-dose. Subjective effects, blood and OF, and breath alcohol concentration (BrAC) were measured at baseline and up to 8.3h post-dose.

Cannabis increased standard deviation of lateral position (SDLP, lane weaving) but not maximum lateral acceleration or lane departures/min. BrAC increased all three; cannabis-alcohol SDLP effects were additive rather than synergistic. Cannabinoids’ time courses and subjective effects exhibited patterns similar to those after smoking, with

THC maximum concentration (Cmax) occurring immediately post-dose and decreasing rapidly thereafter, then subsequent slower excretion. Approximately half of participants self-titrated doses to individual comfort levels via inhalation technique, resulting in similar THC Cmax after low and high doses. Blood THC concentrations 7-10μg/L during driving increased SDLP similarly to 0.05g/210L [impairing] BrAC; 20μg/L, more than

0.08g/210L [US illegal BrAC]. Because SDLP effects were additive, 5μg/L

THC+0.05g/210L alcohol produced similar SDLP to 0.08g/210L. Blood THC during driving generally is much higher than at time of blood draw in authentic DUIC cases. OF cannabinoids documented intake, but THC concentration variability limited interpretation. Data demonstrate significant cannabis effects on driving, provide blood and OF concentrations after vaporized cannabis, and will benefit clinicians and policymakers by improving DUIC and clinical interpretation.

Drug Policy Implications of Inhaled Cannabis: Driving Skills and Subjective Effects, Vaporized Cannabinoid Pharmacokinetics, and Interactions with Alcohol

by Rebecca L. Hartman

Dissertation submitted to the Faculty of the Graduate School of the University of Maryland, Baltimore in partial fulfillment of the requirements for the degree of Doctor of Philosophy 2015

Acknowledgements

This minor section of my dissertation was actually the longest in writing. The lustrum of academic and scientific growth that culminated in this document was fraught with more challenges and opportunities than I could possibly have foreseen. This journey’s successful completion would not have been possible without the tireless efforts, unconditional support, and wisdom of many individuals to whom I am exceedingly grateful.

First and foremost, I would like to thank my principal advisor, Dr. Marilyn Huestis. Your steadfast guidance over the years has helped shape the scientist I am today. Under your leadership I improved as a researcher, as a presenter, and as a writer. Your support and encouragement gave me the tools I needed to build a strong reputation as a forensic toxicologist. We as your students may never know just how much you worked on our behalf, but I will always remember this experience with pride and gratitude.

I also need to thank my first mentor, Dr. Jeanne Beno. You were the person who made me want to be a toxicologist. Your intelligence, strength and charisma served as constant sources of inspiration. My time in your laboratory laid the foundation for the toxicologist I have become, and set me on the path to this doctoral work. I am genuinely honored you agreed to serve on my committee, and have benefited greatly from your continued guidance.

I offer sincere gratitude to Dr. Katherine Squibb, whose kindly direction helped lead me through countless complex university practices. You were always incredibly helpful, regardless of the issue. Learning toxicology from you was a genuine pleasure. The Toxicology program is lucky to have you as its director, and I am no exception.

Thank you also to Drs. Barry Levine and Kenneth Bauer. I have learned so much from you. Your insights have been invaluable throughout this process, and your courses instilled crucial knowledge without which this research could not be completed.

I owe immense thanks to all of my collaborators at NHTSA and particularly the University of Iowa. This project would never have reached fruition without your diligent efforts and patient dedication during endless regulatory document preparation, 16 months of primary study conduct, and unceasingly throughout ongoing data analysis. Thank you to the University of Iowa CRU and NADS staff, especially Rose Schmitt, Cheryl Roe, and Jennifer Henderson. Special thanks to Drs. Tim Brown, Andy Spurgin, Gary Milavetz and Gary Gaffney, for contributing your time and excellent insights to this project. Thank you also to Drs. Dereece Smither and Richard Compton for all of your input and support. It has been a privilege to work with you all.

The NIDA Chemistry and Drug Metabolism Section contributed greatly to this achievement and to my own development as a scientist. To my immediate supervisors, Dr. Karl Scheidweiler, Allan Barnes, and Dr. Marta Concheiro-Guisan, I could never repay all the time, feedback and criticism, advice, expertise, ideas and knowledge you so

iii generously imparted despite your own busy schedules. I have learnt so much from all of you, and am immeasurably appreciative of all you have done. Thank you also to past and current colleagues Drs. Sébastien Anizan, Mateus Bergamaschi, David Gorelick, Moonhee Jang, Garry Milman, Sandrine Pirard, Madeleine Swortwood and especially Ariane Wohlfarth. To my past and present contemporary NIDA graduate students, Marisol Castaneto, Kayla Ellefsen, and Drs. Sarah Himes, Erin Karschner, and Dayong Lee, it was an honor to share this experience with you. I am so grateful to all of you for being such exemplary colleagues and friends.

Two in particular deserve special recognition: Dr. Nathalie Desrosiers—dear friend; 4 years’ flatmate; colleague; fellow NIDA graduate student; and ultimately, sister—and Matt Newmeyer, the friend and confidant on whom I could always rely, the person whose sense of humor made days of misery in -20°C freezers seem fun, and the one who just…understood.

Finally but perhaps most importantly, to all the friends and family that walked this path with me: I can never express how much your love and support has meant to me. To Erika, Lita, and Ger, I could not ask for better friends. Knowing you were there for me made a world of difference. John, the extent of your kindness and the size of your heart never fail to amaze me. I am incredibly fortunate to count you among my closest friends. Andrew, I have spent years contemplating what I would write here; how ironic that despite all the dalliances into philology, I now find myself at a loss for words. I echo you from a time early in our friendship: Thanks. Really, just thanks. You helped me keep the faith in the sunlit uplands when I needed it most. Mom and Dad, you far surpass any daughter’s wildest dreams. You were always there with a friendly voice and a kindly word of sympathy, advice or joy; always ready to twist yourselves into pretzels to help me in every way imaginable. It is because of your love that I am who I am today, and your unfailing support was indispensable in realizing this accomplishment. This dissertation belongs to you.

My deepest gratitude to you all.

R. Hartman

Table of Contents

Acknowledgements ...... iii

Table of Contents ...... v

List of Tables ...... x

List of Figures ...... xiv

List of Abbreviations ...... xvi

Chapter 1 – Introduction ...... 1

Driving Under the Influence of Cannabis (DUIC) ...... 1 Knowledge Gap: Cannabis’ Effects on Driving ...... 3 Knowledge Gap: Cannabis and Alcohol Interactive Effects on Driving ...... 4 Aims and Hypotheses ...... 5 Medical and Recreational Cannabis ...... 6 Vaporized Cannabis ...... 7 Knowledge Gaps...... 10 Aims and Hypotheses ...... 10 Cannabis Pharmacokinetics...... 11 Blood and Plasma ...... 12 Oral fluid (OF) ...... 13 Knowledge Gaps...... 14 Aims and Hypotheses ...... 15

Chapter 2 – Cannabis Effects on Driving Skills ...... 17

Abstract ...... 17 Introduction ...... 18 Objective and Search Methods ...... 19 DUIC: Epidemiological Data ...... 19 DUIC: Experimental Data ...... 30 Combined Alcohol and Cannabis Intake ...... 37 Preventing DUIC ...... 41 Conclusions/Discussion ...... 42

Chapter 3 – Protocol: Effect of Inhaled Cannabis on Driving Performance ...... 78

Précis: ...... 78 Background ...... 80 Rationale: Driving Under the Influence of Cannabis (DUIC) and Alcohol (DUIA) 80 Cannabis’ Subjective Effects ...... 88 Cannabis’ Neurocognitive effects ...... 89 Study Objectives and Hypotheses ...... 90

Subjects ...... 90 Inclusion criteria ...... 91 Exclusion criteria ...... 92 Study Design and Methods ...... 93 Study overview ...... 93 Recruitment ...... 94 Screening methods ...... 94 Study design ...... 96 Study procedures in chronological order ...... 100 Drug Administration ...... 104 Simulated Drive in the NADS ...... 106 Assessment of Risk Perception, Risk Taking, and Impulsivity Traits ...... 108 Assessment of Cannabinoid/Alcohol Intoxication ...... 111 Specimen Collection and Analysis ...... 114 Specimen and Data Storage ...... 115 Follow-up/termination procedures ...... 117 IND/IDE Requirements ...... 118 Risks/ discomforts ...... 118 Outcome measures ...... 123 Statistical Analysis ...... 123 Power Analysis ...... 124 Human Subjects Protection ...... 125 Equity of Subject Selection ...... 125 Exclusion Justification ...... 126 Qualifications of Investigators ...... 126 Data and safety monitoring ...... 129 Adverse event reporting ...... 129 Alternatives to participation or alternative therapies ...... 130 Confidentiality ...... 130

Chapter 4 – Cannabis Effects on Driving Lateral Control With and Without Alcohol .. 132

Abstract ...... 132 1. Introduction ...... 133 2. Methods ...... 135 2.1 Participants ...... 135 2.2 Study Design/Procedures ...... 136 2.3 National Advanced Driving Simulator ...... 137 2.4 Drives...... 137 2.5 Specimen Analysis ...... 139 2.6 Data Analysis ...... 139 3. Results ...... 140 3.1 Participants ...... 140 3.2 Driving ...... 141 3.3 Pre- and Post-drive Blood and OF THC Concentrations ...... 146 4. Discussion ...... 147 4.1 Strengths and limitations ...... 152

5. Conclusion ...... 154

Chapter 5 – Controlled Vaporized Cannabis, With and Without Alcohol: Subjective Effects and Oral Fluid-Blood Cannabinoid Relationships...... 158

Abstract ...... 158 Introduction ...... 159 Methods ...... 160 Participants ...... 160 Study Design...... 161 Specimen Analysis ...... 162 Data Analysis ...... 163 Results ...... 164 Participants ...... 164 Subjective effects ...... 164 OF/Blood and OF/Plasma ...... 173 Discussion ...... 173 Study strengths and limitations...... 183 Conclusion ...... 184

Chapter 6 – Controlled Cannabis Vaporizer Administration: Blood and Plasma Cannabinoids With and Without Alcohol ...... 208

Abstract ...... 208 Introduction ...... 209 Methods ...... 210 Participants ...... 210 Study Design...... 211 Blood and Plasma Analysis ...... 212 Data Analysis ...... 212 Results ...... 213 Participants ...... 213 Blood and plasma cannabinoids ...... 216 Blood/Plasma ratios ...... 232 THCCOOH-glucuronide/THCCOOH ratios ...... 232 Discussion ...... 233

Chapter 7 – Cannabinoid disposition in oral fluid after controlled vaporizer administration with and without alcohol...... 274

Abstract ...... 274 Introduction ...... 275 Materials and Methods ...... 277 Participants ...... 277 Study design ...... 278 Specimen analysis...... 279 Data Analysis ...... 280

vii

Results and Discussion ...... 281 Participants ...... 281 Alcohol ...... 284 QuantisalTM OF cannabinoids ...... 288 Dräger DrugTest® 5000 performance and confirmation comparison ...... 295 Discussion ...... 296 Conclusions ...... 310

Chapter 8 – Conclusions and Future Research ...... 322

Research Summary ...... 322 Research Findings and Conclusions ...... 324 Does cannabis affect lateral control during driving, and at what blood THC concentrations? ...... 324 How does concurrent alcohol interact with cannabis’ effects on driving lateral control? ...... 327 Is cannabis vaporization an effective method of inhaled THC delivery, producing expected subjective effects? ...... 328 How does alcohol interact with cannabis’ subjective effects? ...... 329 Are vaporized cannabis’ blood, plasma, and OF cannabinoid dispositions and pharmacokinetics similar to those observed after smoking? ...... 329 Do cannabis and alcohol interactions alter cannabinoid or alcohol pharmacokinetic dispositions? ...... 331 Does cannabis vaporization or alcohol significantly alter on-site OF test performance or detection windows? ...... 332 Can OF directly predict blood or plasma cannabinoid concentrations? ...... 333 Does alcohol alter OF/blood and OF/plasma cannabinoid relationships? ...... 333 Do OF cannabinoid concentrations correlate to cannabis’ pharmacodynamic effects? ...... 334 Future Directions ...... 334 Finis ...... 335 Additional Research ...... 336

Appendix A ...... 337

Appendix B ...... 350

Appendix C ...... 352

(MDMA) and Metabolite Correlation after Controlled Oral MDMA Administration ...... 352 Synthetic Cannabinoids: Epidemiology, Pharmacodynamics and Clinical Implications ...... 353 Synthetic cannabinoids pharmacokinetics and detection methods in biological matrices ...... 354

Protocol Appendices ...... 356

viii

Protocol Appendix 1. Telephone screening questions ...... 356 Protocol Appendix 2: Driving Survey ...... 361 Protocol Appendix 3: Realism Survey ...... 371 Protocol Appendix 4: Neuromotor Examination ...... 373 Protocol Appendix 5: Risk Perception Questionnaire with Randomization Instructions ...... 374 Protocol Appendix 6: Self-Assessments of Risk Perception, Risk-Taking/Impulsivity and Sensation-Seeking ...... 376 Protocol Appendix 7: Impulsive Sensation-Seeking Subscale from the Zuckerman- Kuhlman Personality Questionnaire...... 377 Protocol Appendix 8: Barratt Impulsiveness Scale (Version 11) ...... 378 Protocol Appendix 9: Subjective Effects ...... 379 Protocol Appendix 10: Diagram of the Balloon Analogue Risk Task (BART) ...... 380 Protocol Appendix 11. Quantity-Frequency-Variability (QFV) scale ...... 381 Protocol Appendix 12. Phone Screening QFV...... 382 Protocol Appendix 13. AUDIT Survey ...... 384 Protocol Appendix 14. CUDIT ...... 385 Protocol Appendix 15.1 Columbia-Suicide Severity Rating Scale (C-SSRS) ...... 386 Protocol Appendix 15.2 Columbia-Suicide Severity Rating Scale (C-SSRS) Permission Letter ...... 387

References ...... 388

List of Tables

Table 1. Self-reported risk of motor vehicle accident (MVA) while driving under the influence of cannabis (DUIC) ...... 21 Table 2. Risk of motor vehicle accident (MVA) after analytical documentation of cannabis exposure ...... 23 Table 3. Risk of driver culpability or responsibility in motor vehicle accidents (MVA) after cannabis exposure ...... 26 Table 4. Summarized effects of cannabis and alcohol on neurocognitive function: laboratory studies...... 31 Table 5. Summarized effects of cannabis and alcohol on simulated and on-road driving...... 32 Table 6 (Supplemental). Self-reported risk of motor vehicle accident (MVA) after cannabis intake ...... 46 Table 7 (Supplemental). Risk of driver culpability or responsibility in motor vehicle accidents (MVA) after cannabis exposure ...... 50 Table 8 (Supplemental). Effect of smoked cannabis on neurocognitive function: laboratory studies ...... 54 Table 9 (Supplemental). Cannabis and alcohol effects on simulated and on-road driving ...... 63 Table 10 (Supplemental). Effect of smoked cannabis and alcohol (EtOH) on neurocognitive function: laboratory studies...... 72 Table 11. Self-reported demographic characteristics, recent cannabis and alcohol consumption and driving history of 19 healthy adult occasional cannabis smokers...... 142 Table 12. General Linear Model (GLM) Select results of effects on lateral control measures in 18 volunteer drivers after controlled vaporized cannabis with or without oral alcohol...... 143 Table 13. GLM Select model estimates for predicted standard deviation of lateral position (SDLP), lane departures/min, and maximum lateral acceleration associated with specific blood ∆9-tetrahydrocannabinol (THC) concentrations and breath alcohol concentrations...... 145 Table 14. Blood and oral fluid THC and variability prior to and after driving (N=19) after controlled vaporized active (2.9% THC and 6.7% THC) cannabis with or without alcohol...... 147 Table 15 (Supplemental). General Linear Model (GLM) Select results of natural log (ln)- transformed standard deviation of lateral position (SDLP) in 18 volunteer drivers after controlled vaporized cannabis with or without oral alcohol...... 155 Table 16 (Supplemental). Standard deviation of lateral position (SDLP) associated with specific blood ∆9-tetrahydrocannabinol (THC) concentrations and breath alcohol concentrations (BrAC) during driving based on transformed ln(SDLP) GLM Select model...... 156 Table 17. Self-reported demographic characteristics and recent cannabis and alcohol consumption history of 19 healthy adult occasional-to-moderate cannabis smokers...... 165

Table 18. Overall effect of blood ∆9-tetrahydrocannabinol (THC) concentration (μg/L), breath alcohol concentration (BrAC, g/210L), time, and interactions (THC*BrAC, time*THC, time*BrAC, time*THC*BrAC) on Visual analogue (VAS) or Likert scales subjective effects in 19 occasional-to-moderate cannabis smokers following controlled vaporized cannabis administration with and without oral alcohol...... 166 Table 19 (Supplemental). Overall effect of blood ∆9-tetrahydrocannabinol (THC) concentration (μg/L), breath alcohol concentration (BrAC, g/210L), time, and interactions (THC*BrAC, time*THC, time*BrAC, time*THC*BrAC) on subjective effects in 19 healthy, occasional-to-moderate cannabis smokers following controlled vaporized cannabis administration with and without oral alcohol...... 185 Table 20 (Supplemental). Overall effect of oral fluid (OF) ∆9-tetrahydrocannabinol (THC) concentration (μg/L), breath alcohol concentration (BrAC, g/210L), time, and interactions (THC*BrAC, time*THC, time*BrAC, time*THC*BrAC) on subjective effects beginning 1.4h post-dose in 19 occasional-to-moderate cannabis smokers following controlled vaporized cannabis administration with and without oral alcohol...... 187 Table 21 (Supplemental). Overall effect of oral fluid (OF) ∆9-tetrahydrocannabinol (THC) concentration (μg/L), breath alcohol concentration (BrAC, g/210L), time, and interactions (THC*BrAC, time*THC, time*BrAC, time*THC*BrAC) on subjective effects in 19 occasional-to-moderate cannabis smokers following controlled vaporized cannabis administration with and without oral alcohol. 194 Table 22 (Supplemental). Median [range] (N) oral fluid (OF)/blood and OF/plasma ∆9- tetrahydrocannabinol (THC), 11-nor-9-carboxy-THC (THCCOOH), cannabidiol (CBD) and cannabinol (CBN) ratios by time point after controlled vaporized cannabis and low-dose oral alcohol to healthy adult occasional-to-moderate cannabis smokers...... 202 Table 23 (Supplemental). (To accompany Figure 8): Regression equations and slope comparisons for (A) oral fluid (OF) ∆9-tetrahydrocannabinol (THC) versus blood THC concentration, and (B) OF versus plasma THC concentration after low (2.9% THC) and high (6.7% THC) dose vaporized cannabis administration with and without oral alcohol to 19 healthy, adult occasional-to-moderate cannabis smokers...... 204 Table 24. Self-reported demographic characteristics and recent cannabis and alcohol consumption history of 32 healthy occasional cannabis smokers...... 214 Table 25. Median [range] blood and plasma pharmacokinetic parameters following controlled vaporized cannabis administration with and without oral alcohol. 217 Table 26. Effects of alcohol, cannabis, and alcohol*cannabis combination on blood cannabinoid pharmacokinetic parameters...... 221 Table 27. Effects of alcohol, cannabis, and alcohol*cannabis combination on plasma cannabinoid pharmacokinetic parameters...... 225 Table 28 (Supplemental). Mean (standard deviation) approximate ∆9- tetrahydrocannabinol (THC), cannabidiol (CBD), and cannabinol (CBN) content in placebo, low-, and high-THC bulk cannabis doses administered by vaporizer balloon...... 240

Table 29 (Supplemental). Median [range] blood and plasma pharmacokinetic parameters (all analytes; including last measured concentration) following controlled vaporized cannabis administration with and without oral alcohol...... 240 Table 30 (Supplemental). Effects of alcohol, cannabis, and alcohol*cannabis combination on blood cannabinoid pharmacokinetic parameters (all analytes; including last measured concentration)...... 247 Table 31 (Supplemental). Effects of alcohol, cannabis, and alcohol*cannabis combination on plasma cannabinoid pharmacokinetic parameters (all analytes; including last measured concentration)...... 255 Table 32 (Supplemental). Median [range] blood and plasma ∆9-THC pharmacokinetic parameters for all 32 participants (full study population, including noncompleters) after controlled cannabis inhalation by vaporizer...... 262 Table 33 (Supplemental). Median [range] blood and plasma 11-OH-THC pharmacokinetic parameters for all 32 participants (full study population, including noncompleters) after controlled cannabis inhalation by vaporizer. . 263 Table 34 (Supplemental). Median [range] blood and plasma THCCOOH pharmacokinetic parameters for all 32 participants (full study population, including noncompleters) after controlled cannabis inhalation by vaporizer...... 264 Table 35 (Supplemental). Median [range] blood and plasma THC-glucuronide pharmacokinetic parameters for all 32 participants (full study population, including noncompleters) after controlled cannabis inhalation by vaporizer. . 265 Table 36 (Supplemental). Median [range] blood and plasma THCCOOH-glucuronide pharmacokinetic parameters for all 32 participants (full study population, including noncompleters) after controlled cannabis inhalation by vaporizer .. 266 Table 37 (Supplemental). Median [range] blood and plasma CBD pharmacokinetic parameters for all 32 participants (full study population, including noncompleters) after controlled cannabis inhalation by vaporizer...... 267 Table 38 (Supplemental). Median [range] blood and plasma cannabinol CBN pharmacokinetic parameters for all 32 participants (full study population, including noncompleters) after controlled cannabis inhalation by vaporizer. . 268 Table 39. Self-reported demographic characteristics and recent cannabis and alcohol consumption history of 43 healthy adult occasional cannabis smokers...... 282 Table 40. Median [range] QuantisalTM oral fluid pharmacokinetic parameters following controlled vaporized placebo, low (2.9%), or high (6.7%) THC cannabis with or without low-dose alcohol for 19 occasional to moderate smokers who completed all six dosing conditions...... 285 Table 41. Overall effects of alcohol, cannabis, and alcohol*cannabis interaction on oral fluid maximum concentration (Cmax), time to maximum concentration (tmax), time of last detection (tlast), and area under the curve (AUC0-8.3h) for cannabinoids ∆9-tetrahydrocannabinol (THC), cannabidiol (CBD), cannabinol (CBN), and 11-nor-9-carboxy-THC (THCCOOH) after inhalation of vaporized cannabis...... 289 Table 42. Median [range] low (2.9% THC) and high (6.7% THC) dose time of last Dräger DrugTest® 5000 on-site test positive detection in 19 completers only (5 µg/L ∆9-tetrahydrocannabinol (THC) oral fluid screening cutoff) with different

xii

oral fluid confirmation cutoffs, following oral inhalation of cannabis by Volcano Medic vaporizer...... 298 Table 43. Performance characteristics for the Dräger DrugTest® 5000 on-site test (5 µg/L ∆9-tetrahydrocannabinol (THC) oral fluid screening cutoff) with different oral fluid confirmation cutoffs, following inhalation of high-dose (6.7% THC) cannabis by Volcano Medic vaporizer, for comparison to smoking a similar- potency cigarette (111)...... 299 Table 44 (Supplemental). Median [range] maximum (Cmax) breath alcohol concentration (BrAC, reported in g/210L or approximate BAC g/dL), time to maximum BrAC (tmax), time of last alcohol detection (tlast), and area under the curve (AUC0-8.3h), following drinking Everclear grain alcohol over 10 min and followed immediately by controlled inhalation of placebo, low (2.9%) and high (6.7%) ∆9-tetrahydrocannabinol (THC) over 10 min...... 312 Table 45 (Supplemental). Median [range] QuantisalTM oral fluid ∆9-tetrahydrocannabinol (THC) maximum concentration (Cmax), baseline concentration (C0), time to maximum concentration (tmax), time of last detection (tlast), and area under the curve (AUC0-8.3h) after controlled cannabis inhalation by vaporizer...... 313 Table 46 (Supplemental). Median [range] oral fluid cannabidiol (CBD) maximum concentration (Cmax), baseline concentration (C0), time to maximum concentration (tmax), time of last detection (tlast), and area under the curve (AUC0-8.3h) after controlled cannabis inhalation by vaporizer...... 314 Table 47 (Supplemental). Median [range] oral fluid cannabinol (CBN) maximum concentration (Cmax), baseline concentration (C0), time to maximum concentration (tmax), time of last detection (tlast), and area under the curve (AUC0-8.3h) after controlled cannabis inhalation by vaporizer...... 316 Table 48 (Supplemental). Median [range] oral fluid 11-nor-9-carboxy-THC (THCCOOH) maximum concentration (Cmax), baseline concentration (C0), time to maximum concentration (tmax), time of last detection (tlast), and area under the curve (AUC0-8.3h) after controlled cannabis inhalation by vaporizer...... 317 Table 49 (Supplemental). Performance characteristics for the Dräger DrugTest® 5000 on- site test (5 µg/L ∆9-tetrahydrocannabinol (THC) oral fluid screening cutoff) with different oral fluid confirmation cutoffs, following inhalation of cannabis by Volcano Medic vaporizer ...... 319 Table 50. Primary research findings ...... 323

xiii

List of Figures

Figure 1. US States with legalized medical cannabis (green) or medical and recreational cannabis (blue) ...... 6 Figure 2. Cannabinoids and Phase I and II metabolic pathways ...... 8 Figure 3. The National Advanced Driving Simulator: A) exterior, dome mounted in room; B) dome interior with car mounted inside; C) view of night-drive simulation...... 138 Figure 4. GLM Select modeled standard deviation of lateral position (SDLP) versus blood ∆9-tetrahydrocannabinol (THC) concentration (lower x-axis) and versus breath alcohol concentration (BrAC, upper x-axis)...... 146 Figure 5 (Supplemental). Median and individual subject model-predicted standard deviation of lateral position (SDLP) for various blood ∆9-tetrahydrocannabinol (THC) concentrations (A) and breath alcohol concentrations (BrAC) (B). ... 157 Figure 6. Median [interquartile range] subjective effects visual-analogue scales (VAS) results versus time in 19 participants after low (2.9% THC) and high (6.7% THC) vaporized cannabis doses with and without low-dose oral alcohol. .... 174 Figure 7. Median subjective effects visual-analogue scales (VAS) results versus median blood ∆9-tetrahydrocannabinol (THC) concentrations, oral fluid (OF) THC, and breath alcohol concentration (BrAC) in 19 participants after placebo, low (2.9% THC) and high (6.7% THC) vaporized cannabis doses with and without low-dose oral alcohol...... 175 Figure 8. Oral fluid (OF) ∆9-tetrahydrocannabinol (THC) concentrations versus blood (A) and plasma (B) THC, and least-squares linear regressions from 19 participants after low (2.9% THC) and high (6.7% THC) vaporized cannabis doses with and without low-dose oral alcohol...... 176 Figure 9. Median [range] oral fluid (OF)/blood (a, c) and OF/plasma (b,d) ∆9- tetrahydrocannabinol (THC) and 11-nor-9-carboxy-THC (THCCOOH) ratios over time in paired-positive specimens from 19 participants after low (2.9% THC) and high (6.7% THC) vaporized cannabis doses with and without low- dose oral alcohol...... 177 Figure 10 (Supplemental). Mean [95% confidence interval] 5-point subjective Likert scale results in 19 participants after placebo, low (2.9% THC) and high (6.7% THC) vaporized cannabis doses with and without low-dose oral alcohol. .... 205 Figure 11 (Supplemental). Median subjective effects visual-analogue scales (VAS) results (“restless” and “sedated”) versus median blood ∆9-tetrahydrocannabinol (THC) concentrations, oral fluid (OF) THC, and breath alcohol concentration (BrAC) in 19 participants after placebo, low (2.9% THC) and high (6.7% THC) vaporized cannabis doses with and without low-dose oral alcohol. Counterclockwise and clockwise arrows represent hysteresis curve progressions over time...... 206 Figure 12 (Supplemental). Individual subjective “high” visual-analogue scales (VAS) results versus blood ∆9-tetrahydrocannabinol (THC) concentrations in 19 participants after low (2.9% THC) and high (6.7% THC) vaporized cannabis doses with and without low-dose oral alcohol. Counterclockwise arrows represent hysteresis curve progressions over time...... 207

xiv

Figure 13. Median [interquartile range] blood and plasma cannabinoids after cannabis vaporization (N=19)...... 229 Figure 14. Percent completers (N=19) positive for ∆9-tetrahydrocannabinol (THC) in whole blood for various cutoffs after controlled administration of vaporized cannabis (placebo, low [2.9%], and high [6.7%] THC) and alcohol (placebo and active)...... 231 Figure 15 (Supplemental). Median [interquartile range] blood cannabinoids in 19 study completers after controlled cannabis vaporization and placebo and active alcohol...... 269 Figure 16 (Supplemental). Median [interquartile range] plasma cannabinoids in 19 study completers after controlled cannabis vaporization and placebo and active alcohol...... 270 Figure 17 (Supplemental). Participant 25 blood and plasma ∆9-tetrahydrocannabinol (THC) concentrations following the placebo THC with active alcohol dose. 271 Figure 18 (Supplemental). Median [interquartile range] blood/plasma cannabinoid ratios after controlled cannabis vaporization and placebo and active alcohol...... 272 Figure 19 (Supplemental). Median [interquartile range] blood and plasma 11-nor-9- carboxy-tetrahydrocannabinol [THCCOOH]-glucuronide to THCCOOH ratios after controlled vaporized cannabis and placebo and active alcohol...... 273 Figure 20. Median [interquartile range] breath alcohol concentration (BrAC) in 19 completers following drinking placebo and three equivalent Everclear grain alcohol doses at separate sessions, with controlled inhalation of placebo, low (2.9%) or high (6.7%) ∆9-tetrahydrocannabinol (THC) vaporized cannabis. 287 Figure 21. ∆9-Tetrahydrocannabinol (THC) maximum oral fluid concentration vs. breath alcohol concentration (BrAC) for placebo, low (2.9%), and high (6.7%) THC doses (all participant data) following drinking alcohol and inhaling controlled cannabis by vaporizer...... 288 Figure 22. Median (interquartile range) oral fluid a) ∆9-tetrahydrocannabinol (THC), b) cannabidiol (CBD), c) cannabinol (CBN) and d) 11-nor-9-carboxy-THC (THCCOOH) vs. time after controlled vaporized cannabis inhalation in 19 completers...... 293 Figure 23. Dräger DrugTest® 5000 oral fluid cannabis detection rates over time in 19 completers with different confirmation cutoff criteria...... 297 Figure 24 (Supplemental). Median (interquartile range) oral fluid a) Δ9- tetrahydrocannabinol (THC) and b) 11-nor-9-carboxy-THC (THCCOOH) vs. time in 19 completers after controlled placebo cannabis inhalation by vaporizer...... 320 Figure 25 (Supplemental). Oral fluid cannabinoids a) ∆9-tetrahydrocannabinol (THC), b) cannabidiol (CBD), c) cannabinol (CBN) and d) 11-nor-9-carboxy-THC (THCCOOH) in participant 30 after controlled inhalation of vaporized placebo-THC cannabis...... 321

List of Abbreviations

11-OH-THC – 11-hydroxy-tetrahydrocannabinol

AE – adverse event

ANOVA – analysis of variance

AUC – area under the curve

AUC0-8.3h – area under the curve from 0-8.3h post-dose

AUC>BL – area under the curve accounting for baseline

AUDIT – Alcohol Use Disorders Identification Test

BAC – blood alcohol concentration

BART – Balloon-Analogue Risk Task

BrAC – breath alcohol concentration

BMI – body mass index

CBD – cannabidiol

CBN – cannabinol

CDM – Chemistry and Drug Metabolism

CI – confidence interval

Clast – last observed concentration

Cmax – maximum concentration

Cmax-BL – Cmax accounting for baseline

CNS – central nervous system

CRU – clinical research unit

CT – critical tracking

CTI – clinical test for impairment

xvi

CUDIT – Cannabis Use Disorders Identification

CYP – cytochrome P450 [enzyme]

DAT – divided attention task

DC – District of Columbia

DRF – driver-related factor

DUI – driving under the influence

DUIA – driving under the influence of alcohol

DUIC – driving under the influence of cannabis

DUID – driving under the influence of drugs

DRUID – Driving under the Influence of Drugs, Alcohol and Medicines

DWI – driving while intoxicated

EtOH – ethanol

FARS – Fatality Analysis Reporting System

FP – false positives

FN – false negatives

GC/MS – gas chromatography-mass spectrometry

HPLC – high-performance liquid chromatography

IDE – investigational device exemption

IND – investigational new drug

IRP – Intramural Research Program

LCMSMS – liquid chromatography-tandem mass spectrometry

LOQ – limit of quantification

MDMA – 3,4-methylenedioxymethamphetamine

xvii

MVA – motor vehicle accident

NADS – National Advanced Driving Simulator

NFDI – non-fatal driver injury

NHTSA – National Highway Traffic Safety Administration

NIDA – National Institute on Drug Abuse

NPS – novel psychoactive substances

OF – oral fluid

ONDCP – Office of National Drug Control Policy

OR – odds ratio

PAH – polycyclic aromatic hydrocarbon

PET – positron emission tomography

PII – personally identifiable information

RR – relative risk

RT – reaction time

SAMHSA – Substance Abuse and Mental Health Services Administration

SC – synthetic cannabinoids

SD – standard deviation

SDLP – standard deviation of lateral (or lane) position

SFST – Standardized Field Sobriety Test

SST – stop signal task

SVA – single vehicle accident

THC – ∆9-tetrahydrocannabinol

THCAA – ∆9-tetrahydrocannabinolic acid A

xviii

THCCOOH – 11-nor-9-carboxy-tetrahydrocannabinol tlast – time of last detection tmax – time of maximum concentration

TOL – time out of lane

TN – true negatives

TP – true positives

UGT – UDP-glucuronosyltransferase [enzyme]

UIA – University of Iowa

VAS – visual-analogue scale

Vd – volume of distribution

xix

Chapter 1 – Introduction

Driving Under the Influence of Cannabis (DUIC)

Driving under the influence of drugs (DUID) increased as driving under the influence of alcohol (DUIA) decreased over the last 40 years (1). Since 2007, licit and illicit drug prevalence in weekend nighttime drivers increased from 3.9% and 12.4% to

4.9% and 15.1%, respectively (1). Thus, drugged driving is a crucial current public safety issue, particularly for illicit drugs. Recently, the US Office of National Drug Control

Policy (ONDCP) prioritized drugged driving among its top research initiatives (2).

Cannabis is the most common illicit substance found in drivers, with 12.6% of weekend nighttime drivers testing positive for ∆9-tetrahydrocannabinol (THC, the primary psychoactive phytocannabinoid) in blood and/or oral fluid (OF) in the 2013-2014 US

National Roadside Survey—a 48% increase from 8.6% in 2007 (1, 3). Several studies demonstrated increased crash risk and driver culpability from DUIC (4-6), including two recent meta-analyses indicating cannabis approximately doubled crash risk (7-8).

However, a recent comprehensive case-control study released by the National Highway

Traffic Safety Administration (NHTSA) (9) indicated no significant difference in adjusted (for age, gender, and ethnicity) crash risk for THC-positive drivers beyond alcohol effects.

Public attitudes toward DUIC are less negative than DUIA. Cannabis is perceived as a less impairing drug than alcohol, and those who drove under the influence (drugs or alcohol) were less likely to be concerned about impaired driving (10). According to the

2007-2011 Monitoring the Future study, a significantly higher percentage of high school

1 seniors who self-reported lifetime alcohol intake believed it caused them to drive unsafely (9.1%) compared to lifetime cannabis intake (6.4%) (11). Among adolescents who indicated lifetime alcohol or cannabis intake ≥10 times, significantly higher odds ratios (OR) were determined for self-reported unsafe driving related to alcohol than cannabis. Similarly, college students considered DUIC more acceptable among peers than

DUIA, and with less likelihood of negative consequences (12). Age of cannabis intake onset is inversely related to risk of lifetime DUIC, with individuals who first took cannabis before age 14 2.23x (adjusted) more likely to DUIC during their lifetime (13).

Although DUIA above 0.08 g/210 L illegal breath alcohol concentration (BrAC, equivalent to approximate blood alcohol concentration [BAC]) decreased 80% since 1973

(1), alcohol remains the most common licit substance identified in drivers. According to the recent NHTSA case-control study, alcohol BrAC ≥0.05 g/210 L was associated with

OR [95% confidence interval, CI] 6.75 [4.20-10.84]x (without drugs) and 5.34 [2.75-

10.37]x (with drugs) higher crash risk relative to alcohol-free matched controls (9). The difference between with and without drugs was not significant. Alcohol-related driving fatalities account for approximately 30% of driving fatalities in the United States. An estimated 12,998 individuals died in 2007 from alcohol-impaired driving collisions (14).

The adoption of 21 as the minimum legal drinking age in all 50 states by 1988 was associated with a decrease in fatalities among young drivers, but DWI remains a public health concern for all ages (15). As many as 14% of nighttime weekend drivers in a recent roadside survey were classified as dependent or abusive drinkers based on self- report, and another 10% were heavy drinkers (16).

Relative risk of crash involvement increases exponentially with increasing BAC:

1.4 for drivers with a BAC between 0.02-0.04%, 11.1 for 0.05-0.09%, 48 for 0.10-0.14%, and approximately 380 for >0.15%. For single-vehicle crashes, each 0.01% increase in

BAC corresponds to a 39% increase in relative risk. Impairment is greater in younger drivers (ages 16-20) than older drivers, due to inexperience, lower tolerance to alcohol’s effects, and a propensity for risk-taking behavior (17-18). All age groups and sexes, however, display increasing impairment from increasing BAC (19). There are no “safe”

BACs for driving, even among those who develop alcohol tolerance (even appearing unimpaired to peers) (20). Alcohol and cannabis are frequently encountered together in crashes and forensic DUID cases (21-23). A more comprehensive review of current knowledge on cannabis’ effects on driving is presented in Chapter 2 – Cannabis Effects on Driving Skills.

Knowledge Gap: Cannabis’ Effects on Driving

Despite extensive research, the degree to which cannabis impairs driving is greatly debated. It was suggested that drivers under the influence of cannabis are aware of impairment and attempt to compensate by slower and less-risky driving (10, 24-29).

Driving is a complex task, involving simultaneous coordination of multiple sub- processes. Cannabis impairs complex neurocognitive and psychomotor functions, and was shown to alter reaction time (RT), perception, short-term memory and attention, motor skills, tracking, and skilled activities (30-32). These abilities are required driving skills. A recent consensus protocol for assessing driving impairment included divided attention, working memory, drowsiness, focused attention, eye tracking, RT (particularly

3 in the context of decision making), and executive function (e.g., planning, risk taking, impulse control and judgment, adaptive problem solving) (33). Road tracking and vehicle control also are important measures of driving ability, and typically assessed in impaired- driving studies (34-39). Maintaining lateral control is a crucial driving skill, necessary to maintain consistent positioning within a lane, accurately follow the roadway, and avoid collisions. Cannabis impairs critical tracking, a validated predictor of driving impairment

(representing road tracking) especially in occasional smokers (40-43). Some studies showed increased standard deviation of lateral position (SDLP, a measure of road tracking) after cannabis (25), while others did not (44) Other lateral control measures

(e.g., lateral acceleration, lane departures) are less studied (45-46). Sophisticated studies are needed to adequately establish cannabis’ effect on these parameters.

Additionally, prior experimental cannabis-driving research, while often showing dose-related impairment relationships (25, 34, 47), did not evaluate driving impairment at specific blood THC concentrations. Because forensic casework evaluates cannabis exposure by blood concentration and current or considered per se legislation is based on blood THC, there is substantial interest in evaluating cannabis-driving effects at specific

THC concentrations.

Knowledge Gap: Cannabis and Alcohol Interactive Effects on Driving

Cannabis and alcohol both impair cognitive and psychomotor function, and interact to exacerbate effects (26, 48). Alcohol increases risky behavior, and impairs RT and driving performance. It remains unclear whether interaction effects of cannabis and alcohol are additive or synergistic (30, 47, 49-51). Because ingesting alcohol and

4 cannabis together is such a common practice, it is clinically and forensically important to elucidate how this combination affects driving performance, compared to placebo and either drug alone.

Aims and Hypotheses

This dissertation presents data from a clinical protocol designed by the National

Institute on Drug Abuse (NIDA) Chemistry and Drug Metabolism section (CDM) in collaboration with NHTSA. The primary aim of the clinical protocol (Chapter 3 –

Protocol: Effect of Inhaled Cannabis on Driving Performance) was to assess cannabis’ effects on driving skills, with and without low-dose alcohol. The sophisticated driving simulator, the National Advanced Driving Simulator (NADS) was utilized to accomplish this. The simulated drive was designed to be sensitive to cannabis’ specific effects.

Particular measures of interest included road tracking and vehicular control, the ability to focus during long stretches on a monotonous drive, decision-making, RT, ability to process changes in driving environment, and divided attention. Another aim was to relate performance to actual blood THC concentrations.

Hypotheses: Cannabis will impair driving in a THC concentration-dependent manner relative to placebo. Lane position variability (SDLP), lane departures, and steering wheel variability will increase after cannabis, indicating detrimental effects on lateral control. Participants will likely decrease overall speed in cannabis conditions compared with placebo, due to possible awareness of impairment and experimental monitoring; but speed variability may increase. More overall mistakes will be made during the driving simulation under cannabis conditions, and errors and RT during the

divided attention task will increase. Visual search (monitored via an eye-tracking system

within the NADS) will decrease with cannabis, and participants may be less likely to

notice potential hazards along the roadway. The greatest effect sizes will be observed

when blood THC concentrations are most elevated. The alcohol-cannabis combination

will increase SDLP and impair performance on divided attention tasks (DATs) more than

placebo, or either drug alone. The tendency to decrease speed when under the influence

of cannabis will not be observed in the alcohol or alcohol + cannabis conditions.

Medical and Recreational Cannabis

Cannabis legalization is currently a heavily debated political issue in the US. To

date, 23 US states and the District of Columbia (DC) approved medical marijuana (Figure

1) (52-53). Additionally, Colorado, Washington, Alaska, and Oregon recently legalized

Figure 1. US States with legalized medical cannabis (green) or medical and recreational cannabis (blue)

recreational cannabis (53-55). Although DC voted to approve it, sales and public smoking

6 legalization were blocked by a Congressional spending bill (56), but as of March 2015 it was legalized for private growing and consumption (53, 57). Since implementing medical marijuana, Colorado observed increased driving under the influence of cannabis (DUIC) law enforcement cases (58). Fatal motor vehicle crashes with cannabis-positive drivers also increased, whereas no significant change was observed in 34 states without medical marijuana (59).

With current national and international debates, and rapidly shifting laws and societal attitudes toward cannabis, the extent to which cannabis affects driving is highly debated. A more complete understanding of cannabis’ impact on driving would help inform legislative policy and improve road safety through education.

Vaporized Cannabis

Typically, cannabis is heated prior to intake, facilitating decarboxylation from inactive precursor ∆9-tetrahydrocannabinolic acid A (THCAA) to active THC (60-62)

(Figure 2). Although smoking is the most common route of cannabis administration (63), alternative routes increased in prevalence over recent years. Smoking is inappropriate pharmacotherapy, due to delivery of harmful substances (64). The advent of medical marijuana was accompanied by an increase in popularity of vaporization. Vaporization as an alternative to smoking reduces the release of toxic pyrolysis products such as particulate “tar”, carbon monoxide, and aromatic hydrocarbons such as benzene, toluene, and naphthalene (65). In a preclinical comparison of cannabis volatilization by combustion and vaporization, THC volatilization was comparable, with slightly higher efficiency by vaporization. However, cannabis smoke contained many compounds not

Figure 2. Cannabinoids and Phase I and II metabolic pathways

8 present in vapor, including ∆8-THC, 1-oxo-cannabinol, 21 unknown cannabinoids, hydrocarbons, phenolic compounds, and nitrogen-containing compounds (66).

Another study comparing combusted and vaporized cannabis by gas chromatography-mass spectrometry (GC/MS) analysis identified only three non- cannabinoid (i.e. THC and cannabinol [CBN]) compounds (including caryophyllene, found in cannabis) in the vapor. Contrastingly, smoke contained over 111 compounds, including 5 known polycyclic aromatic hydrocarbons (PAHs) (67). Several of these classes of thermal degradation byproduct compounds are toxic or carcinogenic (68).

Quantitative high-performance liquid chromatography (HPLC) analysis demonstrated the cannabinoids:byproducts ratio in vapor from vaporizer temperature 185-230°C was significantly higher than in cannabis smoke or vapor from a device at 170°C (69). Thus vaporization produces a less harmful inhalation product than normal cannabis cigarette smoke.

Individuals who took cannabis at least once in the past month were significantly less likely to report respiratory problems with vaporization compared to those who smoked or utilized other inhalation techniques (70). A follow-up investigation evaluated

20 current cannabis smokers reporting at least two respiratory symptoms who utilized a vaporizer device for a month rather than smoking. The 12 participants who did not develop a respiratory illness during the course of the study had significantly decreased subjective respiratory distress and significantly improved pulmonary function (71).

Vaporized cannabis significantly decreases neuropathic pain, and is of interest as medical marijuana legalization expands (72-73). Controlled administration studies also increasingly utilize cannabis vaporization as a smoking alternative (74-78). Multiple

9 studies examined subjective cannabis effects post-vaporization (72, 78-79). However, most limited analyses to subjective “high” and related measures.

Knowledge Gaps

Despite increasing vaporizer prevalence for medical, recreational, and clinical purposes (72-78, 80), its relationship to smoking as an effective cannabinoid delivery route is not fully evaluated. Relatively few clinical studies evaluated vaporized cannabis effects on subjective feelings (72, 78-79), and most limited analyses to subjective “high” and variations. Verifying that vaporization is an effective cannabinoid delivery system that produces similar effects to other inhaled routes is an important knowledge gap to be addressed.

Alcohol interaction with cannabis’ subjective effects also is not yet evaluated, despite the frequency of co-administration. A better understanding of alcohol-cannabis interactions in pharmacodynamic responses, especially the question of whether effects are additive or synergistic, is needed.

Aims and Hypotheses

Specific aims for establishing vaporized cannabis effects included evaluating desirable and undesirable subjective effects after ad libitum inhalation to verify its efficacy as an alternative THC delivery route, and comparing to previous smoked cannabis research. Another specific aim involved determining to what extent cannabis and alcohol interact in subjective effects. These aims were addressed by measuring within-subjects subjective effects following cannabis and alcohol dosing via visual

10 analogue scales (VAS) and by 5-point Likert scales (0 ≡ “None”, 4 ≡ “Severe”) at several points throughout the 8.3 h time course after dosing.

Hypotheses: Cannabis will dose-dependently increase subjective “high”, “good drug effect”, “stimulated” and “stoned”; but also “anxious”, “difficulty concentrating”,

“feel hungry” and “feel thirsty”. Patterns will resemble those observed recently after smoking a 6.8% THC cigarette (81), displaying counterclockwise hysteresis. Alcohol will increase cannabis’ effects.

Cannabis Pharmacokinetics

Inhalation is an efficient administration route that allows rapid absorption and delivery to the brain. Smoking cannabis causes THC loss from degradation (~30%, due to pyrolysis) and side-stream smoke (as much as 40-50%) (82). Although vaporizing cannabis may minimize these losses associated with smoking (due to lower heating temperatures and vapor collection into a balloon), a large proportion of volatilized cannabinoids are not absorbed by the lungs. In a clinical vaporizer study, only ~54% of loaded THC was delivered to the balloon for inhalation, and 30-40% of this was directly exhaled (83). Some THC also may adhere to the plastic balloon interior, particularly if there is extended time between vaporization and inhalation (83). THC is lipophilic and has a high volume of distribution (Vd) once inhaled, rapidly distributing to highly- perfused and adipose tissues (84-85).

THC is rapidly metabolized (Figure 2) by cytochrome P450 (CYP) enzymes 2C9 and 2C19 to active 11-hydroxy-THC (11-OH-THC) and further to inactive 11-nor-9- carboxy-THC (THCCOOH) (84, 86-87). All three compounds can undergo Phase II

11 conjugation (88-89), mainly through glucuronidation, although cannabinoid sulfation also is possible (90-91). Glucuronidation is mediated by UDP-glucuronosyltransferase (UGT) isoenzymes (89). The major Phase II metabolite is THCCOOH-glucuronide (O-ester glucuronide), although THC-glucuronide also was reported in blood, plasma, and urine

(88, 92-94).

More than 100 phytocannabinoids were identified to date (95). Among the primary constituents (apart from THC) are cannabidiol (CBD) and CBN (Figure 2). CBN is a product of THC degradation that occurs within the plant (96), and (along with its metabolite 11-hydroxy-CBN) is pharmacologically active (97). However, it is present in cannabis at low concentrations (95). CBD is a non-psychoactive cannabinoid, of interest for therapeutic purposes including anxiolytic, antiepileptic, anti-inflammatory, neuro- protective, antiemetic, and antipsychotic properties (98-102). It may attenuate some of

THC’s psychoactive and physiological effects (103-106). A 1:1 CBD:THC cannabis extract oromucosal spray formulation did not affect THC pharmacokinetic parameters relative to oral THC (104), and did not significantly affect pharmacodynamic effects

(with a possible exception of subjective “high”) (105). CBN and CBD are both glucuronidated at the 1 position by UGT enzymes (89).

Blood and Plasma

Blood and plasma THC maximum concentrations (Cmax) are reached during smoking, prior to the last puff (107). High initial THC concentrations decrease rapidly as

THC is distributed (84-85). A slower decrease is observed after the first ~2 h, reflecting the gradual re-release of THC from adipose back into blood. Low THC was detected in

12 blood up to 30 days of abstinence in chronic daily smokers (108), but in occasional smokers blood THC time of last detection (tlast) occurred within a day of smoking (93).

Equipotent 11-OH-THC time of Cmax (tmax) occurs within 15 min after start of smoking, but concentrations only reach ~6-10% of concurrent THC (107). THCCOOH and its glucuronide’s Cmax occur after THC and 11-OH-THC, and are detectable for extended periods (93, 107-108). THCCOOH’s extended detection windows exceed those of THC.

THC-glucuronide, CBD and CBN, proposed as recent intake markers, were not detected after 1 h post-dose (93-94).

Oral fluid (OF)

OF is an advantageous matrix for assessing drug exposure, due to relative ease of collection, observable collection (preventing adulteration), facility for on-site testing, and association with recent intake (74, 109-111). Some jurisdictions in the US and elsewhere already adopted OF-based DUIC legislation (112-114). THC, CBD, and CBN enter OF during inhalation primarily due to oromucosal contamination rather than passage from systemic circulation (109-110, 115-118). Although THC and CBN were detected in OF after passive cannabis smoke exposure, metabolite THCCOOH is not present in smoke and helps differentiate active from passive exposure (119). OF THCCOOH also extends cannabis detection windows in chronic, frequent smokers, and differentiates oral THC exposure (for cannabis dependence treatment) from recent smoking (117, 120-121).

Proposed recent intake markers CBD and CBN were detected ≤13.5 h (median, 2.5-4 h and 6-8 h, respectively), shortening detection windows relative to THC (117).

On-site drug screening gained prevalence in the last decade (122). This approach helps evaluate drug exposure roadside, providing law enforcement with real-time responses, and also is utilized as a DUID deterrent (122-123). THC is the target analyte for analysis (111), and on-site detection windows post-smoking were 4-24 h and 1-≥30 h in occasional and frequent smokers, respectively (124).

Knowledge Gaps

Blood, plasma, and OF cannabinoid disposition was recently evaluated in occasional and frequent smokers (93-94, 124). Vaporization is intended to produce efficient THC, CBD, and CBN delivery (125). However, vaporized cannabis disposition is not fully elucidated in any matrix, with few prior clinical studies, short (≤6 h) time courses, and limited metabolite analysis (78-79). Although blood is the more forensically relevant matrix for DUIC cases, most published cannabinoid data are for plasma (126-

128), the appropriate matrix for assessing pharmacotherapy potential. Vaporized OF data are lacking for reference in DUIC and drug monitoring, so it remains unclear whether the increase in vaporizer prevalence requires consideration when evaluating OF results.

Additionally, although several on-site testing devices’ validity were assessed roadside or after cannabis smoking (111, 123-124, 129-131), limited data exist for on-site testing after vaporized cannabis (only one device evaluated, with poor sensitivity by 1.5 h) (74).

Whether subjective and driving effects can be correlated to OF concentrations also is important lacking information, given OF’s utility as a screening matrix. There is interest in determining OF/blood and OF/plasma relationships. Recently a study from the

European Driving under the Influence of Drugs, Alcohol and Medicines (DRUID) project

14 suggested a 44 [95% CI 27-90] μg/L OF THC cutoff to establish similar driving positivity prevalence to 1 μg/L blood THC (132), but this does not reflect specifically equivalent blood and OF concentrations. Although smoked OF cannabinoid variability was high, vaporized OF/blood and OF/plasma relationships are not yet characterized.

Because cannabis and alcohol are frequently co-administered, information regarding cannabis and alcohol’s pharmacokinetic interactions in these matrices also is a critical knowledge gap.

Aims and Hypotheses

The primary aims of this section were to evaluate cannabinoid disposition in blood, plasma, and OF following cannabis vaporization, and to determine to what extent low-dose alcohol effects cannabinoid pharmacokinetics. Blood, plasma, and OF were collected prior to and up to 8.3 h after vaporized cannabis with or without concurrent alcohol, and analyzed for THC, 11-OH-THC, THCCOOH, THC-glucuronide,

THCCOOH-glucuronide, CBD, and CBN. Another aim was to establish OF on-site performance and detection windows after vaporization, and determine whether alcohol presence affects on-site test results. For this aim, on-site testing was performed prior to and throughout 8.3 h post-dosing. The Draeger DrugTest® 5000 (among the more promising devices) (111) was employed instead of the DrugWipe5S® (previously evaluated after vaporization) (74).

In order to determine whether OF correlates to cannabis effects, subjective effects were evaluated relative to OF THC as well as to blood THC, and OF concentrations preceding and immediately following driving are presented.

Hypotheses: Cannabinoid pharmacokinetics after vaporization will be similar to smoking, and alcohol will increase THC Cmax and AUC. On-site testing efficiency also will be similar to smoked cannabis, and alcohol will not affect results. Blood concentrations will be better predictors of cannabis effects than OF.

Chapter 2 – Cannabis Effects on Driving Skills

(As published in Clinical Chemistry, 2013)1

Abstract

Background: Cannabis is the most prevalent illicit drug identified in impaired drivers.

Cannabis effects on driving continue to be debated, making prosecution and legislation difficult. Historically, delays in specimen collection, evaluating the inactive ∆9- tetrahydrocannabinol (THC) metabolite, 11-nor-9-carboxy-THC, and polydrug use have complicated epidemiological evaluation of driver impairment following cannabis use.

Content: We review and evaluate the current literature on cannabis’ effects on driving, highlighting epidemiological and experimental data. Epidemiological data show that the risk of motor vehicle accident (MVA) involvement increases approximately twofold following cannabis smoking. Adjusted driver culpability risk also increases substantially, particularly with increased blood THC concentrations. Studies employing urine as the biological matrix have not shown an association between cannabis and crash risk.

Experimental data show that drivers, after smoking cannabis, attempt to compensate by driving more slowly, but control deteriorates with increasing task complexity. Cannabis increases lane weaving and impaired cognitive function. Critical tracking, reaction times, divided attention tasks and lane-position variability all show cannabis-induced impairment. Despite purported tolerance in frequent smokers, complex tasks still show impairment. Combining cannabis with alcohol enhances impairment, especially lane weaving.

1 Hartman RL, Huestis MA. Clin Chem. 2013;59(3)478-92. 17

Summary: Different study designs frequently account for result inconsistencies between studies. Participant selection bias and confounding factors attenuate ostensible cannabis effects, but the association with MVA often retains significance. Evidence suggests recent smoking and/or blood THC concentrations 2 to 5 ng/mL are associated with significant driving impairment, particularly in occasional smokers. Future cannabis- driving research should emphasize challenging tasks, such as divided attention, and include occasional and chronic daily cannabis smokers.

Introduction

Nearly two thirds of US trauma center admissions are due to motor vehicle accidents (MVA), with almost 60% positive for drugs or alcohol (133). In 2010, 11.4% of

Americans age 12 or older drove under the influence of alcohol, and 10.6 million under the influence of illicit drugs (134). Despite real or perceived impairment, individuals report a willingness to drive if there is a good reason (26, 44) or if they believe they are tolerant (41). Alcohol and cannabis are the most frequently detected drugs (3).

Cannabis is the most widely-consumed illicit substance worldwide (134). In 2009,

125-203 million individuals ages 15-64 ingested cannabis in the past year (135). In the

US in 2010, 6.9% of individuals ≥12 years old smoked cannabis in the last month (134).

The 2007 National Roadside Survey reported cannabis as the most common illicit drug quantified in drivers’ blood or oral fluid (OF) with 8.6% of nighttime drivers positive for

∆9-tetrahydrocannabinol (THC) (3, 136). Thus, driving under the influence of cannabis

(DUIC) is a growing public health concern.

Acute cannabinoid psychological effects include euphoria, dysphoria, sedation, and altered perception (137). Intensity of euphoria/dysphoria varies with dose, administration route and vehicle, expectations of effects, and cannabis smoker’s environment and personality. Cannabis is associated with subjective physical discomfort and effort, and lack of energy (29). Acute cannabis intoxication produces dose-related impairment in cognitive and psychomotor functioning, and risk-taking behavior that can impair driving skills (43, 138). Dose refers to THC content in cannabis preparations in mg or µg/kg. Factors influencing dose include user experience, smoking topography, and cannabis THC concentration, which varies worldwide. Cannabis effects include altered reaction time (RT), perception, short-term memory, attention, motor skills, tracking, and skilled activities (30, 32).

Objective and Search Methods

This review presents relevant published data and evaluates current knowledge of cannabis’ effects on driving. The electronic databases PubMed, Scopus, Web of Science, and Embase were searched through 20 February 2012 for keywords ‘cannabis’,

‘marijuana’, ‘automobile driving’, ‘accidents, traffic’, and ‘motor vehicles’. Additional articles were selected from references in identified sources.

DUIC: Epidemiological Data

Early DUIC epidemiological studies did not provide strong evidence of cannabis causality because individuals with only non-psychoactive 11-nor-9-carboxy-THC

(THCCOOH) in blood were included in cannabis-exposed groups (4). THCCOOH has a long window of detection in blood, well after acute effects dissipate (139). In less-than-

19 daily cannabis smokers, THCCOOH detection was up to 7 days after smoking one joint containing ~38 mg THC (cutoff 0.5 ng/mL) (107). THC blood concentrations decrease rapidly after smoking (139-140). Blood collection occurs about 90 min after arrest (141) and 3-4 h after an accident (142) — long enough that many specimens are cannabinoid- negative, although blood may have been positive at the time of the event. There also were few cannabis-only cases; multiple drugs were found potentially contributing to impairment.

Cannabis smokers share similar demographic characteristics with other high- crash-risk groups, including youth (ages 18-25), male, risk-taking, and high drunk driving incidence (136, 142-145). Cannabis tolerance may develop in frequent smokers, with less impairment than for occasional smokers at similar THC concentrations (146). Sometimes statistically controlling for these potentially confounding variables make results equivocal (147-148).

Ten epidemiological studies from 6 countries have investigated MVA and cannabis (Table 1, Table 2). Varied case-control designs employed self-report or objective biological measurements. Adjusting for confounders lessened apparent effect sizes relative to crude values and sometimes caused loss of significance (6, 143, 149-

150). Six studies evaluated relationships between cannabis exposure and MVA by self- report (Table 1 and Table 6 (Supplemental)). Examining cannabis consumption rather than DUIC has generally produced nonsignificant or lower odds ratios (OR) than targeting DUIC. More frequent (addiction patients >1x/week, >4 days/week) cannabis exposure was associated with significantly increased MVA risk (relative risk [RR] 1.49,

Table 1. Self-reported risk of motor vehicle accident (MVA) while driving under the influence of cannabis (DUIC)

CB exposure 95% CI Crude (C) or Location/ N N CB OR or Cases Controls time prior to Parameter or Adjusted Ref. Country cases controls Only RR driving p-value (factors)

10th & 12th 10th & 12th Atlantic grade 877 grade students, 3,178 ≤1 h -- MVA 4.14a p<0.001 C (143) Canada students, no CB DUIC 10th & 12th 10th & 12th Atlantic grade Demographics, 877 grade students, 3,178 ≤1 h -- MVA 1.84a p<0.001 (143) Canada students, DUIA no CB DUIC Adults, drove Ontario, within 1 h of 2.30- 70 Adult drivers 2,606 ≤1 h -- MVA 3.89a C (149) Canada CB intake in 6.59 past year 21 Adults, drove Ontario, within 1 h of 1.45- 70 Adult drivers 2,606 ≤1 h -- MVA 2.61a Demographics (149) Canada CB intake in 4.68 past year Auckland, Drove within Did not drive New 3 h of CB 37 within 3 h of 1,102 ≤3 h -- MVA-Ic 11.4a 3.6-35.4 C (150) Zealand intake CB intake Demographics, time of day, Auckland, Drove within Did not drive no. passengers, New 3 h of CB 37 within 3 h of 1,102 ≤3 h -- MVA-Ic 0.8a 0.2-3.3 (150) risky behavior Zealand intake CB intake (BAC, speed, seatbelt use) Regular Same cocaine users, population, not NFDI, Spain driving 68 68 ≤1 h -- 7.0b 3.1-16 -- (151) driving within 1 past year within 1 h of h of CB intake CB intake

Table 1. (Continued from previous page) Self-reported risk of motor vehicle accident (MVA) while driving under the influence of cannabis (DUIC)

CB exposure 95% CI Crude (C) or Location/ N N CB OR or Cases Controls time prior to Parameter or Adjusted Ref. Country cases controls Only RR driving p-value (factors)

Regular Same cocaine users, population, not NFDI, Spain driving 68 68 ≤2 h -- 2.2b 1.0-4.9 -- (151) driving within 2 past year within 2 h of h of CB intake CB intake Regular Same cocaine users, population, not CB NFDI, Spain driving 45 45 ≤1 h 5.8b 2.4-14 -- (151) driving within 1 only past year within 1 h of h of CB intake CB intake Regular 22 Same cocaine users, population, not CB NFDI, Spain driving 45 45 ≤2 h 2.2b 0.9-5.2 -- (151) driving within 2 only past year within 2 h of h of CB intake CB intake Regular cocaine users, +Alc Same driving in NFDI, Spain 19 population, not 19 ≤1 h 10.9b 1.3-88 -- (151) within 1 h of past 2 past year DUID CB intake & h 2 h of alcohol aOR (odds ratio) bRR (relative risk) cCars in MVA with ≥1 occupant hospitalized or killed dCase-crossover design Abbreviations: DUIA, driving under the influence of alcohol; CB, cannabis; CI, confidence interval; Ref., reference; NFDI, non-fatal driver injury; BAC, blood alcohol content

Table 2. Risk of motor vehicle accident (MVA) after analytical documentation of cannabis exposure Crude N THC N cases Analytical CB Para- 95% (C) or Country N total Cases Controls controls Cutoff OR Ref. (THC+) Matrix Only meter CI Adjusted (THC +) (ng/mL) (factors) Demo- Randomly- graphics, The selected Injured 110 816 Urine &/or (not MVA 0.55- drugs, Nether- 926 drivers -- 1.22 (152) drivers (13) (49) Blood given) -DI 2.73 time of lands (roadside day, survey) season Randomly- Blood selected Blood Killed 204 10,540 (cases); MVA 6.6- Norway 10,744 drivers 0.6 -- 13.9 C (6) drivers (<24) (≤53) OF -DF 29.2 (roadside OF 5 (controls) survey) Randomly- Demo-

23 Blood selected Blood graphics,

Killed 204 10,540 (cases); MVA 3.9- Norway 10,744 drivers 0.6 -- 8.6 time (6) drivers (<24) (≤53) OF -DF 19.3 (roadside OF 5 period, (controls) survey) season Randomly- Blood selected Blood Killed 204 10,540 (cases); CB MVA 0.3- Norway 10,744 drivers 0.6 1.9 C (6) drivers (<24) (≤53) OF only -DF 13.7 (roadside OF 5 (controls) survey) Randomly- Demo- Blood selected Blood graphics, Killed 204 10,540 (cases); CB MVA 0.1- Norway 10,744 drivers 0.6 0.9 time (6) drivers (<24) (≤53) OF only -DF 7.3 (roadside OF 5 period, (controls) survey) season Randomly- Killed Blood No selected Blood drivers 68 10,540 (cases); CB SVA 6.5- Norway 10,608 drivers 0.6 18.9 C (6) in (<10) (≤53) OF only -DF 54.6 (roadside OF 5 SVA (controls) cases survey)

Table 2. (Continued from previous page) Risk of motor vehicle accident (MVA) after analytical documentation of cannabis exposure Crude N THC N cases Analytical CB Para- 95% (C) or Country N total Cases Controls controls Cutoff OR Ref. (THC+) Matrix Only meter CI Adjusted (THC +) (ng/mL) (factors) Randomly- Demo- Killed Blood No selected Blood graphics, drivers 68 10,540 (cases); CB SVA 2.7- Norway 10,608 drivers 0.6 9.0 time (6) in (<10) (≤53) OF only -DF 30.3 (roadside OF 5 period, SVA (controls) cases survey) season Drivers w/ no traffic Injured 200 injury 849 0.25- Thailand 1,049 Urine 50 -- RTI 0.78 C (153) driversa (4) history in (20) 2.40 past 6 months

24 9 2.07- Multiple Meta-analysis: epidemiological studies MVA 2.66 (7) studies 3.41 9 1.35- Multiple Meta-analysis: epidemiological studies MVA 1.92 (154) studies 2.73 3 1.23- Multiple Meta-analysis: case-control studies MVA 2.79 (154) studies 6.33 6 1.11- Multiple Meta-analysis: culpability studies MVA 1.65 (154) studies 2.46 5 1.31- Multiple Meta-analysis: studies with fatal collisions MVA 2.10 (154) studies 3.36 4 Meta-analysis: studies with non-fatal 0.88- Multiple MVA 1.74 (154) studies collisions 3.46 aAdmitted to hospital within 24 h Abbreviations: THC, ∆9-tetrahydrocannabinol; CB, cannabis; OR, odds ratio; CI, confidence interval; Ref., reference; SVA, single vehicle accident; DI, driver injury; DF, driver fatality; RTI, road traffic injury

OR 2.76, OR 2.5, respectively) (149, 155-156). A crude 11.4 OR for MVA injury within

3 h of cannabis smoking dropped to a nonsignificant 0.8 after adjusting for confounders

(150), while DUIC after past-hour smoking almost doubled crash risk (ORs 1.84 (143) and 2.61 (149)), withstanding adjustment for demographic characteristics (143, 149) and self-reported DUIA (143). Driving within 1 h after smoking produced higher MVA ORs than within 2 h (151). Among 3 case-control studies including objective cannabis measurement (Table 2), 2 utilized urine and did not find significantly increased ORs

(152-153), consistent with cannabis’ extended urine detection window. In the other study,

204 driver fatalities (blood THC ≥0.6 ng/mL) were compared to randomly-selected control drivers (OF THC ≥5 ng/mL) (6). The crude fatality OR was 13.9 for cannabis- positive drivers and retained significance (OR 8.6) after adjusting for demographics, time period, and season. Too few cannabis-only cases had accrued to establish a significant adjusted OR for THC alone. Two recent meta-analyses, each evaluating data from 9 epidemiological studies (2 in common), documented significantly increased MVA risk

(OR 2.66 (7) and 1.92 (154)), even after controlling for confounding variables.

Studies presenting culpability risk associated with cannabis are summarized in

Table 3 (see Table 7 (Supplemental) for additional detail). Increased blood THC concentrations were associated with higher culpability OR. In 2004, Drummer et al (4) demonstrated a statistically significant increase in adjusted driver crash responsibility OR

(2.7) with any measurable blood THC relative to drug-free drivers. The OR increased to

6.6, comparable culpability to 0.15% blood alcohol concentration (BAC), when blood

THC was ≥5 ng/mL. Among alcohol-negative drivers positive for cannabinoids, there was a significant 1.39 unadjusted OR for having at least one driver-related factor (DRF, a

Table 3. Risk of driver culpability or responsibility in motor vehicle accidents (MVA) after cannabis exposure THC CB OR Crude (C) or Location/ N N Analytical Para- 95% Cases Controls Cutoff Onl or Adjusted Ref. Country (cases) (controls) Matrix meter CI (ng/mL) y RR (factors) Demo- Fatally Fatally graphics, Victoria, injured injured Not CB MVA 1.02- crash type, NSW & WA, 58 1,704 Blood 2.7a (4) drivers, drivers, given only -DC 7.0 BAC, year, Australia THC + THC - year x crash type Demo- Fatally Fatally graphics, Victoria, injured injured CB MVA 1.5- crash type, NSW & WA, drivers, 49 1,704 Blood 5 6.6a (4) drivers, only -DC 28 BAC, year, Australia THC ≥5 THC - year x crash ng/mL type Drivers in Drivers

26 fatal crashes, in fatal MVA 1.86- France 298 9,013 Blood 1 -- 2.54a C (5) THC + crashes, -DC 3.48 1-2 ng/mL THC - Demo- Drivers in Drivers graphics, fatal crashes, in fatal MVA 1.09- BAC, Blood France 298 9,013 Blood 1 -- 1.54a (5) THC + crashes, -DC 2.18 THC concen- 1-2 ng/mL THC - tration, time of crash Drivers in Drivers fatal crashes, in fatal MVA 3.04- France 240 9,013 Blood 5 -- 4.72a C (5) THC + crashes, -DC 7.33 ≥5 ng/mL THC - Demo- Drivers in Drivers graphics, fatal crashes, in fatal MVA 1.32- BAC, Blood France 240 9,013 Blood 5 -- 2.12a (5) THC + crashes, -DC 3.38 THC concen- ≥5 ng/mL THC - tration, time of crash

Table 3. (Continued from previous page). Risk of driver culpability or responsibility in motor vehicle accidents (MVA) after cannabis exposure THC CB OR Crude (C) or Location/ N N Analytical Para- 95% Cases Controls Cutoff Onl or Adjusted Ref. Country (cases) (controls) Matrix meter CI (ng/mL) y RR (factors) Drivers Drivers in in fatal Blood or Not 0% MVA 1.21- United States fatal crashes, 1,647 30,896 1.39a C (144) crashes, Urine given BAC -DRF 1.59 THC + THC - Drivers Demo- Drivers in in fatal Blood or Not 0% MVA 1.11- graphics, United States fatal crashes, 1,647 30,896 1.29a (144) crashes, Urine given BAC -DRF 1.50 driving THC + THC - record a Odds ratio (OR) bRelative risk (RR) Abbreviations: THC: ∆9-tetrahydrocannabinol; CB: cannabis; CI: confidence interval; Ref.: reference; NSW: New South Wales; WA: Western Australia; DC: driver was judged culpable; DRF: ≥1 driver related factor (potentially unsafe driving action) contributed to crash; BAC: blood alcohol content

27 Please refer to Table 7 (Supplemental) for complete table, further stratified by THC concentration.

potentially unsafe behavior or action contributing to the collision) in 10 years’ Fatality

Analysis Reporting System (FARS) data (144). After controlling for demographics and driving record, the OR remained significant (1.29). FARS drug test results are based on blood or urine; including urine data may contribute to low ORs due to extended cannabinoid detection windows. In France, drivers in fatal crashes with detectable blood

THC had a 3.17 OR for crash responsibility (1.7 adjusted for demographics, BAC, THC concentration, and crash time) (5). Driver-responsibility OR increased with increasing blood THC. Crude (adjusted) ORs were 2.18 (1.57), 2.54 (1.54), 3.78 (2.13), and 4.72

(2.12) for <1, 1-2, 3-4, and ≥5 ng/mL, respectively. Although relatively few studies evaluated driver responsibility and cannabis intoxication, an increased blood THC concentration was strongly associated with driver MVA culpability.

The debate on cannabis’ effects on driving continues despite these findings, creating challenges for implementing effective drugged driving policies (4-5, 157). To date, 17 states and the District of Columbia have enacted medical marijuana laws (158).

Colorado, which legalized medical marijuana in 2000, has increased DUIC cases and is considering a 5 ng/mL blood THC per se law. This proposal generated strong debate, despite evidence showing increased culpability (159). In Drummer’s 58 cannabis-only culpable MVA cases (4), the median blood THC concentration was 12 ng/mL, with 84%

≥5 ng/mL. Increasing blood THC concentrations predict increasing driving impairment.

In 456 Norwegian suspected drugged drivers (1997-1999) whose blood was positive for cannabis only, the median [range] blood THC concentration was 2.2 [0.3-45.3] ng/mL

(146). A physician performed the clinical test for impairment (CTI), judging 54% of subjects as impaired. Grouping drivers by concentration range and adjusting for gender,

needle marks, and self-reported regular cannabis consumption produced impairment ORs of 2.4, 2.5, and 3.2 for 3.0-4.8, 4.9-10.1, and >10.2 ng/mL blood THC concentrations, respectively. Although some investigators describe a strong linear relationship between serum and OF THC concentrations, linear relationships between performance impairment and serum and OF concentrations are weak (160), and inter-subject variability shows that it is inaccurate to predict plasma concentrations from OF concentrations (121).

In Australia, it is illegal to drive with any detectable blood THC (161). Police randomly test blood or OF for THC. In the first year of testing, median [range] THC OF and blood concentrations were 81 [5-6484] ng/mL and 6 [3-19] ng/mL, respectively

(112). One year later, median OF THC was 66.5 ng/mL (median blood 6 ng/mL) (161).

Mean OF and blood concentrations were 274.3 and 7.6 ng/mL. In 2005, Switzerland imposed a punishable blood THC limit of 2.2 ng/mL (162). Of 1704 drivers confirmed

THC-positive (≥1.0 ng/mL), 1292 (76%) exceeded 2.2 ng/mL. Mean, median, and maximum blood concentrations were 5.8, 3.8, and 62 ng/mL, respectively. In cannabis- only cases (57.7%), mean (8.1 vs. 5.9 ng/mL) and median (5.8 vs. 4.1 ng/mL) concentrations were significantly higher for single vs. poly-drug users, respectively. A ten-year DUID study in Sweden (8794 THC-positive cases) revealed 2.1 and 1.0 ng/mL

(mean, median) blood THC concentrations (141). Drivers claiming regular cannabis consumption (177/456) were significantly less often (32% vs. 55%) judged impaired by

CTI than occasional smokers, with no difference in median blood THC concentration

(146). A multiple regression model controlling for THC concentration revealed a 1.8 impairment OR for occasional vs. regular cannabis smokers.

A 2007 international expert evaluation of epidemiological and experimental evidence concluded that risk-based legal limits were unsupportable (157). The consensus proposal was a 7-10 ng/mL serum lower limit based on meta-analyses of experimental data, with an added safety margin for individual variability and laboratory error. In contrast, Jones et al (141) advocated that zero-tolerance, based on THC limits of quantification (LOQ), is more pragmatic because any non-zero science-based per se laws would allow many individuals to evade prosecution. The debate is complicated by temporal dissociation between THC concentrations and acute driving impairment.

Karschner et al (163) recently reported THC in whole blood and plasma (≥0.25 ng/mL) from chronic, daily cannabis smokers for more than 7 days of continuously monitored abstinence. Recently Bosker et al (164) documented psychomotor impairment in chronic daily cannabis smokers relative to matched occasional drug users on critical tracking

(CT) and divided attention, validated driving impairment indicators, through 21-23 days of abstinence. Residual cognitive impairment (165) and withdrawal effects such as sleep disruption (166) were reported after chronic cannabis smoking. These effects may impair driving performance.

DUIC: Experimental Data

Experimental studies of driving performance under the influence of cannabis are the most rigorous way to evaluate impairment causality. Table 4 and Table 5 summarize laboratory studies on cannabis-induced neurocognitive function and driving (simulator and on-road). Study details including THC dose, participants’ cannabis history,

Table 4. Summarized effects of cannabis and alcohol on neurocognitive function: laboratory studies. Cannabis Intake Studies Studies History: Task/ Studies Studies Not Showing Showing Occasionalb (O), Outcome Showing THC Showing THC Cannabis- No Cannabis- Frequentc (F) or Measurea Impairment Impairment Alcohol Alcohol Not Specifiedd Interaction Interaction (NS) Free recall NS -- (167) -- (167) Time/Distance F -- (168) -- -- Perception NS (167) -- (167) -- O (41) ------RT F (41) (169) -- (169) NS (160) ------Standing Steadiness/Equil NS (170) (167) -- (167, 170) ibrium Wisconsin Card F (168) ------Sorting Task Digit-Symbol NS (167) -- -- (167) Substitution Test Backward Digit NS -- (167) -- -- Span Logical NS -- (167) -- (167) Reasoning Gambling Taske F (168) ------O -- (41) -- -- Tower of F -- (41, 169) -- (169) London NS (160) ------Virtual Maze F (168) ------O (41) ------CTT F -- (41, 169) -- (169) NS (160) ------O (41, 44) -- (44) -- DAT F (169) (41) (169) -- NS (167) -- -- (167) aPlease refer to Table 8 (Supplemental) (cannabis only) and Table 9 (Supplemental) (cannabis and alcohol) for specific outcome measures, THC doses, study details, and results. bOccasional denotes <12x/month cFrequent denotes ≥3x/week or ≥12x/month d Not specified denotes history not given, or various. Please refer to full Table 8 (Supplemental) and Table 9 (Supplemental) for specific cannabis history. eTHC effect on percentage of choosing less likely outcomes Abbreviations: THC, ∆9-tetrahydrocannabinol; RT, reaction time; CTT, critical tracking task; DAT, divided attention task

Table 5. Summarized effects of cannabis and alcohol on simulated and on-road driving. On- Cannabis Simulator Studies Simulator On-Road Road Studies Intake Studies Showing Studies Studies Studies Showing History: Not No Outcome Showing Showing Not Cannabis- Occasionalb Showing Cannabis- Measurea THC THC Showing Alcohol (O) or Not THC Alcohol Impair- Impair- THC Interact- Specifiedc Impair- Interact- ment ment Impair- ion (NS) ment ion ment O (29, 171) (172) ------RT NS (25, 173) (170) -- (47) (34, 47) (25, 170) Headway (25, 34, Mainten- NS (25) -- -- (47) -- 47) ance Headway NS (25) -- (34, 47) -- (34, 47) (25) Variability

Road O (29) (44, 172) -- -- (44) -- Tracking, SDLP NS (25) -- (34, 47) -- (34, 47) (25)

Road O (174) ------Tracking, Other NS (45) (173) -- (34, 47) (34, 47) -- (29, 44, O (173) -- -- (44) -- Speed 172) NS (25) -- -- (47) -- (25) Speed O (29, 171) (172) (44) ------(44) Variability NS (25) (173) -- (47) -- (25) Divided O (172) ------Attention NS (25) ------(25) Visual Search/ Processing O (174) -- -- (26) (26) -- Speed/ Short-term Memory Collisionsd O (29, 44) ------(44) -- a Please refer to Table 9 (Supplemental) for specific outcome measures, THC doses, study details, and results. bOccasional denotes <12x/month cNot specified denotes history not given, or various. Please refer to full Table 9 (Supplemental) for specific cannabis history. dToo few collisions in the studies for statistical analysis; data are reported as studywide number of collisions Abbreviations: THC, ∆9-tetrahydrocannabinol; RT, reaction time

performance measures and results are presented in Table 8 (Supplemental) and Table 9

(Supplemental). Past experimental studies often were inconclusive because outcome measures lacked sensitivity and were not tailored to specific THC effects (175-176).

Cannabis drivers appeared to be aware of impairment and attempted to compensate by driving more slowly and taking fewer risks (24-29). Perceived driving effort increased under the influence of THC (26). Others reported that it was not possible to fully compensate due to a control cost (25). THC’s impairing effects increase with task complexity; a realistic driving task involves subtasks requiring simultaneous attention.

Subjects performed worse on divided attention tasks (DATs, two or more subtasks performed simultaneously) (24-25, 167), when faced with unexpected circumstances and choices, and during long monotonous drives. Cannabis-associated impairment may manifest as failure to demonstrate expected practice effects, suggesting drivers lose some benefit afforded by prior experiences (28). Increased RT is among the most common cannabis-associated impairments (25, 29, 43, 173, 175, 177). Road tracking (maintaining correct road position) is one of the most sensitive, dose-dependent measures (47). THC increases lane position variability (weave) or standard deviation of lateral position

(SDLP) (25, 29, 47, 174), and steering wheel variability (25, 29). A recent study demonstrated significant THC-induced cognitive performance decrements (immediate recall, attention, working memory, executive function) (178) in occasional smokers with a wide range of prior cannabis exposure, 2-1000 lifetime episodes. One 3.95% THC cigarette produced similar body sway and brake latency to that observed with a 0.05% breath alcohol content (173).

Five controlled cannabis administration studies examined smoked cannabis’ effects on neurocognitive function (Table 4and Table 8 (Supplemental)). THC 13 and 17 mg doses did not produce time or distance perception effects in chronic daily cannabis smokers ~1.25 h after smoking (168). However, ≥1x/month smokers underestimated 60 and 120 sec intervals 1.25 h after two 3.6% THC cigarettes (4 puffs/cigarette) administered 2 h apart (167). THC (13 and 17 mg) produced minor but significant dose- dependent impairment on a card-sorting task ~0.75 h after smoking (168) and on a digit- symbol substitution test (167). An hour post-smoking, neither 13 nor 17 mg THC significantly affected decision-making speed in a gambling task (168), although after the higher dose the percentage of participants choosing least-likely outcomes was significantly higher than after placebo. Tower of London (decision-making) RT after 400 and 500 µg/kg THC (~28 and ~35 mg) was not significantly affected in occasional (≤1 day/week) or frequent (>4 days/week) smokers (41, 169). The number of correct decisions in the Tower of London Task was significantly decreased 0.75-5.75 h after 500

µg/kg THC among recreational (≥5x in past year) smokers (160), but was not significantly affected after 1 h among frequent (>4 days/week) smokers (41). Complex tasks requiring multiple neurocognitive and/or neuromotor skills were particularly sensitive to THC’s impairing effects and displayed less tolerance. In a virtual maze, 17 mg THC significantly increased wall collisions (5.5) relative to placebo (2.9) and 13 mg

THC (3.2) (168). Significant CT performance decrements were observed in occasional smokers 0.25-5.25 h after 250 µg/kg (~17.5 mg) THC (160) and 0.17-7.08 h after 500

µg/kg (41, 160). In these experiments THC did not significantly affect CT in frequent (>4

days/week) smokers (41, 169), but DATs (41, 167, 169) and RT (stop signal task, SST)

(41, 160) reflected impairment in frequent as well as occasional smokers.

Simulator and on-road studies are briefly summarized in Table 5 and fully characterized in Table 9 (Supplemental). Driving simulator studies offer greater face validity than laboratory studies for measuring THC driving effects, and less risk for participants. Simulators also allow measurement of specific performance decrements in ways unachievable in real-road driving experiments. Nine simulator experiments were examined. RT, road tracking, speed, and SD speed were the most commonly measured outcomes. THC dose-dependently increased RT in 4 of 6 studies (25, 29, 170-173). Low

13 and 17 mg THC doses produced significant and dose-dependent increases in RT in a

DAT (29), suggesting particular sensitivity to THC effects. Only 1 simulator experiment included headway maintenance (25). Smoked THC (19 and 38 mg) significantly and dose-dependently increased mean and SD headway relative to placebo. The most sensitive road tracking measure was SDLP, showing THC-associated impairment in 2 of

4 studies. Relatively low-dose (13 and 17 mg) smoked THC increased SDLP relative to placebo in occasional (1-4x/month) smokers (29), and 19 and 38 mg also produced significant (4 and 7 cm, respectively) increases (25). No significant SDLP increase after

13 mg in 1-4x/month smokers (44) or 22.9 mg in 1-10x/month smokers (28) was reported. Other road tracking outcomes were cones knocked over (173), percent time in lane (174), and “straddled line” variables (45). Significant THC-induced impairment was demonstrated 60-330 min (174) and 80 min (45) after 14-52 mg THC. There were only trends toward impairment in “straddling the solid line” (dividing different direction lanes,

p=0.09) and “straddling the barrier line” (broken line dividing same direction lanes, p=0.08) 30 min post-smoking (45).

Standardized Field Sobriety Tests (SFSTs) 55 and 105 min after smoking corresponded to 80 min simulator results in >65% of cases for 14 and 52 mg THC doses.

Several participants were classified “impaired” on SFSTs when driving performance suggested otherwise, but line straddling is a relatively insensitive impairment measure.

Four of 5 studies detected compensatory decreases in mean speed after THC doses ≥13 mg THC (25, 28-29, 44). When cannabis history was described, all participants were occasional smokers. No significant THC effect was observed (1.77% or 3.95% THC cigarettes) for participants who smoked at least weekly but not daily (173). Speed variability increased after THC relative to placebo in 3 of 6 studies (25, 29, 171), suggesting less vehicular control. DATs were employed in only 2 simulator studies. After

22.9 mg THC, participants failed to demonstrate practice effects observed under placebo on a paced auditory serial addition test during an otherwise uneventful drive (28). During combined car-following and sign detection tasks, 38 mg THC resulted in increased mean and SD headway (25). In occasional smokers, 45.7 mg THC also decreased visual search and processing speed (174).

A series of on-road studies (26, 34, 47) conducted in the Netherlands evaluated smoked THC effects on actual driving performance (Table 5). In a 22-km road-tracking closed course test, 100, 200, and 300 µg/kg (~7, ~14, and ~21 mg) smoked THC increased SDLP relative to placebo with no significant differences in mean or SD speed

(26). SDLP impairment was the same 40 and 100 min after starting smoking. In a highway escalating-dose experiment (100, 200, 300 µg/kg THC), 16 participants drove

beginning 45 min after commencing smoking (47). The drive included a 16 km car- following task (~15 min), 64 km road tracking (~50 min), and a second 16 km car- following task. THC increased SDLP dose-dependently: lowest dose produced slight and non-significant SDLP elevation, medium dose a significant but modest increase, and highest dose a significant and substantial increase. Mean headway in the car-following test increased 8, 6, and 2 m for the 100, 200, and 300 µg/kg doses, respectively. The authors suggested the inverse headway-dose relationship was a practice effect from decreasing driver caution with increasing task experience, rather than pharmacodynamic tolerance. THC (100 and 200 µg/kg) impaired driving performance on 40 km car- following and road tracking tasks (34). Drives were conducted 30 and 75 min after smoking. SD headway and SDLP significantly increased relative to placebo after each active dose (SD headway, by 2.9 and 3.8 m for 100 and 200 µg/kg; SDLP, by 2.7 and 3.5 cm, respectively). Participants’ SDLP was higher in the second than first drive. The final on-road study administered placebo or 100 µg/kg smoked THC 25 min prior to driving

45 min through a city (26). Performance was evaluated with the Driving Proficiency Test.

THC had no significant effect on total score, vehicle checks, handling, action in traffic, traffic observation, or turning.

Combined Alcohol and Cannabis Intake

Cannabis and alcohol share some cognitive and psychomotor effects (29, 160,

179). Both are central nervous system (CNS) depressants, and alcohol activates the cannabinoid CB1 receptor pathway (180); however, different effects on driving behavior were noted at the THC doses evaluated. Alcohol resulted in faster driving (44), while

typically-studied cannabis doses reduced driving speed. Alcohol inflates self-confidence, causing underestimation of impairment (29). In contrast, cannabis-influenced drivers occasionally appear more cautious in experimental settings.

Alcohol and cannabis are commonly identified together in MVA victims. DUIC is more common among people who also drive drunk (181). Among 322 MVA victims,

30% THC-positive individuals had been drinking alcohol also (133). A larger French case-control study found over 40% of 681 THC-positive drivers involved in fatal crashes with BACs above the 0.05% legal limit (5). Over a 90-day period, 30.6% of 108 drivers admitted to the University of Maryland Medical Center Shock Trauma Center were alcohol-positive (93.5% with BAC ≥0.07%); one third also tested positive (urinalysis, 50 ng/mL cutoff) for cannabis (analyte not specified) (145). Alcohol was detected with THC in nearly 20% of Swiss cases where THC exceeded the 2.2 ng/mL legal limit in blood

(162).

Among 727 French drivers in fatal accidents with blood THC ≥1 ng/mL, 40% also had an illegal BAC of ≥0.05% (142). Drunk driving (with or without cannabis) produced a higher single-vehicle accident incidence (62%) than did DUIC (34%). Drivers whose blood contained only cannabis were 2.3 times more likely to be culpable than those without cannabis or alcohol. This responsibility index (% responsible/% not responsible) increased to 9.4 for those with alcohol only, and to 14.1 for combined alcohol and cannabis. THC-positive drivers with BAC ≥0.05% had a 2.9 culpability OR relative to those with BAC ≥0.05% alone (4), implying that THC enhanced alcohol’s impairing effects. In a case-controlled logistic regression analysis, patients in Ontario,

Canada seeking treatment for combined alcohol and cannabis abuse had significantly

higher prior driving while intoxicated (DWI) conviction likelihood (OR 3.65) relative to random driver controls matched by age and sex (182). Cannabis-only patients were not significantly different from controls for DWI convictions, while alcohol-only patients had a 5.19 prior DWI conviction OR. DWI convictions were not necessarily concurrent with consumption of drug(s) for which individuals were subsequently treated. A significant

5.8 RR for driving-related injury within an hour following cannabis exposure (case- crossover self-report study) nearly doubled to 10.9 for alcohol and cannabis combined

(151).

Four studies included laboratory data on alcohol and cannabis interactions (Table

4 and Table 10 (Supplemental)). In a time estimation task, two 3.6% THC cigarettes (4 puffs each, 2 h apart) yielded underestimated time targets (167). Alcohol (0.6/0.5 g/kg

[male/female]) produced overestimations. In combination, these effects canceled each other. Two of 3 studies showed cannabis-alcohol interaction on DATs, suggesting this was a sensitive measure. In occasional (1-4x/month) smokers, some THC (13 mg) or alcohol (target BAC 0.05%) impairment effects occurred 15 to 75 min after smoking

(44). Although each substance increased false alarm responses, the greatest effect occurred in combination. Performance impairment and subjective effects were generally strongest after consuming both drugs. Frequent (≥4 days/week) smokers showed increased control losses after 400 µg/kg THC and 0.05 and 0.07% BACs (169).

Combinations produced the greatest effects, although it was unclear whether alcohol and

THC produced additive or synergistic effects. The percentage of drivers judged impaired increased with increasing blood THC concentrations and BAC (48). When neither alcohol (blood cutoff 0.001%) nor THC (blood LOD 0.2 ng/mL) was detected, 14% of

CTI observations showed impairment. Alcohol alone at BACs 0.001-0.05% (low) and

>0.05% (high) was associated with 77 and 95% impairment, respectively. THC concentrations between 0.30-1.6 ng/mL were associated with 45, 91, and 97% impairment for 0, low, and high BACs, respectively. THC ≥1.6 ng/mL was associated with 53, 93, and 100% impairment. These CTI data depict progressive and increasing impairment with alcohol and THC combinations.

One simulator study showed a cannabis-alcohol interaction (Table 5). In occasional (1-4x per month) smokers, 0.05% target BAC and 13 mg smoked THC increased SDLP relative to either drug alone or placebo, which did not differ (44).

Alcohol alone increased drive speed relative to THC; THC alone decreased speed. The combination produced a borderline-significant speed increase relative to THC alone. No significant effect was observed on SD speed. Alcohol and THC increased the total number of collisions (5/12) relative to either drug alone (2/12 and 3/12 for alcohol and

THC, respectively).

The most straightforward cannabis-alcohol effect appeared in a study that administered 0, 100 (~7 mg), or 200 (~14 mg) µg/kg THC and alcohol (target BACs 0 or

0.04%) (Table 5) (34, 47). Alcohol plus the high THC dose increased RT 36%; this was the only dosing condition to affect RT. Alcohol or low THC alone slightly increased

SDLP; the higher THC dose caused moderate impairment. Neither drug alone significantly increased time out of lane. Combining either THC dose with alcohol severely increased SDLP and dose-dependently increased time out of lane. Combining the 100 and 200 µg/kg doses with 0.04% BAC created impairment equivalent to 0.09 and

0.14% BAC, respectively. Visual search for traffic at intersections was significantly

decreased 3% relative to placebo following 0.04% BAC and 100 µg/kg THC, an effect not observed with either drug alone (26).

Although several studies reported additive or synergistic cannabis-alcohol effects, some studies reported no interactions. Inconsistencies likely arose from differing procedures, outcome measures, and cannabis history in study populations. In occasional smokers, after combining moderate drug doses no interaction effect was observed on free word recall, digit-symbol substitution, logical reasoning (167), standing steadiness or equilibrium (167, 170). Frequent smokers did not show cannabis-alcohol interaction on a

Tower of London task, SST, or CT after 400 µg/kg (~28 mg) THC and 0.05% or 0.07%

BAC (169). A 3.3% (~30 mg) THC cigarette decreased equilibrium scores and 0.5 g/kg alcohol increased brake latency, with neither effect significantly altered with the combination (170). Authors speculated that the lack of interaction was an impairment awareness artifact, coupled with expectation of an “emergency.” A recent 9-session simulator experiment investigating 0, 19, and 38 mg THC and 0, 0.4, and 0.6 g/kg alcohol in all possible combinations demonstrated substantial impairment from THC alone (both active doses) (25). Both alcohol doses increased SDLP; the lower alcohol dose also increased mean and SD speed. No interactive effects were observed, most likely because alcohol doses were low. Mean achieved BACs were 0, 0.02 and 0.05% for placebo, low and high alcohol conditions, respectively.

Preventing DUIC

One of the greatest challenges is dealing with public attitudes regarding DUIC.

One fourth (26.3%) of 320 drivers who smoked cannabis in the previous year indicated

>90% DUIC likelihood even after being shown data on increased crash risk (183). Only

7.5% reported they would be unlikely (0-10% likelihood) to drive. The majority indicated

>50% DUIC probability in the future, even given higher MVA risk. Regular smokers who had previously DUIC emphasized that publicity campaigns would not deter them from future DUIC (184). Past experience convinced them that they could compensate for cannabis-associated performance decrements. Most believed cannabis caused minimal driving impairment; a few considered it to have no, or even a positive, effect upon driving. High likelihood of apprehension and punishment was a better deterrent. Given a hypothetical scenario with no chance of punishment, three quarters indicated >50% chance for DUIC, and half indicated >90% likelihood (183). In contrast, given a hypothetical scenario with high punishment certainty, participants were significantly less willing to DUIC (OR, 0.2, p<0.001). However, in a small study on DUIC attitudes, none who reported being stopped by police while DUIC indicated being deterred by this experience (none were charged) (184). Study findings suggested that random roadside testing (with arrest of those found cannabis-positive) would be a better deterrent than advertising campaigns promoting the hazards of DUIC.

Conclusions/Discussion

Many epidemiology studies involved selection bias. Some only looked at specific populations, such as those being treated for substance abuse or addiction, or deceased drivers. Case-control studies are highly useful, but may be biased by control selection.

Case and control populations may come from different time periods or include different cannabis detection cutoffs or matrices (e.g., blood analysis for killed vs. OF for living

drivers). The accuracy of self-reported information varies, depending on data collection methods. Self-reported prevalence estimates are often underestimated due to the sensitivity of illicit drug-related information (149-150). Even when objective measures of cannabis exposure were employed, detection cutoffs varied and were not always reported.

Urine testing has an extended cannabis detection window and cannot establish a valid temporal association with crash risk.

Several studies showed increased crash and culpability risks, even after adjusting for confounders such as age, sex, risky behaviors, and polypharmacy. Increased blood

THC concentrations and driving within an hour after smoking were strongly associated with higher crash and culpability risks. Human laboratory controlled drug administration studies showed THC-induced driving performance decrements within the first hour lasting ≥2 h after smoking, largely consistent with epidemiological data. Laboratory- based impairment experiments identified DATs and executive function tasks as most sensitive to cannabis’ effects. Evaluating actual driving performance demonstrated dose- dependent THC impairment in road tracking, although low to moderate THC doses were administered due to safety concerns. Simulator technology improved since first employed for impairment experiments, progressing from rudimentary controls and a projected cyclorama (171) to full passenger cars, interactive screens with more complete fields of view, advanced driver monitoring through built-in cameras, realistic three-dimensional audio output to enhance the simulated drive, and motion platforms simulating physical driving sensations (25, 28-29). Such advanced driving simulators are an ideal platform for future research, combining realistic driving scenarios with highly-controlled and measurable environments and providing a degree of safety not possible in on-road

experiments. Driving simulators are particularly necessary when challenging drivers with difficult tasks, such as DATs and real-time decision-making processes, and when evaluating higher THC doses. Simulator studies from the last decade demonstrated significant THC impairment on RT and SDLP. THC (22.9 and 38 mg) inhibited expected practice effects on DATs and produced deleterious performance effects, respectively (25,

28). Depending upon THC dose, driving speed may have been reduced (25, 29, 44), particularly while multitasking (28), and headway increased (25). These results suggest impairment awareness and compensation, but did not preclude control decrements.

Inconsistencies in findings likely result from differences in study design, population and setting (simulator vs. road), sophistication of equipment, sensitivity and specificity of tasks (including drive length), THC dosage and time after smoking.

Differences in driver compensation while under observation may contribute to variability.

Future work should focus on extended segments of monotonous driving (drawing drivers into a state of complacency or sleepiness) followed by sudden changes requiring reaction, realistic situations presenting decision dilemmas, and DATs. These constructs appear most sensitive to THC’s impairing influence.

Combining alcohol with THC exacerbated observed effects, especially on RT and

SDLP. Low (100 µg/kg) and moderate (200 µg/kg) THC doses combined with 0.04%

BAC produced road tracking impairment to a degree associated with 0.09% and 0.14%

BAC (47). Because consuming alcohol and cannabis together is common, fully evaluating their combined impact on driving performance is essential. Employing clinically relevant THC and alcohol doses in future studies is necessary to generate findings that better inform public policy. In past research, administering low THC and

alcohol doses often accounted, at least in part, for lack of observed effects. Cannabis smokers typically self-titrate doses, and alcohol drinkers consume enough alcohol to attain BACs ≥0.05%. Most self-administered cannabis and alcohol doses are higher than doses administered in many research studies.

Tolerance to acute impairment is an important consideration for future research and policy debate. DAT and tracking tasks demonstrated impairment in chronic cannabis smokers, while other parameters did not (41, 169). The debate regarding per se and zero tolerance drugged driving laws is a prominent issue. Increased blood THC concentrations are strongly associated with increased crash risk, but there is not a direct correlation between driving impairment and THC concentrations.

DUIC is an important public health and safety concern requiring development of evidence-based drugs and driving policy and legislation. Impaired driving endangers individuals inside and outside the vehicle. Consuming cannabis prior to driving, with or without alcohol, is a common occurrence resulting in significant morbidity and mortality on the roadway. Research is needed to further define cannabis effects on driving performance and to provide the scientific basis for laws to improve road safety.

Table 6 (Supplemental). Self-reported risk of motor vehicle accident (MVA) after cannabis intake Cannabis CB Crude (C) 95% CI Location/ N N Intake exposure CB Para- OR or or N total Cases Controls or Ref. Country cases controls History or time prior Only? meter RR Adjusted p-value DUIC to driving (factors) 10th & 12th 10th & grade Atlantic 12th grade students, no 4,055 877 3,178 DUIC ≤1 h -- MVA 4.14a p<0.001 C (143) Canada students, CB DUIC consumption 10th & 12th 10th & grade Demogra- Atlantic 12th grade students, no 4,055 877 3,178 DUIC ≤1 h -- MVA 1.84a p<0.001 phics, (143) Canada students, CB DUIA DUIC consumption Frequency- CB Demogra- 46 matched Chronic/ Ontario, addiction b 1.17- phics,

699 181 (age, sex) 518 abuse -- -- MVA 1.49 (155) Canada treatment 1.89 cocaine licensed history patients abuse drivers Adult Adult drivers drivers who Ontario, 1.08- Demogra- 2,676 who con- 1,097 never 1,579 History -- -- MVA 1.47a (149) Canada 1.99 phics sumed CB consumed in lifetime CB Adult Adult drivers w/ drivers w/ Ontario, 0.80- Demogra- 2,611 past year 254 no past- 2,357 History -- -- MVA 1.24a (149) Canada 1.93 phics CB intake year CB ≤1x/wk intake Adult Adult drivers w/ drivers w/ Ontario, 1.50- Demogra- 2,422 past year 65 no past- 2,357 History -- -- MVA 2.76a (149) Canada 5.08 phics CB intake year CB >1x/wk intake

Table 6 (Supplemental). (Continued from previous page) Self-reported risk of motor vehicle accident (MVA) after cannabis intake Cannabis CB Crude (C) 95% Location/ N N Intake exposure CB Para- OR or or N total Cases Controls CI or Ref. Country cases controls History or time prior Only? meter RR Adjusted p-value DUIC to driving (factors) Adult Adults, drivers, did drove not drive Ontario, within 1 h 2.30- 2,676 70 within 1 h 2,606 DUIC ≤1 h -- MVA 3.89a C (149) Canada of CB 6.59 of CB intake in intake in past year past year Adult Adults, drivers, did drove not drive Ontario, within 1 h 1.45- Demogra- 2,676 70 within 1 h 2,606 DUIC ≤1 h -- MVA 2.61a (149) Canada of CB 4.68 phics of CB intake in intake in past year 47 past year

Did not Drove Auckland drive within 3 h 3.6- New 1,139 37 within 3 h 1,102 DUIC ≤3 h -- MVA-Ic 11.4a C (150) of CB 35.4 Zealand of CB intake intake Demogra- phics, time of day, no. Did not Drove passen- Auckland drive within 3 h gers, risky New 1,139 37 within 3 h 1,102 DUIC ≤3 h -- MVA-Ic 0.8a 0.2-3.3 (150) of CB behavior Zealand of CB intake (BAC, intake speed, seatbelt use) Car Demogra- drivers, Car drivers, phics, con- MVA- Spain 155 11 no CB con- 144 History -- -- 1.2a 0.5-2.8 distance (156) sumed CB NFI sumption driven, 1-4 other drugs days/wk

Table 6 (Supplemental). (Continued from previous page) Self-reported risk of motor vehicle accident (MVA) after cannabis intake Cannabis CB Crude (C) 95% CI Location/ N N Intake exposure CB Para- OR or or N total Cases Controls or Ref. Country cases controls History or time prior Only? meter RR Adjusted p-value DUIC to driving (factors) Car Demogra- drivers, Car drivers, phics, con- MVA- Spain 156 12 no CB con- 144 History -- -- 2.5a 1.2-5.1 distance (156) sumed NFI sumption driven, CB >4 other drugs days/wk Regular Same pop- cocaine ulation, not users, NFDI, driving Spain 68d driving 68 68 DUIC ≤1 h -- past 7.0b 3.1-16 -- (151) within 1 h within 1 year of CB h of CB intake intake

48 Regular Same pop-

cocaine ulation, not users, NFDI, driving Spain 68d driving 68 68 DUIC ≤2 h -- past 2.2b 1.0-4.9 -- (151) within 2 h within 2 year of CB h of CB intake intake Regular Same pop- cocaine ulation, not users, NFDI, driving CB Spain 45d driving 45 45 DUIC ≤1 h past 5.8b 2.4-14 -- (151) within 1 h only within 1 year of CB h of CB intake intake Regular Same pop- cocaine ulation, not users, NFDI, driving CB Spain 45d driving 45 45 DUIC ≤2 h past 2.2b 0.9-5.2 -- (151) within 2 h only within 2 year of CB h of CB intake intake

Table 6 (Supplemental). (Continued from previous page) Self-reported risk of motor vehicle accident (MVA) after cannabis intake Cannabis CB Crude (C) 95% CI Location/ N N Intake exposure CB Para- OR or or N total Cases Controls or Ref. Country cases controls History or time prior Only? meter RR Adjusted p-value DUIC to driving (factors) Regular cocaine users, driving Same pop- NFDI, +Alc in Spain 19d within 1 h 19 ulation, not 19 DUIC ≤1 h past 10.9b 1.3-88 -- (151) past 2 h of CB DUID year intake & 2 h of alcohol aOR (odds ratio) bRR (relative risk) cCars in MVA with ≥1 occupant hospitalized or killed dCase-crossover design

49 Abbreviations: DUIA, driving under the influence of alcohol; CB, cannabis; CI, confidence interval; Ref., reference; NFI, non-fatal injury; NFDI, non-fatal driver injury; BAC,

blood alcohol content

Table 7 (Supplemental). Risk of driver culpability or responsibility in motor vehicle accidents (MVA) after cannabis exposure Crude (C) THC OR Location/ N N Analytical CB Para- 95% or N (total) Cases Controls Cutoff or Ref. Country (cases) (controls) Matrix Only meter CI Adjusted (ng/mL) RR (factors) Demogra- Victoria, Fatally Fatally phics, crash NSW & injured injured Not CB MVA 1.02- 1,762 58 1,704 Blood 2.7a type, BAC, (4) WA, drivers, drivers, given only -DC 7.0 year, year x Australia THC + THC - crash type Fatally Demogra- Victoria, Fatally injured phics, crash NSW & injured CB MVA 1.5- 1,753 drivers, 49 1,704 Blood 5 6.6a type, BAC, (4) WA, drivers, only -DC 28 THC ≥5 year, year x Australia THC - ng/mL crash type Canna- Fre- bis quency-

50 N/A: Demogra- addict- matched

Ontario, Collision/ MVA 1.21- phics, 699 ion 181 (age, 518 -- -- 1.68b (155) Canada conviction -DC 2.34 cocaine treat- sex) records abuse ment licensed patients drivers Drivers Drivers in fatal in fatal Not MVA 2.56- France 9,772 759 9,013 Blood -- 3.17a C (5) crashes, crashes, given -DC 3.94 THC + THC - Demogra- phics, Drivers Drivers BAC, in fatal in fatal Not MVA 1.40- France 9,091 759 9,013 Blood -- 1.78a blood THC (5) crashes, crashes, given -DC 2.25 concentra- THC + THC - tion, time of crash

Table 7 (Supplemental). (Continued from previous page) Risk of driver culpability or responsibility in motor vehicle accidents (MVA) after cannabis exposure Crude (C) THC OR Location/ N N Analytical CB Para- 95% or N (total) Cases Controls Cutoff or Ref. Country (cases) (controls) Matrix Only meter CI Adjusted (ng/mL) RR (factors) Drivers in fatal Drivers crashes, in fatal Not MVA 1.22- France 9,091 78 9,013 Blood -- 2.18a C (5) THC + crashes, given -DC 3.89 <1 THC - ng/mL Demogra- Drivers phics, in fatal Drivers BAC, crashes, in fatal Not MVA 0.84- France 9,091 78 9,013 Blood -- 1.57a blood THC (5) THC + crashes, given -DC 2.95 concentra- 51 <1 THC - tion, time ng/mL of crash Drivers in fatal Drivers crashes, in fatal MVA 1.86- France 9,311 298 9,013 Blood 1 -- 2.54a C (5) THC + crashes, -DC 3.48 1-2 THC - ng/mL Demogra- Drivers phics, in fatal Drivers BAC, crashes, in fatal MVA 1.09- France 9,311 298 9,013 Blood 1 -- 1.54a blood THC (5) THC + crashes, -DC 2.18 concentra- 1-2 THC - tion, time ng/mL of crash

Table 7 (Supplemental). (Continued from previous page) Risk of driver culpability or responsibility in motor vehicle accidents (MVA) after cannabis exposure Crude (C) THC OR Location/ N N Analytical CB Para- 95% or N (total) Cases Controls Cutoff or Ref. Country (cases) (controls) Matrix Only meter CI Adjusted (ng/mL) RR (factors) Drivers in fatal Drivers crashes, in fatal MVA 2.24- France 9,156 143 9,013 Blood 3 -- 3.78a C (5) THC + crashes, -DC 6.37 3-4 THC - ng/mL Demogra- Drivers phics, in fatal Drivers BAC, crashes, in fatal MVA 1.22- France 9,156 143 9,013 Blood 3 -- 2.13a blood THC (5) THC + crashes, -DC 3.73 concentra- 52 3-4 THC - tion, time ng/mL of crash Drivers in fatal Drivers crashes, in fatal MVA 3.04- France 9,253 240 9,013 Blood 5 -- 4.72a C (5) THC + crashes, -DC 7.33 ≥5 THC - ng/mL Demogra- Drivers phics, in fatal Drivers BAC, crashes, in fatal MVA 1.32- France 9,253 240 9,013 Blood 5 -- 2.12a blood THC (5) THC + crashes, -DC 3.38 concentra- ≥5 THC - tion, time ng/mL of crash Drivers Drivers United in fatal in fatal Blood or Not 0% MVA 1.21- 32,543 1,647 30,896 1.39a C (144) States crashes, crashes, Urine given BAC -DRF 1.59 THC + THC -

Table 7 (Supplemental). (Continued from previous page) Risk of driver culpability or responsibility in motor vehicle accidents (MVA) after cannabis exposure Crude (C) THC OR Location/ N N Analytical CB Para- 95% or N (total) Cases Controls Cutoff or Ref. Country (cases) (controls) Matrix Only meter CI Adjusted (ng/mL) RR (factors) Drivers Drivers Demogra- United in fatal in fatal Blood or Not 0% MVA 1.11- phics, 32,543 1,647 30,896 1.29a (144) States crashes, crashes, Urine given BAC -DRF 1.50 driving THC + THC - record a Odds ratio (OR) bRelative risk (RR) Abbreviations: THC, ∆9-tetrahydrocannabinol; CB, cannabis; CI, confidence interval; Ref., reference; NSW, New South Wales; WA, Western Australia; DC, driver was judged culpable; DRF, ≥1 driver related factor (potentially unsafe driving action) contributed to crash; BAC, blood alcohol content

Table 8 (Supplemental). Effect of smoked cannabis on neurocognitive function: laboratory studies Age (years) Time (h) Cannabis Task THC Dose N Mean±SD, Measure After THC Outcome Relative to Ref. History [Range] Dosing Memory Two 3.6% THC ≥1x/Month; joints; 10/14: >100 # Words Free recall 14 24.5, [21-34] ~1.25-1.5b ns Placebo (167) 4 puffs/joint lifetime recalled spaced by 2 h occasions Distance/Time Perception Estimated time & ≥1 Joint/day for Distance distance to 13, 17 mg 14 27±7.45 ~1.25a ns Placebo (168) ≥5 years estimation approaching car Estimated time & ≥1 Joint/day for Time a

54 distance to 13, 17 mg 14 27±7.45 ~1.25 ns Placebo (168) ≥5 years estimation

approaching car Two 3.6% THC ≥1x/Month; Respond Underproduction Time joints (4 10/14: >100 when believe (estimated time 14 24.5, [21-34] ~1.25b Placebo (167) production puffs/joint) lifetime 30, 60, 120 intervals shorter spaced by 2 h occasions sec elapsed than targets) Logical/Mathematical Reasoning Skills # Blocks ≥1 Joint/day for -1.2 WCS 13 mg 14 27±7.45 completed ~0.75a Placebo (168) ≥5 years (16.6 vs. 17.8) without error # Blocks ≥1 Joint/day for -1.6 Placebo WCS 17 mg 14 27±7.45 completed ~0.75a (168) ≥5 years (16.2 vs. 17.8) without error # Non- ≥1 Joint/day for +1.1 WCS 13 mg 14 27±7.45 perseverance ~0.75a Placebo (168) ≥5 years (4.3 vs. 3.2) errors # Non- ≥1 Joint/day for +2.1 WCS 17 mg 14 27±7.45 perseverance ~0.75a Placebo (168) ≥5 years (5.3 vs. 3.2) errors

Table 8 (Supplemental). (Continued from previous page) Effect of smoked cannabis on neurocognitive function: laboratory studies Age (years) Time (h) Cannabis Task THC Dose N Mean±SD, Measure After THC Outcome Relative to Ref. History [Range] Dosing ≥1 Joint/day for +1.2 Placebo WCS 13 mg 14 27±7.45 # Total errors ~0.75a (168) ≥5 years (5.0 vs. 3.8) ≥1 Joint/day for +1.9 WCS 17 mg 14 27±7.45 # Total errors ~0.75a Placebo (168) ≥5 years (5.7 vs. 3.8) Two 3.6% THC ≥1x/Month; joints; 10/14: >100 Percentage -4.8% Before drug DSST 14 24.5, [21-34] ~1.25-1.5b (167) 4 puffs/joint lifetime correct (91.3 vs. 96.1%) dose spaced by 2 h occasions Two 3.6% THC ≥1x/Month; Backward joints; 10/14: >100 Before drug 14 24.5, [21-34] -- ~1.25-1.5b ns (167) digit span 4 puffs/joint lifetime dose spaced by 2 h occasions Two 3.6% THC ≥1x/Month; True/False Logical joints; 10/14: >100 statements Before drug 55 14 24.5, [21-34] ~1.25-1.5b ns (167) reasoning 4 puffs/joint lifetime about letter dose

spaced by 2 h occasions pairs Decision-Making ≥1 Joint/day for Decision Gambling 13, 17 mg 14 27±7.45 ~1.0a ns Placebo (168) ≥5 years making speed Percentage ≥1 Joint/day for choices of less Gambling 13 mg 14 27±7.45 ~1.0a ns Placebo (168) ≥5 years likely outcomes Percentage ≥1 Joint/day for choices of less Gambling 17 mg 14 27±7.45 ~1.0a significantly higher Placebo (168) ≥5 years likely outcomes Tower of 250 µg/kg # Correct 20 19-29 ≥5x in past year 0.75 -2.1±1.1 Placebo (160) London (~17.5 mg) decisions Tower of 250 µg/kg # Correct 20 19-29 ≥5x in past year 1.75 -2.2±1.1 Placebo (160) London (~17.5 mg) decisions Tower of 250 µg/kg # Correct 20 19-29 ≥5x in past year 5.75 +0.3±0.8 Placebo (160) London (~17.5 mg) decisions

joints/occasion >4 Days/week in past year, Tower of 500 µg/kg (~35 340±86 # Correct 12 23.2±3.3 1.0 ns Placebo (41) London mg) occasions/year, decisions 2.3±1.2 joints/occasion >4 Days/week in past year, Tower of 500 µg/kg (~35 340±86 12 23.2±3.3 RT 1.0 ns Placebo (41) London mg) occasions/year, 2.3±1.2 joints/occasion

57 ≥1 Joint/day for Completion Immediately Virtual Maze 13, 17 mg 14 27±7.45 ns Placebo (168)

≥5 years time after vitals Mean # wall ≥1 Joint/day for Immediately Virtual Maze 13 mg 14 27±7.45 collisions per ns Placebo (168) ≥5 years after vitals trial Mean # wall ≥1 Joint/day for Immediately +2.6 Virtual Maze 17 mg 14 27±7.45 collisions per Placebo (168) ≥5 years after vitals (5.5 vs. 2.9) trial Mean # wall ≥1 Joint/day for Immediately +2.3 Virtual Maze 17 mg 14 27±7.45 collisions per 13 mg THC (168) ≥5 years after vitals (5.5 vs. 3.2) trial 250 µg/kg CTT 20 19-29 ≥5x in past year λcc 0.25 -0.21±0.13 rad/sec Placebo (160) (~17.5 mg) 250 µg/kg CTT 20 19-29 ≥5x in past year λc 1.25 -0.40±0.16 rad/sec Placebo (160) (~17.5 mg) 250 µg/kg Placebo CTT 20 19-29 ≥5x in past year λc 3.25 -0.18±0.13 rad/sec (160) (~17.5 mg) 250 µg/kg CTT 20 19-29 ≥5x in past year λc 5.25 -0.17±0.13 rad/sec Placebo (160) (~17.5 mg)

Table 8 (Supplemental). (Continued from previous page) Effect of smoked cannabis on neurocognitive function: laboratory studies Age (years) Time (h) Cannabis Task THC Dose N Mean±SD, Measure After THC Outcome Relative to Ref. History [Range] Dosing 500 µg/kg (~35 CTT 20 19-29 ≥5x in past year λc 0.25 -0.60±0.21 rad/sec Placebo (160) mg) 500 µg/kg (~35 CTT 20 19-29 ≥5x in past year λc 1.25 -0.55±0.17 rad/sec Placebo (160) mg) 500 µg/kg (~35 CTT 20 19-29 ≥5x in past year λc 3.25 -0.27±0.13 rad/sec Placebo (160) mg) 500 µg/kg (~35 CTT 20 19-29 ≥5x in past year λc 5.25 -0.48±0.08 rad/sec Placebo (160) mg) ≤1Day/week, 55±36 500 µg/kg (~35 CTT 12 22.8±2.3 occasions/year, λc 0.17 ~ -0.7 rad/sec Placebo (41) mg) 1.2±0.5 joints/occasion ≤1Day/week,

58 55±36 500 µg/kg (~35 CTT 11 22.8±2.3 occasions/year, λc 3.08 ~ -0.6 rad/sec Placebo (41) mg) 1.2±0.5 joints/occasion ≤1Day/week, 55±36 500 µg/kg (~35 CTT 11 22.8±2.3 occasions/year, λc 5.08 ~ -0.2 rad/sec Placebo (41) mg) 1.2±0.5 joints/occasion ≤1Day/week, 55±36 500 µg/kg (~35 CTT 11 22.8±2.3 occasions/year, λc 7.08 ~ -0.4 rad/sec Placebo (41) mg) 1.2±0.5 joints/occasion

Table 8 (Supplemental). (Continued from previous page) Effect of smoked cannabis on neurocognitive function: laboratory studies Age (years) Time (h) Cannabis Task THC Dose N Mean±SD, Measure After THC Outcome Relative to Ref. History [Range] Dosing >4 Days/week in past year, 500 µg/kg (~35 340±86 CTT 12 23.2±3.3 -- 0.17-7.08 ns Placebo (41) mg) occasions/year, 2.3±1.2 joints/occasion >4 Days/week in past year, 400 µg/kg (~28 23.2±8.4 373.7±101.6 Before THC CTT 21 λc 0.3-3.3 ns (169) mg) [19-38] occasions/year, dose 5.0±3.9 joints/occasion 2 0% or 3.6% ≥1x/Month; Response to THC joints (4 10/14: >100 Before drug 59 DAT 14 24.5, [21-34] primary ~1.25-1.5b ns (167) puffs/joint) lifetime dose target: hit rate spaced by 2 h occasions 2 0% or 3.6% ≥1x/Month; Response to THC joints (4 10/14: >100 Before drug DAT 14 24.5, [21-34] primary ~1.25-1.5b ns (167) puffs/joint) lifetime dose target: RT spaced by 2 h occasions 2 0% or 3.6% ≥1x/Month; THC joints (4 10/14: >100 # False alarm +0.7 Before drug DAT 14 24.5, [21-34] ~1.25-1.5b (167) puffs/joint) lifetime responses (1.6 vs. 0.9) dose spaced by 2 h occasions ≤1Day/week, 55±36 500 µg/kg (~35 DAT 12 22.8±2.3 occasions/year, Tracking error 0.33 ~ +5.0 mm Placebo (41) mg) 1.2±0.5 joints/occasion

60 2.3±1.2

joints/occasion >4 Days/week in past year, % Correct 500 µg/kg (~35 340±86 detections on DAT 12 23.2±3.3 0.33 ns Placebo (41) mg) occasions/year, secondary 2.3±1.2 task joints/occasion >4 Days/week in past year, 500 µg/kg (~35 340±86 # Control DAT 12 23.2±3.3 0.33 ns Placebo (41) mg) occasions/year, losses 2.3±1.2 joints/occasion

61 400 µg/kg (~28 23.2±8.4 373.7±101.6 Before THC DAT 21 RT 0.5-2.5 Increased (169) mg) [19-38] occasions/year, dose 5.0±3.9 joints/occasion Reaction Time/Motor Impulsivity 250 µg/kg SST 11 19-29 ≥5x in past year Stop RT 0.5 +17.5±17.2 msec Placebo (160) (~17.5 mg) 250 µg/kg SST 11 19-29 ≥5x in past year Stop RT 1.5 +10.9±22.9 msec Placebo (160) (~17.5 mg) 250 µg/kg SST 11 19-29 ≥5x in past year Stop RT 3.5 -3.3±20.1 msec Placebo (160) (~17.5 mg) 250 µg/kg SST 11 19-29 ≥5x in past year Stop RT 5.5 +4.8±11.4 msec Placebo (160) (~17.5 mg) 500 µg/kg (~35 SST 11 19-29 ≥5x in past year Stop RT 0.5 +60.6±21.6 msec Placebo (160) mg) 500 µg/kg (~35 SST 11 19-29 ≥5x in past year Stop RT 1.5 +64.6±25.0 msec Placebo (160) mg) 500 µg/kg (~35 SST 11 19-29 ≥5x in past year Stop RT 3.5 +12.9±12.0 msec Placebo (160) mg)

Table 8 (Supplemental). (Continued from previous page) Effect of smoked cannabis on neurocognitive function: laboratory studies Age (years) Time (h) Cannabis Task THC Dose N Mean±SD, Measure After THC Outcome Relative to Ref. History [Range] Dosing 500 µg/kg (~35 SST 11 19-29 ≥5x in past year Stop RT 5.5 +7.0±17.9 msec Placebo (160) mg) ≤1Day/week, 55±36 500 µg/kg (~35 SST 12 22.8±2.3 occasions/year, Stop RT 0.58 ~ +25 msec Placebo (41) mg) 1.2±0.5 joints/occasion >4 Days/week in past year, 500 µg/kg (~35 340±86 SST 12 23.2±3.3 Stop RT 0.58 ~ +110 msec Placebo (41) mg) occasions/year, 2.3±1.2 joints/occasion >4 Days/week in

62 past year,

400 µg/kg (~28 23.2±8.4 373.7±101.6 Before THC SST 21 Stop RT 1-3 ns (169) mg) [19-38] occasions/year, dose 5.0±3.9 joints/occasion >4 Days/week in past year, 400 µg/kg (~28 23.2±8.4 373.7±101.6 Commission Before THC SST 21 1-3 ns (169) mg) [19-38] occasions/year, errors dose 5.0±3.9 joints/occasion aDenotes times after smoking initiated bDenotes times after smoking completed cλc: Critical frequency (tracking frequency at which control loss occurs) Abbreviations: THC, ∆9-tetrahydrocannabinol; SD, standard deviation; ns, not significant; WCS, Wisconsin Card Sorting; DSST, Digit Symbol Substitution Test; RT, reaction time; CTT, critical tracking task; DAT, divided attention task; SST, stop signal task

Table 9 (Supplemental). Cannabis and alcohol effects on simulated and on-road driving Time Dose Age (min) Drive Outcome: Outcome: Measured THC EtOH (years) Cannabis Outcome: Measure N after Length/ EtOH Combina- Ref. by Dose (g/kg or Mean±SD, History THC only THC Time only tion target BAC) [Range] dose Simulator 0, 8, 12, 16 mg 3 Naïve; Med, Brake 8 Not dosed RT ∆1-THC 0.095% 21-29 all >105 10 min High THC ↑ 44% (171) latency M together (ate ≤1x/week ↑ 16, 66% cake) Random High THC 0, 1.77, Brake ≥Weekly, test ↑ 0.054 RT 3.95%; -- 10 21-45 ≤3.5 km -- -- (173) latency

63 interact- 0, 1.75, 2-21 of Brake ↑ ~0.06- ion: RT 3.33%; 0, 0.25, 0.5 12 21-45 past 30 25a 10 min ns (170) latency 0.07 sec ↑ ns from 10 puffs days EtOH alone Low THC ↑ RT to 0, 13, 17 1-4x/ 54.5 ↑ Same as Not dosed RT secondary 0, 0.5 14 26.1 ± 1.3 30 High THC (29) mg Month kmc Low THC together task ↑ more than Low THC 50 M Incursion 0, 22.9 1-10x/ RT -- 18-31 ~15b ~15 min ns -- -- (172) avoidance mg Month 35 F

Table 9 (Supplemental). (Continued from previous page) Cannabis and alcohol effects on simulated and on-road driving Time Dose Age (min) Drive Outcome: Outcome: Measured THC EtOH (years) Cannabis Outcome: Measure N after Length/ EtOH Combina- Ref. by Dose (g/kg or Mean±SD, History THC only THC Time only tion target BAC) [Range] dose High THC Sign 0, 19, 38 18-21, Not No RT 0, 0.4, 0.6 47 5b 6.6 km ↑ 0.051 ns (25) detection mg 25-40 specified interaction sec Low, Mean Maintain 0, 19, 38 18-21, Not High THC No Headway 0, 0.4, 0.6 47 5b 6.6 km ns (25) Headway mg 25-40 specified ↑ 8.76, interaction (m) 16.76 Low, Headway SD 0, 19, 38 18-21, Not High THC No Varia- Headway 0, 0.4, 0.6 47 5b 6.6 km ns (25) mg 25-40 specified ↑ 4.29, interaction bility (m) 7.78 Low & 64 Road 0, 13, 17 1-4x/ 54.5 Not dosed SDLP 0, 0.5 14 26.1 ± 1.3 30 c High THC ns (29) Tracking mg Month km together ↑ 50 M Road 0, 22.9 1-10x/ SDLP -- 18-31 ~15b ~15 min ns -- -- (172) Tracking mg Month 35 F Low, High THC ↑ Low, Road SDLP 0, 19, 38 18-21, Not 4, 7; High No 0, 0.4, 0.6 47 5b 6.6 km (25) Tracking (cm) mg 25-40 specified EtOH ↑ interaction Low-High 5, 5 difference ns

65 45.7 mg; High THC % Time in 8 Occasiona Not Tracking 20 mg -- 22-30 60-330b & -- -- (174) lane M l smokers given dronabin dronabino ol l impaired “Straddled Low, line” High THC variables Road 0, 14, 52 Not impaired, (lane -- 40 21-35 Not given 30, 80b -- -- (45) Tracking mg given only at 2nd division & drive center (t=80 min) division) Random 0, 1.77, Mean ≥Weekly, test Not Speed 3.95%; -- 10 21-45 ns -- -- (173) speed

Table 9 (Supplemental). (Continued from previous page) Cannabis and alcohol effects on simulated and on-road driving Time Dose Age (min) Drive Outcome: Outcome: Measured THC EtOH (years) Cannabis Outcome: Measure N after Length/ EtOH Combina- Ref. by Dose (g/kg or Mean±SD, History THC only THC Time only tion target BAC) [Range] dose THC ↓ 50 only M Mean 0, 22.9 1-10x/ during Speed -- 18-31 ~15b ~15 min -- -- (172) speed mg Month multi- 35 tasking F (PASAT) Mean High THC Low 0, 19, 38 18-21, Not No Speed speed 0, 0.4, 0.6 47 5b 6.6 km ↓ EtOH ↑ (25) mg 25-40 specified interaction (km/h) 0.97 0.72 THC+ EtOH ↑, 37.7 ↑ Relative bor- 66 Mean 1-4x/ Speed 0, 13 mg 0.05% 12 24-29 30 km, THC ↓ to THC derline- (44) speed Month rurald only significant relative to THC only 0, 8, 12, Speed 16 mg 3 Naïve; 8 High THC Not dosed vari- SD speed ∆1-THC 0.095% 21-29 all >105 10 min ns (171) M ↑ ~110% together ability (ate ≤1x/week cake) Random Speed 0, 1.77, ≥Weekly, test Not varia- SD speed 3.95%; -- 10 21-45 ns -- -- (173)

Table 9 (Supplemental). (Continued from previous page) Cannabis and alcohol effects on simulated and on-road driving Time Dose Age (min) Drive Outcome: Outcome: Measured THC EtOH (years) Cannabis Outcome: Measure N after Length/ EtOH Combina- Ref. by Dose (g/kg or Mean±SD, History THC only THC Time only tion target BAC) [Range] dose 50 Speed M 0, 22.9 1-10x/ varia- SD speed -- 18-31 ~15 C ~15 min ns -- -- (172) mg Month bility 35 F Speed High THC Low SD speed 0, 19, 38 18-21, Not No varia- 0, 0.4, 0.6 47 5 C 6.6 km ↑ EtOH ↑ (25) (km/h) mg 25-40 specified interaction bility 0.62 0.44 Speed 37.7 1-4x/ No varia- SD speed 0, 13 mg 0.05% 12 24-29 30 km, ns ns (44) Month interaction bility rurald PASAT 67 Failure to distracter 50 show Impaired task M 0, 22.9 1-10x/ practice Divided during -- 18-31 ~15b 110 sec -- -- (172) mg Month effects Attention uneventful 35 seen under drive F placebo segment High THC Car ↑ Impaired Following 0, 19, 38 18-21, Not mean No Divided + Sign 0, 0.4, 0.6 47 5b 6.6 km ns (25) mg 25-40 specified headway interaction Attention Detection & ↑ SD Tasks headway

THC: 6 # 37.7 Placebo: 2 Col- Collisions 1-4x/ Placebo: 2 Placebo: 2 0, 13 mg 0.05% 12 24-29 30 km, THC+ (44) lisions across Month THC: 3 EtOH:2 rurald EtOH: 5 study Placebo: 2 # drivers Low Col- involved 0, 13, 17 1-4x/ 54.5 Placebo: 2 Not dosed 0, 0.5 14 26.1 ± 1.3 30 THC: 3 (29) lisions in mg Month kmc EtOH: 3 together High collisions THC: 6 On-Road 0, 100, Mean RT 200, 300 to >1x/ µg/kg RT preceding -- 16 21-40 Month; 45a 96 kme ns -- -- (47) (0, ~7, vehicle’s not daily ~14, ~21 movement mg)

69 Mean 200 >1x/ Maintain No (34, Headway µg/kg 0, 0.04% 18 21-40 Month; 30, 75a 80 kmf ns ns Headway interaction 47) (m) (0, ~7, not daily ~14 mg) 0, 100, Low, Headway SD 200 >1x/ Low, High (34, Varia- Headway µg/kg 0, 0.04% 18 21-40 Month; 30, 75a 80 kmf High THC ↑ 0.9 THC+ 47) bility (m) (0, ~7, not daily ↑ 2.9, 3.8 EtOH ↑ ~14 mg) 2.5, 3 0, 100, Low, 200, 300 >1x/ Med, Road µg/kg SDLP -- 24 21-40 Month; 40, 100a 22 km High THC -- -- (47) Tracking (0, ~7, not daily ↑, both ~14, ~21 drives mg)

70 Low, 0, 100, High 200 >1x/ Road Time Out THC+ (34, µg/kg 0, 0.04% 18 21-40 Month; 30, 75a 80 kmf ns ns Tracking of Lane EtOH ↑ to 47) (0, ~7, not daily >0.6%, to ~14 mg) ~1.1% 0, 100, 200, 300 >1x/ Mean µg/kg Speed -- 24 21-40 Month; 40, 100a 22 km ns -- -- (47) speed (0, ~7, not daily ~14, ~21 mg) 0, 100, 200, 300 >1x/ Mean µg/kg Speed -- 16 21-40 Month; 45a 96 kme ns -- -- (47) speed (0, ~7, not daily ~14, ~21 mg)

71 Visual Search for 0, 100 0, 0.05% >1x/ Search/ traffic at µg/kg 40-45 (reached 16 21-40 Month; 25a/15c ns ns ↓ 3% (26) Proces- inter- (0, ~7 min 0.04%) not daily sing sections mg) aDenotes times after smoking initiated bDenotes times after smoking completed cMostly-straight road, maintain 55 mph:17.2 km; downhill winding road, limit 45 mph: 12.7 km; follow behind a lead car: 11.6 km; 4 unexpected events, limit 55 mph: 13.0 km dMonotonic road with 5 unexpected events: 14.8 km; sharp curves, limit 45 mph, foggy with unexpected events: 12.1 km; downhill, limit 45 mph, moderate & sharp curves, 2 unexpected events: 10.8 km eCar following: 16 km; road tracking: 64 km; 2nd car following, 16 km fCar following: 40 km (~25 min); road tracking: 40 km (~25 min) Abbreviations: THC, ∆9-tetrahydrocannabinol; BAC, blood alcohol content (%); SD, standard deviation; RT, reaction time; ns, not significant; SDLP, standard deviation of lateral position; PASAT, Paced Auditory Serial Additions Test

Table 10 (Supplemental). Effect of smoked cannabis and alcohol (EtOH) on neurocognitive function: laboratory studies EtOH Age Time Outcome: Outcome: (g/kg or (years) Cannabis (h) Outcome: Relative Task THC Dose N Measure EtOH Combina- Ref. target Mean±SD, History After THC only to only tion BAC) [Range] THC Two 0% or 3.6% THC ≥1x/Month; 0 or -2.7 Free joints; 4 24.5, 10/14: >100 # Words ~1.25- No Before 0.6 [M]/ 14 No effect (6.6 vs. (167) recall puffs/joint [21-34] lifetime recalled 1.5b interaction dosing 0.5 [F] 9.3) spaced by 2 occasions h Two 0% or 3.6% THC ≥1x/Month; Time 0 or Respond at Interaction: joints; 4 24.5, 10/14: >100 Under- Over- Esti- 0.6 [M]/ 14 estimated ~1.25b canceled Placebo (167) puffs/joint [21-34] lifetime estimation estimation mation 0.5 [F] time targets effect spaced by 2 occasions h Before 72 0%, 400 µg/kg 23.2±8.4, ≥4 No THC; SST 0.05%, 21 Stop RT 1-3 No effect Increased (169) (~28 mg) [19-38] Days/week interaction placebo 0.07% alcohol Two 0% or Stand- 3.6% THC ≥1x/Month; Time (sec) 0 or -11.8 sec ing joints; 4 24.5, 10/14: >100 before foot ~1.25- No Before 0.6 [M]/ 14 No effect (45.9 vs. (167) Steadi- puffs/joint [21-34] lifetime down in 1.5b interaction dosing 0.5 [F] 57.7 sec) ness spaced by 2 occasions OLS h 0, 1.75% Sway/ (~16 Equilib- 0, 0.25, 24±3 2-21 Days in Composite No THC mg)THC, 12 ~0.25a Decreased No effect (170) rium 0.5 [21-45] past month Equilibriu interaction alone 3.3% THC m (~30 mg)

Table 10 (Supplemental). (Continued from previous page) Effect of smoked cannabis and alcohol (EtOH) on neurocognitive function: laboratory studies EtOH Age Time Outcome: Outcome: (g/kg or (years) Cannabis (h) Outcome: Relative Task THC Dose N Measure EtOH Combina- Ref. target Mean±SD, History After THC only to only tion BAC) [Range] THC Two 0% or 3.6% THC ≥1x/Month; 0 or joints; 4 24.5, 10/14: >100 # ~1.25- No Before DSST 0.6 [M]/ 14 No effect Decreased (167) puffs/joint [21-34] lifetime Attempted 1.5b interaction dosing 0.5 [F] spaced by 2 occasions h Two 0% or 3.6% THC ≥1x/Month; 0 or -6.5 joints; 4 24.5, 10/14: >100 ~1.25- No Before DSST 0.6 [M]/ 14 # Correct No effect (37.4 vs. (167) puffs/joint [21-34] lifetime 1.5b interaction dosing 0.5 [F] 43.9) spaced by 2 occasions

73 h

Two 0% or 3.6% THC ≥1x/Month; 0 or -4.8% -5.4% joints; 4 24.5, 10/14: >100 ~1.25- No Before DSST 0.6 [M]/ 14 % Correct (91.3% vs. (91.1% vs. (167) puffs/joint [21-34] lifetime 1.5b interaction dosing 0.5 [F] 96.1%) 96.5%) spaced by 2 occasions h Two 0% or 3.6% THC ≥1x/Month; Logical 0 or joints; 4 24.5, 10/14: >100 True/False ~1.25- No Before Reason- 0.6 [M]/ 14 No effect No effect (167) puffs/joint [21-34] lifetime statements 1.5b interaction dosing ing 0.5 [F] spaced by 2 occasions h 0%, Before Tower 400 µg/kg 0.05%, 23.2±8.4, ≥4 # Correct 1.17- No THC; of 21 No effect No effect (169) (~28 mg) 0.07% [19-38] Days/week decisions 3.17 interaction placebo London alcohol

Table 10 (Supplemental). (Continued from previous page) Effect of smoked cannabis and alcohol (EtOH) on neurocognitive function: laboratory studies EtOH Age Time Outcome: Outcome: (g/kg or (years) Cannabis (h) Outcome: Relative Task THC Dose N Measure EtOH Combina- Ref. target Mean±SD, History After THC only to only tion BAC) [Range] THC Before Tower 0%, 400 µg/kg 23.2±8.4, ≥4 1.17- No THC; of 0.05%, 21 RT No effect No effect (169) (~28 mg) [19-38] Days/week 3.17 interaction placebo London 0.07% alcohol Before 0%, 400 µg/kg 23.2±8.4, ≥4 No THC; CTT 0.05%, 21 λc 0.3-3.3 No effect Decreased (169) (~28 mg) [19-38] Days/week interaction placebo 0.07% alcohol Two 0% or 3.6% THC ≥1x/Month; Response 0 or joints; 4 24.5, 10/14: >100 to primary ~1.25- No Before DAT 0.6 [M]/ 14 b No effect No effect (167)

74 puffs/joint [21-34] lifetime target: hit 1.5 interaction THC 0.5 [F]

spaced by 2 occasions rate h Two 0% or 3.6% THC ≥1x/Month; 0 or Response joints; 4 24.5, 10/14: >100 ~1.25- No Before DAT 0.6 [M]/ 14 to primary No effect No effect (167) puffs/joint [21-34] lifetime 1.5b interaction THC 0.5 [F] target: RT spaced by 2 occasions h Two 0% or 3.6% THC ≥1x/Month; 0 or # False +0.7 joints; 4 24.5, 10/14: >100 ~1.25- No Before DAT 0.6 [M]/ 14 alarm (1.6 vs. No effect (167) puffs/joint [21-34] lifetime 1.5b interaction dosing 0.5 [F] responses 0.9) spaced by 2 occasions h

Table 10 (Supplemental). (Continued from previous page) Effect of smoked cannabis and alcohol (EtOH) on neurocognitive function: laboratory studies EtOH Age Time Outcome: Outcome: (g/kg or (years) Cannabis (h) Outcome: Relative Task THC Dose N Measure EtOH Combina- Ref. target Mean±SD, History After THC only to only tion BAC) [Range] THC Two 0% or # Errors 3.6% THC ≥1x/Month; 0 or estimating +0.6 joints; 4 24.5, 10/14: >100 ~1.25- No Before DAT 0.6 [M]/ 14 # of No effect (1.6 vs. (167) puffs/joint [21-34] lifetime 1.5b interaction dosing 0.5 [F] secondary 1.0) spaced by 2 occasions targets h 26.1 # Exercises DAT 0, 13 mg 0, 0.05% 12 1-4x/Month 0.25 No effect Decreased Decreased Placebo (44) [24-29] completed % Success 26.1 in No DAT 0, 13 mg 0, 0.05% 12 1-4x/Month 0.25 Decreased No effect Placebo (44) [24-29] arithmetic interaction

75 task

# Exercises 26.1 DAT 0, 13 mg 0, 0.05% 12 1-4x/Month with no 0.25 Increased Increased Increased Placebo (44) [24-29] response RT to 26.1 Slightly Slightly No DAT 0, 13 mg 0, 0.05% 12 1-4x/Month secondary 0.25 Placebo (44) [24-29] increased increased interaction task Increased; Interaction: 26.1 False alarm Slightly Slightly greatest DAT 0, 13 mg 0, 0.05% 12 1-4x/Month 0.25 Placebo (44) [24-29] responses increased increased effect after combination

Table 10 (Supplemental). (Continued from previous page) Effect of smoked cannabis and alcohol (EtOH) on neurocognitive function: laboratory studies EtOH Age Time Outcome: Outcome: (g/kg or (years) Cannabis (h) Outcome: Relative Task THC Dose N Measure EtOH Combina- Ref. target Mean±SD, History After THC only to only tion BAC) [Range] THC Decreased; Interaction: alcohol effects 26.1 # Exercises DAT 0, 13 mg 0, 0.05% 12 1-4x/Month 1.25 No effect No effect persisted to Placebo (44) [24-29] completed second trial when combined w/ THC % Success Decreased 26.1 in ; smaller No DAT 0, 13 mg 0, 0.05% 12 1-4x/Month 1.25 No effect Placebo (44)

76 [24-29] arithmetic effect than interaction

task 1st trial Increased; Interaction: effects # Exercises persisted to 26.1 DAT 0, 13 mg 0, 0.05% 12 1-4x/Month with no 1.25 No effect No effect second trial Placebo (44) [24-29] response w/combina -tion, time effect p=0.06 RT to 26.1 Slightly DAT 0, 13 mg 0, 0.05% 12 1-4x/Month secondary 1.25 No effect No effect Placebo (44) [24-29] increased task

Table 10 (Supplemental). (Continued from previous page) Effect of smoked cannabis and alcohol (EtOH) on neurocognitive function: laboratory studies EtOH Age Time Outcome: Outcome: (g/kg or (years) Cannabis (h) Outcome: Rela- Task THC Dose N Measure EtOH Combina- Ref. target Mean±SD, History After THC only tive to only tion BAC) [Range] THC Increased; Interaction: greatest effect 26.1 1-4x/ False alarm Slightly DAT 0, 13 mg 0, 0.05% 12 1.25 No effect after combi- Placebo (44) [24-29] month responses increased nation; persisted to second trial Increased; Interaction: Before 0%, [statistical] 400 µg/kg 23.2±8.4, ≥4 days/ # Control THC; DAT 0.05%, 21 0.5-2.5 Increased Increased combination (169) (~28 mg) [19-38] week losses placebo

77 0.07% produced alcohol

greatest impairment # Correct Before 0%, 400 µg/kg 23.2±8.4, ≥4 days/ detections on No THC; DAT 0.05%, 21 0.5-2.5 Decreased Decreased (169) (~28 mg) [19-38] week secondary interaction placebo 0.07% task alcohol Before 0%, 400 µg/kg 23.2±8.4, ≥4 days/ No THC; DAT 0.05%, 21 RT 0.5-2.5 Increased Increased (169) (~28 mg) [19-38] week interaction placebo 0.07% alcohol Before 0%, 400 µg/kg 23.2±8.4, ≥4 days/ Tracking No THC; DAT 0.05%, 21 0.5-2.5 No effect Increased (169) (~28 mg) [19-38] week error interaction placebo 0.07% alcohol aDenotes time after smoking initiated bDenotes time after smoking completed Abbreviations: THC, ∆9-tetrahydrocannabinol; EtOH, ethanol (alcohol); BAC, blood alcohol content; SD, standard deviation; SST, stop signal task; RT, reaction time; DSST, digit symbol substitution test; CTT, critical tracking task; DAT, divided attention task

Chapter 3 – Protocol: Effect of Inhaled Cannabis on Driving Performance

Précis:

Background: Cannabis is the most commonly detected illicit drug among drivers (NHTSA

2007 National Roadside Survey). Drivers killed in traffic accidents with ∆9- tetrahydrocannabinol (THC) in their blood had a significantly increased odds ratio for culpability (2.7) compared to those who were drug- and alcohol-free (Drummer et al., 2004).

There was an even higher (6.6) ratio if the blood THC was >5 ng/mL. A recent meta-analysis of nine epidemiological studies showed an odds ratio of 2.66 for motor vehicle crash after taking cannabis (Li et al 2011). Experimental studies of cannabis’ effects on driving performance are needed. This study determines driving, psychomotor and cognitive impairment of cannabis, alone and in combination with low-dose alcohol and characterizes the disposition of cannabinoids in blood, plasma, and oral fluid.

Objectives: 1) evaluate mechanisms of driving impairment by inhaled cannabis on crucial driving skills including the ability to focus during long stretches of monotonous driving, to process changes in the driving environment, reaction time, decision-making, risk-taking, and performance of multiple tasks at once (divided attention); 2) evaluate interaction of low-dose alcohol and cannabis on driving performance; 3) correlate driving impairment to whole blood, plasma, and oral fluid cannabinoid concentrations.

Hypothesis: We hypothesize that cannabis will dose-dependently impair driving performance compared to placebo, and that low-dose alcohol and cannabis will further impair performance.

Study population: Up to 60 healthy current cannabis smokers (ages 21-55) will be enrolled to obtain 20 completers.

Design and Methods: This is a within-subject, randomized, double-blind study of the effects of cannabis and alcohol on driving. Eligible subjects attend a training session followed by six experimental sessions (placebo THC + placebo alcohol, placebo THC + alcohol

(approximately 0.05 g/dL [refers to the goal BAC for the simulator drive (0.05%).

Subjects will be dosed to an approximate peak BAC of 0.065%. Subjects will be tested on the decline such that subjects will be at or above the goal BAC throughout the drive]), low-dose THC + placebo alcohol, high-dose THC + placebo alcohol, low-dose THC + alcohol, high-dose THC + alcohol), with at least 1 week between sessions. Each session includes an overnight inpatient stay at the University of Iowa clinical research unit the night prior to cannabis administration, followed by transportation to the National Advanced

Driving Simulator (NADS) for dosing and completing a prepared driving scenario 0.5-1 h later. Alcohol will be administered in the form of a mixed drink over 10 min. A Volcano® hot air generator will vaporize cannabis plant material into a balloon for smokeless inhalation over 10 min, immediately following alcohol dosing. Breath alcohol concentrations will be measured and blood and oral fluid will be collected for cannabinoid analyses prior to and up to 8h after dosing. Additional neurocognitive and neuromotor tests will be administered outside of the simulator for up to 8 h.

Outcome Measures: Primary outcome measures include driving performance and cannabinoid concentrations in whole blood, plasma, and oral fluid. Secondary outcome measures include performance on neurocognitive tasks and subjective assessments.

Benefits: This study offers no direct benefit to participants, but is likely to yield generalizable knowledge about the effects of cannabis alone and combined with alcohol on driving performance.

Risks: Participation in this study represents more than minimal risk because of the administration of cannabis and alcohol, and the operation of an advanced driving simulator capable of motion.

Background

Cannabis is the most widely-used illicit drug (185), and alcohol or ethanol, the most common licit drug taken in the United States and most areas of the world. More than half of Americans age 12 or older are current alcohol drinkers, with highest prevalence among those aged 21-34 (185); 12.8 million age 21 and older are current

(prior month) cannabis consumers (134). In current driving studies in the US (NHTSA), the prevalence of cannabis in drivers’ blood and oral fluid is only exceeded by the prevalence of ethanol. Thus, many motorists’ driving performance may be influenced by the presence of cannabis and/or alcohol. Concurrent ethanol and cannabis intake produces significant impairment (34).

Rationale: Driving Under the Influence of Cannabis (DUIC) and Alcohol (DUIA)

Epidemiology

Nearly two-thirds of trauma center admissions in the US are due to motor vehicle accidents, with almost 60% of these victims testing positive for drugs or alcohol (133). In

2009, 12% of Americans age 12 or older drove under the influence of alcohol at least once in the past year with the highest rate among those aged 21-34, and 10.5 million people reported driving under the influence of illicit drugs (185). Despite real or perceived impairment, individuals are willing to drive if they believe they have a good

reason to do so (26, 44). Alcohol and cannabis were the drugs most often detected in drivers in 2007 (3).

Cannabis and Driving—Epidemiology

Despite the abundance of research on this topic, cannabis’ effects on driver performance continue to be debated, creating challenges for implementing effective driving under the influence of drugs policies (4-5, 157, 186). Colorado, a medical marijuana state, experienced increased driving under the influence of cannabis cases and proposed a liberal 5 ng/mL THC in whole blood per se law; however the proposal met strong resistance and debate, thwarting enactment to date (159, 187-191).

Early epidemiological studies of culpability while driving under the influence of cannabis did not provide conclusive evidence of causality. A major contributing factor was inclusion within the cannabis exposure group of individuals with only the non- psychoactive metabolite of ∆9-tetrahydrocannabinol (THC), 11-nor-9-carboxy-THC

(THCCOOH), in their blood (4). This inactive metabolite has a long window of detection in whole blood and plasma, long after the euphoria or “high” and acute cognitive and psychomotor impairment associated with cannabis smoking has dissipated (139). In a study of occasional to moderate cannabis smokers, THCCOOH was quantifiable an average of 84 and 152 h after smoking a single 1.75 or 3.55% THC cigarette, respectively, with a 0.5 ng/mL cutoff, and for some subjects up to 168 h post-smoking

(107). Karschner et al recently reported detection of parent compound THC in whole blood and plasma of chronic, daily cannabis smokers for more than 7 days of continuously monitored abstinence with a 0.25 ng/mL cutoff concentration (163).

Furthermore, cannabinoid blood concentrations decrease rapidly after smoking (139-

140). Blood collection occurs, on average, up to 90 minutes after arrest for DUI (141) and

4 h after an accident—a long enough delay that many individuals’ specimens were negative for cannabinoids even though they may have been positive and impaired at the time of the event. Another major contributing factor is that it was difficult to find cannabis-only cases; many drivers took multiple drugs that may have contributed to impairment. In 2004, Drummer et al accrued sufficient cannabis-only cases to demonstrate a statistically significant increase in crash risk odds ratio following cannabis intake. THC-positive drivers had a 2.7 culpability odds ratio relative to drug-free drivers, a relationship that increased to 6.6 if their blood THC concentration was >5 ng/mL (4).

An analysis of drivers evaluated for DUI found higher blood THC concentrations among those judged impaired relative to those judged not impaired. The percentage of impaired drivers increased from 38% among those with <0.7 ng/mL THC to 57% for those with

THC >10 ng/mL; there was a significant positive relationship between THC concentration and risk of being judged impaired (146).

Confounding variables often make it difficult to define the effect of cannabis on driving. Cannabis smokers often have the same demographic characteristics as other high-crash-risk groups, including youthful age, male, tendency toward risk-taking or thrill-seeking, and driving under the influence of alcohol (143, 186, 192). Cannabis tolerance may occur in frequent smokers, resulting in less apparent impairment from the same THC concentrations relative to occasional cannabis intake (146). In some epidemiological studies, statistically controlling for some variables led to equivocal results (147-148). Nevertheless, after controlling for BAC, age, vehicle type, and crash

time, drivers in fatal road collisions had a higher odds ratio for culpability when THC positive; those with blood THC >5 ng/mL had higher odds ratios (2.12) than those with

<1 ng/mL THC (1.57) (5). Similarly, the odds ratio of being judged impaired increases with increasing THC concentration after adjusting for gender, needle marks, and regular cannabis intake (146). A recent meta-analysis by Li et al (2011) found that odds of a motor vehicle accident more than doubled, to 2.66, after smoking cannabis. Several of the studies included in the meta-analysis controlled for some confounding variables, which tended to lessen effect size; however, in most cases the odds increase was statistically significant (193).

Cannabis and Driving—Experimental Studies

Experimental studies of actual driver performance under the influence of cannabis are the most rigorous way to evaluate causality of impairment. Past experimental studies often were inconclusive because outcome measures were not tailored to specific THC effects (175-176, 194). Several studies concluded that cannabis impairment is different from that following alcohol consumption, in that drivers behaved more cautiously while under THC’s influence. Drivers appeared aware of their impairment and attempted to compensate by driving more slowly and taking fewer risks (24-29, 195-196). Perceived effort of driving increased under the influence of THC (26).

However, it is not possible to compensate for all aspects of impairment. There may be a control cost (25). Most importantly, THC’s impairing effects tend to increase with task complexity; a realistic driving task involves a number of smaller subtasks requiring simultaneous attention. Subjects under the influence of cannabis perform worse

on divided attention tasks (24-25, 167), when faced with unexpected circumstances and choices, and during long stretches of monotonous driving (24). Ramaekers et al noted that THC decreased critical tracking performance (perceptual motor control), increased stop reaction time in a stop signal task and decreased the number of correct decisions on a

Tower of London cognitive function and planning task (160). That study did not directly correlate magnitude of performance impairment with serum THC concentrations; but the proportion of observations showing impairment increased with increasing serum THC concentration. Cannabis-associated impairment also may manifest as failure to demonstrate expected practice effects, suggesting that drivers may lose some benefit afforded by prior experiences (28).

Increased reaction time is one of the most common cannabis-associated impairments (25, 29, 43, 173, 175, 177), while road tracking is one of the most sensitive measures, displaying dose-dependent effects (47). THC also increases lane position variability (measured as standard deviation of lateral position, or SDLP) (25, 29, 47, 174) and steering wheel variability (25, 29). A recent study demonstrated significant THC- induced decrements in cognitive performance (immediate recall, attention and working memory, and executive function tasks). The study population had a wide range of prior cannabis exposure, from 2 to approximately 1000 lifetime episodes. On average, participants were light users (178). One 3.95% THC cigarette produced similar equilibrium deficits (body sway) and brake latency to that observed while driving with a breath alcohol content (BrAC) of 0.05% (173).

Alcohol and Driving—Epidemiology

Alcohol-related driving fatalities account for approximately 30% of driving fatalities in the United States. An estimated 12,998 individuals died in 2007 from alcohol-impaired driving collisions (14). The adoption of 21 as the minimum legal drinking age in all 50 states by 1988 was associated with a decrease in fatalities among young drivers, but DWI remains a public health concern for all ages (15). As many as

14% of nighttime weekend drivers in a recent roadside survey were classified as dependent or abusive drinkers based on self-report, and another 10% were heavy drinkers

(16).

Relative risk of crash involvement increases exponentially with increasing BAC:

1.4 for drivers with a BAC between 0.02-0.04%, 11.1 for 0.05-0.09%, 48 for 0.10-0.14%, and approximately 380 for >0.15%. For single-vehicle crashes, each 0.01% increase in

BACs for driving, even among those who develop alcohol tolerance (even appearing unimpaired to peers) due to frequent and heavy alcohol intake (20).

Alcohol and Driving—Experimental Studies

Blood alcohol levels peak within approximately 30 minutes to 1 h after drinking

(34, 197). During the post-absorptive phase, elimination follows zero-order kinetics (due to enzyme saturation) until BAC decreases below 0.02%; at that point it shifts to first-

order (197-198). The average blood ethanol concentration decreases at a rate of 15 mg/dL/h (range 10-35 mg/dL/h) in moderate drinkers, but in apprehended drivers it is closer to 19 mg/dL/h due to a higher tendency toward binge drinking in this population

(198). The wide individual variability in elimination rates and alcohol effects is due to multiple factors including fasting state, demographic characteristics and metabolism induction or impairment (197-198).

The impairment caused by alcohol is well demonstrated in experimental studies.

Alcohol dose-dependently impairs divided attention, judgment, tracking and visual search, attention and vigilance, perception/signal detection/sensory response, coordination and balance, information processing, vision, eye movement, cognition, psychomotor skills, and reaction time (17, 199-200). The range of functions affected increases with increasing BAC. Driving impairment was documented at BACs <0.02%.

The majority of alcohol studies indicated impairment at a 0.05% BAC (199-200). Under the influence of alcohol, individuals were more susceptible to impaired driving precision, especially in distracted or divided attention tasks. They tended to drive faster, swerve more (demonstrating a higher SDLP), and increasingly failed to stop for red lights (201).

DUI Alcohol and Cannabis Combined

Driving under the influence of cannabis is more common among people who also drive under the influence of alcohol (181). In a study of 322 motor vehicle crash victims,

30% of those positive for THC also were alcohol positive. This study did not distinguish drivers from other crash victims (133). However, a larger French study was able to

examine drivers specifically, and found that over 40% of 681 THC-positive drivers involved in fatal crashes also had a BAC above 0.5 g/L (0.05%) (5).

While some of the cognitive and psychomotor effects of cannabis and alcohol are similar (29, 160, 179) (both being central nervous system (CNS) depressants, with some evidence suggesting that alcohol activates the cannabinoid CB1 receptor pathway (180), the impact that they have on driving behavior has some differences. Alcohol causes subjects to drive faster (44, 201-202), the opposite effect of cannabis—and concurrently impairs reaction time and increases steering variability (29). It also affects confidence, causing those under its influence to underestimate their own impairment (29).

Some findings suggested that the combined effects of alcohol and THC are greater than those of either drug alone (34, 47-48, 167, 196). Combining the two led to a greater subjective sensation of potency (44). THC-positive drivers with a BAC ≥0.05 g% had 2.9 higher odds of culpability as compared to THC-negative drivers with BAC ≤0.05 g% (4). Some studies found driving impairment when THC and alcohol were combined at low doses that did not produce impairment when given separately. Visual search frequency and ability to detect peripheral traffic were significantly decreased by a combination of 100 µg/kg THC and 0.04% BAC, although neither condition alone influenced these functions (26). Lane position variability increased significantly when drivers were under the influence of cannabis and alcohol, while either drug alone did not significantly alter this parameter under some conditions (44). In a real-road driving experiment, 0.04% BAC alone increased lane position variability by 2.2 cm, while 100 or

200 µg/kg THC alone increased SDLP by 2.7 and 3.5 cm, respectively. When THC was combined with alcohol, SDLP increased to 5.3 and 8.5 cm for 100 and 200 µg/kg,

respectively. These combinations of alcohol and THC elevated time out of lane, and increased SDLP to levels associated with a 0.09% and 0.14% BAC, respectively. The difference in reaction time from alcohol and 200 µg/kg THC (relative to placebo) was 1.6 sec, corresponding to an additional 42 m traveled prior to deceleration (34). A recent within-subject driving simulation experiment on 12 volunteers revealed increasing numbers of total collisions among participants under the influence of 0.05% BAC (2 collisions), 13 mg THC (3 collisions), and both (5 collisions); under placebo conditions, there were no collisions (44).

Prior laboratory studies did not always agree on the manner of impairment caused by alcohol and THC. In one study, THC diminished equilibrium (increased body sway), but did not affect brake latency (a measure of reaction time). Alcohol increased brake latency, and there were no significant additive effects from combining the two drugs. The authors speculated that the lack of additive effects was an artifact of awareness of impairment, coupled with expectation of the “emergency” (170). Although both drugs caused similar impairment in digit-symbol substitution and word recall neurocognitive tasks, Heishman et al found that neither impaired time perception or reaction time at doses comparable to other studies’ (179).

Cannabis’ Subjective Effects

Acute psychological effects of cannabinoids include euphoria, dysphoria, sedation, and altered perception (137, 203-204). Cannabis also is associated with subjective physical discomfort, lack of energy, and higher subjective physical effort by

SOFI questionnaire (29). Intensity of euphoria/dysphoria varies with dose, route of

administration, vehicle in which cannabis is administered (137, 205), expectations of effects, and environment and personality of the cannabis smoker (137).

Cannabis’ Neurocognitive effects

Acute cannabis intoxication produces dose-related impairments in cognitive and psychomotor functioning, as well as risk-taking behavior (43, 138, 206-209). Reaction time, perception, short-term memory and attention, motor skills, tracking, and skilled activities are altered (30-32).

While several observational studies evaluated the relationship between past cannabis use and risk-taking behavior, e.g., criminal behavior (210-212), risky sexual behavior (213-215), and driving under the influence of cannabis (216-217), few studies examined whether acute cannabis exposure directly affected risk-taking. Lane et al (138) observed that participants selected the risky option in a binary risk-taking gambling task significantly more often after smoking a 3.58% THC cigarette than prior to smoking. A medium dose (15 mg) of oral synthetic THC (dronabinol) yielded mixed effects on four different impulsivity tasks (209). This difference may reflect decreased bioavailability and lower peak blood THC concentrations following oral administration compared to smoked cannabis. Further characterization of cannabis-induced risk taking following cannabis inhalation will provide additional insight into impulsivity and reward/punishment constructs, allowing for greater understanding of mechanisms responsible for cannabis abuse and resulting choices detrimental to public safety.

Study Objectives and Hypotheses

The primary study objective is to determine mechanisms of cannabis and alcohol driving impairment (33). We seek to clarify the contentious issue of driving performance under the influence of cannabis alone and in combination with low dose alcohol.

Specifically, we aim to:

1) evaluate how cannabis alters or impairs crucial driving skills including the ability

to focus during long stretches of a monotonous drive, process changes in the

driving environment, reaction time, decision-making, and perform multiple tasks at

once (divided attention);

2) evaluate the interaction of low-dose alcohol and cannabis on driving performance,

risk-taking, and cognitive performance;

3) correlate whole blood, plasma, and oral fluid cannabinoid concentrations with

driving and other performance measures.

We hypothesize that cannabis will impair driving performance compared to placebo in a dose-dependent manner, and that combination with alcohol will enhance these effects.

Subjects

It is anticipated that approximately 60 healthy cannabis and alcohol users aged

21-55 will be enrolled and consented. It is expected that 20 will be lost during screening

(due to not meeting all eligibility criteria), 20 will not complete all dosing sessions and 20 will complete all sessions.

Inclusion criteria

1. Healthy adult (age 21-55) men and women, based on medical and psychological

evaluation

2. Currently valid unrestricted (except for vision correction) US driver's license

3. Licensed driver for at least the past two years

4. Drove at least 1300 miles in the past year, by self-report

5. Live within a 80 mile radius of NADS

6. Available for an overnight stay followed by a full-day study session for six

sessions

7. Must be considered a light or moderate drinker according to Quantity-Frequency-

Variability Scale (QFV) or, if a heavy drinker, not drink more than 1-2 times a

week and not have a modal quantity of 5-6 drinks (Protocol Appendix 11)

8. Cannabis use with a minimum frequency averaging at least one day per quarter and

no more than three days a week during the three months prior to study entry

9. Peripheral veins suitable for repeated venipuncture and/or placement of an

intravenous catheter

10. Systolic blood pressure within a clinically normal range (120 ± 30 mmHg) and

diastolic blood pressure of 80 ± 20 mmHg.

11. Good command of written and spoken English

12. Female subjects with reproductive potential must agree to use (and/or have their

partner use) one (1) acceptable method of birth control beginning at the screening

visit throughout the study (including washout intervals between treatment

periods/panels) and until 2 weeks after the last dose of study drug in the last

treatment period. Acceptable methods of birth control include the following:

intrauterine device (IUD-with or without local hormone release), diaphragm,

spermicides, cervical cap, contraceptive sponge, oral contraceptives or condoms.

Abstinence is an alternative lifestyle and subjects practicing abstinence may be

included in the study.

Exclusion criteria

1. Presence of any clinically significant illness, as detected by history, physical

examination, and/or laboratory tests, that might influence driving performance

(e.g., seizures, sleep apnea, narcolepsy, vertigo, chronic fatigue syndrome) or put

the subject at increased risk of adverse events (e.g., cardiac arrhythmia,

hypertension)

2. History of a clinically significant adverse event associated with cannabis or alcohol

intoxication

3. Donation of 450 mL or more of blood in the 2 weeks preceding study drug

administration

4. If female, pregnant or nursing

5. Currently interested in or participating in drug abuse treatment, or participated in

drug abuse treatment within 60 days preceding study enrollment

6. Currently taking drugs that are contraindicated for use with study drugs

7. Requires any special equipment to aid in driving (ex. pedal extensions, hand brake

or throttle, spinner wheel knobs or other non-standard equipment)

8. Significant history of motion sickness or demonstrates significant simulator

sickness during practice drives at screening (SSQ). Subjects must have scores

below the following values on the SSQ: Nausea <21, Oculomotor <32,

Disorientation <15, and Total Score <32.

9. Current alcohol or cannabis use disorder, as identified by the Alcohol Use

Disorders Identification Test (AUDIT, Protocol Appendix 13. AUDIT Survey) for

alcohol or Cannabis Use Disorders Identification Test (CUDIT; Protocol Appendix

14. CUDIT) for cannabis.

a. Alcohol—AUDIT score greater than 15

b. Cannabis-CUDIT score greater than 12.

10. History of any illness that, in the opinion of the study investigator, might confound

the results of the study or pose an additional risk to the subject from study

participation

11. Prior participation in a driver impairment or distraction-related research study

conducted at NADS that uses the same base drive.

Study Design and Methods

Study overview

According to a recent NHTSA consensus protocol, an appropriate test battery for assessment of driving impairment includes a driving simulator measuring speed, lane position, steering variability, and reaction time (33). Additionally, tests of focused, shifting, and divided attention, decision making, coordination, risk taking/risk assessment, subjective effects, and eye tracking are recommended. Divided attention and

distraction are particularly important; the hallmark of cannabis impairment is the decreased ability to react to changes in the environment (25, 28).

The study employs a six-session, double-blind, placebo-controlled, randomized, within-subject design to address the variables listed above. Each session involves an overnight stay at the Clinical Research Unit (CRU) followed by dosing and experimental procedures the next day. Participants ingest a mixed drink and inhale vaporized cannabis from a balloon filled by heating cannabis in a Volcano® Medic hot air generator. Blood and oral fluid specimens are collected, a simulated drive in the NADS-1 is performed, and other neurocognitive and subjective effects tasks and questionnaires are completed.

Recruitment

Applicants who might potentially be eligible are identified from the existing

NADS database. Recruiters perform a database query for the prospective age groups of people who drink at least one alcoholic beverage per week. Potential applicants are screened by telephone to determine if they meet study criteria. Additional participants may be recruited from the community with print, radio, television, and web-based advertisements.

Screening methods

Screening:

Applicants are screened by telephone to determine potential eligibility for the study (see Protocol Appendix 1. Telephone screening questions).

Those who may be eligible based on the telephone interview will be scheduled for a screening session. Prior to the screening visit, staff will email or mail a copy of the informed consent for applicant to review. At the start of the screening visit, staff will perform an in-person consent (see consent process below) and review the eligibility criteria with participants.

Participants will undergo a screening drive in the simulator to assess susceptibility to simulator sickness and provide initial practice to drivers. The drive will reduce practice effects, in addition to allowing for evaluation of the propensity to motion sickness. This drive also provides familiarity with operating the simulator and with secondary tasks. If the participant remains eligible based on the exclusion criteria for simulator sickness, they will proceed to the remainder of the screening.

Participants provide a comprehensive medical history and complete a physical and psychological evaluation. Physical examination will include urine collection for drug tests and pregnancy screening, a 12-lead ECG, and an assessment of suicidality

(Columbia-Suicide Severity Rating Scale (218)). The applicant must show a current, valid driver’s license.

Participants are asked to complete a questionnaire (Protocol Appendix 2: Driving

Survey) about their demographic information, health, driving behavior, driving records, driving history, and consumption of alcohol and cannabis.

Participants only provide information about themselves.

Participants no longer meeting study criteria are excluded from the study and paid for their time and effort.

Consent Process:

Study staff will consent eligible subjects. Staff will provide an oral overview of the informed consent and will provide applicants with an informed consent document and ample time to read and ask questions. Screening procedures do not continue until applicants sign the document.

Pre-Session Evaluation:

All participants undergo the following procedures at each session to determine continuing study eligibility and to characterize baseline condition:

On admission to the CRU the night prior to dosing, female subjects have a urine pregnancy test. Those testing positive will be discharged from the study.

Study design

The study will take place at the National Advanced Driving Simulator (NADS) at the University of Iowa Research Park, with an inpatient stay the night prior to each session at the CRU. An optional second night stay is at the discretion of the clinical staff only if necessary for subject safety. Enrolled participants undergo an initial training session to ensure familiarity with research procedures before participating in study sessions.

There are six experimental sessions per participant. Each session involves the following: 1) Overnight stay at the UIA CRU. 2) Collection of biological samples (blood and oral fluid specimens; real-time breath alcohol measurements) before and periodically after dosing. 3) Administration of study drugs (oral alcohol, vaporized cannabis) about 30 min prior to start of driving simulation. 4) 25-40 min simulated driving session. 5)

Completion of psychological tests and questionnaires before and periodically after drug administration (see Appendices 5-10). 6) After completion of the session, a safety evaluation to confirm eligibility for discharge. 7) Participant may be asked to stay a second night for safety reasons by the on-site physician.

Each participant receives each of the 6 drug combinations once, in random order.

Cannabis vapor is produced from 500 mg either placebo (0% THC), approximately 2.5-

3.5% THC (low dose), or approximately 6.0-7.5% THC (high dose) bulk cannabis plant material to yield doses of approximately 0, 12.5-17.5, or 30-37.5 mg THC. Alcohol dose is either placebo or an amount calculated from age, gender, height, and weight (see

Ingested Alcohol section below) to produce approximately 0.065 g/dL BAC. Subjects will be tested on the decline such that subjects will be at or above the goal BAC (0.05%) throughout the drive. No more than 486 mL blood (for whole blood and plasma specimens) is collected during the study (all sessions).

Participants conduct a scripted drive in the NADS, perform other neurocognitive tasks, and complete subjective effects questionnaires during the session. Participants may be discharged with normal vital signs and normal neuromotor exam no earlier than 8h after dosing. Participants may be requested to stay overnight in the CRU in the event that there are safety concerns.

National Advanced Driving Simulator (NADS)

The National Advanced Driving Simulator (NADS) consists of a large dome in which entire cars and cabs of heavy trucks, buses, farm/construction equipment can be mounted. Vehicle cabs are mounted on high-frequency vibration actuators located at each

wheel and provide road feel cues. The dome sits on a turntable that can rotate 330 degrees in either direction. Underneath the turntable is a hexapod that provides roll, pitch, and yaw. This entire structure can move laterally and longitudinally within an area of

4000 sq. feet. This entire motion base was designed to recreate the motion cues associated with driving an actual vehicle over various road surfaces under different driving conditions. Virtual scenery is projected 360 degrees around the driver on the interior dome walls. Drivers experience realistic driving scenes, traffic sounds, and road conditions such as gravel and potholes (219).

The NADS was effectively employed in a number of different driving studies, both past and ongoing, assessing many issues of public health concern. NADS study topics include driver distraction, impairment detection, active safety systems, older driver safety systems, younger driver risk assessment, nighttime driving, and the effects of pharmaceutical drugs.

Volcano® Medic Vaporizer for Cannabis Inhalation

The most common route of administration for cannabis is inhalation—particularly smoking in cigarettes (joints), pipes, or cigars (blunts)—which is associated with health hazards. Chronic cannabis smoking is associated with bronchitis and emphysema (220).

Furthermore, the combustion process creates toxic pyrolysis compounds, such as carcinogenic polycyclic aromatic hydrocarbons (PAHs) and carbon monoxide, that are inhaled with the cannabis vapor (67). The Volcano® vaporization system heats cannabis to a temperature that vaporizes THC without pyrolysis of harmful cannabis plant constituents, reducing exposure to unsafe byproducts including PAHs (78, 83). THC

vaporization with the Volcano also is more efficient than burning the plant material, due to reduced THC degradation (66). Subjective effects and plasma THC concentrations are similar in both administration techniques, and prior studies have shown a participant preference for vaporization (78).

Assessment of Risk Perception, Risk Taking, and Impulsivity Traits

The influence of acute and chronic cannabis intake and the relationships among risk perception, impulsivity, risk-taking or risk propensity, and risk-reward decision- making are not well known. Due to potentially multiple involved processes, the effect of

THC on impulsivity is mixed and is not well-characterized (177). Multiple constructs

(brain functions) may be differentially affected by different licit and illicit drugs, including cannabis (221). These constructs are often assessed with a variety of validated self-report questionnaires (222-225) and computerized tasks, e.g., Balloon Analogue Risk

Task (BART) (226-227) and Rogers Decision-Making task (228).

Rogers Decision-Making Task

Acute social alcohol doses impair risky decision-making as measured by the

Rogers Decision-Making task; the ability to differentiate between large and small potential gains decreases under alcohol conditions relative to placebo, impairing integration of information between probability of winning and magnitude of potential gain (229). Low doses of THC decreases one’s likelihood to choose a gamble with varying gains and losses, and increases the choice of gambles with a zero-expected value, possibly reflecting some driving studies’ finding of a decrease in risk-taking behavior.

THC also alters the integration and processing of potential losses when the probability of losing is high and that of winning is low, suggesting that adaptive decision-making may be impaired in high-risk situations with potential for negative consequences (230).

Behavioral Risk Assessment (Balloon Analogue Risk Task [BART])

The BART assesses risk-taking behavior in an objective manner. BART scores correlate with various self-reports of risk-taking, sensation seeking behavior, impulsivity, and drug use (226). The BART also demonstrated acceptable test-retest reliability and stability over time within individual subjects (227). This task assessed risk-taking behavior in crack/cocaine dependent individuals (231), 3,4-methylenedioxymethamphetamine (MDMA) users (232) and individuals after amphetamine administration (233). The BART is currently being employed to assess risk-taking propensity following acute cannabis smoking in a study conducted at the NIDA

Intramural Research Program (IRP). Significant increased behavioral risk-taking, as measured by the BART, was directly related to increased alcohol use (234). Behavior on the BART under the influence of alcohol changes such that people make more risky decisions in beginning trial blocks than those who did not ingest alcohol. Upon discovering that their strategy does not work, these alcohol-positive individuals alter behavior, but do not approach the optimal balance as do those without alcohol (235).

Study procedures in chronological order

This is a six-session study, requiring up to eight visits (1 screening, 1 training, 6 experimental sessions; screening and training may be combined into one visit). The

100

screening/consenting should take approximately 2 to 3 hours, and training less than 2 h.

Each experimental session involves approximately 24 h (inpatient admission to CRU the night prior to dosing, and patient monitoring until at least 8 h after the morning dosing

(with the exceptions of days in which patients remain at NADS after the overnight stay at the CRU as the only way to collect all data for the visit)). The participant may stay a second night at the CRU, only at the discretion of the investigators, if necessary for subject safety. There must be at least 1 week between drug administrations to avoid carryover from residual cannabinoid excretion. Subjects who have an incomplete

Simulator Drive or other study procedures for reasons outside their control (e.g., equipment or software malfunction, investigator error) will be able to return to repeat the incomplete session. There must still be a week between sessions, and the repeated session must occur before continuing with the next dosage condition.

Visit 1 (Screening) – Screening Procedures are described in section iii above.

Visit 1 or 2 (Training):

Participants complete several questionnaires and psychological tests, receive training on computer tasks (gambling, BART) and additional training about the simulator tasks.

The Self-Assessments of Risk Perception, Risk-Taking/Impulsivity and

Sensation-Seeking, Barratt Impulsiveness Scale, and Risk Perception Questionnaire are administered at the training session for later comparison to results after drug administrations.

101

Visits 3-8 (Dosing Sessions):

Subject arrives by 8 PM the evening prior to dosing, and is admitted to the CRU.

A urine pregnancy test is administered to females. A breath alcohol measurement, and oral fluid samples by QuantisalTM and Draeger 5000 are collected on admission. The subject spends the night on the CRU under medical supervision to ensure that s/he does not self-administer any drugs and will not be intoxicated on admission. Subject will be provided food and non-caffeinated beverages during the session. After breakfast, the neuromotor exam is performed, baseline performance on neurocognitive tasks and subjective parameters are assessed, and vital signs are recorded. The participant, at their option, may have an intravenous catheter placed to facilitate repeated blood collection.

Up to three subjects are dosed in one day at staggered dosing times. After being provided with a light breakfast, the subject will be transported by shuttle (approximately

20-minute drive) to the NADS facility. Participants have whole blood, plasma, and oral fluid collected along with a breath alcohol measurement prior to dosing.

Subject drinks a prepared mixed alcohol drink ad libitum over the course of 10 minutes. The drink is either placebo (only enough alcohol to suggest smell/taste, ≤ 10 mL) or an ethanol dose calculated to produce an approximate peak BAC of 0.065% and an approximate minimum BAC of 0.05% at the end of the driving visit (based on age, height, weight, and drinking practices. Both drinks consist of a mixture of Everclear grain alcohol and juice. Subjects will have the choice of several juices, but will be required to use the same juice across all sessions.

After alcohol consumption, the subject inhales vaporized cannabis from a filled

Volcano® balloon ad libitum over 10 minutes. The cannabis is either placebo, 2.5-3.5%

102

THC, or 6.0-7.5% THC cannabis plant material. Approximately half a gram (500 mg) dried cannabis is placed in the filling chamber of the Volcano® Medic device and vaporized into the associated Volcano balloon, delivering approximately 0, 17.5, or 33 mg THC for inhalation (69, 168, 236-238).

Over the course of the study, the subject receives each of the following dosing conditions once (one per session), in random order: placebo THC and placebo alcohol, placebo THC and alcohol, low-dose THC and placebo alcohol, high-dose THC and placebo alcohol, low-dose THC and alcohol, high-dose THC and alcohol. The entire dosing process lasts approximately 30 minutes.

Biological specimens (blood, plasma, oral fluid, breath alcohol measurement) are collected up to 8 h after the start of dosing. No more than 486 mL total blood is collected during the whole study, and no more than 81 mL during any session (no more than 9 mL at any single collection) will occur. Oral fluid specimens are collected periodically with the QuantisalTM collection device and Draeger DrugTest® 5000. Breath analysis of alcohol occurs at the same collection times as well, measured on-site using the Alco-

Sensor IV (Intoximeters, Inc., St. Louis, MO).

After dosing, the subject is escorted to the driving simulator. Eye movements in the simulator are tracked continuously in real time by a Face LabTM 5.0 (Seeing

Machines, Canberra, Australia) eye-tracking system. The scripted driving task is conducted approximately 30 to 60 min post-dose. (See description of simulation). Upon completion of the driving simulation, the subject is escorted from the simulator.

Neurocognitive task performance and subjective effects outside the simulator are assessed periodically up to 8 h post-dose. Subjective effects such as “high,” “euphoria,”

103

strength of drug effect, and sedation are assessed periodically with 100 mm visual analog scales (VAS) and the 5-point Likert scale. After the simulator session, subjects complete the Realism Survey (to rate the driving experience, Protocol Appendix 3: Realism

Survey).

Prior to discharge, participants will be evaluated for normal psychomotor function, as evidenced by adequate performance on a standardized neuromotor examination that evaluates mental status and gross motor coordination (Protocol

Appendix 4: Neuromotor Examination). Orthostatic vital signs will be assessed before subjects are discharged. Subjects stay a second night in the CRU only if investigators consider it necessary for safety purposes. Transportation home is provided or a relative or friend may pick the subject up. Subjects are instructed not to drive for the rest of the day.

Drug Administration

Ingested Alcohol

Subjects will consume the alcohol drink ad libitum over a span of 10 minutes. For alcohol administration sessions, the drink will contain Everclear grain alcohol and juice.

For alcohol sessions, subjects will be dosed to an approximate peak BAC of 0.065%.

Subjects will be tested on the decline such that subjects will be at or above the goal BAC

(0.05%) throughout the drive. On placebo alcohol days, the drink contains one part water and 1.5 parts juice. The glass has its rim swabbed with alcohol, and 10 mL is floated on top to produce the initial odor and taste of alcohol (202).

104

Inhaled Cannabis by Volcano® Vaporizer

The Volcano® Medic (Vapormed GmbH & Co., Tuttlingen, Germany), approved for medical use in Germany, the Netherlands, and Canada, is employed for administration of cannabis. It includes fuse circuitry to prevent overheating and potential burns.

Since this study employs a vaporizer rather than a smoking method, the harmful effects associated with cannabis smoking will be greatly reduced. This study’s inhalation procedure is thus safer than typical recreational use (78, 83). The Volcano is a healthier alternative for pulmonary intake of cannabis, and involves less risk to subjects than their normal cannabis intake by cigarette.

The higher the temperature of vaporization, the higher the efficiency of THC volatilization (238). However, exceeding heating levels of 230°C would cause combustion (78), and higher temperatures generally result in harsher taste and consequently throat irritation in some users. It was determined that 210°C is ideal for both comfort and efficiency (238).

Dried bulk cannabis plant material obtained from the NIDA Drug Supply

Program is ground (herb mill/grinder provided by Volcano manufacturer) and 500 mg is heated to a temperature of 210°C by the hot air generator to vaporize THC (69, 238). The vapor is collected in a plastic balloon; a valve shuts when detached to prevent escape of vapor. The balloon is attached to a mouthpiece that allows subject-controlled inhalation of cannabis vapor; the valve closes when the subject’s mouth is not pressing on it. Each session utilizes a new balloon, to ensure hygienic administration. The generator will be operated according to the manufacturer’s specifications.

105

After completing alcohol intake, subjects will inhale the vapor ad libitum over a span of 10 min.

Simulated Drive in the NADS

The National Advanced Driving Simulator (NADS) contains a 1996 Malibu sedan mounted in a 24-foot dome. The motion system, on which the dome is mounted, provides

400 square meters of horizontal and longitudinal travel and ±330 degrees of rotation. The driver feels acceleration, braking, and steering cues as if he or she were actually driving a real vehicle. Each of the three front projectors has a resolution of 1600 x 1200; the five rear projectors have a resolution of 1024 x 768. The edge blending between projectors is five degrees horizontal. The NADS produces a complete record of vehicle state (e.g., lane position) and driver inputs (e.g., steering wheel position), sampled at up to 240 Hz (202).

The cab is equipped with a Face Lab™ 5.0 (Seeing Machines, Canberra,

Australia) eye-tracking system mounted on the dash in front of the driver’s seat above the steering wheel. The worst-case head-pose accuracy is estimated to be about 5º. In the best case, where the head is motionless and both eyes are visible, a fixated gaze may be measured with an RMS error of 2º (202).

An Alco-Sensor IV (Intoximeters, Inc., St. Louis, MO) breath alcohol-testing instrument measures BAC. The hand-held sensor uses a fuel cell to determine BAC level.

The system is approved by the US DOT for evidential use and exceeds the federal model specification for traffic enforcement and Omnibus Breath Alcohol Testing. The system is designed to measure BAC levels from 0.00% to 0.40% with drift of less than 0.005%

106

BAC over several months. The system is checked at least every other day for calibration and recalibrated by an approved dry gas method (202).

The driving scenarios are modifications of those utilized in the IMPACT study

(202) and the ongoing Advanced Countermeasures for Multiple Impairments project.

Each drive includes three connected nighttime driving segments. The drives start with an urban segment composed of a two-lane roadway through a city with posted speed limits of 25 to 45 mph as well as signal-controlled and uncontrolled intersections. An interstate segment follows that consists of a four-lane divided expressway with a posted speed limit of 70 mph. The interstate segment begins with a 4-minute period in which drivers encounter two slower moving heavy trucks. This period overlaps a secondary task involving inserting CD into a CD-player and navigating to specific tracks. The drives conclude with a rural segment featuring a two-lane undivided road with curves onto a gravel road. The final segment of the drive includes an extension of the original gravel roadway from IMPACT and then a 300 second straight paved roadway. Overall, these three segments mimic a drive home from an urban parking spot to a rural location via an interstate. Events in each of the three segments combine to provide a representative trip home in which drivers encounter situations that might be encountered in a real drive, based on FARS data. Throughout the urban section, a series of potential hazards requires drivers to scan the roadside, but not necessarily physically respond to avoid collisions.

These hazards include pedestrians, motorbikes, and cars entering and exiting parking and oncoming lanes. These hazards have paths that would cross the driver’s path if they were to remain on their initial headings. There is an instance where a pedestrian crossed the driver’s path well in front of the driver.

107

Additional changes to the drive to improve sensitivity will be made during the development and pilot testing phase of this project.

Assessment of Risk Perception, Risk Taking, and Impulsivity Traits

Risk Perception

Risk Perception Questionnaire

The risk perception questionnaire (222) consists of 40 different events of three types: controllable, uncontrollable, and neutral (Protocol Appendix 5: Risk Perception

Questionnaire with Randomization Instructions). Participants rate the likelihood that they will experience each event with a 7-point Likert scale. Participants answer the question,

“Compared to other adults of your age and gender, how likely is it that you will at some point in your lifetime?” The response options are numbers between –

3 and 3, in which 0 represents “equally likely”; –1, –2, and –3 represents “a little less likely,” “less likely,” and “much less likely,” respectively; and 1, 2, and 3 represent “a little more likely,” “more likely,” and “much more likely,” respectively. The 40 items and scales are printed in a booklet, the first page of which features task instructions and practice items. The degree to which participants judge themselves to be “less likely” than others to experience the individual events reflects their level of optimism. Differences in levels of optimism across item types reflect the influence of controllability upon participants’ optimistic biases. These are the primary outcome measures for this task.

During a brief introduction and practice phase, participants are instructed on how to complete the questionnaire. Two practice events are completed with the help of the interviewer before the participants begin to answer the questionnaire. When the practice

108

trials are completed, the interviewer reads each event on the questionnaire aloud to each participant to control for differences in reading ability and to provide opportunities for the participants to ask questions about the events. Participants are assured they do not have to report responses aloud and do not have to answer any questions that make them feel uncomfortable.

Participants receive different versions of the questionnaire. The versions differ only in the order of the questions, which is randomized for each participant to control for potential order effects. Test duration is approximately 10 min.

Self-Assessment of Risk Perception

The self-assessment of risk perception is one of three short-form assessments (the others are in the Risk-taking and Impulsivity Assessments section) designed by Cherpitel and colleagues through factor analyses of data from a larger national alcohol survey (223)

(Protocol Appendix 6: Self-Assessments of Risk Perception, Risk-Taking/Impulsivity and Sensation-Seeking). It has 6 items answered on a 5-point Likert scale; subjects respond to the question “How likely is it that something bad will happen to you if you…”

(e.g., “…had sex with someone you just met”) (1=Very Unlikely, 5=Very Likely). Test duration is approximately 2 min.

Risk-Taking and Impulsivity

Self-Assessment of Risk-Taking/Impulsivity

The risk-taking/impulsivity scale has 5 items that are answered on a 5-point Likert scale (223) (Protocol Appendix 6: Self-Assessments of Risk Perception, Risk-

109

Taking/Impulsivity and Sensation-Seeking). Subjects respond to the question “How much does each of the following statements describe you?” for 5 items (e.g., “I often act on the spur of the moment without stopping to think”) (1=Not at all, 5=Quite a lot). Test duration is 2 min.

Self-Assessment of Sensation-Seeking

The 4-item sensation-seeking scale has 4 items answered on a 5-point Likert scale

(223) (Protocol Appendix 6: Self-Assessments of Risk Perception, Risk-

Taking/Impulsivity and Sensation-Seeking). Subjects respond to the question “How much does each of the following statements describe you?” for 4 items (e.g., “I like to try new things for the excitement”) (1=Not at all, 5=Quite a lot). Test duration is approximately 2 min.

Impulsive Sensation-Seeking subscale from the Zuckerman-Kuhlman Personality

Questionnaire

The Impulsive Sensation-Seeking subscale of the Zuckerman-Kuhlman

Personality Questionnaire (ZKPQ) (224) predicts future risky behaviors (Protocol

Appendix 7: Impulsive Sensation-Seeking Subscale from the Zuckerman-Kuhlman

Personality Questionnaire), based on responses to 19 True/False questions (e.g., “I like doing things just for the thrill of it.”; “Before I begin a complicated job, I make careful plans.”) Test duration is approximately 5 min.

110

Barratt Impulsiveness Scale Version 11

The Barratt Impulsiveness Scale Version 11 (BIS-11) (225) is a widely-used 30- question self-report questionnaire designed to assess impulsiveness (Protocol Appendix

8: Barratt Impulsiveness Scale (Version 11)). Questions are answered on a 4-point Likert scale: Never/Rarely (1 point), Occasionally (2 points), Often (3 points), or Almost

Always/Always (4 points). Individual item scores are summed to give an overall impulsivity score, with higher scores suggesting greater impulsivity. To minimize response bias, selected items are worded so that a positive response suggests less impulsivity. Test duration is approximately 5 min.

Assessment of Cannabinoid/Alcohol Intoxication

Symptoms of cannabinoid and/or alcohol intoxication are assessed prior to and for up to 8 h following alcohol and cannabis administration.

Subjective Effects

Subjective effects are assessed with seven 100-mm visual-analogue scales, taking about 1.5 min to complete, and 13 five-point Likert scale items, taking about two min to complete (Protocol Appendix 9: Subjective Effects). Both types of scales measure psychological and physical symptoms associated with cannabinoid intoxication.

Neurocognitive Performance

The effects of cannabis and alcohol on neurocognitive performance are assessed with the Rogers Decision-Making (gambling) task and the BART.

111

Rogers Decision-Making (Gambling) Task

The Rogers Decision-Making task assesses effects of cannabis and/or alcohol on risk-taking behavior. Subjects are presented with a series of choices between gambles.

The subject begins with 100 points and is told to attempt to increase this total. This

Rogers version has been modified in that the odds stimuli are displayed as pie charts instead of the histogram bars created by the Rogers group. The odds, wins and losses structure of the task remains the same; there are 10 trial types (including all-win and all- loss trials), but the way that feedback is displayed will be different. The subject must choose which gamble to play by pressing the corresponding button. There are 30 trials and each trial lasts for 12 seconds. Jitter was placed before and after the choice stimuli and the feedback stimuli in order to allow resolution of reward/punishments from the decision making process. Feedback will be given on a single screen and the feedback stimulus slide is only on for 2 seconds. The feedback will appear as a green arrow displaying positive gains, or a red down-ward facing arrow displaying losses. The final point total is displayed at the end of the 30 trials.

Specific gamble stimuli are presented as described in Rogers et al. (228), and are designed to distinguish effects on probabilistic discrimination between expected rewards or losses, and biasing towards risk-taking or risk-aversion. . A monetary incentive increases the salience of each choice and maintains participants’ interest. The incentive will be $1 per 100 points total score at the completion of the task. It is expected that participants will earn approximately $10 to $15 per session (or $60 to $90 total) as a result of this incentive.

112

Behavioral Risk Assessment (Balloon Analogue Risk Task [BART])

The BART is a computerized task in which participants view a computer screen displaying a small, simulated balloon and a simulated balloon pump (Protocol Appendix

10: Diagram of the Balloon Analogue Risk Task (BART)). Participants use a mouse to click on the pump in order to inflate the balloon by approximately one-eighth of an inch.

With each click of the pump, participants earn 1 cent. This money is deposited into a temporary cache, visible to the participant. Participants are asked to inflate the balloon as much as possible, earning as much money as possible, before the balloon pops. If the balloon pops, the money held in the temporary cache is lost. However, if participants want to stop inflating the balloon, they click on the “Press to Collect $$$” button in order to move money from the temporary cache to a permanent bank. The amount of money in the permanent bank is displayed continuously on the screen. After the balloon pops or money is collected, the next balloon appears. The task is repeated for a total of 20 balloons (“trials”). The number of pumps, amount of money earned and number of popped balloons serve as primary outcomes for quantifying risk-taking behavior.

In order to make the task as personally relevant as possible, participants earn money (1 cent/pump, averaging 40 cents/balloon, $8 per session or about $48 over six sessions) as a result of the choices they make. Participants cannot lose money. Test duration is approximately 5 min for 20 balloons.

113

Specimen Collection and Analysis

The University of Iowa NADS research program has extensive experience conducting controlled drug administration clinical trials, including administering alcohol and assessing BAC with an appropriate breath alcohol testing device (202).

The Chemistry and Drug Metabolism Section (CDM) of the National Institute on

Drug Abuse (NIDA) Intramural Research Program has a long history of investigation of the pharmacodynamics and pharmacokinetics of drugs of abuse. As part of each investigation, the disposition of drugs and metabolites in traditional and alternative matrices is included. This is especially important in this research that focuses on the effects of cannabis and alcohol on driving ability, given current debates on per se limits for THC impairment (157, 160, 239). This study evaluates the effects of THC and alcohol alone and together on driving performance, and on the potential ability to detect performance impairment and correlate with plasma, blood, and oral fluid (Draeger on-site and Quantisal collection) testing and subsequent analysis. CDM has multiple analytical tools for quantifying drugs and metabolites in biological matrices, currently including seven GCMS instruments, 5 equipped with two-dimensional (2D) GCMS and 5 with negative and positive chemical ionization. Additionally, the Section has four liquid chromatography tandem mass spectrometers (LCMSMS). CDM is a recognized leader in the field of sensitive and specific drug and metabolite analysis in blood, urine, oral fluid, sweat and hair, and this expertise will be applied to the quantification of cannabinoids collected in this study.

114

Whole Blood/Plasma Specimens

Peripheral venous whole blood is collected throughout the dosing session.

Approximately 81 mL of whole blood is collected during each session, with no more than

9 mL blood at a single time point. Blood specimens (3 mL for blood; 6 mL for plasma) are collected and stored on ice.

Oral Fluid Specimens

OF specimens are collected with the Quantisal™ oral fluid collection device and for the on-site Draeger DrugTest® 5000. Up to 12 OF specimens are collected with each device during each session, for a total of up to 144 oral fluid specimens per participant.

Breath Specimens

Blood alcohol concentration (BAC) is determined in real time with a standard, on- site breath testing device, the Alco-Sensor® IV. Measurement times correspond to collection of blood and oral fluid specimens.

Specimen and Data Storage

University of Iowa

Biological specimens are stored on ice until freezing. No more than 2 h after collection, 6 mL blood specimens are centrifuged to separate plasma that is transferred into Nunc cryotubes; whole blood and OF also are stored in Nunc cryotubes. Samples

115

must be refrigerated immediately after processing and frozen no later than the end of each session.

Prior to shipment, all biological specimens (blood, oral fluid, plasma) are stored in the dark in a secure freezer with proper identification labels. Labels include subject identification number, which indicates the study, the subject’s gender, session, and collection time, but no personally identifiable information (PII). Biological specimens are frozen and batch shipped on dry ice by certified courier to the NIDA IRP for analysis.

All clinical and analytical data are stored on a secure, password protected server.

All data collected in the simulator suite are maintained on the secure server per

UIA/NADS protocol.

Session records of collection times, results of real-time analyses, and research data from the neurocognitive tasks and questionnaires are labeled with encoded subject information only. Physical (on paper rather than electronic) data should be kept locked in a secure location, accessible only to staff.

NIDA IRP

All clinical and analytical data from the Chemistry and Drug Metabolism Section

(CDMS) of NIDA IRP are ultimately archived on a secure, password-protected server located in the server room in the NIH Biomedical Research Center (BRC) building in

Baltimore, MD. The entire server contents are backed up to Digital Analog Tapes (DAT) twice weekly. Data may from time to time be stored on password protected desk-top computers located in locked offices of the BRC building.

116

Specimen Handling

Specimens are labeled with encoded subject information and collection time, but no PII.

NIDA IRP

The NIDA IRP does not have access to PII. Data and biological specimens are fully coded to protect subjects’ information prior to delivery to NIDA.

Electronic data are stored on a secure server or security-enabled computers. Data transfer between the University of Iowa and the NIDA IRP will be through a secure network connection.

Specimens are stored in a secure freezer until fully depleted.

Hard copy and electronic data and biological specimens are retained indefinitely for possible future analysis. Data may be combined with data obtained in other studies.

Follow-up/termination procedures

Study procedures are for research purposes and not clinical care. Individual results will not be provided to participants; however, results of the study will be published and publically available.

Re-contact of subjects during study

Study personnel will contact participants by phone to schedule sessions. After multiple attempts by phone, if a participant is unreachable, s/he will be considered a noncompleter and discharged from the study.

117

Re-contact of subjects after study

The study is finite in scope and design; upon completion of six dosing sessions, participants will have completed the study. If there are no adverse events, there is no need for follow-up contact.

IND/IDE Requirements

This study requires approval of an IND/IDE for administration of cannabis by vaporizer. The Volcano® Medic is manufactured by Vapormed, a division of Storz &

Bickel, Tuttlingen, Germany. NIDA Drug Supply Program supplies the bulk cannabis material that is shipped directly to the DEA-schedule 1 licensee at the University of Iowa.

Adequate facilities are available to store the cannabis securely under appropriate conditions (frozen at -20°C) (238).

Risks/ discomforts

Adverse Events Associated with Vaporized Cannabis

Cannabis is administered as inhaled vapor rather than cannabis smoke; adverse effects may be reduced due to limited vaporization of nitrogenous amines, etc present in cannabis smoke (78, 83).

Psychological effects

Based on the clinical literature describing recreational smoked cannabis and the scientific literature on human experimental studies with smoked, oral, and inhaled vaporized cannabis and THC, we expect the following dose-dependent psychological 118

effects: euphoria and dysphoric reactions, including anxiety and panic, paranoia and psychosis. These reactions are more common in naïve users, anxious subjects and psychologically vulnerable individuals. Cannabis also may produce perceptual changes, including distorted perception of space and time and hallucinations (220, 240). Cognitive and psychomotor effects may include slowing of reaction time, motor incoordination, deficits in short-term memory, difficulty in concentration and impairment in complex tasks that require divided attention.

Physiological effects

Cannabinoids produce dose-related tachycardia reaching rates of 160 beats per minute or more, although tolerance may develop with around-the-clock use. Widespread vasodilation with reddening of the conjunctivae are common (241). Postural hypotension and fainting may occur.

Adverse Events Associated with Ingested Alcohol

Psychological effects

Psychological effects of alcohol at typical recreational doses include a feeling of well-being and reduced problems, a greater tendency toward instinctive behavior, overconfidence, reduced inhibitions, and greater sociability (242). There is a slight risk to subjects of embarrassment because of disinhibited or inappropriate behavior.

119

Physiological effects

Ethyl alcohol is a CNS-depressant. Moderate recreational doses and associated

BAC lead to dulled and slowed sensory input. Mental and motor functions also may be impaired, such that reaction time is slowed and an increase in body sway is observed among drinkers. It is possible that a subject may experience temporary nausea, dizziness, headache, and/or hangover. A target BAC of 0.05% is employed in this protocol, less than the legal limit for DUI and frequently surpassed in normal, recreational settings.

Therefore, no adverse events (AEs) are anticipated at this alcohol concentration. Risk should be comparable to that typically experienced by the study population during normal drinking sessions.

Alcohol, even in small to moderate amounts, may damage a fetus. Pregnancy tests are not 100% accurate and it is possible, although unlikely, that a pregnancy test taken at

NADS may result in a false positive or false negative result.

Adverse Events Associated with Study Measures

Blood collection

Venous blood sampling may cause pain, tenderness, bruising, or bleeding at the needle puncture site. Some subjects may feel transient lightheadedness or dizziness, or lose consciousness (syncope), because of anxiety and vasovagal reaction. This risk is minimized by performing venipuncture while subjects are seated and having them under staff observation. The risk of infection is negligible due to standard sterile collection technique. Placement of indwelling venous catheters poses a risk of infection or thrombophlebitis, increasing with duration of placement. This risk is minimized by use of

120

sterile techniques, with prompt removal if there are clinically significant signs or symptoms of tenderness, swelling, or redness. The risk of anemia is negligible because subjects must have a normal hematocrit to participate and the total amount of blood to be collected in one session (approximately 81 mL) is less than the amount collected during a routine blood donation. Because there must be at least 1 week between sessions, subjects will replace some lost blood prior to donating in subsequent sessions.

Oral Fluid Collection

The only risk associated with oral fluid collection is dryness of the mouth.

Breath Collection

There is no risk associated with breath collection.

Simulator Driving

Risks associated with driving in the NADS include motion sickness, nausea, and disorientation. Previous studies with similar driving intensities and simulator setups have produced few disorientation effects. When effects are reported, they are usually mild to moderate and consist of slight uneasiness, warmth, or eyestrain for a small number of subjects. These effects are believed to last for only a short time, usually 10-15 minutes, after leaving the simulator.

In the rare event that normal exiting of the simulator is not available, subjects would be assisted down a small ladder and escorted to a waiting room. This could pose a slight risk if a subject has difficulty negotiating a ladder or walkway in the simulator bay.

121

Cognitive Testing

Risks of cognitive testing include boredom and anxiety while taking the tests.

Emotional, Psychological, Legal Risks

There is a slight emotional/psychological risk stemming from potential loss of confidentiality. Should specific participant information be released, participants could face legal repercussions. However, this risk is minimized by the federal certificate of confidentiality protecting against release of this information (see confidentiality section).

Applicants will be informed of the measures taken to protect their confidentiality.

Prevention/Minimization of Adverse Events

Participants are carefully screened medically and psychologically to exclude individuals who may be at increased risk for adverse events. Subjects are carefully monitored by trained staff throughout the sessions as described below. Clinical staff are experienced and will maintain confidentiality, minimizing the risk of embarrassment and/or repercussions. Ample opportunity to take breaks during the questionnaire sessions will minimize the risk of restlessness. Any abnormality, adverse event, or matter of concern is promptly evaluated as appropriate. The simulator may be stopped on demand if necessary.

Subject monitoring

Participants’ responses to study drugs are closely monitored by clinical staff and study personnel. A physician is available on-call. Participants remain under staff 122

observation. Participants will be withdrawn from the study at their request, if they are unable or unwilling to comply with protocol requirements, or if the physician deems it necessary.

Subjects will be informed that they do not have to answer any questions or are not obligated to answer any questions that make them feel uncomfortable. However, if a subject chooses not to answer questions about current habits that pertains to eligibility criteria or does not meet criteria, participation in the study may be terminated.

Outcome measures

Primary outcome measures are driving performance in the simulator (as measured by speed, lane position, steering wheel position, reaction time, headway to lead vehicle, eye tracking, etc) and alcohol and cannabinoid concentrations in blood, plasma, and oral fluid.

Secondary outcome measures include performance on the BART and Rogers gambling task, subjective effects (VAS, Likert scales), and correlations between cannabinoid concentrations and driving performance and effects on the BART and

Rogers gambling tasks.

Statistical Analysis

Statistical analyses include descriptive statistics (mean, median, standard deviation, standard error, range, frequency distribution) for all outcome measures and

123

comparisons of measures by drug conditions. Analyses will seek associations among variables. Significance will be assigned at two-tailed p < 0.05.

Analysis of pharmacodynamic effects will consist of one-way repeated-measures analyses of variance (ANOVAs) to assess the effect of time on each outcome. A significant time effect is evidence for impairment, and will be followed with post-hoc testing to determine which time points differ from baseline. Modified Dunnett’s tests will be used, and adjustments will be made to constrain the overall type I error to 0.05.

Probability levels will be expressed as two-tailed, and effects will be considered significant if p <0.05. Comparison of the pharmacodynamic response across different doses will use mixed model ANOVA with 3 terms (session [dose], time [within session], time x session), with the dependent variable change from the session baseline value.

Pharmacokinetic parameters calculated for cannabinoid biomarkers and breath alcohol measurements include maximum concentrations (Cmax), time to maximum concentrations (Tmax), areas under the curve (AUC) and metabolite/parent ratios over time. Cannabinoid concentrations are correlated with ongoing pharmacodynamic effects.

Power Analysis

Prior research (34) demonstrated the ability to detect differences in standard deviation of lane position for detecting alcohol and THC effects with a sample of 18 subjects. Prior work at NADS (202) showed the ability to detect differences in driving performance measures across alcohol conditions with a sample of 18 subjects per experimental condition. Assuming the desire to detect a difference in SDLP of 2 cm, which is similar to differences found previously, and a standard deviation of 3 cm, a

124

sample of twenty subjects for a within subject comparison would provide a power of

0.8073. With an anticipated twenty subjects completing all sessions sufficient power should be present to detect the anticipated differences.

Accrual Number Request

We require 20 participants to complete the study. To reach that number, it is anticipated that 60 participants will need to be enrolled. Because screening procedures are performed after enrollment into this protocol, we expect approximately 20 enrolled participants to fail to meet eligibility criteria after screening, while another 20 participants are expected to not complete all six sessions.

Human Subjects Protection

Equity of Subject Selection

Participants will be recruited without regard to ethnic origin, sex, race, or religion.

The recruitment target is based on the national population aged 21 years or older who were current (in the past three months) illicit drug takers and alcohol drinkers (derived from the 2009 National Household Survey on Drug Use and Health (185) and the 2000

Iowa City census. Target enrollment is: 63% male; 88% Caucasian, 4% African

American, 2% Asian, and 3% other; and 3% Hispanic.

125

Exclusion Justification

Pregnant or nursing women are excluded because of possible risks of alcohol and cannabis to the fetus or nursing child. Urine pregnancy tests are administered on admission prior to each session for females of childbearing potential. Although contraceptives are not required, study staff may strongly encourage female participants to remain on birth control for the duration of the study. A positive pregnancy test is grounds for exclusion from the remainder of the study.

Subjects must be at least 21 years old to be administered alcohol legally in the state of Iowa. Individuals older than 55 years old are excluded because of the increased likelihood of medical problems that could be exacerbated by the effects of cannabis or alcohol and increase in crash risk per mile driven associated with drivers over 55.

Lack of a valid driver’s license or less than two years of holding a license are grounds for exclusion because this is a driving study examining the effects of psychoactive drugs on driving performance. Driver inexperience/inability to drive would confound the data.

Qualifications of Investigators

Gary Gaffney, M.D. – Principal Investigator, UIA; Medically Accountable Investigator -

As an Associate Professor in the Division of Child and Adolescent Psychiatry, in the

Department Psychiatry, Dr. Gaffney retains board certification in Child & Adolescent

Psychiatry, and Adult Psychiatry. His degrees include an M.D. from the University of

Iowa with adult psychiatry residency at the Johns Hopkins Hospital and a Child and

Adolescent Psychiatry fellowship at the University of Iowa Hospitals and Clinics. Dr

126

Gaffney has authored over 50 academic publications. chapters, and abstracts, as well as well over 150 academic presentations in his career. Dr. Gaffney’s research endeavors include clinical and epidemiological studies, as well as brain imaging studies in aspects of childhood development He has engaged in several randomized clinical trials of various pharmaceutical agents in children and adults. Recent research endeavors include driving in Autism, and diving simulation in adults administered alcohol.

Omar Ahmad, M.S. - Project Manager, UIA; Assistant Director – Mr. Omar Ahmad is

Assistant Director at the National Advanced Driving Simulator (NADS). Mr. Ahmad has been working in the field of driving simulation for the past 16 years. He received his

Masters degree in Computer Science from the University of Iowa in 1995 and served as part of the team that designed and wrote the software for the simulators at the National

Advanced Driving Simulator. Mr. Ahmad has served as the project manager for a number of research studies related to studying vehicle safety and human performance in the areas of active safety, impairment, driver distraction and novice drivers. He is also actively engaged in matching simulator fidelity requirements to applications in roadway design, clinical assessments, and driver training. Mr. Ahmad is a member of the National

Academies Transportation Research Board Committee on Simulation and Measurement of Vehicle and Operator Performance

Timothy L. Brown, Co-Investigator - Sr. Researcher, H.F. – Timothy L. Brown has more than a decade of research experience in transportation human factors. His academic and

127

research career has focused on in-vehicle systems, warning systems, medical considerations in driving, and modeling driver performance. Dr. Brown is the author of over fifty journal articles, conference papers, and technical reports, and a recipient of the

Jerome H. Ely Award for the best article published in Volume 44 of Human Factors for

2002. He has also taught classes in engineering economics and cognitive human factors.

Particular areas of expertise include function analysis, analysis of driver behavior, simulation and modeling, in-vehicle warning systems, evaluation of in-vehicle systems, and driving simulation.

Study Benefits

This study offers no direct benefit to participants, but is likely to yield generalizable knowledge about the effects of cannabis alone and combined with alcohol on driving performance that will benefit society as a whole. Experimental studies of actual driver performance under the influence of cannabis may help in development of better methods for prevention, detection, and amelioration of cannabis-associated driving impairment.

Summary/classification of risk: for the study as a whole

This study poses more than minimal risk to participants. The primary risks are administration of an illicit drug and breach of confidentiality.

128

We assert that there is a favorable risk-benefit ratio in terms of the knowledge this study will contribute to society, since driving while impaired in any capacity is a potentially fatal danger not just to the individual but to people and property nearby (14,

30, 243-244).

Consent documents and process

Consent is obtained from participants by qualified investigators. The consent form will be mailed or emailed to applicants prior to screening visit. At the start of the screening visit, prior to any data collection, staff will orally review the informed consent document with applicant, allow the applicant time to read, and discuss, and resolve any questions.

Data and safety monitoring

This is a single-site study with a relatively small number of subjects. The on-site investigators and study staff have prompt knowledge of and access to all information regarding adverse events. Therefore, the principal investigator can appropriately be responsible for overall data and safety monitoring.

Adverse event reporting

All AEs, including but not limited to adverse drug reactions and/or motion sickness, shall be documented in accordance with Federal and IRB requirements and good clinical practice. This information shall be stored organized by subject and experimental session.

129

Alternatives to participation or alternative therapies

Subjects do not receive any clinical treatment in this study or forego any clinical treatment in order to participate in this study. The alternative, therefore, is not to participate.

Confidentiality

Confidentiality of participants is maintained at all times. Study records, data, and specimens identify subjects by ID number, not by name or other PII. The code linking ID number with subjects’ identity is kept confidential by investigators, and under lock and key when not in use. No information will be disseminated to participants’ health care providers except as necessary under emergency conditions.

A federal certificate of confidentiality will be obtained. This allows investigators to withhold PII from outside requestors such as police, courts, and subjects’ relatives. No identifiable subject information will be released outside the research team without written permission, with the following exceptions: 1) information needed by other clinical personnel or facilities to provide medical care to subjects; 2) audits or reviews of the research study by UIA, FDA, or other authorized federal agencies. No information will be disseminated to participants’ health care providers except as necessary under emergency conditions.

Conflict of Interest

No investigators have reported a conflict of interest.

130

The Volcano Medic vaporizer from Vapormed/Storz & Bickel is employed for administration of cannabis. This product is not commercially available in the United

States and will be used under an IDE. The Quantisal Oral Fluid Collection Device from

Immunalysis is employed for specimen collections. There are no conflicts of interest to report.

Research and Travel Compensation

Volunteers are compensated for time and research-related inconveniences.

Participant Remuneration

Subjects will be compensated for their time and effort. Subjects who complete all study procedures successfully will receive up to $980, including the monetary incentives associated with the Rogers Decision-Making task and BART. For subjects that do not complete, prorated compensation will be provided. For subjects that repeat a session, they will receive compensation for both the original and the repeat visit, which may result in total compensation greater than $980.

131

Chapter 4 – Cannabis Effects on Driving Lateral Control With and Without Alcohol

(As submitted to Drug and Alcohol Dependence, March 2015)

Abstract

Background: Effects of cannabis, the most commonly encountered non-alcohol drug in driving under the influence cases, on driving are unclear. We aimed to determine how blood ∆9-tetrahydrocannabinol (THC) concentrations relate to driving lateral control impairment, with and without alcohol.

Methods: Current occasional (≥1x/last 3 months, ≤3days/week) cannabis smokers drank placebo or low-dose alcohol, and inhaled 500 mg placebo, low (2.9%)-THC, or high

(6.7%)-THC vaporized cannabis over 10 min ad libitum in separate sessions (institutional review board-approved study, within-subject design, 6 conditions). Participants drove

(National Advanced Driving Simulator, University of Iowa) simulated drives (~0.8 h duration). Blood and oral fluid (OF) were collected before (0.17 h, 0.42 h) and after (1.4 h, 2.3 h) driving at 0.5 h, with simultaneous breath alcohol measurements. We evaluated standard deviations of lateral position (lane weave, SDLP) and steering angle, lane departures/min, and maximum lateral acceleration.

Results: In N=18 completers (13 men, ages 21-37 years), cannabis and alcohol increased

SDLP 0.26 cm per 1 μg/L blood THC and 0.42 cm per 0.01 g/210L alcohol. No interaction term indicated additive (not synergistic) effects. During-drive 8.2 and 13.1

μg/L blood THC increased SDLP similar to 0.05 and 0.08 g/210L breath alcohol, illegal in most countries. Because cannabis-alcohol SDLP effects were additive, 5 μg/L THC +

132

0.05 g/210L alcohol showed similar SDLP to 0.08 g/210L. Only alcohol increased lane departures/min (0.030 per 0.01 g/210L) and lateral acceleration (0.0023 m/s2 per 0.01 g/210L). OF THC variability was 2-5-fold higher than blood.

Conclusions: SDLP was a sensitive cannabis-related lateral control impairment measure.

During-drive blood THC ≥8.2 μg/L increased SDLP similar to notably-impairing alcohol concentrations. Despite screening value, OF variability poses challenges in concentration-based effects interpretation.

1. Introduction

Reducing drugged driving is a priority worldwide, and for the US Office of

National Drug Control Policy (2). Cannabis is the most frequently detected illicit drug in drivers (1, 21, 136, 245), and a 48% increase in weekend nighttime drivers positive for

∆9-tetrahydrocannabinol (THC, the primary psychoactive compound in cannabis), from

8.6% in 2007 to 12.6% in 2013-2014 was recently noted (1). Although increased crash risk and driver culpability in crashes associated with blood THC were detected (6-8, 246-

247), cannabis effects on driving are heavily debated. Road tracking and ability to remain within the lane is a crucial driving skill. Lane weaving, an observable effect of drug- impaired driving, is a common diagnostic indicator for assessing driving performance.

Standard deviation of lateral position (SDLP) is a sensitive indicator of vehicular control, often employed in simulated and on-road drugged driving research (25, 172, 248-249).

Previously, cannabis increased SDLP, but results were assessed by dose rather than blood

THC concentrations (34, 47).

133

To date, 23 states and the District of Columbia (DC) approved medical marijuana;

4 states and DC additionally legalized recreational cannabis for adults (250). The consequence of increasing cannabis legalization is a crucial road safety issue. Since legalizing medical marijuana (2000), Colorado observed increased driving under the influence of cannabis (DUIC) cases (58), and fatal motor vehicle crashes with cannabis- positive drivers; whereas no significant change was observed in 34 states without legalized medical marijuana (59). Establishing evidence-based per se laws for DUIC remains challenging and highly debated, with varying laws across the US (157, 239,

251). Those in favor of medical or recreational cannabis are concerned that implementing concentration-based cannabis-driving legislation will unfairly target individuals not acutely intoxicated, because residual THC can be detected in blood for up to a month of sustained abstinence in some chronic frequent smokers (108). Appropriate blood THC concentrations to universally reflect driving impairment remain elusive. Determining blood THC concentrations associated with lateral control impairment in occasional users would benefit forensic interpretation.

There is also interest in linking driving impairment with oral fluid (OF) THC concentrations. OF is easy to collect, non-invasive, and associated with recent cannabis intake (74, 109-110). OF-based DUIC legislation was established in some jurisdictions

(112-114); however, limited simultaneous driving and OF concentration data preclude direct association with impairment.

Alcohol is the most common drug identified in drivers (1, 21). Cannabis and alcohol, frequently detected together (21), produced greater impairing effects together than either separately (44, 47), but it is unclear whether effects are additive or synergistic.

134

We aimed to assess cannabis’ effects relative to THC concentrations, with and without alcohol, on drivers’ lateral control. We hypothesized cannabis and alcohol would each impair lateral control, alcohol effects would be greater than those of cannabis, and the combination would be synergistic.

2. Methods

2.1 Participants

Healthy adults provided written informed consent for this Institutional Review

Board-approved study. Inclusion criteria were ages 21-55 years; self-reported cannabis consumption ≥1x/3 months but ≤3 days/week over the past 3 months (Cannabis Use

Disorders Identification Test [CUDIT] (252)); self-reported “light” or “moderate” alcohol consumption according to a Quantity-Frequency-Variability (QFV) scale (253); or if

“heavy”, not more than 3-4 servings on a typical drinking occasion; licensed driver for ≥2 years with currently valid unrestricted license; and self-reported driving ≥1300 miles in the past year. Exclusion criteria included past or current clinically significant medical illness; history of clinically significant adverse event associated with cannabis/alcohol intoxication or motion sickness; ≥450 mL blood donation in 2 weeks preceding drug administration; pregnant/nursing; interest in drug abuse treatment within past 60 days; currently taking drugs contraindicated with cannabis or alcohol or known to impact driving; requirements for nonstandard driving equipment; and prior participation in a similar driving simulator study.

135

2.2 Study Design/Procedures

Participants entered the clinical research unit 10-16 h prior to drug administration to preclude acute intoxication. Participants drank 90% grain alcohol to reach approximately 0.065% peak breath alcohol concentration [BrAC] mixed with fruit juice, or placebo (juice with alcohol-swabbed rim and topped with 1 mL alcohol to mimic alcohol taste and odor) ad libitum over 10 min. After drinking, they inhaled 500 mg placebo (0.008±0.002% THC), low (2.9±0.14% THC)-, or high (6.7±0.05% THC)-THC vaporized (Volcano® Medic, Storz & Bickel, Tuttlingen, Germany) cannabis ad libitum over 10 min. Cannabis was from NIDA’s Chemistry and Physiological Systems Research

Branch. Participants received all six alcohol/cannabis combinations in randomized order, sessions separated by ≥1 week.

Simulated drives occurred 0.5-1.3 h after start of cannabis dosing. Blood collection times were 0.17, 0.42, 1.4, and 2.3 h post-inhalation. Blood was collected via indwelling peripheral venous catheter into grey-top (potassium oxalate/sodium fluoride)

Vacutainer® tubes (Becton, Dickinson and Company), and stored on ice ≤2 h. Specimens were transferred into 3.6 mL Nunc® cryotubes (Thomas Scientific), stored at -20°C, and analyzed for cannabinoids within 3 months, based on our previous stability study (254).

OF was collected at the same times as blood (except 0.42 h), with the QuantisalTM collection device (Immunalysis, Pomona, CA). BrAC was measured via Alco-Sensor® IV

(Intoximeters, St. Louis, MO) at the same times as blood, reporting alcohol in g/210L breath (limit of quantification [LOQ] 0.006 g/210L), equivalent to blood alcohol concentration (BAC).

136

2.3 National Advanced Driving Simulator

All driving simulations were conducted in NADS-1, the high-fidelity, full-motion simulator at the National Advanced Driving Simulator (NADS), located at University of

Iowa Research Park, Iowa City, IA (Figure 3). A 1996 Malibu sedan is mounted in a 7.3 m-diameter dome with a motion system providing 400 m2 acceleration space, ±330° rotation, and high-frequency motion (37). Drivers experience acceleration, braking, steering cues, sensations of road conditions (e.g., gravel), and realistic sounds (e.g., wind, motor). NADS-1 produces a complete record of vehicle state (e.g., lane position) and driver inputs (e.g., steering wheel position).

2.4 Drives

The 45 min drive focused on multiple driving skills affected by cannabis, including SDLP. Each drive had urban, interstate and rural nighttime segments. The urban segment involved a two-lane city roadway with posted speed limits 25-45 miles/h

(40-72 km/h) and signal-controlled and uncontrolled intersections; interstate, a four-lane divided expressway with posted 70 miles/h (113 km/h) speed limit; rural, two-lane undivided road with curves, a gravel portion, and a 10 min timed straightaway. This represented driving from an urban center to a rural home via interstate. Because each participant drove six times, three scenarios with varied event orders were utilized to minimize practice effects. Each scenario contained the same number of curves and turns, in varied order and position.

137

Figure 3. The National Advanced Driving Simulator: A) exterior, dome mounted in room; B) dome interior with car mounted inside; C) view of night-drive simulation.

138

2.5 Specimen Analysis

Blood THC concentration was quantified by a previously-published method

(255). Briefly, 0.5 mL blood was protein-precipitated with ice-cold acetonitrile, and supernatants diluted and solid-phase extracted. THC’s linear range was 1-100 μg/L. Inter- assay (n=30) analytical bias and imprecision were ≤3.7% and ≤8.7%, respectively. OF

THC quantification is described in detail elsewhere (256). We utilized a published validated method (257), modified by adding 0.4 mL hexane to solid-phase extraction columns before loading the initial elution solvent. THC’s linear range was 0.5-50 µg/L.

Inter- and intra-assay imprecision were ≤6.6%; analytical bias, ≤14.4% (n=21). If concentrations exceeded the upper LOQ, OF specimens were diluted with drug-free

QuantisalTM buffer to achieve concentrations within the method’s linear range.

2.6 Data Analysis

Blood THC concentrations during drives were modeled via individual power- curve regression from pre-drive (0.17 and 0.42 h) and post-drive (1.4 and 2.3 h) specimens. BrAC concentrations during drives were modeled by linear interpolation.

Driving data were analyzed by participants’ modeled concentrations during drives.

Prior to analysis, data were reviewed to determine which events were suitable for analysis. Events for which a dependent measure were not meaningful (e.g., SDLP during a turn), were excluded. For each dependent measure, events with similar means were grouped for analytic purposes. Data were analyzed using SAS v9.4 General Linear Model

(GLM) Select function to identify appropriate regression models to describe the observed data. This procedure was selected due to its ability to accommodate continuous

139

dependent measures and combinations of continuous and categorical independent measures (258) (required to develop a regression model including subject effects). The stepwise selection method was chosen, and the Schwarz Bayesian Information Criterion determined model entry/removal (optimal to prevent subjective bias) (259). Effect hierarchy was not enforced on model parameters. Available model parameters were blood

THC, BrAC, interaction term THC*BrAC, speed limit, inverse curvature (accounting for road curve effect on lateral control), and subject. Dependent measures of drivers’ lateral control included: SDLP, standard deviation of steering wheel angle, lane departures/min

(“lane departure” defined as edge of vehicle crossing a lane boundary; per minute allowed for normalization across drive events), and maximum lateral acceleration in events without sharp turns. For the final regression models, the analysis of variance for the fit of the model is presented, along with estimates, t-values, and p-values for the model parameters.

3. Results

3.1 Participants

Nineteen healthy adults (13 men, ages 21-37 years, 74% white) participated

(Table 11). Most consumed cannabis ≥2x/month (but ≤3 days/week), and reported last intake within a week prior to admission. Participants self-reported driving 6-23 years, and all reported driving ≥1x/week. Data review revealed one subject (#12) was consistently an extreme outlier in almost every measure and dosing condition, including placebo cannabis/placebo alcohol. Further review of driving videos indicated markedly erratic

140

behavior in all conditions. These data were excluded from all driving analyses, yielding

N=18 completing drivers.

3.2 Driving

GLM Select model results are depicted in Table 12. THC concentration and BrAC were significantly associated with SDLP, but the interaction (THC*BrAC) was not selected into the model. This indicates additive, rather than synergistic, cannabis and alcohol effects. To account for a possible ceiling effect of increasing concentrations, quadratic terms THC2 and BrAC2 were added to the list of potential predictors; neither was included in the resultant (unchanged) model. Table 13 indicates model-predicted results at specific blood THC concentrations and BrAC. Blood THC and BrAC increased

SDLP 0.26 cm per µg/L THC and 0.42 cm per 0.01 g/210L BrAC), representing 0.8% and 1.3% increases relative to the median baseline (drug-free) SDLP per µg/L THC or

0.01 g/210L BrAC, respectively. Participants displayed high inter-individual variability in baseline (drug free) SDLP (Figure 5 (Supplemental)). Figure 4 presents superimposed graphs of SDLP versus blood THC and SDLP versus BrAC, illustrating THC and BrAC concentrations that produced the same SDLP. BrAC concentrations of 0.05% and 0.08%, the most common per se alcohol limits worldwide, were associated with similar SDLP to

8.2 and 13.1 µg/L THC concentrations, respectively.

Natural-log SDLP transformation is common analytical practice; for completeness we evaluated this transformation as well as the untransformed SDLP. Results obtained from ln(SDLP) were similar to untransformed SDLP (Table 15 (Supplemental),

141

Table 11. Self-reported demographic characteristics, recent cannabis and alcohol consumption and driving history of 19 healthy adult occasional cannabis smokers. Race Alcohol Hours “stoned” Time since Amount last Typical Cannabis Years of Driving Partici- Age and BMI intake on typical last cannabis consumedb Sex drinks per intake driving frequen- pant (years) ethni- (kg/m2) frequen- cannabis consumed (joint or joint occasion frequency experience cy city cy occasiona (days) equivalent) 1 M 23.7 W 24.3 2-3x/wk 2-4 2-4x/m 1-2 1 1 7 ≥1x/d 2 F 28.4 AA 23.8 ≥4x/wk 2-4 2-4x/m 3-4 14 1 --c --c 3 M 21.9 W 24.7 2-3x/wk 5-6 2-4x/m 1-2 6 1 7 ≥1x/d 4 M 37.8 W 26.1 2-3x/wk 2-4 2-3x/wk 1-2 3 2.5 23 ≥1x/d 5 M 26.6 W 21.6 ≤1x/m 2-4 ≤1x/m 1-2 11 3.5 12 ≥1x/d 6 F 26.3 W 20.0 2-3x/wk 2-4 2-3x/wk 3-4 1 0.25 12 ≥1x/d 7 M 25.8 W 40.6 2-4x/m 2-4 2-3x/wk 1-2 0.3 0.5 11 ≥1x/d 8 M 26.1 H 31.5 2-4x/m 1-2 2-3x/wk 1-2 3 1 10 ≥1x/d 9 M 23.2 W 19.5 2-3x/wk 2-4 2-3x/wk 3-4 2 1 7 ≥1x/wk 10 M 23.1 W 23.9 2-4x/m 2-4 ≤1x/m 1-2 2 0.25 9 ≥1x/d 11 M 32.3 O, H 28.9 2-3x/wk 2-4 2-3x/wk 1-2 4 1 16 ≥1x/d 142 12d F 23.4 W 23.3 2-3x/wk 2-4 2-4x/m 3-4 4 1 8 ≥1x/wk

13 F 30.3 AA 24.1 2-3x/wk 2-4 ≤1x/m <1 120 1 14 ≥1x/d 14 M 24.6 W 23.3 2-3x/wk 2-4 2-4x/m 1-2 7 0.8 8 ≥1x/wk 15 M 21.8 W 32.7 ≤1x/m 1-2 2-4x/m 1-2 7 0.13 6 ≥1x/d 16 F 21.7 AA, W 23.0 2-4x/m 1-2 2-3x/wk 1-2 1.1 1.5 7 ≥1x/d 17 M 28.7 W 18.3 2-3x/wk 2-4 ≤1x/m 3-4 45 0.5 12 ≥1x/wk 18 M 28.1 W 48.3 2-4x/m 2-4 2-4x/m 3-4 5 1 12 ≥1x/d 19 F 22.9 W 21.6 2-4x/m 5-6 2-3x/wk 3-4 1 1 6 ≥1x/d Median (all) 25.8 23.9 4.0 1.0 10 Mean (all) 26.1 26.3 12.5 1.0 10 StDev (all) 4.1 7.5 27.9 0.8 4 Median (N=18) 25.9 24.0 3.5 1.0 10 Mean (N=18) 26.3 26.5 13.0 1.1 11 StDev (N=18) 4.2 7.7 28.6 0.8 4 a‘Hours “stoned” ’ wording originates from Cannabis Use Disorders Identification Test, source of self-reported cannabis frequency data bCannabis amount last consumed is based on empirically-normalized joint consumption, to account for various administration routes and self-reported “sharing” between multiple individuals cParticipant did not provide response dParticipant excluded from driving analyses due to consistently outlying behavior Abbreviations: W, White; AA, African American; H, Hispanic or Latino; As, Asian; O, Other; AI, American Indian/Native American; StDev, standard deviation

Table 12. General Linear Model (GLM) Select results of effects on lateral control measures in 18 volunteer drivers after controlled vaporized cannabis with or without oral alcohol. Standard p-value Parameter DF Estimate (b) t Error Standard Deviation of Lateral Position (SDLP) THC 1 0.26 3.6 0.07 0.0004 BrAC 1 0.42 2.9 0.15 0.0037 THC*BrAC Speed Limit 1 0.50 19 0.03 <0.0001 Inverse Curvature 1 464 9.5 49 <0.0001 Intercept 1 17.3 8.3 2.1 <0.0001 Subject 17 Model df: 21 Model F-value 28.24 Error df: 1916 Standard Deviation of Steering Angle (Curvy) THC BrAC THC*BrAC Speed Limit 1 0.07 5.4 0.01 <0.0001 Inverse Curvature 1 -122 -7.7 16 <0.0001 Intercept 1 5.2 9.0 0.6 <0.0001 Subject Model df: 2 Model F-value 29.59 Error df: 427 Standard Deviation of Steering Angle (Straight) THC BrAC THC*BrAC Speed Limit 1 -0.40 -17 0.02 <0.0001 Inverse Curvature 1 1389 27 51 <0.0001 Intercept 1 25 21 1.2 <0.0001 Subject Model df: 2 Model F-value 657.9 Error df: 1936

143

Table 12. (Continued from previous page) Standard p-value Parameter DF Estimate (b) t Error Lane Departures/min THC BrAC 1 0.030 2.8 0.009 0.0055 THC*BrAC Speed Limit 1 0.010 6.8 0.001 <0.0001 Inverse Curvature 1 10.9 5.2 2.1 <0.0001 Intercept 1 1.4 10.3 0.14 <0.0001 Subject 17 Model df: 20 Model F-value 19.59 Error df: 840 Maximum Lateral Acceleration (Non-Sharp Events) THC BrAC 1 0.0023 3.5 0.0007 0.0005 THC*BrAC Speed Limit 1 0.0012 11.4 0.0001 <0.0001 Inverse Curvature Intercept 1 0.091 10.0 0.0091 <0.0001 Subject 17 Model df: 19 Model F-value 17.37 Error df: 2026 Maximum Lateral Acceleration (Sharp Events) THC BrAC THC*BrAC Speed Limit Inverse Curvature 1 -1.8 -4.3 0.43 <0.0001 Intercept 1 0.45 17 0.027 <0.0001 Subject 17 Model df: 18 Model F-value 8.61 Error df: 304 Driving occurred 0.5h after drinking placebo or active alcohol (calculated to produce approximate peak 0.065% BrAC) and inhaling placebo, 2.9% THC, or 6.7% THC vaporized bulk cannabis (500 mg, Volcano® Medic vaporizer). Estimate represents parameter (coefficient) estimate [effect size scaled to the unit] for each factor (negative b indicates the parameter decreases the effect; positive b indicates the parameter increases the overall effect). Boldface indicates parameter included in the final GLM Select model. All p-values for final overall analysis of variance of model fits were <0.0001. Abbreviations: DF, degrees of freedom; THC, blood ∆9-tetrahydrocannabinol concentration; BrAC, breath alcohol concentration

144

Table 16 (Supplemental)), so we report the more straightforward and conservative SDLP results.

BrAC significantly increased lane departures/min and maximum lateral acceleration, but these measures were not sensitive to cannabis. Neither THC nor BrAC affected standard deviation of steering wheel angle.

Table 13. GLM Select model estimates for predicted standard deviation of lateral position (SDLP), lane departures/min, and maximum lateral acceleration associated with specific blood ∆9-tetrahydrocannabinol (THC) concentrations and breath alcohol concentrations. Maximum Lateral During-Drive Standard Deviation of Lane Departures/min Acceleration Concentration Lateral Position (SDLP) (Non-Sharp Events) Median Median Median [range] Percent [range] Percent Percent BrAC [range] Differ Differ predicted Differ- THC In- predict-ted In- In- (g/ predicted -ence -ence maximum ence (μg/L) creasea lane depar- creasea creasea 210L) SDLP (cm) (N) lateral accel- (m/s2) (%) tures/ (%) (%) (cm) eration min (N) (m/s2) 31.4 0.38 1.17 0 0 ------[24.7-44.8] [0.05-1.95] [0.87-1.54] 31.7 0.38 1.17 1 0 0.26 0.8 0 0 0 0 [25.0-45.1] [0.05-1.95] [0.87-1.54] 32.0 0.38 1.17 2 0 0.52 1.6 0 0 0 0 [25.3-45.4] [0.05-1.95] [0.87-1.54] 32.7 0.38 1.17 5 0 1.3 4.1 0 0 0 0 [26.0-46.1] [0.05-1.95] [0.87-1.54] 33.3 0.38 1.17 7 0 1.8 5.8 0 0 0 0 [26.5-46.7] [0.05-1.95] [0.87-1.54] 34.0 0.38 1.17 10 0 2.6 8.2 0 0 0 0 [27.3-47.4] [0.05-1.95] [0.87-1.54] 36.6 0.38 1.17 20 0 5.2 16 0 0 0 0 [29.9-50.0] [0.05-1.95] [0.87-1.54] 31.9 0.41 1.19 0 0.01 0.42 1.3 0.026 6.9 0.022 1.9 [25.2-45.3] [0.08-1.97] [0.90-1.56] 32.3 0.43 1.21 0 0.02 0.84 2.7 0.053 14 0.045 3.8 [25.6-45.7] [0.11-2.00] [0.92-1.58] 33.6 0.51 1.28 0 0.05 2.1 6.7 0.13 35 0.11 9.5 [26.8-47.0] [0.19-2.08] [0.98-1.65] 34.8 0.59 1.35 0 0.08 3.4 11 0.21 55 0.18 15 [28.1-48.2] [0.26-2.16] [1.05-1.72] 35.7 0.64 1.39 0 0.10 4.2 13 0.26 69 0.22 19 [29.0-49.1] [0.32-2.21] [1.10-1.76] 34.1 0.51 1.28 2 0.05 2.6 8.4 0.13 35 0.11 9.5 [27.4-47.5] [0.19-2.08] [0.98-1.65] 34.9 0.51 1.28 5 0.05 3.4 11 0.13 35 0.11 9.5 [28.1-48.3] [0.19-2.08] [0.98-1.65] Data generated from 18 healthy occasional cannabis smokers 0.5-1.3 h after ingesting placebo or active oral alcohol and inhaling placebo or active vaporized bulk cannabis. Values obtained by assessing general linear model (GLM) Select results of each measure at specific THC concentrations and BrAC. All estimates are for speed 55 miles/h (89 km/h), straight road. aRelative to median baseline (blood THC 0 μg/L, BrAC 0 g/210L) value

145

Figure 4. GLM Select modeled standard deviation of lateral position (SDLP) versus blood ∆9-tetrahydrocannabinol (THC) concentration (lower x-axis) and versus breath alcohol concentration (BrAC, upper x-axis). Note x-axis scales are different so slopes cannot be directly compared; dotted lines indicate THC concentrations producing equivalent SDLP to 0.02, 0.05, and 0.08 g/210L BrAC.

3.3 Pre- and Post-drive Blood and OF THC Concentrations

Pre- and post-drive blood and OF concentrations are presented in Table 14. Full blood and OF pharmacokinetic data are presented in (260) and (256), respectively.

Between-subject blood concentration variability (coefficient of variation) was substantially lower than matched OF concentration variability at all time points: 45-65%

146

vs. 125-207%, respectively, immediately post-dose, 39-69% vs. 129-184% at 1.4 h, and

61-82% vs. 139-174% at 2.3 h (Table 14).

Table 14. Blood and oral fluid THC and variability prior to and after driving (N=19) after controlled vaporized active (2.9% THC and 6.7% THC) cannabis with or without alcohol. Time Blood THC (μg/L) OF THC (μg/L) post-dose No Alcohol Alcohol No Alcohol Alcohol (h) 2.9% 6.7% 2.9% 6.7% 2.9% 6.7% 2.9% 6.7% Median 0 0 0 0 0.5 0 0 0.6 [range] [0-6.2] [0-5.4] [0-4.9] [0-6.3] [0-30.7] [0-11.7] [0-72.9] [0-34.2] -0.8 Mean 0.5 0.4 0.5 0.6 4.6 2.6 6.3 4.7 (baseline) (SD) (1.5) (1.3) (1.2) (1.5) (8.7) (4.0) (17.0) (8.9) %CV 284% 332% 245% 282% 191% 157% 272% 189% 32.7 42.2 35.3 67.5 848 764 735 952 Median [11.4- [15.2- [13.0- [18.1- [32.1- [25.1- [72.9- [22.7- 0.17 [range] 66.2] 137] 71.4] 210] 18,230] 23,680] 7,494] 66,200] (pre- Mean 35.9 56.2 40.5 75.0 2,101 3,220 1,599 7,652 drive 1) (SD) (16.7) (36.4) (18.2) (48.1) (4,142) (5,645) (2,005) (15,842) %CV 46% 65% 45% 64% 197% 175% 125% 207% 10.0 13.2 10.6 16.2 Median [1.6- [2.4- [5.5- [5.3------[range] 0.42 17.9] 40.8] 17.4] 43.9] (pre- Mean 10.0 16.8 10.4 19.0 ------drive 2) (SD) (4.5) (10.9) (3.4) (11.9) %CV 45% 65% 33% 63% ------6.2 52.5 91.0 69.5 138 Median 3.7 4.6 3.6 [1.3- [3.0- [9.3- [7.0- [5.2- 1.4 [range] [0-10.7] [0-14.7] [1.4-6.3] 18.4] 662] 1,028] 1,822] 3,940] (post- Mean 3.9 5.7 3.6 6.8 91.3 213 228 637 drive 1) (SD) (2.3) (3.9) (1.4) (4.6) (145) (275) (418) (1,097) %CV 59% 69% 39% 68% 159% 129% 184% 172% 33.1 46.9 35.4 91.0 Median 1.9 2.6 1.8 3.2 [1.8- [1.9- [8.7- [1.6- 2.3 [range] [0-8.5] [0-9.6] [0-4.9] [0-9.5] 374] 542] 473] 1,541] (post- Mean 2.2 3.2 1.8 3.2 47.7 92.1 86.4 263 drive 2) (SD) (1.8) (2.6) (1.1) (2.5) (81.1) (128) (124) (458) %CV 82% 82% 61% 77% 170% 139% 144% 174% Abbreviations: THC, ∆9-tetrahydrocannabinol; OF, oral fluid; SD, standard deviation; CV, coefficient of variation

4. Discussion

Using a sophisticated driving simulator and rigorous placebo-controlled, within- subject design, we found a positive association between blood THC concentration and one (SDLP) of 3 sensitive driving impairment measures; alcohol concentration was

147

associated with impairment on all 3 (including normalized lane departures and maximum acceleration). Impairing effects of cannabis-alcohol combinations were additive, not synergistic.

Decreased lateral control was associated with blood THC concentrations and

BrAC (Table 13), based on the descriptive models generated. SDLP is among the most sensitive and consistently utilized driving impairment measures (37, 249, 261-262).

SDLPs were similar between 0.02, 0.05, 0.08 g/210L BrAC and 3.3, 8.2, 13.1 μg/L THC, respectively (Figure 4). Low (1 and 2 μg/L) blood THC concentrations were associated with SDLP increases similar to 0.01 g/210L BrAC. At 5 μg/L THC, a 4.1% increase

(relative to no THC, no alcohol) in SDLP was observed; at 10 μg/L, SDLP increased

8.2%. This change was comparable to 0.05 g/210L BrAC (6.7% increase) and 0.08 g/210L BrAC (11% increase). Given that most countries have 0.05 or 0.08% BAC per se laws, this degree of SDLP increase may be substantial enough to be considered impairment. This 10 μg/L THC increase also was similar to 2 μg/L THC + 0.05 g/210L

BrAC (8.4% increase). At higher 20 μg/L THC, SDLP increased 16%, comparable to

0.10 g/210L BrAC (13% increase). In an on-road study (34, 47), 100, 200 and 300 μg/kg

THC doses (~7 mg, ~14 mg, ~21 mg) significantly increased SDLP 1.7-3.5 cm relative to placebo. These increases are consistent with our 7-10 μg/L during-drive THC (5.8-8.2% increase) or 0.05-0.08 g/210L BrAC (6.7-10.7% increase, Table 13). Our final lane departures/min and maximum lateral acceleration GLM Select models did not include

THC, indicating increasing THC concentrations did not increase these measures. BrAC increased lane departures/min and maximum lateral acceleration dose-dependently, with

0.05 g/210L corresponding to 35% and 9.5% increases, respectively.

148

Combining cannabis with alcohol produced an additive—rather than synergistic— effect on SDLP. No interaction terms were part of the GLM Select models. Past simulator studies showed inconsistent findings regarding a possible cannabis-alcohol interaction effect on SDLP. Ronen et al (44) observed significant increases in lane position variability (relative to placebo, THC-only, and alcohol-only) when 13 mg THC and

0.05% (BAC) alcohol were combined, despite neither producing an independent significant effect. Conversely, Lenné et al (25) observed significant main effects of cannabis and alcohol independently, but no interaction. Our results more closely resemble the latter finding. Combining 100 or 200 μg/kg THC with 0.04% target BAC in the on-road study described above significantly increased SDLP by 5.3 and 8.5 cm, classified as “severe” performance decrements (34, 47). In our model, this increase is similar to ≥20 μg/L blood THC alone.

Unlike cannabis, alcohol affected additional lateral control parameters besides

SDLP. Lane departures/min and maximum lateral acceleration also increased with BrAC.

This suggests more extreme reaction to lateral position when alcohol driving is compared to DUIC. Cannabis-influenced drivers may attempt to drive more cautiously to compensate for potential impairing effects, whereas alcohol-drivers often underestimate their impairment and take more risks (49). Alcohol’s strong effects on driving are well- established (261, 263-265). A recent report indicated alcohol increased center and edge lane crossings, as well as time over the edge line, in a simulated drive (261). In one on- road study, only cannabis-alcohol combinations significantly increased time out of lane

(34, 47); neither cannabis nor alcohol (0.04% BAC) alone produced a significant effect.

Since increasing lane departures and “time out of lane” requires more substantial lane

149

weaving than SDLP, this discrepancy may result from the low alcohol dose administered in that study. SDLP is more sensitive, with observable impairment at BACs as low as

0.04% (263).

Neither cannabis nor alcohol affected standard deviation of steering angle. To our knowledge, only one prior simulator study found a significant alcohol effect on this parameter (25). Lenné et al administered 0.4 and 0.6 g/kg alcohol to attain approximate peak BACs ≤0.025% and ~0.05%; only the high dose produced a significant but small increase in standard deviation of steering angle. Although cannabis alone (19 and 38 mg doses) did not significantly increase steering angle variability (main effect), there was significant interaction with driver experience. Experienced drivers (≥7 years driving) showed unchanged or decreased steering angle variability with increasing cannabis dose relative to placebo, whereas inexperienced drivers (<2 years) had increased variability

(25). All of our participants had ≥6 years of driving experience, perhaps accounting for this discrepancy. The prior simulator study analyzed effects by administered dose rather than by concentration (25), possibly resulting in greater apparent effect size because dose-wise (categorical) variables analyses generally have higher power than continuous variables. Multiple other studies found no cannabis-only effect on steering wheel position variability (44, 172), although one observed increased steering variability in occasional smokers after alcohol alone and alcohol-cannabis combination (44). Overall, standard deviation of steering angle appears an insensitive measure of these effects, due to the amplifying effect of steering mechanisms. Minor steering adjustments can substantially alter course and change lane position due to forward motion, despite re-straightening the wheel.

150

Because smokers self-titrate cannabis doses to individual comfort levels (266), concentration-based analysis was preferable to analysis by dose. In our sample, approximately half of participants showed evidence of self-titration in blood concentration (260) Even so, substantial concentration variability was observed, consistent with prior cannabis research (93). This further underscores the robustness of concentration-based—rather than dose-based—driving analysis.

There is substantial interest in relating driving performance directly to OF concentrations due to advantages as a screening tool. When positive, OF THC identifies recent cannabis intake. THC enters OF primarily by oral mucosa contamination during inhalation, and consequently is less representative of systemic concentrations shortly after cannabis intake. OF concentration variability was 2-5-fold higher than blood concentration for the same participants and times, so it is more challenging to interpret effects from OF concentrations. Similar to blood, low OF THC concentrations are difficult to interpret because intake history and individual variability affect detection time and later concentrations. However, in this sample, OF THC >1600 μg/L indicated intake within the last 1.4 h, and >600 μg/L indicated intake within the last 2.3 h. In a roadside study, the percentage of people displaying observable cannabis-impairment signs increased with increasing OF concentrations when aggregated into wide ranges (≤3.00

μg/L, 3.01-25.00 μg/L, 25.01-100 μg/L, >100 μg/L) (267). The 2-5-fold higher OF concentration variability relative to paired blood suggests OF concentrations are less suitable for forensic inference of impairment than blood concentrations.

151

4.1 Strengths and limitations

Major strengths of this study include the double-blind, placebo-controlled, within- subject design; drive scenarios controlling for other road conditions (speed limit and curvature), which can potentially affect drivers’ lateral control and road tracking performance; administration of multiple doses of cannabis (THC) and alcohol; concentration (rather than dose)-based analysis; and specimen collection during actual driving, which allows for modeling pharmacokinetics during driving, to better relate driving impairment to THC concentrations.

In authentic DUIC cases, measured THC concentrations do not reflect those present during driving. Blood collection is typically delayed 90 min to 4 h after the event

(141-142). During this delay period, there is rapid THC distribution from blood into highly-perfused and adipose tissues, resulting in rapid blood THC concentration decrease in the first hour post-inhalation. After this initial phase, the THC concentration continues to decrease, albeit more slowly. The result is that measured THC concentrations are lower than they were during driving. In contrast, in this controlled-administration study, we examined driving performance relative to THC concentrations and BrAC that were present during the drive. Thus, to our knowledge, the current study is among the most robust analyses of cannabis and alcohol effects on lateral control at specific THC concentrations. For reference, we report driving performance results at concentrations typically considered or established for per se laws around the world (1, 2, 5, 7 μg/L THC;

0.02, 0.05, 0.08% BrAC) (40, 239, 251, 268-270). However, as mentioned above, these per se limits are applied to THC concentrations that may be substantially lower than those present during actual driving. Thus, our reported THC 1-5 μg/L SDLP changes may

152

be understated compared to forensic DUIC cases. In the present study, median blood and

OF THC concentrations immediately post-dose were >30 μg/L and >700 μg/L, respectively. Blood THC ≥20 μg/L indicated intake within the last 0.42 h and THC ≥10

μg/L indicated intake within the last 1.4 h. Thus, if people drive during or soon after cannabis inhalation, during-drive THC concentrations could substantially exceed 5, 10, or even 20 μg/L. Our SDLP increase associated with THC ≥20μg/L (~5.2 cm) was considered “severe” by other researchers (34, 47), representing a 16% increase in our observed lane position variability.

This study has several potential limitations. We approached data analyses via a stepwise GLM Select procedure, with the goal of describing data in an unbiased manner—without assumptions of which parameters (THC, BrAC, other) would produce fixed effects. In research settings, participants are aware their driving is constantly under observation, and consequently may drive with greater caution or focus than outside the research setting. Also, selection bias may have yielded study participants with a desire to demonstrate that cannabis does not affect driving, possibly making our study sample less representative of the general driving public. Many individuals believe cannabis does not substantially impair driving, and attitudes toward DUIC are less negative than DUI alcohol (184, 271). However, self-perceptions of driving performance or impairment— even without drugs—may be unreliable (262, 264). It is unclear whether participant #12’s behavior was fundamentally different to normal, or merely represents a different subpopulation of drivers. If the latter is the case, there was not a high enough N to detect other cases. For the sake of most accurately describing the data from the population recruited, it was necessary to remove those data. Finally, this study was limited to

153

occasional smokers (41), who do not represent the entire population of cannabis smokers.

Although frequent smokers demonstrate tolerance to some acute cannabis intoxication effects (169), tolerance may not compensate for all effects. There is currently substantial interest in comparing occasional to frequent smokers and assessing potential tolerance effects (41, 272-273), especially as medical and recreational cannabis becomes more commonplace.

We do not believe that conducting this study in a driving simulator, rather than on the road, represents a significant limitation. Simulators offer advantages for assessing impaired driving. Participants can engage in risky driving behavior without endangering themselves or others. Additionally, simulators provide controlled environments for research and the ability to make highly detailed real-time measurements. Simulator technology improved greatly in the last few decades, producing highly realistic driving scenarios (274). The NADS-1 is the world’s most sophisticated simulator, and was successfully utilized to assess distracted and drugged driving (202, 219).

5. Conclusion

In this robust double-blind placebo controlled study, cannabis and alcohol were significantly associated with impaired driving lateral control. Cannabis only affected

SDLP; whereas alcohol affected SDLP, lane departures/min, and maximum acceleration.

During-drive 7-10 μg/L blood THC was associated with SDLP increases similar to 0.05 g/210L BrAC (~0.05% BAC), and SDLP at 20 μg/L THC was higher than at 0.10 g/210L

BrAC. Combining alcohol and cannabis produced an additive effect on SDLP; 5 μg/L

THC with 0.05 g/210L BrAC was associated with SDLP impairment similar to 0.08

154

g/210L. Because blood collection in DUIC cases is delayed relative to the time of apparently impaired driving, our reported SDLP increases associated with 10 and 20 μg/L

THC (during driving) may be better indices of cannabis-related driving impairment than lower concentrations. OF concentration variability was substantially greater than blood concentration variability, suggesting it performs better as a screening tool than as an impairment gauge.

Table 15 (Supplemental). General Linear Model (GLM) Select results of natural log (ln)- transformed standard deviation of lateral position (SDLP) in 18 volunteer drivers after controlled vaporized cannabis with or without oral alcohol. ln(SDLP) ln Estimate Standard Parameter DF t p-value (b) Error THC 1 0.008 3.79 0.002 0.0002 BrAC 1 0.014 3.30 0.004 0.0010 THC*BrAC Speed Limit 1 0.013 17 0.001 <0.0001 Inverse 1 15 10 1.4 <0.0001 Curvature Intercept 1 3.0 48 0.062 <0.0001 Subject 17 Model df: 21 Model F-value 26.02 Error df: 1916 Driving occurred 0.5 h after drinking placebo or active alcohol (calculated to produce approximate peak 0.065% BrAC) and inhaling placebo, 2.9% THC, or 6.7% THC vaporized bulk cannabis (500 mg, Volcano® Medic vaporizer). Estimate represents parameter (coefficient) estimate for each factor (negative b indicates the parameter decreases the effect; positive b indicates the parameter increases the overall effect). The t-values estimate effect size. Boldface indicates parameter included in the final GLM Select model of ln(SDLP). The p-value for the final overall analysis of variance model was <0.0001 Abbreviations: DF, degrees of freedom; THC, blood ∆9-tetrahydrocannabinol concentration; BrAC, breath alcohol concentration

155

Table 16 (Supplemental). Standard deviation of lateral position (SDLP) associated with specific blood ∆9-tetrahydrocannabinol (THC) concentrations and breath alcohol concentrations (BrAC) during driving based on transformed ln(SDLP) GLM Select model. During-Drive Standard Deviation of Lateral Position (SDLP) Concentration Median [range] Percent THC BrAC Differencea Predicted SDLP Increasea (%) (μg/L) (g/210L) (cm) (cm) 0 0 26.5 [20.6-40.5] -- -- 1 0 26.8 [20.8-40.8] 0.2 0.8 2 0 27.0 [21.0-41.2] 0.4 1.6 5 0 27.6 [21.5-42.2] 1.1 4.2 7 0 28.1 [21.8-42.9] 1.6 5.9 10 0 28.8 [22.4-43.9] 2.3 8.5 20 0 31.2 [24.3-47.7] 4.7 18 0 0.01 26.9 [20.9-41.1] 0.4 1.4 0 0.02 27.3 [21.2-41.7] 0.8 2.9 0 0.05 28.5 [22.1-43.5] 2.0 7.4 0 0.08 29.7 [23.1-45.4] 3.2 12 0 0.10 30.6 [23.8-46.7] 4.1 15 2 0.05 29.0 [22.5-44.2] 2.4 9.1 5 0.05 29.7 [23.1-45.3] 3.1 12 Data generated from 18 healthy occasional cannabis smokers 0.5-1.3 h after ingesting placebo or active oral alcohol and inhaling placebo or active vaporized bulk cannabis. Values obtained by assessing GLM Select results at specific THC concentrations and BrAC, speed limit 55 miles/h (89 km/h), straight road. aRelative to median.

156

Figure 5 (Supplemental). Median and individual subject model-predicted standard deviation of lateral position (SDLP) for various blood ∆9-tetrahydrocannabinol (THC) concentrations (A) and breath alcohol concentrations (BrAC) (B). Data generated from 18 healthy occasional cannabis smokers 0.5-1.3 h after ingesting placebo or active oral alcohol and inhaling placebo or active vaporized bulk cannabis. Values obtained by assessing GLM Select results at specific THC concentrations and BrAC, speed limit 55 miles/h (89 km/h), straight road.

157

Chapter 5 – Controlled Vaporized Cannabis, With and Without Alcohol: Subjective Effects and Oral Fluid-Blood Cannabinoid Relationships

(As submitted to Drug Testing and Analysis, May 2015)

Abstract

Aims: This study aimed to evaluate cannabis’ subjective effects, with and without alcohol, relative to blood and oral fluid (OF, advantageous for cannabis exposure screening) cannabinoid concentrations and OF/blood and OF/plasma vaporized- cannabinoid relationships.

Design: Participants received vaporized placebo or active cannabis (2.9% and 6.7% ∆9- tetrahydrocannabinol, THC) with or without oral low-dose alcohol (~0.065 g/210L peak breath alcohol concentration [BrAC]) (within-subject).

Setting: National Advanced Driving Simulator, University of Iowa.

Participants: Healthy adult occasional-to-moderate cannabis smokers.

Measurements: Blood and OF were collected up to 8.3h post-dose and subjective effects measured at matched time points with visual-analogue scales and 5-point Likert scales.

Subjective “high” and other effects were evaluated by THC concentration, BrAC and interactions (linear mixed models), and by time point (analysis of variance). OF versus blood or plasma cannabinoid ratios and correlations were evaluated in paired-positive specimens.

Findings: Nineteen participants (13 men) completed the study. Blood THC concentration or BrAC significantly associated with subjective effects including “high,” while OF

158

contamination prevented significant OF concentration associations <1.4 h post-dose.

Subjective effects persisted through 3.3-4.3 h, with alcohol potentiating cannabis effects’ duration. Effect-versus-THC concentration and effect-versus-alcohol concentration hystereses were counterclockwise and clockwise, respectively. OF/blood and OF/plasma

THC significantly correlated (all Spearman r ≥0.71), but variability was high.

Conclusions: Vaporized cannabis subjective effects were similar to those previously reported after smoking, with duration extended by concurrent alcohol. Cannabis intake was identified by OF testing, but OF concentration variability limited interpretation.

Blood THC concentrations were more consistent across subjects and more accurate at predicting cannabis’ subjective effects.

Introduction

Twenty-three US states and the District of Columbia legalized medical marijuana

(250), with Colorado, Washington, Oregon, and Alaska decriminalizing recreational cannabis. Smoking, the most common administration route (63), is disadvantageous as pharmacotherapy, delivering hazardous pyrolytic byproducts (64). Volatilizing cannabinoids at sub-combustion temperatures (vaporizing) should provide similar subjective effects (78-79, 83), with decreased pyrolytic byproducts (69, 71) leading to decreased reports of respiratory symptoms (70). As cannabis vaporization prevalence increases, it is important for clinical and forensic purposes to fully evaluate subjective effects, blood and oral fluid (OF) concentrations, and their relationships.

Cannabis is the most common illicit drug identified in driving under the influence

(DUI) cases (274). States with legalized medical or recreational cannabis had increased

DUI-cannabis (DUIC) cases (275-276), with enforcement complicated by changing

159

cannabis laws. Blood ∆9-tetrahydrocannabinol (THC) and its non-psychoactive metabolite (11-nor-9-carboxy-THC, THCCOOH) concentrations may provide information regarding time since last intake and cannabis consumption frequency (277-

278). Oral fluid (OF), a valuable alternative sampling matrix, is non-invasively collected, more difficult to adulterate than urine, and provides information about recent intake (74,

109-110). Some jurisdictions already adopted OF-specific legislation for DUIC (112-

114). However, OF correlation with cannabis effects or blood concentrations is not fully understood. Cannabis and alcohol are often identified together in DUI cases (274), making understanding their combined effects critical.

In this vaporized cannabis and oral alcohol controlled administration study, we evaluated subjective effects and OF and blood/plasma cannabinoid concentration relationships, with and without low-dose alcohol.

Methods

This protocol was approved by the University of Iowa Institutional Review Board.

The study was performed at the University of Iowa Hospitals and Clinics Clinical

Research Unit (UIHC-CRU) and National Advanced Driving Simulator (NADS).

Participants

Participants were recruited from the NADS subject database and provided written informed consent for the study. Inclusion criteria were ages 21-55 years; self-reported average cannabis consumption ≥1x/3 months but ≤3 days/week over the past 3 months

(Cannabis Use Disorders Identification Test [CUDIT] (252)); self-reported “light” or

160

“moderate” alcohol consumption according to a Quantity-Frequency-Variability (QFV) scale (253); or if “heavy”, not more than 3-4 servings in a typical drinking occasion.

Exclusion criteria included past or current clinically significant medical illness; history of clinically significant adverse event associated with cannabis or alcohol intoxication; ≥450 mL blood donation in 2 weeks preceding drug administration; pregnant or nursing; interest in drug abuse treatment within past 60days; and currently taking drugs contraindicated with cannabis or alcohol or known to impact driving.

Study Design

Participants entered the clinical research unit 10-16 h before dosing to preclude intoxication. Participants drank 90% grain alcohol (to ~0.065% peak breath alcohol concentration [BrAC] (263)) mixed with juice or placebo-alcohol (juice with alcohol- swabbed rim, topped with 1 mL alcohol to mimic taste and odor) ad libitum over 10 min; then inhaled 500 mg placebo (0.008±0.002% THC), low (2.9±0.14% THC, ~14.5 mg)-, or high (6.7±0.05% THC, ~33.5 mg)-dose vaporized ground bulk cannabis (210°C,

Volcano® Medic, Storz & Bickel, Tuttlingen, Germany) ad libitum over 10 min.

Cannabis was obtained from NIDA Chemistry and Physiological Systems Research

Branch. Participants received all six alcohol/cannabis combinations in randomized order, one combination per session, separated by ≥1 week.

OF was collected with QuantisalTM collection devices (Immunalysis, Pomona,

CA) -0.8, 0.17, 1.4, 2.3, 3.3, 4.3, 5.3, 6.3, 7.3, and 8.3 h after start of cannabis dosing

(256). Devices were placed under the tongue until indicators turned blue (collecting

1.0±0.1mL OF) or for 10 min maximum, and placed into the stabilizing buffer. OF was

161

stored in Nunc® cryotubes (Thomas Scientific, Swedesboro, NJ) at 4°C for analysis within a month (279). Oral intake was prohibited 10 min prior to OF collection. Blood was collected via indwelling peripheral venous catheter into grey-top potassium oxalate/sodium fluoride Vacutainer® tubes (BD, Franklin Lakes, NY) concurrently with

OF (except 4.3 and 5.3 h due to blood volume limits), with a second sample centrifuged at 1600×g, 15 min. Blood and plasma were stored at -20°C in 3.6 mL Nunc cryotubes, and analyzed within 3 months (254).

Subjective effects were measured at the same times as OF collection by 100 mm visual-analogue scales (VAS; “high”, “good drug effect”, “stimulated”, “stoned”,

“anxious”, “sedated”, and “restless”) anchored by “Not At All”-“Most Ever”; and 5-point

(“none”, “slight”, “mild”, “moderate”, “severe”) Likert scales (“difficulty concentrating”,

“altered sense of time”, “slowed or slurred speech”, “body feels sluggish/heavy”, “feel hungry”, “feel thirsty”, “shakiness/tremulousness”, “nausea”, “headache”, “palpitations”,

“upset stomach”, “dizzy”, and “dry mouth or throat”).

Specimen Analysis

OF was quantified for THC, THCCOOH, cannabidiol (CBD), and cannabinol

(CBN) by two-dimensional gas chromatography-mass spectrometry (257), modified by adding 0.4 mL hexane to solid-phase extraction columns before loading the initial elution solvent. THC, THCCOOH, CBD and CBN linear ranges were 0.5-50 µg/L, 15-500 ng/L,

1-50 µg/L and 1-50 µg/L, respectively. Inter- and intra-assay imprecision were ≤12.3%; analytical bias, ≤14.4% (n=21). For concentrations >upper limit of quantification (LOQ),

OF was diluted with drug-free Quantisal buffer. Blood and plasma cannabinoids were

162

quantified by liquid chromatography-tandem mass spectrometry (LCMSMS) (255).

Briefly, 0.5 mL blood or plasma was protein-precipitated with ice-cold acetonitrile, supernatants diluted and solid-phase extracted with Bond-Elut Plexa cartridges (Agilent

Technologies, Santa Clara, CA). THC, THCCOOH, CBD, and CBN linear ranges were

1-100 μg/L. Inter-assay (n=30) analytical bias and imprecision were ≤9.3% and ≤10.0%.

Data Analysis

VAS and Likert results were assessed via linear mixed models in SPSS® version

19 for Windows (IBM, Armonk, NY). Initial data review and analyses indicated insufficiently different low-versus-high cannabis-dose THC concentrations; consequently, mixed-model analyses utilized blood THC and BrAC concentrations

(continuous variables), producing the best-fit models. THC, BrAC, time, THC*BrAC, time*THC, time*BrAC, and time*THC*BrAC were evaluated as fixed effects; subject*THC and intercepts as random effects (heterogeneous (1) autoregressive). Two- tailed p<0.05 indicated significance. The same analyses were conducted with OF THC concentrations, including and excluding t=0.17h. For analytical purposes, concentrations

Likert responses to 5-point numerical scales (0≡“None”-4≡“Severe”). Likert linear mixed models for “feel hungry” and “feel thirsty” were only evaluated through 3.3h due to lunch. Friedman’s [factorial] repeated measures analysis of variance (ANOVA, factors: cannabis, alcohol; cannabis*alcohol interaction term, pairwise post-hoc comparisons) evaluated within-subject dose differences by time point. The Greenhouse-Geisser correction evaluated sphericity violations (Mauchly’s test). For time point analyses, the

163

conservative Bonferroni correction was utilized for multiple comparisons (p<0.005 significance level), and Bonferroni post-hoc testing for subjective effects differences from baseline by dosing condition at each time point. OF versus blood and plasma correlations and regression comparisons were performed with GraphPad Prism®6 (La

Jolla, CA). OF/blood and OF/plasma cannabinoid ratios were calculated when quantifiable data were available for both. Dose and baseline differences were calculated via ANOVA.

Results

Participants

Nineteen cannabis smokers (13 men, ages 21-37 years, 74% white) reported cannabis consumption ≥2x/month (but ≤3 days/week), and last use within a week prior to admission (Table 17). One participant (#13) self-reported last intake 4 months ago, despite reporting overall average consumption ≥1x/3 months.

Subjective effects

Table 18 presents linear mixed models subjective effects by THC and alcohol concentrations. In these models, b is the coefficient estimate for each contributing factor

(negative or positive b indicates parameter decreases or increases effect, respectively).

Blood THC was positively associated with “high”, “good drug effect”, “stimulated”,

“stoned”, “anxious”, and “restless”, and feelings of altered time, “slowed/slurred speech”,

164

Table 17. Self-reported demographic characteristics and recent cannabis and alcohol consumption history of 19 healthy adult occasional-to-moderate cannabis smokers. Hours Time Typical Amount last Race Alcohol Cannabis “stoned” since last Partici- Age BMI drinks consumedb Sex and intake intake on typical cannabis pant (years) (kg/m2) per (joint or joint ethnicity frequency frequency cannabis consumed occasion equivalent) occasiona (days) 1 M 23.7 W 24.3 2-3x/wk 2-4 2-4x/m 1-2 1 1 2 F 28.4 AA 23.8 ≥4x/wk 2-4 2-4x/m 3-4 14 1 3 M 21.9 W 24.7 2-3x/wk 5-6 2-4x/m 1-2 6 1 4 M 37.8 W 26.1 2-3x/wk 2-4 2-3x/wk 1-2 3 2.5 5 M 26.6 W 21.6 ≤1x/m 2-4 ≤1x/m 1-2 11 3.5 6 F 26.3 W 20.0 2-3x/wk 2-4 2-3x/wk 3-4 1 0.25 7 M 25.8 W 40.6 2-4x/m 2-4 2-3x/wk 1-2 0.3 0.5 8 M 26.1 H 31.5 2-4x/m 1-2 2-3x/wk 1-2 3 1

165 9 M 23.2 W 19.5 2-3x/wk 2-4 2-3x/wk 3-4 2 1 10 M 23.1 W 23.9 2-4x/m 2-4 ≤1x/m 1-2 2 0.25

11 M 32.3 O, H 28.9 2-3x/wk 2-4 2-3x/wk 1-2 4 1 12 F 23.4 W 23.3 2-3x/wk 2-4 2-4x/m 3-4 4 1 13 F 30.3 AA 24.1 2-3x/wk 2-4 ≤1x/m <1 120 1 14 M 24.6 W 23.3 2-3x/wk 2-4 2-4x/m 1-2 7 0.8 15 M 21.8 W 32.7 ≤1x/m 1-2 2-4x/m 1-2 7 0.13 16 F 21.7 AA, W 23.0 2-4x/m 1-2 2-3x/wk 1-2 1.1 1.5 17 M 28.7 W 18.3 2-3x/wk 2-4 ≤1x/m 3-4 45 0.5 18 M 28.1 W 48.3 2-4x/m 2-4 2-4x/m 3-4 5 1 19 F 22.9 W 21.6 2-4x/m 5-6 2-3x/wk 3-4 1 1 Median 25.8 23.9 4.0 1.0 Mean 26.1 26.3 12.5 1.0 StDev 4.1 7.5 27.9 0.8 a‘Hours “stoned” ’ wording originates from Cannabis Use Disorders Identification Test, source of self-reported cannabis frequency data bCannabis amount last consumed is based on empirically-normalized joint consumption, to account for various administration routes and self-reported “sharing” between multiple individuals Abbreviations: W, White; AA, African American; H, Hispanic or Latino; As, Asian; O, Other; AI, American Indian/Native American; StDev, standard deviation

166

Table 18. (Continued from previous page) 95% a Parameter b SEb df t p Confidence Interval of b Time*THC*BrAC 22.343 5.665 633.4 3.9 <0.001 11.220 33.467 Subject Variance in 131.518 46.704 0.005 65.570 263.79 Intercepts (THC) Subject variance in Slopes 0.093 0.043 0.029 0.038 0.229 (THC) ARH1 rho (slope- 0.588 0.190 0.002 0.105 0.847 intercept covariance) VAS Restless Intercept 11.466 2.266 36.6 5.1 <0.001 6.873 16.059 Blood THC 0.156 0.064 25.4 2.4 0.022 0.024 0.288 BrAC 635.9 0.1 0.903 Time 643.2 -0.2 0.860 THC*BrAC 648.3 -1.5 0.136 Time*THC 593.6 0.8 0.436 Time*BrAC 638.2 -0.1 0.952 Time*THC*BrAC 623.8 0.8 0.439 Subject Variance in 63.440 22.696 0.005 31.467 127.90 Intercepts (THC) Subject variance in Slopes 0.146 (THC) ARH1 rho (slope- 0.549 intercept covariance) VAS Sedated Intercept 17.942 2.893 28.9 6.2 <0.001 12.023 23.860 Blood THC 15.9 0.2 0.879 BrAC 632.9 0.0 0.984 Time -1.444 0.253 639.0 -5.7 <0.001 -1.942 -0.947 THC*BrAC 647.7 -0.4 0.701 Time*THC 593.0 1.6 0.119 Time*BrAC 634.8 1.3 0.186 Time*THC*BrAC 627.4 0.1 0.941 Subject Variance in 120.808 41.818 0.004 61.298 238.09 Intercepts (THC) Subject variance in Slopes 0.149 0.074 0.043 0.057 0.392 (THC) ARH1 rho (slope- -0.749 0.118 <0.001 -0.905 -0.417 intercept covariance) VAS Stimulated Intercept 21.682 2.838 27.1 7.6 <0.001 15.860 27.503 Blood THC 0.297 0.080 19.2 3.7 0.001 0.130 0.464 BrAC 168.759 43.163 633.6 3.9 <0.001 84.000 253.51 Time -2.827 0.232 645.0 -12.2 <0.001 -3.284 -2.371

167

Table 18. (Continued from previous page) 95% a Parameter b SEb df t p Confidence Interval of b THC*BrAC -3.620 1.624 652.8 -2.2 0.026 -6.809 -0.431 Time*THC 0.568 0.097 611.4 5.8 <0.001 0.377 0.759 Time*BrAC -60.558 23.436 637.7 -2.6 0.010 -106.58 -14.54 Time*THC*BrAC 628.8 1.5 0.133 Subject Variance in 120.357 41.508 0.004 61.223 236.61 Intercepts (THC) Subject variance in Slopes 0.069 0.037 0.060 0.024 0.196 (THC) ARH1 rho (slope- 0.469 0.214 0.028 -0.028 0.780 intercept covariance) VAS Stoned Intercept 19.446 2.790 28.8 7.0 <0.001 13.737 25.154 Blood THC 0.398 0.109 18.1 3.7 0.002 0.170 0.627 BrAC 630.2 1.2 0.236 Time -2.875 0.254 634.7 -11.3 <0.001 -3.374 -2.375 THC*BrAC -5.502 1.780 640.5 -3.1 0.002 -8.997 -2.007 Time*THC 0.687 0.107 634.4 6.4 <0.001 0.477 0.896 Time*BrAC 632.1 -0.7 0.489 Time*THC*BrAC 19.712 5.496 625.9 3.6 <0.001 8.919 30.505 Subject Variance in 109.037 38.954 0.005 54.135 219.620 Intercepts (THC) Subject variance in Slopes 0.159 0.073 0.030 0.065 0.392 (THC) ARH1 rho (slope- 0.631 intercept covariance) Likert Difficulty Concentrating Intercept 0.554 0.111 27.3 5.0 <0.001 0.327 0.781 Blood THC 21.8 1.9 0.077 BrAC 5.151 1.672 645.3 3.1 0.002 1.868 8.434 Time -0.062 0.009 652.2 -6.9 <0.001 -0.080 -0.044 THC*BrAC 657.5 0.4 0.658 Time*THC 0.012 0.004 441.4 3.2 0.001 0.005 0.019 Time*BrAC 648.6 -0.9 0.390 Time*THC*BrAC 638.0 0.5 0.611 Subject Variance in 0.183 0.063 0.004 0.093 0.358 Intercepts (THC) Subject variance in Slopes 0.000 0.000 0.019 0.000 0.000 (THC) ARH1 rho (slope- 0.787 0.130 <0.001 0.373 0.940 intercept covariance)

168

Table 18. (Continued from previous page) 95% a Parameter b SEb df t p Confidence Interval of b Likert Altered Sense of Time Intercept 0.406 0.106 29.6 3.8 0.001 0.190 0.623 Blood THC 0.010 0.004 18.5 2.6 0.018 0.002 0.018 BrAC 652.8 1.2 0.227 Time -0.059 0.009 656.0 -6.3 <0.001 -0.078 -0.041 THC*BrAC 603.0 -0.2 0.806 Time*THC 0.013 0.003 83.3 3.7 <0.001 0.006 0.020 Time*BrAC 654.0 0.9 0.386 Time*THC*BrAC 645.5 0.1 0.917 Subject Variance in 0.159 0.055 0.004 0.080 0.314 Intercepts (THC) Subject variance in Slopes 0.000 0.000 0.021 0.000 0.000 (THC) ARH1 rho (slope- 0.987 0.047 <0.001 -0.807 1.000 intercept covariance) Likert Slowed/Slurred Speech Intercept 0.163 0.071 32.4 2.3 0.030 0.017 0.308 Blood THC 0.008 0.003 20.2 2.7 0.015 0.002 0.014 BrAC 3.774 1.280 647.3 2.9 0.003 1.261 6.287 Time -0.028 0.007 652.8 -4.1 <0.001 -0.042 -0.015 THC*BrAC -0.095 0.048 650.5 -2.0 0.049 -0.189 -0.001 Time*THC 0.009 0.003 210.0 3.6 <0.001 0.004 0.015 Time*BrAC 649.9 0.1 0.903 Time*THC*BrAC 641.5 1.2 0.228 Subject Variance in 0.068 0.024 0.005 0.034 0.136 Intercepts (THC) Subject variance in Slopes 0.000 0.000 0.015 0.000 0.000 (THC) ARH1 rho (slope- 0.912 0.075 <0.001 0.585 0.984 intercept covariance) Likert Body Feels Sluggish/Heavy Intercept 0.600 0.101 34.5 6.0 <0.001 0.396 0.805 Blood THC 28.8 1.6 0.115 BrAC 4.568 1.936 645.0 2.4 0.019 0.767 8.369 Time -0.066 0.010 651.1 -6.3 <0.001 -0.087 -0.046 THC*BrAC 658.3 0.0 0.965 Time*THC 0.015 0.004 558.1 3.6 <0.001 0.007 0.024 Time*BrAC 647.6 0.4 0.691 Time*THC*BrAC 638.7 0.0 0.993 Subject Variance in 0.127 0.046 0.006 0.062 0.259 Intercepts (THC)

169

Table 18. (Continued from previous page) 95% a Parameter b SEb df t p Confidence Interval of b Subject variance in Slopes 0.000 0.000 0.038 0.000 0.000 (THC) ARH1 rho (slope- 0.322 intercept covariance) Likert Feel Thirsty Intercept 0.728 0.166 64.2 4.4 <0.001 0.396 1.059 Blood THC 45.3 -0.1 0.949 BrAC 419.1 -0.1 0.944 Time 432.8 0.9 0.377 THC*BrAC 433.4 0.6 0.524 Time*THC 0.083 0.014 429.9 6.1 <0.001 0.057 0.110 Time*BrAC 6.077 1.836 416.1 3.3 0.001 2.468 9.687 Time*THC*BrAC 412.8 -1.3 0.181 Subject Variance in 0.236 0.091 0.009 0.111 0.501 Intercepts (THC) Subject variance in Slopes 0.000 0.000 0.135 0.000 0.000 (THC) ARH1 rho (slope- 0.407 intercept covariance) Likert Dizzy Intercept 0.125 0.040 55.1 3.1 0.003 0.045 0.206 Blood THC 0.007 0.002 25.9 2.8 0.009 0.002 0.011 BrAC 646.1 1.5 0.141 Time -0.017 0.005 651.9 -3.2 0.001 -0.027 -0.006 THC*BrAC 656.1 -1.8 0.065 Time*THC 144.2 1.0 0.318 Time*BrAC 649.3 -0.4 0.717 Time*THC*BrAC 645.6 -0.1 0.899 Subject Variance in 0.014 0.006 0.012 0.007 0.032 Intercepts (THC) Subject variance in Slopes 0.000 0.000 0.006 0.000 0.000 (THC) ARH1 rho (slope- 0.829 0.147 <0.001 0.257 0.971 intercept covariance) Likert Dry Mouth or Throat Intercept 0.917 0.131 34.1 7.0 <0.001 0.651 1.183 Blood THC 0.008 0.003 20.5 2.3 0.034 0.001 0.015 BrAC 646.3 -0.8 0.414 Time -0.120 0.014 654.5 -8.7 <0.001 -0.147 -0.093 THC*BrAC 624.2 -1.9 0.057 Time*THC 0.033 0.006 399.8 5.8 <0.001 0.022 0.044 Time*BrAC 649.1 1.4 0.157

170

Table 18. (Continued from previous page) 95% a Parameter b SEb df t p Confidence Interval of b Time*THC*BrAC 1.308 0.301 626.2 4.3 <0.001 0.716 1.900 Subject Variance in 0.211 0.079 0.007 0.102 0.438 Intercepts (THC) Subject variance in Slopes 0.212 (THC) ARH1 rho (slope- 0.147 intercept covariance) Data from 19 healthy, adult cannabis smokers who participated in all dosing sessions. Subjective effects were measured by 100 mm VAS or 5-point Likert scales with choices 0≡“none”, 1≡“slight”, 2≡“mild”, 3≡“moderate”, 4≡“severe” after drinking placebo or active alcohol (calculated to produce approximate peak 0.065% BrAC) and inhaling placebo, 2.9% THC, or 6.7% THC vaporized cannabis (500 mg, Volcano® Medic vaporizer). Linear mixed model results; b is parameter (coefficient) estimate for each factor (negative b indicates the parameter decreases the subjective effect; positive b indicates the parameter increases the overall effect). Overall equation: [Subjective Effect Result] = Intercept +bBlood THC*[THC]blood + bBrAC*BrAC + bTime*post-dose time + bTHC*BrAC*[THC]blood*BrAC + bTime*THC*post-dose time*[THC]blood + bTime*BrAC*post-dose time*BrAC + bTime*THC*BrAC*post-dose time*[THC]blood*BrAC aValues in bold are statistically significant (p<0.05); only significant predictors are considered in the final model. Abbreviations: SE, standard error; df, degrees of freedom; VAS, 100 mm visual-analogue scale; Likert, 5-point Likert scale; THC, ∆9-tetrahydrocannabinol; BrAC, breath alcohol concentration.

“dizziness”, and “dry mouth/throat” (Figure 10 (Supplemental)). BrAC was positively associated with “high”, “good drug effect”, and “stimulated” and “difficulty concentrating”, “slowed/slurred speech”, and “body feels sluggish/heavy”. Most models contained negative time terms, indicating effects generally were highest immediately post-dose, decreasing over time. Significant negative THC*BrAC interactions were observed for “high”, “good drug effect”, “stoned”, “stimulated”, “anxious”, and

“slowed/slurred speech”, but the first three contained additional significant positive time*THC*BrAC interactions. Table 19 (Supplemental) provides model results where subject covariance parameters could not be calculated (thus resultant model is less certain). Models produced from OF THC were different than for blood (Table 20

171

(Supplemental) and Table 21 (Supplemental)). For multiple subjective effects, significant main effects for blood THC were not detected in OF when the time course included

<1.4h; but “high”, “good drug effect”, “anxious”, “stimulated”, “stoned”, “altered sense of time”, “feel thirsty”, and “dry mouth/throat” had significant main OF effects for models that only included times ≥1.4 h, after oromucosal contamination cleared. For

“anxious” and “sedated”, significant (but small) OF THC*time effects were present but blood THC*time effects were not significant. Several models (“good drug effect”,

“high”, “stimulated”, “stoned”, “difficulty concentrating”, “altered sense of time”, “body feels sluggish/heavy”, “feel thirsty”, “dry mouth/throat”) had significant THC*time interactions common to blood and OF.

All active-drug interventions were positively associated with subjective “good drug effect” 0.17 and 1.4 h post-dose relative to baseline (time-point analyses, Figure 6).

Although alcohol only displayed a significant main dose effect at 0.17 h, significant increases from baseline persisted 3.3 and 4.3 h with combined cannabis and alcohol. Both low (2.9%-THC) and high (6.7%-THC) cannabis doses were positively associated with

“high”, “good drug effect”, “stimulated”, and “stoned” over the first 3.3h. Significant alcohol-dose effects were detected 0.17 h after cannabis dosing initiation (0.24h after drinking initiation) for “good drug effect” and “stimulated”. We observed only two significant low-versus-high cannabis differences by time point: “stoned” 1.4 h post-dose and “anxious” 0.17 h post-dose. Significant cannabis effects on “sedated” occurred at time points 2.3-4.3 h post-dose. Cannabis also affected “altered sense of time” (1.7-2.3 h), “feel thirsty” (0.17-2.3 h), and “dry mouth/throat” (0.17-3.3 h) (Figure 10

(Supplemental)). Subjective effects versus blood and OF THC concentrations displayed

172

counterclockwise hysteresis; whereas subjective effects versus BrAC showed clockwise hysteresis (Figure 7, Figure 11 (Supplemental)).

OF/Blood and OF/Plasma

OF/blood and OF/plasma ratios showed large variability. Median [range] paired- positive OF/blood ratios were 9.4 [0.3-887, N=413] THC and 3.7 [0.6-20.9] ng/μg,

N=339] THCCOOH (Table 22 (Supplemental)). Median [range] OF/plasma ratios were

7.3 [0.2-585, N=455] THC and 2.4 [0.4-13.3] ng/μg, N=341] THCCOOH. Paired- positive CBD and CBN specimens occurred only 0.17 h post-dose (9-12 pairs) and showed high variability. OF THC concentration significantly correlated (p<0.001) with blood THC concentration (Figure 8, Spearman r [95%CI]=0.7469 [0.6574-0.8156] and

0.8057 [0.7339-0.8598] for low- and high-dose cannabis without alcohol, r=0.7321

[0.6389-0.8042] and 0.8447 [0.7858-0.8884] for cannabis with alcohol) and with plasma

THC (Spearman r≥0.7066 in either matrix for every dose) (Table 23 (Supplemental)).

Alcohol presence did not significantly affect ratios. Due to high variability, the only significant dose effect by time point was an overall cannabis effect on OF/plasma 8.3 h post-dose (Figure 9). Ratio differences between time points could not be statistically evaluated because ratio variability was high with few paired-positives (Figure 9, Table 22

(Supplemental)).

Discussion

Blood THC concentration after vaporization was significantly and positively associated with subjective effects (Table 18), while there generally was no significant differentiation between effects of low (2.9% THC) and high (6.7% THC) dose cannabis.

173

Figure 6. Median [interquartile range] subjective effects visual-analogue scales (VAS) results versus time in 19 participants after low (2.9% THC) and high (6.7% THC) vaporized cannabis doses with and without low-dose oral alcohol. Note y-axis scales vary by measure; all VAS were out of 100. *Overall cannabis p≤0.005 (significant with Bonferroni correction for ten measurements); #Overall cannabis p≤0.01 (trend with Bonferroni correction for ten measurements); @Overall alcohol p≤0.005; &Overall cannabis*alcohol p≤0.005; αp≤0.005, low versus placebo cannabis; βp≤0.005, high versus placebo cannabis; γP≤0.005, high versus low cannabis; δp≤0.005, high versus placebo alcohol*cannabis; εp≤0.005, high versus low alcohol*cannabis; 1p≤0.005 versus baseline, placebo cannabis without alcohol; 2p≤0.005 versus baseline, low cannabis without alcohol; 3p≤0.005 versus baseline, high cannabis without alcohol; 4p≤0.005 versus baseline, placebo cannabis with alcohol; 5p≤0.005 versus baseline, low cannabis with alcohol; 6p≤0.005 versus baseline, high cannabis with alcohol. 174

Figure 7. Median subjective effects visual-analogue scales (VAS) results versus median blood ∆9- tetrahydrocannabinol (THC) concentrations, oral fluid (OF) THC, and breath alcohol concentration (BrAC) in 19 participants after placebo, low (2.9% THC) and high (6.7% THC) vaporized cannabis doses with and without low-dose oral alcohol. Counterclockwise and clockwise arrows represent hysteresis curve progressions over time.

175

Figure 8. Oral fluid (OF) ∆9-tetrahydrocannabinol (THC) concentrations versus blood (A) and plasma (B) THC, and least-squares linear regressions from 19 participants after low (2.9% THC) and high (6.7% THC) vaporized cannabis doses with and without low-dose oral alcohol. Insets illustrate (zoom) densest regions; note graph scales. OF significantly correlated (p<0.001) with blood and plasma (Spearman r≥0.7066 in either matrix for every dose). See Table 23 (Supplemental) for regression equations and comparisons.

176

177

Figure 9. Median [range] oral fluid (OF)/blood (a, c) and OF/plasma (b,d) ∆9-tetrahydrocannabinol (THC) and 11-nor-9-carboxy-THC (THCCOOH) ratios over time in paired-positive specimens from 19 participants after low (2.9% THC) and high (6.7% THC) vaporized cannabis doses with and without low-dose oral alcohol.

This is consistent with pharmacokinetic results from these participants (256, 260), and supports previous findings that THC concentrations are a better predictor of subjective effects than cannabis dose (169). Observed effect sizes (represented by coefficient b) for most Likert measures generally were much lower than VAS for the same factors, possibly because of the shorter Likert measurement scale. Blood THC concentration was not significantly associated with “sedated” in the overall linear mixed model, although time point dose-wise ANOVA showed significant increases over 1.4-4.4 h (Figure 6). This may result from higher variability and less-consistent results throughout the time course, or possibly other study procedures (e.g., simulated driving). “High”, “good drug effect”, and “stimulated” are likely desirable effects for recreational intake, whereas “anxious” and “restless” are likely undesirable. “Stoned” and “sedated” could be either, but would be undesirable for pharmacotherapy. Vaporized cannabis significantly increased these measures immediately post-dose, lasting 3.3 or 4.3 h. “Anxious” showed significant cannabis-dose effects through 1.4 h. Undesirable effects including “feel thirsty” and “dry mouth/throat” increased for the first 3.3 h post-cannabis. “Difficulty concentrating” and

“altered sense of time” produced mixed effects over 2.3 h. Only time significantly increased “feel hungry” in the hours prior to lunch, unexpectedly with no significant THC effect. Another study found cannabis significantly increased “feel hungry” relative to baseline on a 5-point Likert scale after smoking a 6.8% THC cigarette (81); however, as there was no placebo, possibly the observed effect was due to time since last eating.

There is growing interest in correlating cannabis’ subjective effects directly to OF

THC concentrations, due to OF advantages as a sampling matrix (74, 109-110). However, our results indicate caution in interpreting effects from OF concentrations. Unlike blood

178

models, OF regression models (full time course, Table 21 (Supplemental)) had low b- values even when main effects or interactions were statistically significant, probably due to high inter-individual variability in OF THC concentrations and a time course influenced by OF oral contamination rather than systemic cannabinoid concentrations

(109-110). THC concentrations after active doses ranged from 22.7-66,200 μg/L (256).

OF THC b-values represented concentration coefficients, so b in the thousandths (order of magnitude) would indicate clinically significant effects for OF THC >1000 μg/L.

Considering only times ≥1.4 h post-dose (Table 20 (Supplemental)) produced models with more robust significant OF main effects, as initial OF contamination decreased.

However, active-dose OF THC concentrations still ranged ~1000-fold, 3.0-3940 μg/L at

1.4 h and 1.6-1541 μg/L at 2.3 h. The ≥1000-fold concentration differences impose challenges to reliably assess effects based on OF; blood THC at 0.17, 1.4, and 2.4 h ranged only 11.4-210 μg/L, 0-18.4 μg/L, and 0-9.6 μg/L, respectively. Additionally, this may account for the high variability of OF/blood and OF/plasma THC ratios (Figure 9), although the influence of OF contamination should be greatest immediately post- inhalation. In other words, OF did not closely track blood or plasma THC changes during this 8.3 h time course. Overall, OF THC concentrations were not reliable indices of blood and plasma THC concentration, accounting for the former’s weak association with subjective effects. The relationship between subjective effects and blood or OF THC concentration showed counterclockwise hysteresis (Figure 7), consistent with previous findings (81). During cannabis inhalation, maximum blood and OF THC concentrations

(Cmax) occurred immediately prior to last inhalation, then decreased rapidly (280), while peak subjective effects occur over the first 2h (41, 43, 169, 281). Subjective effects are

179

related to brain THC concentration, with THC equilibration time in brain accounting for the lag between blood THC Cmax and maximum subjective effects (282). Blood THC rapidly decreases during distribution to highly-perfused and adipose tissues (140), producing maximum subjective effects after blood tmax, explaining the counterclockwise hysteresis. In contrast, alcohol’s slower absorption and later Cmax (256) led to observed clockwise hysteresis. Clockwise hysteresis may be caused by tachyphylaxis (acute tolerance to an effect happening within a single dose time course, possibly due to receptor down-regulation) or feedback regulation (283).

BrAC was significantly associated, albeit not robustly, with “good drug effect”,

“high”, and “stimulated”. The THC*BrAC interaction was less-than-additive (i.e., significant negative interaction term), suggesting that THC+BrAC effects were less than the sum of each individual substance effect (i.e., partial mitigation of simple main effects). However, models for several subjective effects (“high”, “good drug effect”,

“stoned”) included positive time*THC*BrAC interaction terms that yielded overall approximately-additive THC+BrAC effects immediately post-dose and more-than- additive (synergistic) effects as time progressed, prolonging subjective effects.

Significant increases from baseline in these effects longer in cannabis-alcohol combinations (extending effects beyond those of either drug alone) corroborate this finding (Figure 6).

Alcohol-alone produced hystereses shifted lower than curves for cannabis+alcohol combinations (Figure 7), indicating that participants experienced more effects after alcohol combined with active cannabis compared to alcohol-alone. Low- and high-dose cannabis combined with alcohol produced superimposed curves for “high”,

180

“good drug effect”, “stimulated”, and “stoned”, suggesting no dose differential in cannabis effects when combined with alcohol, although individual variability was high

(Figure 12 (Supplemental) subjective “high” (N=19)). A previous study found similar variability in individual hysteresis curves (81), albeit with just one dosing condition. Only high-THC cannabis combined with alcohol produced substantially higher blood THC

Cmax. This possibly resulted from increased THC-absorption rates during inhalation (due to alcohol-induced increased cardiac output (284) and pulmonary capillary flow) or less- careful cannabis self-titration during alcohol intoxication.

Vaporized cannabis produced subjective effects and time courses similar to smoking, consistent with prior findings (81, 281). Few studies examined combined cannabis-alcohol subjective effects (169, 285-287), and none as comprehensively as reported herein. In one study, mean subjective “high” post-cannabis intake did not significantly increase with prior alcohol relative to without (285). Although alcohol-only increased subjective cannabis-specific “high” (corroborating our findings (Figure 6)), overall, participants correctly distinguished cannabis’ from ethanol’s “high”. Participants who drank alcohol before cannabis smoking also were aware of this distinction (169): subjective “drunkenness” was dominant before smoking, subjective [cannabis] “high” thereafter. Alcohol pretreatment significantly decreased latency to smoked-cannabis effects and increased euphoria duration (287). In the current study, subjective effects significantly >baseline persisted longer post-cannabis dosing with alcohol than post- cannabis dosing without alcohol (“high”, “good drug effect”, “stimulated”, “stoned”,

“sedated”, “difficulty concentrating”, “dry mouth/throat”).

181

Prior studies directly compared THC and THCCOOH relationships between OF and blood (288-289) or OF and plasma/serum (40, 74, 116, 121, 290-291). However, few included concurrent alcohol administration (116), and none provided within-subject blood and plasma data. Plasma is more commonly used for clinical and pharmacokinetic purposes, but blood is more common in forensic settings. A forensic OF-blood THC linear regression study in suspected drugged drivers had negligible (albeit statistically significant) correlation (R2=0.030) (289), likely caused by high variability in time since last intake and unknown food or drink ingestion. Our controlled-administration fits were stronger for all doses (Table 23 (Supplemental)), and we observed higher correlations

(Spearman r=0.7321-0.8447 among all active-cannabis conditions). However, consistent with prior research, we observed high variability in OF/blood and OF/plasma ratios

(Figure 9), particularly for THC. Recently reported OF/serum THC ratios showed similar ranges (116), reiterating (291-292) that OF/blood or OF/serum ratios are too variable to predict one concentration from the other. Recently, 44 [95%CI 27-90] μg/L OF THC produced the same cannabis driving prevalence as 1 μg/L blood THC (132), but as we showed, there is too much variability to predict blood or oral fluid THC from the other matrix concentration. OF retains its value in identifying recent cannabis exposure (289), but is more limited in predicting cannabis effects. There were no significant alcohol effects on OF/blood or OF/plasma THC (consistent with other findings (116)).

Our study found narrower OF/blood and OF/plasma THCCOOH ratio ranges because THCCOOH enters OF from systemic circulation rather than oromucosal contamination. THCCOOH was not always detected in OF in this occasional-to- moderate-smokers cohort and, when present, was in low ng/L concentrations. OF

182

THCCOOH distinguishes passive environmental smoke exposure from cannabis intake

(293) [although chronic passive exposure was not studied] (256). Alcohol did not affect

OF/blood or OF/plasma THCCOOH. THCCOOH is non-psychoactive and cannot be related to subjective effects. Its value remains as a cannabis use marker. Although OF

CBD and CBN persisted for hours due to oral contamination (256), they were not present in blood and plasma after 0.42 h. When present, these markers help identify recent intake, but are more likely to be detected in OF than blood in forensic settings, where blood collection lag times often exceed detection windows (141-142).

Study strengths and limitations

This is the most comprehensive evaluation of which we are aware of vaporized cannabis subjective effects time courses, with and without alcohol, over an extended period. We observed significant cannabis subjective effects for most measures through

3.3 or 4.3 h. Our robust within-subjects design, evaluation of multiple subjective effects utilizing two different types of measurement scales, and concentration-based linear mixed models approach provided in-depth analyses of cannabis, alcohol, and interaction effects over time, also comparing blood and OF concentrations. Study limitations include lack of an explicit “bad drug effect” measure, although we did measure potential negative side effects (“anxious”, “difficulty concentrating”, “body feels sluggish/heavy”), and exclusion of frequent cannabis users (>3x/week) as participants. The latter may limit the external validity of our findings, as a prior study found different subjective effects patterns in frequent versus occasional cannabis smokers (288). To our knowledge, only one other study compared OF/serum THC concentrations after controlled vaporized

183

cannabis in frequent smokers (74). Authors found similar broad variability in OF/serum

THC, but did not report OF/serum THCCOOH ratios.

Conclusion

We delineated subjective psychological effects of inhaled THC, with and without oral alcohol, concomitantly comparing blood and plasma to OF cannabinoid concentrations during the treatment period. Vaporized cannabis produced a notable

“high” and other subjective effects through 4.3 h post-dose, similar to the effect of smoked cannabis. Alcohol prolonged the duration of cannabis’ effects. Subjective effect- versus-cannabinoid concentration curves displayed counterclockwise hysteresis, but subjective effect-versus-alcohol concentration produced clockwise hysteresis possibly due to slower alcohol absorption. We observed robust OF/blood and OF/plasma correlations, but high OF cannabinoid variability challenged reliable cannabis-effects predictions. Although OF retains strong cannabis exposure screening validity, blood THC demonstrated considerably more consistent results for predicting intoxicating effects of cannabis inhalation.

184

Table 19 (Supplemental). Overall effect of blood ∆9-tetrahydrocannabinol (THC) concentration (μg/L), breath alcohol concentration (BrAC, g/210L), time, and interactions (THC*BrAC, time*THC, time*BrAC, time*THC*BrAC) on subjective effects in 19 healthy, occasional-to-moderate cannabis smokers following controlled vaporized cannabis administration with and without oral alcohol. 95% Confidence Parameter b SE df t p b Interval of b Lower Upper Likert Feel Hungry Bound Bound Intercept 0.435 0.197 32.9 2.2 0.035 0.034 0.836 Blood THC 406.9 -1.6 0.103 BrAC 433.4 0.1 0.902 Time 0.473 0.046 433.9 10.4 <0.001 0.384 0.563 THC*BrAC 433.2 0.9 0.379 Time*THC 445.2 1.3 0.210 Time*BrAC 433.5 -0.3 0.753 Time*THC* 433.8 0.4 0.659 BrAC Likert Shakiness/Tremulousness Intercept 0.072 0.036 140.9 2.0 0.044 0.002 0.143 Blood THC 0.004 0.002 18.5 2.5 0.023 0.001 0.008 BrAC 547.2 -0.4 0.682 Time -0.016 0.008 547.6 -2.1 0.037 -0.031 -0.001 THC*BrAC 549.4 0.2 0.805 Time*THC 550.2 1.5 0.129 Time*BrAC 546.4 1.6 0.114 Time*THC* 545.4 -0.2 0.846 BrAC Likert Nausea Intercept 0.116 0.043 36.2 2.7 0.010 0.030 0.202 Blood THC -0.002 0.001 666.0 -2.2 0.031 -0.004 0.000 BrAC 657.0 -1.4 0.151 Time -0.019 0.004 657.4 -4.2 <0.001 -0.028 -0.010 THC*BrAC 658.6 0.9 0.369 Time*THC 0.005 0.002 623.2 2.5 0.011 0.001 0.008 Time*BrAC 656.6 1.8 0.067 Time*THC* 659.2 -0.8 0.396 BrAC Likert Headache Intercept 0.172 0.057 42.4 3.0 0.004 0.057 0.286 Blood THC 191.2 -1.6 0.110 BrAC -2.438 1.224 659.1 -2.0 0.047 -4.841 -0.035 Time 659.4 -0.7 0.471 THC*BrAC 661.2 0.0 0.964 Time*THC 618.3 -0.3 0.782 Time*BrAC 658.8 -0.1 0.891

185

Table 19 (Supplemental). (Continued from previous page) 95% Confidence Parameter b SE df t p b Interval of b Time*THC* 659.4 1.1 0.251 BrAC Likert Palpitations Intercept 539.2 0.4 0.657 Blood THC 0.93 1.6 0.377 BrAC 654.8 0.4 0.698 Time 654.9 -0.4 0.662 THC*BrAC 654.4 -1.3 0.207 Time*THC 654.5 0.0 0.981 Time*BrAC 654.8 -0.3 0.728 Time*THC* 654.4 1.6 0.115 BrAC Likert Upset Stomach Intercept 48.6 2.0 0.057 Blood THC 140.9 -0.5 0.637 BrAC 657.4 -0.8 0.400 Time -0.008 0.004 656.8 -2.3 0.019 -0.015 -0.001 THC*BrAC 658.3 0.4 0.692 Time*THC 671.3 1.4 0.153 Time*BrAC 656.6 0.9 0.365 Time*THC* 657.6 0.4 0.712 BrAC Data from 19 occasional-to-moderate cannabis smokers who participated in all dosing sessions. Subjective effects were measured by 100 mm visual-analogue scales (VAS) or 5-point Likert scales (Likert) with choices 0≡“none”, 1≡“slight”, 2≡“mild”, 3≡“moderate”, 4≡“severe” after drinking placebo or active alcohol (calculated to produce approximate peak 0.065% BrAC) and inhaling placebo, 2.9% THC, or 6.7% THC vaporized cannabis (500 mg, Volcano® Medic vaporizer). Linear mixed model results; b is parameter (coefficient) estimate for each factor (negative b indicates the parameter decreases the subjective effect; positive b indicates the parameter increases the overall effect). Overall equation: [Subjective Effect Result] = Intercept +bBlood THC*[THC]blood + bBrAC*BrAC + bTime*post-dose time + bTHC*BrAC*[THC]blood*BrAC + bTime*THC*post-dose time*[THC]blood + bTime*BrAC*post-dose time*BrAC + bTime*THC*BrAC*post-dose time*[THC]blood*BrAC aValues in bold are statistically significant (p<0.05); only significant predictors are considered in the final model. Abbreviations: SE, standard error; df, degrees of freedom; VAS, visual-analogue scale; THC, ∆9- tetrahydrocannabinol; BrAC, breath alcohol concentration.

186

Table 20 (Supplemental). Overall effect of oral fluid (OF) ∆9-tetrahydrocannabinol (THC) concentration (μg/L), breath alcohol concentration (BrAC, g/210L), time, and interactions (THC*BrAC, time*THC, time*BrAC, time*THC*BrAC) on subjective effects beginning 1.4h post-dose in 19 occasional-to-moderate cannabis smokers following controlled vaporized cannabis administration with and without oral alcohol. 95% Confidence Parameter b SE df t pa b Interval of b Lower Upper VAS Anxious Bound Bound Intercept 5.332 1.102 58.09 4.84 <0.001 3.127 7.537 OF THC 0.045 0.014 11.06 3.32 0.007 0.015 0.075 BrAC 869.16 1.50 0.133 Time -0.370 0.128 866.25 -2.88 0.004 -0.621 -0.118 THC*BrAC 861.97 -1.17 0.244 Time*THC 821.08 -0.15 0.884 Time*BrAC 869.48 0.25 0.805 Time*THC*BrAC 852.98 -0.78 0.433 Subject Variance in 11.928 4.245 0.005 5.938 23.960 Intercepts (THC) Subject variance in 0.176 Slopes (THC) ARH1 rho (slope- 0.682 0.174 <0.001 0.191 0.900 intercept covariance) VAS Good Drug Effect Intercept 16.427 1.904 52.38 8.63 <0.001 12.608 20.247 OF THC 0.222 0.057 7.31 3.92 0.005 0.089 0.354 BrAC 376.67 68.01 869.93 5.54 <0.001 243.18 510.15 Time -2.081 0.213 864.72 -9.75 <0.001 -2.500 -1.662 THC*BrAC -0.698 0.299 860.75 -2.33 0.020 -1.285 -0.111 Time*THC 820.97 -0.32 0.749 Time*BrAC -75.40 29.06 853.06 -2.59 0.010 -132.44 -18.35 Time*THC*BrAC -0.129 0.177 861.05 -0.73 0.468 -0.477 0.219 Subject Variance in 37.956 13.375 0.005 19.03 75.73 Intercepts (THC) Subject variance in 0.082 Slopes (THC) ARH1 rho (slope- 0.805 0.128 <0.001 0.381 0.950 intercept covariance) VAS High Intercept 15.33 1.80 59.87 8.53 <0.001 11.73 18.92 OF THC 0.373 0.117 10.50 3.20 0.009 0.115 0.632 BrAC 296.40 66.95 854.38 4.43 <0.001 164.99 427.81 Time -1.994 0.211 860.70 -9.47 <0.001 -2.407 -1.581 THC*BrAC -0.583 0.295 858.67 -1.98 0.048 -1.162 -0.004 Time*THC 862.55 -0.55 0.580 Time*BrAC -65.08 28.50 849.15 -2.28 0.023 -121.02 -9.15 Time*THC*BrAC 866.18 -0.26 0.794 Subject Variance in 31.076 11.110 0.005 15.421 62.622 Intercepts (THC) Subject variance in 0.242 0.111 0.030 0.098 0.597 Slopes (THC) ARH1 rho (slope- 0.555 0.186 0.003 0.099 0.819 intercept covariance)

187

Table 20 (Supplemental). (Continued from previous page) 95% Confidence Parameter b SE df t pa b Interval of b VAS Restless Intercept 10.97 2.36 41.71 4.65 <0.001 6.21 15.73 OF THC 93.04 1.84 0.069 BrAC 810.57 -0.29 0.770 Time 869.52 0.16 0.871 THC*BrAC 819.76 -0.73 0.467 Time*THC 887.55 -0.38 0.704 Time*BrAC 872.48 0.57 0.566 Time*THC*BrAC 875.57 0.03 0.977 Subject Variance in 67.08 23.06 0.004 34.20 131.59 Intercepts (THC) Subject variance in 0.311 Slopes (THC) ARH1 rho (slope- 0.541 intercept covariance) VAS Sedated Intercept 17.21 3.08 29.43 5.60 <0.001 10.92 23.50 OF THC 59.47 0.57 0.570 BrAC 830.58 0.63 0.528 Time -1.26 0.24 875.41 -5.16 <0.001 -1.73 -0.78 THC*BrAC 831.80 0.48 0.631 Time*THC 886.81 0.58 0.563 Time*BrAC 877.81 0.71 0.478 Time*THC*BrAC 882.50 -1.75 0.081 Subject Variance in 140.12 46.80 0.003 72.81 269.65 Intercepts (THC) Subject variance in 0.333 Slopes (THC) ARH1 rho (slope- -0.699 0.222 0.002 -0.938 -0.015 intercept covariance) VAS Stimulated Intercept 16.37 2.05 37.02 7.97 <0.001 12.21 20.53 OF THC 0.219 0.075 5.10 2.93 0.032 0.028 0.411 BrAC 222.54 61.77 840.77 3.60 <0.001 101.29 343.79 Time -1.799 0.194 855.59 -9.27 <0.001 -2.180 -1.418 THC*BrAC 824.85 -1.53 0.126 Time*THC 840.19 0.19 0.853 Time*BrAC 827.69 -1.92 0.055 Time*THC*BrAC 833.83 -1.26 0.209 Subject Variance in 54.51 18.57 0.003 27.96 106.28 Intercepts (THC) Subject variance in 0.141 Slopes (THC) ARH1 rho (slope- 0.564 0.176 0.001 0.133 0.816 intercept covariance) VAS Stoned Intercept 14.67 2.21 39.74 6.63 <0.001 10.20 19.13 OF THC 0.280 0.093 6.46 3.01 0.022 0.057 0.503 BrAC 283.22 69.37 841.69 4.08 <0.001 147.05 419.38 Time -1.740 0.217 853.57 -8.00 <0.001 -2.167 -1.314 THC*BrAC 833.62 -1.63 0.102 Time*THC 844.07 -0.80 0.426

188

Table 20 (Supplemental). (Continued from previous page) 95% Confidence Parameter b SE df t pa b Interval of b Time*BrAC -76.84 29.67 834.59 -2.59 0.010 -135.07 -18.60 Time*THC*BrAC 839.33 -0.68 0.499 Subject Variance in 60.63 20.77 0.004 30.98 118.67 Intercepts (THC) Subject variance in 0.100 Slopes (THC) ARH1 rho (slope- 0.615 intercept covariance) Likert Difficulty Concentrating Intercept 0.514 0.092 32.69 5.57 <0.001 0.326 0.702 OF THC 6.91 2.12 0.072 BrAC 7.86 2.55 876.67 3.08 0.002 2.86 12.87 Time -0.052 0.008 870.90 -6.46 <0.001 -0.067 -0.036 THC*BrAC 850.03 0.48 0.632 Time*THC 857.74 0.79 0.432 Time*BrAC 863.12 -1.88 0.061 Time*THC*BrAC 852.98 -0.40 0.693 Subject Variance in 0.119 0.040 0.003 0.061 0.230 Intercepts (THC) Subject variance in 0.147 Slopes (THC) ARH1 rho (slope- 0.549 0.178 0.002 0.116 0.807 intercept covariance) Likert Altered Sense of Time Intercept 0.334 0.071 49.98 4.68 <0.001 0.191 0.478 OF THC 0.005 0.002 12.06 2.35 0.037 0.000 0.010 BrAC 5.605 2.503 881.85 2.24 0.025 0.693 10.517 Time -0.044 0.008 874.51 -5.56 <0.001 -0.059 -0.028 THC*BrAC 875.97 -1.03 0.302 Time*THC 719.54 0.24 0.807 Time*BrAC 874.10 -0.95 0.343 Time*THC*BrAC 839.79 -0.10 0.919 Subject Variance in 0.055 0.019 0.004 0.028 0.109 Intercepts (THC) Subject variance in 0.000 0.000 0.022 0.000 0.000 Slopes (THC) ARH1 rho (slope- 0.922 0.060 <0.001 0.674 0.983 intercept covariance) Likert Slowed/Slurred Speech Intercept 0.141 0.045 85.37 3.14 0.002 0.052 0.230 OF THC 14.87 1.73 0.104 BrAC 5.748 1.859 875.26 3.09 0.002 2.100 9.396 Time -0.019 0.006 875.46 -3.25 0.001 -0.030 -0.008 THC*BrAC 874.56 1.18 0.239 Time*THC 875.67 0.01 0.993 Time*BrAC 874.96 -1.22 0.223 Time*THC*BrAC 874.66 -1.45 0.147 Subject Variance in 0.015 0.006 0.009 0.007 0.032 Intercepts (THC) Subject variance in 0.00014 0.00006 0.021 0.00006 0.00032 Slopes (THC)

189

Table 20 (Supplemental). (Continued from previous page) 95% Confidence Parameter b SE df t pa b Interval of b ARH1 rho (slope- -- intercept covariance) Likert Body Feels Sluggish/Heavy Intercept 0.574 0.091 48.11 6.32 <0.001 0.391 0.757 OF THC 11.29 1.71 0.115 BrAC 7.502 3.152 881.42 2.38 0.018 1.316 13.687 Time -0.058 0.010 874.12 -5.91 <0.001 -0.078 -0.039 THC*BrAC 873.41 -0.20 0.842 Time*THC 875.71 1.32 0.188 Time*BrAC 877.21 -0.60 0.546 Time*THC*BrAC 867.94 -0.96 0.338 Subject Variance in 0.091 0.032 0.004 0.046 0.181 Intercepts (THC) Subject variance in 0.225 Slopes (THC) ARH1 rho (slope- 0.386 intercept covariance) Likert Feel Thirsty Intercept 1.142 0.112 48.47 10.22 <0.001 0.918 1.367 OF THC 0.004 0.001 12.40 2.88 0.013 0.001 0.007 BrAC 880.91 -0.09 0.926 Time -0.128 0.012 875.64 -10.6 <0.001 -0.152 -0.104 THC*BrAC 878.53 0.55 0.581 Time*THC 0.001 0.000 863.23 3.20 0.001 0.000 0.002 Time*BrAC 5.565 1.660 878.88 3.35 0.001 2.307 8.822 Time*THC*BrAC -0.024 0.010 873.11 -2.35 0.019 -0.044 -0.004 Subject Variance in 0.138 0.048 0.004 0.069 0.272 Intercepts (THC) Subject variance in 0.136 Slopes (THC) ARH1 rho (slope- 0.598 0.256 0.020 -0.092 0.900 intercept covariance) Likert Dizzy Intercept 0.082 0.027 197.43 3.01 0.003 0.028 0.135 OF THC 15.59 1.43 0.174 BrAC 4.408 1.277 864.60 3.45 0.001 1.902 6.914 Time -0.009 0.004 863.05 -2.22 0.027 -0.017 -0.001 THC*BrAC 875.78 0.13 0.895 Time*THC 782.83 -0.04 0.968 Time*BrAC -1.481 0.543 864.55 -2.73 0.007 -2.547 -0.415 Time*THC*BrAC 860.09 -0.18 0.857 Subject Variance in 0.003 0.001 0.029 0.001 0.007 Intercepts (THC) Subject variance in 0.0003 0.0001 0.007 0.0001 0.0005 Slopes (THC) ARH1 rho (slope- 0.509 intercept covariance) Likert Dry Mouth or Throat Intercept 0.727 0.093 71.08 7.83 <0.001 0.542 0.913 OF THC 0.013 0.005 9.79 2.54 0.030 0.002 0.024 BrAC 863.27 1.60 0.110 Time -0.084 0.011 868.02 -7.33 <0.001 -0.107 -0.062

190

Table 20 (Supplemental). (Continued from previous page) 95% Confidence Parameter b SE df t pa b Interval of b THC*BrAC 864.66 0.47 0.637 Time*THC 0.001 0.000 842.04 3.06 0.002 0.000 0.002 Time*BrAC 856.97 -0.54 0.591 Time*THC*BrAC 865.39 -0.98 0.326 Subject Variance in 0.074 0.027 0.006 0.036 0.152 Intercepts (THC) Subject variance in 0.000 0.000 0.039 0.000 0.001 Slopes (THC) ARH1 rho (slope- 0.628 0.169 0.000 0.190 0.857 intercept covariance) Likert Feel Hungry Intercept 1.771 0.123 57.82 14.41 <0.001 1.525 2.017 OF THC 896.47 -1.67 0.096 BrAC -14.84 4.41 882.70 -3.37 0.001 -23.49 -6.19 Time -0.238 0.014 883.14 -16.6 <0.001 -0.266 -0.210 THC*BrAC 889.20 1.45 0.148 Time*THC 0.001 0.000 899.57 2.83 0.005 0.000 0.002 Time*BrAC 9.144 1.944 884.02 4.70 <0.001 5.328 12.959 Time*THC*BrAC 895.64 -1.88 0.060 Subject Variance in 0.150 0.053 0.005 0.075 0.299 Intercepts (THC) Subject variance in 0.479 Slopes (THC) ARH1 rho (slope- -- intercept covariance) Likert Feel Shakiness/Tremulousness Intercept 810.68 1.66 0.097 OF THC 0.27 1.37 0.650 BrAC 879.36 -0.38 0.702 Time 876.77 -1.45 0.149 THC*BrAC 878.33 -1.88 0.061 Time*THC 878.42 -0.48 0.628 Time*BrAC 879.56 0.86 0.390 Time*THC*BrAC 0.0059 0.0027 878.26 2.14 0.033 0.0005 0.0113 Subject Variance in 0.721 Intercepts (THC) Subject variance in 0.000 0.000 0.006 0.000 0.000 Slopes (THC) ARH1 rho (slope- -- intercept covariance) Likert Nausea Intercept 0.126 0.032 93.30 3.88 <0.001 0.061 0.190 OF THC 251.26 1.01 0.315 BrAC 886.90 -0.82 0.414 Time -0.017 0.004 884.95 -3.97 <0.001 -0.026 -0.009 THC*BrAC 891.21 -0.89 0.374 Time*THC 896.37 -0.35 0.728 Time*BrAC 886.58 0.54 0.591 Time*THC*BrAC 890.11 1.04 0.299 Subject Variance in 0.007 0.003 0.009 0.004 0.016 Intercepts (THC)

191

Table 20 (Supplemental). (Continued from previous page) 95% Confidence Parameter b SE df t pa b Interval of b Subject variance in 0.000 0.000 0.040 0.000 0.000 Slopes (THC) ARH1 rho (slope- -- intercept covariance) Likert Headache Intercept 0.168 0.056 70.65 3.01 0.004 0.057 0.280 OF THC 892.32 -0.56 0.578 BrAC 881.93 -1.38 0.169 Time 882.26 -1.17 0.241 THC*BrAC 892.35 -0.54 0.586 Time*THC 898.44 0.85 0.393 Time*BrAC 884.07 0.30 0.761 Time*THC*BrAC 898.61 0.60 0.549 Subject Variance in 0.027 0.010 0.006 0.013 0.055 Intercepts (THC) Subject variance in 0.923 Slopes (THC) ARH1 rho (slope- -- intercept covariance) Likert Palpitations Intercept -0.012 0.005 411.13 -2.27 0.024 -0.022 -0.002 OF THC 7.33 1.52 0.171 BrAC 0.819 0.268 887.10 3.06 0.002 0.294 1.345 Time 0.0018 0.0008 889.95 2.13 0.034 0.0001 0.0034 THC*BrAC 883.82 -1.17 0.243 Time*THC 886.13 -1.31 0.192 Time*BrAC -0.255 0.114 885.69 -2.24 0.025 -0.478 -0.032 Time*THC*BrAC 884.81 -0.07 0.947 Subject Variance in 0.083 Intercepts (THC) Subject variance in 0.000 0.000 0.004 0.000 0.000 Slopes (THC) ARH1 rho (slope- -- . . intercept covariance)

192

Table 20 (Supplemental). (Continued from previous page) 95% Confidence Parameter b SE df t pa b Interval of b Likert Upset Stomach Intercept 0.061 0.021 173.16 2.95 0.004 0.020 0.102 OF THC 47.87 1.37 0.177 BrAC 889.33 -0.61 0.540 Time -0.007 0.003 886.65 -2.42 0.016 -0.013 -0.001 THC*BrAC -0.014 0.004 889.86 -3.41 0.001 -0.023 -0.006 Time*THC -0.0002 0.0001 894.36 -1.97 0.0497 -0.0004 0.0000 Time*BrAC 888.49 0.17 0.868 Time*THC*BrAC 0.010 0.002 888.86 4.05 <0.001 0.005 0.015 Subject Variance in 0.002 0.001 0.020 0.001 0.004 Intercepts (THC) Subject variance in 0.000 0.000 0.010 0.000 0.000 Slopes (THC) ARH1 rho (slope- -- intercept covariance) Data from 19 occasional-to-moderate cannabis smokers who participated in all dosing sessions, excluding times <1.4h post-dose. Subjective effects were measured by 100 mm visual-analogue scales (VAS) or 5-point Likert scales (Likert) with choices 0≡“none”, 1≡“slight”, 2≡“mild”, 3≡“moderate”, 4≡“severe” after drinking placebo or active alcohol (calculated to produce approximate peak 0.065% BrAC) and inhaling placebo, 2.9% THC, or 6.7% THC vaporized cannabis (500 mg, Volcano® Medic vaporizer). Linear mixed model results; b is parameter (coefficient) estimate for each factor (negative b indicates the parameter decreases the subjective effect; positive b indicates the parameter increases the overall effect). Overall equation: [Subjective Effect Result] = Intercept +bOF THC*[THC]blood + bBrAC*BrAC + bTime*post-dose time + bTHC*BrAC*[THC]OF*BrAC + bTime*THC*post-dose time*[THC]OF + bTime*BrAC*post-dose time*BrAC + bTime*THC*BrAC*post-dose time*[THC]OF*BrAC aValues in bold are statistically significant (p<0.05); only significant predictors are considered in the final model. Abbreviations: SE, standard error; df, degrees of freedom; VAS, visual-analogue scale; THC, ∆9- tetrahydrocannabinol; BrAC, breath alcohol concentration.

193

Table 21 (Supplemental). Overall effect of oral fluid (OF) ∆9-tetrahydrocannabinol (THC) concentration (μg/L), breath alcohol concentration (BrAC, g/210L), time, and interactions (THC*BrAC, time*THC, time*BrAC, time*THC*BrAC) on subjective effects in 19 occasional-to-moderate cannabis smokers following controlled vaporized cannabis administration with and without oral alcohol. 95% Confidence Parameter b SE df t pa b Interval of b Lower Upper VAS Anxious Bound Bound Intercept 9.003 1.123 45.88 8.0 <0.001 6.742 11.264 OF THC 6.97 0.6 0.585 BrAC 978.55 0.6 0.559 Time -0.965 0.125 970.53 -7.7 <0.001 -1.210 -0.719 THC*BrAC 942.52 1.0 0.322 Time*THC 0.0063 0.0025 810.20 2.5 0.011 0.0014 0.0111 Time*BrAC 974.93 0.4 0.718 Time*THC*BrAC -0.103 0.050 876.28 -2.1 0.037 -0.201 -0.006 Subject Variance in 13.890 4.984 0.005 6.875 28.062 Intercepts (THC) Subject variance in 0.528 Slopes (THC) ARH1 rho (slope- 0.646 0.310 0.037 -0.268 0.948 intercept covariance) VAS Good Drug Effect Intercept 24.610 2.114 36.63 11.6 <0.001 20.325 28.895 OF THC 0.0073 0.0026 10.54 2.8 0.017 0.0016 0.0130 BrAC 264.584 39.073 990.22 6.8 <0.001 187.909 341.259 - Time -3.401 0.208 987.76 <0.001 -3.809 -2.992 16.3 THC*BrAC 977.73 0.4 0.687 Time*THC 0.0182 0.0040 867.47 4.5 <0.001 0.0103 0.0260 Time*BrAC -57.421 19.932 988.98 -2.9 0.004 -96.535 -18.306 Time*THC*BrAC -0.230 0.077 937.40 -3.0 0.003 -0.380 -0.079 Subject Variance in 56.978 19.796 0.004 28.838 112.576 Intercepts (THC) Subject variance in 0.058 0.000 0.000 Slopes (THC) ARH1 rho (slope- 0.830 0.142 0.000 0.285 0.970 intercept covariance) VAS High Intercept 25.628 2.050 42.01 12.5 <0.001 21.491 29.765 OF THC 0.0115 0.0036 11.00 3.2 0.008 0.0037 0.0193 BrAC 195.297 40.951 991.81 4.8 <0.001 114.935 275.658 - Time -3.631 0.219 988.84 <0.001 -4.060 -3.201 16.6 THC*BrAC 935.55 -0.5 0.620 Time*THC 0.0187 0.0041 638.56 4.5 <0.001 0.0106 0.0268 Time*BrAC -47.901 20.878 991.03 -2.3 0.022 -88.870 -6.931

194

Table 21 (Supplemental). (Continued from previous page) 95% Confidence Parameter b SE df t pa b Interval of b Time*THC*BrAC -0.196 0.079 787.41 -2.5 0.013 -0.351 -0.041 Subject Variance in 49.024 17.353 0.005 24.497 98.108 Intercepts (THC) Subject variance in 0.000 0.000 0.037 0.000 0.001 Slopes (THC) ARH1 rho (slope- 0.923 0.070 <0.001 0.596 0.987 intercept covariance) VAS Restless Intercept 12.717 2.121 33.43 6.0 <0.001 8.405 17.029 OF THC 55.54 0.0 0.982 BrAC 977.92 -0.2 0.821 Time 971.19 -1.3 0.181 THC*BrAC 946.79 1.0 0.338 Time*THC 994.51 1.4 0.173 Time*BrAC 965.82 0.3 0.780 Time*THC*BrAC 995.77 -1.1 0.266 Subject Variance in 60.781 20.866 0.004 31.014 119.120 Intercepts (THC) Subject variance in 0.312 Slopes (THC) ARH1 rho (slope- 0.653 intercept covariance) VAS Sedated Intercept 18.366 2.715 27.42 6.8 <0.001 12.799 23.933 OF THC 998.87 -1.4 0.174 BrAC 990.41 0.4 0.659 Time -1.448 0.208 990.22 -7.0 <0.001 -1.856 -1.041 THC*BrAC 994.46 1.3 0.197 Time*THC 0.0087 0.0041 1003.97 2.1 0.037 0.0005 0.0168 Time*BrAC 990.19 1.3 0.209 Time*THC*BrAC -0.159 0.079 1000.95 -2.0 0.044 -0.313 -0.004 Subject Variance in 18.366 2.715 0.003 12.799 23.933 Intercepts (THC) Subject variance in ------Slopes (THC) ARH1 rho (slope------intercept covariance) VAS Stimulated Intercept 23.298 2.282 29.53 10.2 <0.001 18.634 27.962 OF THC 3.50 1.5 0.221 BrAC 226.775 35.601 945.07 6.4 <0.001 156.910 296.641 - Time -2.930 0.188 978.90 <0.001 -3.299 -2.561 15.6 THC*BrAC 0.044 0.016 953.00 2.7 0.006 0.012 0.076 Time*THC 0.0213 0.0036 936.20 5.9 <0.001 0.0142 0.0285

195

Table 21 (Supplemental). (Continued from previous page) 95% Confidence Parameter b SE df t pa b Interval of b - Time*BrAC -77.518 18.029 950.94 -4.3 <0.001 -42.137 112.898 Time*THC*BrAC -0.324 0.069 936.73 -4.7 <0.001 -0.460 -0.187 Subject Variance in 76.099 25.727 0.003 39.229 147.622 Intercepts (THC) Subject variance in 0.245 Slopes (THC) ARH1 rho (slope- 0.628 0.162 <0.001 0.211 0.851 intercept covariance) VAS Stoned Intercept 23.581 2.219 36.32 10.6 <0.001 19.081 28.081 OF THC 8.10 1.1 0.300 BrAC 159.548 40.963 979.29 3.9 <0.001 79.163 239.933 - Time -3.178 0.218 984.66 <0.001 -3.605 -2.751 14.6 THC*BrAC 976.63 0.0 0.987 Time*THC 0.0166 0.0042 975.54 3.9 <0.001 0.0083 0.0250 Time*BrAC -55.741 20.971 978.65 -2.7 0.008 -96.895 -14.586 Time*THC*BrAC -0.212 0.081 974.10 -2.6 0.009 -0.370 -0.053 Subject Variance in 63.059 21.915 0.004 31.909 124.615 Intercepts (THC) Subject variance in 0.097 Slopes (THC) ARH1 rho (slope- 0.249 intercept covariance) Likert Difficulty Concentrating Intercept 0.594 0.097 27.13 6.1 <0.001 0.394 0.794 OF THC 8.39 1.5 0.161 BrAC 7.252 1.376 989.34 5.3 <0.001 4.552 9.953 Time -0.064 0.007 991.00 -8.8 <0.001 -0.078 -0.050 THC*BrAC 989.10 0.1 0.885 Time*THC 0.0003 0.0001 981.98 2.1 0.034 0.0000 0.0006 Time*BrAC -1.560 0.701 988.02 -2.2 0.026 -2.937 -0.184 Time*THC*BrAC -0.006 0.003 980.15 -2.2 0.025 -0.011 -0.001 Subject Variance in 0.146 0.049 0.003 0.076 0.281 Intercepts (THC) Subject variance in 0.087 Slopes (THC) ARH1 rho (slope- 0.659 0.155 <0.001 0.247 0.869 intercept covariance) Likert Altered Sense of Time Intercept 0.473 0.084 32.21 5.6 <0.001 0.301 0.645 OF THC 10.37 1.8 0.102 BrAC 3.087 1.421 995.32 2.2 0.030 0.297 5.876 Time -0.066 0.008 995.52 -8.7 <0.001 -0.081 -0.051

196

Table 21 (Supplemental). (Continued from previous page) 95% Confidence Parameter b SE df t pa b Interval of b THC*BrAC 921.42 0.8 0.444 Time*THC 0.0004 0.0001 524.49 2.6 0.008 0.0001 0.0006 Time*BrAC 995.25 0.1 0.886 Time*THC*BrAC -0.007 0.003 688.97 -2.6 0.010 -0.012 -0.002 Subject Variance in 0.098 0.034 0.003 0.050 0.192 Intercepts (THC) Subject variance in 0.000 0.000 0.034 0.000 0.000 Slopes (THC) ARH1 rho (slope- 0.981 0.024 <0.001 0.776 0.999 intercept covariance) Likert Slowed/Slurred Speech Intercept 0.268 0.056 38.18 4.8 <0.001 0.154 0.382 OF THC 18.85 1.8 0.091 BrAC 2.907 1.065 989.38 2.7 0.006 0.816 4.997 Time -0.038 0.006 990.28 -6.7 <0.001 -0.049 -0.027 THC*BrAC 990.31 -0.8 0.430 Time*THC 992.84 0.2 0.803 Time*BrAC 989.37 -0.1 0.904 Time*THC*BrAC 991.08 0.5 0.604 Subject Variance in ------Intercepts (THC) Subject variance in ------Slopes (THC) ARH1 rho (slope------intercept covariance) Likert Body Feels Sluggish/Heavy Intercept 0.677 0.089 33.81 7.6 <0.001 0.496 0.858 OF THC -0.0001 0.0000 958.74 -3.4 0.001 -0.0002 0.0000 BrAC 6.043 1.595 998.29 3.8 <0.001 2.912 9.174 Time -0.075 0.008 996.81 -9.0 <0.001 -0.092 -0.059 THC*BrAC 0.002 0.001 1007.01 3.0 0.003 0.001 0.004 Time*THC 0.0007 0.0002 1014.76 4.0 <0.001 0.0003 0.0010 Time*BrAC 998.17 -0.7 0.503 Time*THC*BrAC -0.011 0.003 1012.50 -3.4 0.001 -0.017 -0.005 Subject Variance in ------Intercepts (THC) Subject variance in ------Slopes (THC) ARH1 rho (slope------intercept covariance) Likert Feel Thirsty Intercept 1.036 0.132 82.86 7.9 <0.001 0.773 1.298 - - OF THC -0.0002 0.0001 25.13 -3.0 0.006 0.00032 0.00006 BrAC 434.92 -0.7 0.455

197

Table 21 (Supplemental). (Continued from previous page) 95% Confidence Parameter b SE df t pa b Interval of b Time 437.45 0.2 0.803 THC*BrAC 0.005 0.001 411.88 4.2 <0.001 0.003 0.007 Time*THC 0.0015 0.0003 407.93 5.4 <0.001 0.0010 0.0021 Time*BrAC 5.808 1.648 430.90 3.5 <0.001 2.569 9.048 Time*THC*BrAC -0.023 0.005 428.57 -4.3 <0.001 -0.034 -0.013 Subject Variance in 0.122 0.050 0.014 0.055 0.271 Intercepts (THC) Subject variance in 0.299 Slopes (THC) ARH1 rho (slope- 0.581 intercept covariance) Likert Dizzy Intercept 0.180 0.032 77.56 5.6 <0.001 0.117 0.244 OF THC 15.89 1.3 0.208 BrAC 976.30 0.1 0.895 Time -0.024 0.004 977.01 -5.7 <0.001 -0.032 -0.016 THC*BrAC 991.05 -0.6 0.522 Time*THC 925.95 -0.6 0.517 Time*BrAC 979.24 -0.4 0.659 Time*THC*BrAC 969.93 0.9 0.353 Subject Variance in 0.008 0.003 0.009 0.004 0.017 Intercepts (THC) Subject variance in 0.000 0.000 0.005 0.000 0.000 Slopes (THC) ARH1 rho (slope- 0.478 0.210 0.023 -0.013 0.783 intercept covariance) Likert Dry Mouth or Throat Intercept 0.936 0.096 44.50 9.7 <0.001 0.742 1.130 OF THC 5.95 -1.4 0.199 BrAC 994.59 1.4 0.157 - Time -0.117 0.011 992.21 <0.001 -0.138 -0.096 11.1 THC*BrAC 0.003 0.001 955.85 3.5 0.001 0.001 0.005 Time*THC 0.0016 0.0002 827.72 7.8 <0.001 0.0012 0.0020 Time*BrAC 994.25 1.1 0.274 Time*THC*BrAC -0.020 0.004 893.64 -5.1 <0.001 -0.028 -0.012 Subject Variance in 0.106 0.037 0.005 0.053 0.211 Intercepts (THC) Subject variance in 0.480 Slopes (THC) ARH1 rho (slope- 0.652 0.316 0.039 -0.289 0.952 intercept covariance) Likert Feel Hungry Intercept 0.398 0.199 65.02 2.0 0.050 0.000 0.796 OF THC 435.03 -1.5 0.147

198

Table 21 (Supplemental). (Continued from previous page) 95% Confidence Parameter b SE df t pa b Interval of b BrAC 431.91 1.0 0.298 Time 0.498 0.040 431.36 12.6 <0.001 0.420 0.576 THC*BrAC 433.29 1.3 0.208 Time*THC 439.20 1.2 0.212 Time*BrAC 431.62 -0.7 0.481 Time*THC*BrAC 437.25 -0.9 0.348 Subject Variance in ------Intercepts (THC) Subject variance in ------Slopes (THC) ARH1 rho (slope------intercept covariance) Likert Feel Shakiness/Tremulousness Intercept 0.105 0.022 241.61 4.7 <0.001 0.061 0.149 OF THC 7.36 1.5 0.164 BrAC 975.85 -0.8 0.396 Time -0.016 0.004 977.34 -4.5 <0.001 -0.023 -0.009 THC*BrAC 975.86 -0.2 0.854 Time*THC 976.80 0.5 0.606 Time*BrAC 976.13 1.1 0.256 Time*THC*BrAC 976.30 0.6 0.560 Subject Variance in ------Intercepts (THC) Subject variance in ------Slopes (THC) ARH1 rho (slope------intercept covariance) Likert Nausea Intercept 0.099 0.032 46.07 3.1 0.003 0.035 0.162 OF THC 1011.88 -1.8 0.070 BrAC 997.10 -1.5 0.142 Time -0.014 0.003 996.65 -4.1 0.000 -0.020 -0.007 THC*BrAC 1005.44 1.4 0.158 Time*THC 1015.98 1.7 0.087 Time*BrAC 996.83 1.9 0.061 Time*THC*BrAC 1015.12 -1.5 0.133 Subject Variance in ------Intercepts (THC) Subject variance in ------Slopes (THC) ARH1 rho (slope------intercept covariance) Likert Headache Intercept 0.120 0.048 45.24 2.5 0.017 0.023 0.217 OF THC 1012.48 -1.8 0.073

199

Table 21 (Supplemental). (Continued from previous page) 95% Confidence Parameter b SE df t pa b Interval of b BrAC 997.27 -1.9 0.061 Time 996.05 -0.1 0.901 THC*BrAC 1007.83 1.3 0.200 Time*THC 1009.41 1.7 0.094 Time*BrAC 997.05 0.5 0.606 Time*THC*BrAC 1013.23 -1.1 0.255 Subject Variance in ------Intercepts (THC) Subject variance in ------Slopes (THC) ARH1 rho (slope------intercept covariance) Likert Palpitations Intercept 0.018 0.009 440.53 2.1 0.039 0.001 0.036 OF THC 3.33 1.1 0.342 BrAC 994.30 1.5 0.135 Time -0.003 0.001 987.49 -2.1 0.035 -0.006 0.000 THC*BrAC 994.68 0.1 0.900 Time*THC 990.39 0.9 0.382 Time*BrAC 994.39 -1.2 0.213 Time*THC*BrAC 994.15 -0.6 0.521 Subject Variance in ------Intercepts (THC) Subject variance in ------Slopes (THC) ARH1 rho (slope------intercept covariance)

200

Table 21 (Supplemental). (Continued from previous page) 95% Confidence Parameter b SE df t pa b Interval of b Likert Upset Stomach Intercept 0.050 0.023 48.34 2.2 0.033 0.004 0.095 OF THC 1011.55 -1.6 0.109 BrAC 997.08 -0.4 0.677 Time -0.007 0.003 997.06 -2.7 0.007 -0.012 -0.002 THC*BrAC 1005.32 0.7 0.456 Time*THC 1008.66 1.7 0.087 Time*BrAC 996.78 0.8 0.453 Time*THC*BrAC 1012.68 -0.9 0.361 Subject Variance in ------Intercepts (THC) Subject variance in ------Slopes (THC) ARH1 rho (slope------intercept covariance) Data from 19 occasional-to-moderate cannabis smokers who participated in all dosing sessions. Subjective effects were measured by 100 mm visual-analogue scales (VAS) or 5-point Likert scales (Likert) with choices 0≡“none”, 1≡“slight”, 2≡“mild”, 3≡“moderate”, 4≡“severe” after drinking placebo or active alcohol (calculated to produce approximate peak 0.065% BrAC) and inhaling placebo, 2.9% THC, or 6.7% THC vaporized cannabis (500 mg, Volcano® Medic vaporizer). Linear mixed model results; b is parameter (coefficient) estimate for each factor (negative b indicates the parameter decreases the subjective effect; positive b indicates the parameter increases the overall effect). Overall equation: [Subjective Effect Result] = Intercept +bOF THC*[THC]blood + bBrAC*BrAC + bTime*post-dose time + bTHC*BrAC*[THC]OF*BrAC + bTime*THC*post-dose time*[THC]OF + bTime*BrAC*post-dose time*BrAC + bTime*THC*BrAC*post-dose time*[THC]OF*BrAC aValues in bold are statistically significant (p<0.05); only significant predictors are considered in the final model. Abbreviations: SE, standard error; df, degrees of freedom; VAS, visual-analogue scale; THC, ∆9- tetrahydrocannabinol; BrAC, breath alcohol concentration.

201

Table 22 (Supplemental). Median [range] (N) oral fluid (OF)/blood and OF/plasma ∆9- tetrahydrocannabinol (THC), 11-nor-9-carboxy-THC (THCCOOH), cannabidiol (CBD) and cannabinol (CBN) ratios by time point after controlled vaporized cannabis and low- dose oral alcohol to healthy adult occasional-to-moderate cannabis smokers. No Alcohol Alcohol Time Post- Low High Low High Placebo Placebo Dose 2.9% THC 6.7% THC 2.9% THC 6.7% THC (h) OF/Blood THC 4.0 [0.5-8.2] 4.9 [3.3-7.1] 1.6 [1.1-2.2] 2.2 [0.8-9.0] 5.2 [1.3-6.4] 2.6 [2.0-8.3] -0.8 (5) (3) (2) (3) (4) (3) 2.9 [0.4- 19.6 [0.8-840] 27.1 [1.6-887] 2.9 [1.4-30.7] 23.0 [1.9-256] 16.5 [0.7-723] 0.17 10.0] (19) (19) (9) (19) (19) (17) 1.7 [0.8-7.9] 15.4 [0.9-225] 27.8 [1.7-227] 1.8 [1.2-36.4] 20.8 [1.5-417] 26.7 [1.3-444] 1.4 (6) (18) (18) (5) (19) (19) 1.6 [0.7-8.9] 18.9 [0.8-133] 21.5 [0.8-253] 2.0 [1.8-4.4] 16.4 [4.0-214] 24.2 [1.0-452] 2.3 (6) (17) (15) (3) (16) (16) 1.7 [1.5-2.7] 15.7 [2.3-110] 9.6 [0.9-174] 1.5 [1.3-4.6] 11.9 [2.8-168] 33.5 [3.2-203] 3.3 (4) (12) (15) (4) (14) (13) 0.8 [0.3-1.7] 3.7 [0.7-15.3] 7.1 [1.1-15.7] 1.2 [0.3-1.4] 4.9 [0.9-31.1] 8.5 [2.1-55.9] 6.3 (5) (8) (7) (3) (7) (9) 0.5 [0.3-1.8] 1.6 [0.7-49.4] 1.8 [1.3-10.1] 0.9 [0.3-2.0] 3.7 [1.2-29.9] 5.3 [0.8-28.5] 8.3 (3) (8) (5) (4) (5) (7) OF/Plasma THC 5.1 [0.6-8.2] 5.9 [3.1-13.2] 1.4 [1.0-8.4] 2.5 [0.6-7.7] 2.5 [0.7-5.5] 4.8 [1.9-20.1] -0.8 (6) (5) (4) (5) (5) (4) 1.9 [0.3-7.5] 15.5 [0.5-505] 21.0 [1.0-585] 2.8 [0.4-22.2] 14.8 [1.5-156] 11.5 [0.4-463] 0.17 (18) (19) (19) (11) (19) (19) 1.9 [1.1-6.8] 11.3 [0.6-159] 17.6 [1.4-166] 2.0 [1.1-32.5] 16.2 [0.9-359] 18.6 [0.9-319] 1.4 (5) (18) (18) (5) (19) (19) 1.8 [1.4-6.0] 13.0 [0.5-90.1] 15.0 [0.5-144] 1.5 [1.2-4.9] 12.5 [2.3-186] 20.7 [0.8-315] 2.3 (5) (17) (16) (5) (17) (18) 1.6 [1.3-4.3] 7.7 [1.2-80.5] 8.7 [0.7-139] 1.3 [0.9-1.4] 9.8 [2.1-188] 19.7 [0.5-129] 3.3 (5) (13) (15) (3) (17) (16) 0.9 [0.3-1.2] 1.8 [0.5-15.3] 4.3 [0.8-62.2] 1.2 [0.2-1.3] 4.6 [0.5-35.0] 5.7 [0.6-47.7] 6.3 (6) (7) (11) (4) (10) (10) 0.5 [0.2-1.5] 1.5 [0.4-33.7] 4.0 [0.9-46.3] 0.5 [0.2-0.8] 2.6 [0.4-21.8] 4.6 [1.0-20.6] 8.3 (5) (9) (7) (4) (8) (9) OF/Blood THCCOOH 4.0 [1.5- 3.0 [2.5-7.7] 3.8 [1.2-7.7] 3.9 [2.1-8.0] 4.0 [1.8-12.1] 3.7 [1.9-8.8] -0.8 10.5] (7) (7) (6) (6) (7) (8) 2.5 [1.2-7.9] 3.2 [2.8-4.2] 2.8 [0.7-3.8] 3.8 [2.0-7.0] 3.6 [1.3-12.5] 2.9 [0.6-11.6] 0.17 (6) (7) (8) (7) (8) (9) 3.4 [1.2-9.0] 2.6 [1.9-9.7] 3.0 [1.2-8.7] 4.9 [2.2-8.9] 4.6 [2.1-17.0] 4.8 [1.9-20.9] 1.4 (7) (8) (9) (9) (12) (11) 4.8 [1.4-7.2] 3.4 [1.8-7.9] 3.5 [2.4-11.1] 6.2 [3.2-10.1] 4.7 [2.6-13.6] 4.5 [2.2-17.9] 2.3 (6) (9) (8) (7) (11) (12) 3.1 [1.3-5.7] 3.2 [2.1-5.6] 2.8 [1.3-7.2] 4.7 [2.1-5.8] 4.3 [2.4-14.3] 4.4 [1.3-19.6] 3.3 (6) (10) (10) (7) (10) (9) 3.8 [2.1-5.9] 4.3 [1.7-11.5] 3.2 [1.9-5.3] 4.3 [3.0-13.4] 4.5 [1.5-12.3] 3.5 [2.7-9.7] 6.3 (6) (9) (9) (7) (11) (8) 3.7 [2.8- 4.7 [2.0-5.8] 3.1 [1.5-6.3] 4.0 [2.1-8.1] 5.0 [1.5-9.8] 2.9 [2.2-6.1] 8.3 11.7] (8) (8) (6) (8) (8) (4)

202

Table 22 (Supplemental). (Continued from previous page) No Alcohol Alcohol Time Post- Low High Low High Placebo Placebo Dose 2.9% THC 6.7% THC 2.9% THC 6.7% THC (h) OF/Plasma THCCOOH 2.7 [1.0-6.3] 1.7 [1.2-4.7] 2.0 [0.7-3.6] 2.7 [1.5-5.0] 2.7 [1.1-6.8] 2.3 [1.7-5.1] -0.8 (8) (7) (7) (7) (6) (7) 1.6 [1.0-4.5] 2.1 [1.7-2.8] 1.9 [0.5-2.6] 2.6 [1.6-4.8] 2.1 [0.8-7.1] 1.8 [0.4-7.6] 0.17 (6) (7) (8) (7) (8) (9) 2.7 [0.9-5.7] 1.8 [1.3-6.6] 1.9 [0.8-4.3] 3.2 [1.4-5.9] 3.2 [1.1-10.6] 3.5 [1.1-13.3] 1.4 (7) (8) (9) (9) (12) (11) 3.3 [1.1-4.6] 2.3 [1.1-4.6] 2.8 [1.4-5.5] 3.9 [2.0-7.0] 3.1 [1.8-8.8] 3.0 [1.2-11.1] 2.3 (6) (9) (8) (7) (11) (12) 2.0 [1.0-3.6] 2.0 [0.9-4.0] 2.0 [0.7-3.4] 3.2 [1.2-4.8] 3.0 [1.6-7.9] 2.5 [0.8-12.4] 3.3 (6) (10) (10) (7) (10) (9) 2.5 [1.4-4.0] 3.5 [1.1-8.3] 2.0 [1.0-3.4] 2.8 [2.0-9.3] 2.8 [1.0-8.8] 2.4 [1.9-6.4] 6.3 (6) (9) (9) (7) (11) (8) 2.3 [1.9-7.0] 3.2 [1.3-3.7] 1.9 [0.8-4.0] 2.7 [1.6-5.1] 3.3 [1.0-5.8] 2.0 [1.4-3.8] 8.3 (5) (8) (8) (6) (8) (8) OF/Blood CBD -0.8 ------32.8 [10.6- 45.2 [2.5-183] 0.17 1607] (10) (13) 1.4 ------2.3 ------3.3 ------6.3 ------8.3 ------OF/Plasma CBD -0.8 ------56.5 [2.2-1015] 25.6 [8.6-1060] 0.17 ------(12) (16) 1.4 ------2.3 ------3.3 ------6.3 ------8.3 ------OF/Blood CBN -0.8 ------1.5 20.7 [1.0-173] 36.7 [2.0-120] 38.9 29.0 [1.4-192] 28.2 [5.1-947] 0.17 (1) (11) (9) (1) (16) (14) 1.4 ------2.3 ------3.3 ------6.3 ------8.3 ------OF/Plasma CBN -0.8 ------1.3 10.3 [0.8-640] 60.5 [1.7-118] 43.5 29.8 [1.7-166] 20.2 [5.3-828] 0.17 (1) (14) (10) (1) (15) (15) 1.4 ------2.3 ------3.3 ------6.3 ------8.3 ------Alcohol administration: oral placebo or active alcohol (calculated to produce approximate peak 0.065% BrAC). Cannabis administration: inhaled placebo, 2.9% THC, or 6.7% THC vaporized cannabis (500 mg, Volcano® Medic vaporizer). Data from 19 occasional-to-moderate cannabis smokers who participated in all dosing sessions. N represents number of paired-positive specimens contributing to the result. Abbreviations: OF, oral fluid; THC, ∆9-tetrahydrocannabinol; THCCOOH, 11-nor-9-carboxy-THC; CBD, cannabidiol; CBN, cannabinol.

203

Table 23 (Supplemental). (To accompany Figure 8): Regression equations and slope comparisons for (A) oral fluid (OF) ∆9-tetrahydrocannabinol (THC) versus blood THC concentration, and (B) OF versus plasma THC concentration after low (2.9% THC) and high (6.7% THC) dose vaporized cannabis administration with and without oral alcohol to 19 healthy, adult occasional-to-moderate cannabis smokers. A) OF versus Blood THC Figure 7 Regression Equations

Low (2.9% THC) High (6.7% THC) phigh-low (slopes) No y = 34.12x + 104.5 y = 43.40x + 85.52 0.5123 Alcohol R2 = 0.075, N = 130 R2 = 0.184, N = 130 y = 34.60x + 40.58 y = 80.65x + 210.3 Alcohol 0.0866 R2 = 0.324, N = 131 R2 = 0.153, N = 131 palcohol- Overall p no alcohol 0.9646 0.0745 (slopes, all doses): (slopes) 0.0171

B) OF versus Plasma THC Figure 7 Regression Equations

Low (2.9% THC) High (6.7% THC) phigh-low (slopes) No y = 25.20x + 91.54 y = 28.98x + 92.24 0.6925 Alcohol R2 = 0.087, N = 130 R2 = 0.192, N = 130 y = 24.76x + 35.60 y = 50.41x + 282.3 Alcohol 0.1725 R2 = 0.333, N = 131 R2 = 0.141, N = 131 palcohol- Overall p no alcohol 0.9534 0.1183 (slopes, all doses): (slopes) 0.0668 N represents number of specimen pairs.

204

Figure 10 (Supplemental). Mean [95% confidence interval] 5-point subjective Likert scale results in 19 participants after placebo, low (2.9% THC) and high (6.7% THC) vaporized cannabis doses with and without low-dose oral alcohol. In participant responses, 0≡“None”, 1≡“Slight”, 2≡“Mild”, 3≡“Moderate”, and 4≡“Severe”. # *Overall cannabis p≤0.005 (significant with Bonferroni correction for ten measurements); Overall cannabis p≤0.01 (trend with Bonferroni correction for ten measurements); @Overall alcohol p≤0.005; &Overall cannabis*alcohol p≤0.005; αp≤0.005, low versus placebo cannabis; βp≤0.005, high versus placebo cannabis; γP≤0.005, high versus low cannabis; δp≤0.005, high versus placebo alcohol*cannabis; εp≤0.005, high versus low alcohol*cannabis; 1p≤0.005 versus baseline, placebo cannabis without alcohol; 2p≤0.005 versus baseline, low cannabis without alcohol; 3p≤0.005 versus baseline, high cannabis without alcohol; 4p≤0.005 versus baseline, placebo cannabis with alcohol; 5p≤0.005 versus baseline, low cannabis with alcohol; 6p≤0.005 versus baseline, high cannabis with alcohol. 205

Figure 11 (Supplemental). Median subjective effects visual-analogue scales (VAS) results (“restless” and “sedated”) versus median blood ∆9-tetrahydrocannabinol (THC) concentrations, oral fluid (OF) THC, and breath alcohol concentration (BrAC) in 19 participants after placebo, low (2.9% THC) and high (6.7% THC) vaporized cannabis doses with and without low-dose oral alcohol. Counterclockwise and clockwise arrows represent hysteresis curve progressions over time.

206

207

Figure 12 (Supplemental). Individual subjective “high” visual-analogue scales (VAS) results versus blood ∆9- tetrahydrocannabinol (THC) concentrations in 19 participants after low (2.9% THC) and high (6.7% THC) vaporized cannabis doses with and without low-dose oral alcohol. Counterclockwise arrows represent hysteresis curve progressions over time.

Chapter 6 – Controlled Cannabis Vaporizer Administration: Blood and Plasma Cannabinoids With and Without Alcohol

(As accepted for publication in Clinical Chemistry, 2015)2

Abstract

Background: Increased medical and legal cannabis intake is accompanied by greater use of cannabis vaporization and more driving under the influence of cannabis cases. Despite frequent simultaneous ∆9-tetrahydrocannabinol (THC) and alcohol use, potential pharmacokinetic interactions are poorly understood. Here we studied blood and plasma vaporized cannabinoid disposition, with and without simultaneous oral low-dose alcohol.

Methods: Thirty-two adult cannabis smokers (≥1x/3 months, ≤3 days/week) drank placebo or low-dose alcohol (target ~0.065% peak breath-alcohol concentration) 10 min prior to inhaling 500 mg placebo, low-dose (2.9% THC), or high-dose (6.7% THC) vaporized cannabis (within-subjects 6 alcohol-cannabis combinations). Blood and plasma were obtained before and up to 8.3 h post-dose.

Results: Nineteen participants completed all sessions. Median [range] maximum blood concentrations (Cmax) for low and high THC doses (no alcohol) were 32.7 [11.4-

66.2] and 42.2 [15.2-137] µg/L THC; and 2.8 [0-9.1] and 5.0 [0-14.2] µg/L 11-OH-THC.

With alcohol, low and high dose Cmax were 35.3 [13.0-71.4] and 67.5 [18.1-210] µg/L

THC; and 3.7 [1.4-6.0] and 6.0 [0-23.3] µg/L 11-OH-THC, significantly higher than

2 Hartman RL et al. Clin Chem 2015; in press. 208

without alcohol. With detection cutoff THC ≥1 μg/L, ≥16.7% of participants remained positive 8.3 h post-dose; whereas ≤21.1% were positive by 2.3 h with THC ≥5 μg/L.

Conclusions: Vaporization is an effective THC delivery route. The significantly higher blood THC and 11-OH-THC Cmax with alcohol possibly explain increased performance impairment observed from cannabis-alcohol combinations. Chosen driving-related THC cutoffs should be considered carefully to best reflect performance impairment windows.

Our results will help facilitate forensic interpretation and inform the debate on drugged driving legislation.

Introduction

Currently, 23 states and the District of Columbia have legalized medical cannabis, and Colorado, Washington, Oregon, and Alaska have decriminalized recreational cannabis intake (250). Per se cannabinoid blood cutoffs for driving under the influence

(DUI) include zero-tolerance or 1, 2 or 5 μg/L ∆9-tetrahydrocannabinol (THC) (239), with the District of Columbia enacting a 5 µg/L per se law and Colorado a 5 μg/L

“permissible inference” law. These legal changes resulted in increased DUI cannabis cases (275-276) and more complicated enforcement of cannabinoid drugged driving laws

(108, 268, 294). A major confounding factor is extended cannabinoid excretion with chronic frequent intake (108). Cannabis plus alcohol is among the most frequent drug combinations identified in driving cases worldwide, with evidence of increased performance impairment (294). Despite frequent concomitant THC and alcohol intake, little is known about a potential pharmacokinetic interaction. Thus, understanding

209

cannabinoid blood disposition, with and without simultaneous alcohol, is critical for proper test interpretation (21).

Although smoking is the most common cannabis administration route (63), vaporization is growing rapidly, providing similar effects (78-79) while reducing exposure to harmful pyrolytic byproducts (69) and decreasing adverse respiratory symptoms (70). THC is highly lipophilic, rapidly distributing to highly perfused tissues, and later to fat (84). Blood and plasma smoked cannabinoid disposition was recently evaluated in occasional and frequent cannabis smokers (93-94), but vaporized cannabis disposition is not yet fully characterized. The few prior clinical studies had short (≤6 h) time courses and limited metabolite analyses (78-79). As medical cannabis use increases, plasma cannabinoid data following vaporized cannabis are needed for therapeutic optimization, while blood cannabinoid data are needed for forensic DUI cannabis cases

(126).

In this study we simultaneously evaluated phase I and II cannabinoid disposition in blood and plasma after controlled vaporized cannabis administration, with and without low-dose oral alcohol administration. We hypothesized that cannabinoid delivery and disposition would be similar to that observed with smoking, and that alcohol would not substantially impact cannabinoid pharmacokinetics.

Methods

Participants

Healthy adults provided written informed consent for this University of Iowa

Institutional Review Board-approved study. Inclusion criteria were ages 21-55 y; self-

210

reported mean cannabis consumption ≥1x/3 months but ≤3 days/week over the past 3 months (Cannabis Use Disorders Identification Test (252)); self-reported “light” or

“moderate” alcohol consumption according to a Quantity-Frequency-Variability scale

(253); or if “heavy”, not more than 3-4 servings in a typical drinking occasion. Exclusion criteria included past or current clinically significant medical illness; history of clinically significant adverse event associated with cannabis/alcohol intoxication; ≥450 mL blood donation in 2 weeks preceding drug administration; pregnant/nursing; interest in drug abuse treatment within past 60 d; and currently taking drugs contraindicated with cannabis or alcohol (ethanol) or known to impact driving.

Study Design

Participants entered the clinical research unit 10-16 h prior to drug administration to preclude intoxication at dosing. Over 10 min, participants drank ad libitum placebo [-]

([apple, orange or cranberry juice, consistent within-subject] with ethanol-swabbed rim and topped with 1 mL ethanol to mimic alcohol taste and odor) or low-dose [+] 90% grain ethanol (to produce approximately 0.065% peak breath alcohol concentration) mixed with juice. After drinking, they orally inhaled 500 mg placebo ([P], 0.008±0.002%

THC/0.001±0.001% cannabidiol [CBD]/0.009±0.003% cannabinol [CBN]), low ([L],

2.9±0.14% THC/0.05±0.00% CBD/0.22±0.02% CBN)-, or high ([H], 6.7±0.05%

THC/0.19±0.01% CBD/0.37±0.03% CBN)-THC vaporized ground bulk cannabis

(210°C, standard (~8 L) balloon volume, Volcano® Medic, Storz & Bickel, Tuttlingen,

Germany, www.vapormed.com) ad libitum over 10 min (Table 28 (Supplemental)). Bulk cannabis was obtained through the NIDA Chemistry and Physiological Systems Research

211

Branch (Research Triangle Institute, Oxford, MS). In this within-subjects design, participants received all six alcohol/cannabis doses in randomized order, in sessions separated by ≥1 week to prevent cannabinoid carryover from study interventions. Blood was collected via indwelling peripheral venous catheter into grey-top potassium oxalate/sodium fluoride Vacutainer® tubes (VWR Scientific) at baseline (-0.8 h) and

0.17, 0.42, 1.4, 2.3, 3.3, 4.8, 6.3 and 8.3 h, stored on ice ≤2 h, with a second sample centrifuged at 1600×g for 10 min. Blood and plasma were transferred into 3.6 mL Nunc® cryotubes (Thomas Scientific), stored at -20°C, and analyzed within 3 months, based on our previous stability study (254).

Blood and Plasma Analysis

Blood and plasma cannabinoids were quantified by a previously published liquid chromatography-tandem mass spectrometry (LC-MS/MS) method (255). Briefly, 0.5 mL blood or plasma was protein-precipitated with ice-cold acetonitrile, supernatants diluted and solid-phase extracted with Bond-Elut Plexa cartridges (Agilent Technologies). Linear ranges were 1-100 μg/L for THC, 11-OH-THC, THCCOOH, CBD, and CBN; 5-250

μg/L for THCCOOH-glucuronide, and 0.5-50 μg/L for THC-glucuronide. Inter-assay

(n=30) analytical accuracy and imprecision were 93.1-109.3% and 4.5-12.8%, respectively.

Data Analysis

Non-compartmental analyses were performed with Phoenix WinNonLin® 6.3 for

Windows (Pharsight). Maximum concentration (Cmax), Cmax accounting for baseline

212

(Cmax-BL), time to maximum concentration (tmax), area under the curve from 0 to 8.3 h post-dose (AUC0-8.3h), AUC0-8.3h accounting for baseline (AUC>BL-8.3h), time of last observed concentration (tlast), and last observed concentration (Clast) were compared with

SPSS® Statistics version 19 for Windows (IBM). For statistical purposes, concentrations

(ANOVA, factors: cannabis, alcohol, cannabis-alcohol) with Bonferroni correction for individuals who completed all 6 sessions. When Mauchly’s sphericity test was violated, the Greenhouse-Geisser correction was used. Friedman’s ANOVA with pairwise post- hoc comparisons was used to determine within-subject dose differences, overall and by time point. For time point analyses, the conservative Bonferroni correction was used for multiple comparisons (p<0.05/9 measurements = p<0.006). Blood/plasma or metabolite ratios were calculated when quantifiable (positive) data were available. We assessed

THCCOOH-glucuronide/THCCOOH ratios with factorial repeated-measures ANOVA in

SPSS, with factors alcohol and cannabis, and covariate time.

Results

Participants

Thirty-two healthy adults (22 men, ages 21-42 y, 72% Caucasian) participated in the study (Table 24). Most participants consumed cannabis ≥2x/month and reported intake within a week prior to admission. Two individuals (participants 17 and 20) self- reported last intake 4 and 6 months ago, respectively, despite reporting overall mean

213

Table 24. Self-reported demographic characteristics and recent cannabis and alcohol consumption history of 32 healthy occasional cannabis smokers. Hours Amount last Race Typical Time since Doses “stoned” on consumedb Parti- Age and BMI Alcohol drinks Cannabis last cannabis received Sex typical (joint or cipant (years) ethnicit (kg/m2) frequency per frequency consumed (Reason for cannabis joint y occasion (days) withdrawal) occasiona equivalent) 1 F 30.6 W 21.4 2-4x/m 2-4 2-3x/wk 1-2 1 2 2 (P) 2 M 23.7 W 24.3 2-3x/wk 2-4 2-4x/m 1-2 1 1 6 3 F 28.4 AA 23.8 ≥4x/wk 2-4 2-4x/m 3-4 14 1 6 4 M 27.8 W 33.2 2-3x/wk 2-4 2-3x/wk 1-2 1 1 3 (P) 5 M 21.9 W 24.7 2-3x/wk 5-6 2-4x/m 1-2 6 1 6 6 M 37.8 W 26.1 2-3x/wk 2-4 2-3x/wk 1-2 3 2.5 6 7 M 26.6 W 21.6 ≤1x/m 2-4 ≤1x/m 1-2 11 3.5 6 8 F 26.3 W 20.0 2-3x/wk 2-4 2-3x/wk 3-4 1 0.25 6 9 M 25.8 W 40.6 2-4x/m 2-4 2-3x/wk 1-2 0.3 0.5 6 10 M 26.1 H 31.5 2-4x/m 1-2 2-3x/wk 1-2 3 1 6

214 11 M 26.9 W 22.9 2-3x/wk 1-2 ≤1x/m 3-4 2 1 3 (P) 12 M 23.2 W 19.5 2-3x/wk 2-4 2-3x/wk 3-4 2 1 6

13 M 23.1 W 23.9 2-4x/m 2-4 ≤1x/m 1-2 2 0.25 6 14 M 21.1 W 20.6 2-3x/wk 5-6 2-3x/wk 1-2 2 2 3 (P) 15 M 32.3 O, H 28.9 2-3x/wk 2-4 2-3x/wk 1-2 4 1 6 16 F 23.4 W 23.3 2-3x/wk 2-4 2-4x/m 3-4 4 1 6 17 F 30.3 AA 24.1 2-3x/wk 2-4 ≤1x/m <1 120 1 6 18 M 24.6 W 23.3 2-3x/wk 2-4 2-4x/m 1-2 7 0.8 6 19 M 40.8 W 31.7 2-3x/wk 2-4 2-4x/m 3-4 5 3 2 (P) 20 F 21.8 W 30.8 2-4x/m 2-4 2-3x/wk 1-2 183 0.5 4 (P) 21 M 42.1 W 24.2 2-4x/m 1-2 ≤1x/m 1-2 45 2 2 (P) 22 M 39.4 W, As 34.6 2-4x/m 2-4 2-4x/m 3-4 1 4.5 4 (P) AI, As, 23 M 21.1 24.0 2-4x/m 2-4 2-3x/wk 5-6 2 1 2 (P) AA, W 24 F 24.6 W, H 19.1 2-3x/wk 2-4 2-4x/m 3-4 28 0.5 3 (AE) 25c M 21.8 W 32.7 ≤1x/m 1-2 2-4x/m 1-2 7 0.13 6 26 M 29.0 O 28.0 2-3x/wk 2-4 ≤1x/m <1 30 0.2 2 (P) 27 F 23.0 W 21.0 2-3x/wk 2-4 2-4x/m 5-6 7 0.3 2 (P) 28 F 21.7 AA, W 23.0 2-4x/m 1-2 2-3x/wk 1-2 1.1 1.5 6

Table 24. (Continued from previous page) Hours Amount last Race Typical Time since Doses “stoned” on consumedb Parti- Age and BMI Alcohol drinks Cannabis last cannabis received Sex typical (joint or cipant (years) ethnicit (kg/m2) frequency per frequency consumed (Reason for cannabis joint y occasion (days) withdrawal) occasiona equivalent) 29 M 28.7 W 18.3 2-3x/wk 2-4 ≤1x/m 3-4 45 0.5 6 30 M 28.1 W 48.3 2-4x/m 2-4 2-4x/m 3-4 5 1 6 31 F 22.9 W 21.6 2-4x/m 5-6 2-3x/wk 3-4 1 1 6 32 M 22.7 W 26.1 2-4x/m 1-2 2-4x/m 1-2 8 1 3 (P) Median (com- 25.8 23.9 4.0 1.0 pleters) Mean (com- 26.1 26.3 12.5 1.0 pleters) StDev 215 (com- 4.1 7.5 27.9 0.8

pleters) Median 26.0 24.0 4.0 1.0 (all) Mean 27.1 26.2 17.3 1.2 (all) StDev 5.8 6.6 38.0 1.0 (all) a‘Hours “stoned”’ wording originates from Cannabis Use Disorders Identification Test, source of self-reported cannabis frequency data bCannabis amount last consumed is based on empirically-normalized joint consumption, to account for various administration routes and self-reported “sharing” between multiple individuals cMay have consumed active cannabis during placebo-alcohol session Abbreviations: W, White; AA, African American; H, Hispanic or Latino; As, Asian; O, Other; AI, American Indian/Native American; P, withdrew for personal reasons (job obligations/scheduling/choice); AE, withdrew due to adverse event (nausea/emesis or dizziness, related to study drugs or other study procedures); StDev, standard deviation

consumption at ≥1x/3 months. Nineteen participants completed all 6 sessions (P-/+, L-/+,

H-/+); there were no significant differences in cannabis history, age, or body mass index between these and the 13 non-completers (Mann-Whitney U (exact) test). One participant

(24) withdrew due to nausea/emesis from cannabis administration; other noncompleters withdrew for personal reasons.

Blood and plasma cannabinoids

In total, 1324 blood and 1327 plasma specimens were quantified for cannabinoids. Blood and plasma pharmacokinetic parameters for 19 completers are presented in Table 25, Table 26, Table 27, Table 29 (Supplemental), Table 30

(Supplemental), and Table 31 (Supplemental). Blood Cmax, with and without accounting for baseline concentrations for THC, 11-OH-THC, and CBN, were significantly higher when alcohol was co-administered with cannabis; also blood THCCOOH-glucuronide tmax was earlier and blood CBN AUC0-8.3h higher with concomitant alcohol (Table 26 and

Table 30 (Supplemental)). In plasma, alcohol significantly increased THC, 11-OH-THC, and CBN Cmax, and CBD Cmax-BL (Table 27 and Table 31 (Supplemental)). Significant additional alcohol-cannabis interactions were observed for 11-OH-THC (Cmax, Clast, tlast in blood; Clast in plasma) and plasma CBD (Cmax). Blood and plasma THC AUC0-8.3h L and

H doses were significantly higher than P (p<0.001 and p=0.036 respectively, in both blood and plasma). Accounting for baseline revealed a significant overall cannabis effect on AUC>BL-8.3h and significantly higher AUC>BL-8.3h after H- vs. L-cannabis. Significant overall cannabinoid concentration differences (p<0.006) were observed in blood through

3.3 h (Figure 13, Figure 15 (Supplemental)) for THC and 11-OH-THC, throughout the

216

Table 25. Median [range] blood and plasma pharmacokinetic parameters following controlled vaporized cannabis administration with and without oral alcohol. THC Blood Plasma (LOQ 1 μg/L) No Alcohol Alcohol No Alcohol Alcohol 2.1 0.6 3.2 1.4 C placebo max [0-7.6]*# [0-5.2]*# [0-9.8]*# [0-9.6]*# 32.7 35.3 46.5 48.6 μg/L low [11.4-66.2]*# [13.0-71.4]*# [16.6-114]*# [21.7-102]*# 42.2 67.5 62.1 97.8 high # # # # [15.2-137]* [18.1-210]* [23.6-196]* [24.5-339]* 1.7 0 2.4 0.7 C placebo max-BL [0-5.3]*# [0-3.2]*# [0-9.2]*# [-0.7-9.6]*# 32.7 35.3 45.7 48.6 μg/L low [11.4-66.2]*# [8.1-71.4]*# [16.6-113]*# [2.3-102]*# 42.2 67.5 62.1 96.1 high # # # # [15.2-137]* [18.1-204]* [23.6-196]* [24.5-332]* 0.17 0.18 0.17 0.22 t placebo max [0.15-6.3] [0.07-≥8.3] [0.15-6.3] [0.07-≥8.3] 0.17 0.17 0.17 0.17 h low [0.15-0.33] [0.15-0.25] [0.15-0.33] [0.15-0.25] 0.17 0.17 0.17 0.17 high [0.15-0.3] [0.12-0.37] [0.15-0.30] [0.12-0.37] 1.1 0.3 3.4 1.3 AUC placebo 0-8.3h [0-53.2]~ [0-36.4]~ [0-66.6]~ [0-103,623]~ 31.9 36.2 44.6 49.4 h*μg/L low [10.6-84.2]~ [18.0-52.2]~ [14.1-124]~ [26.9-80]~ 43.1 62.2 56.2 93.2 high ~ ~ ~ ~ [10.6-113] [13.2-1445] [15.9-182] [19.4-2370] 0.6 0 1.4 0.3 AUC placebo >BL-8.3h [0-7.0]* [0-19.6]* [0-7.7]* [0-7.4]* 21.7 18.7 29.2 24.9 h*μg/L low [6.9-38.4]* [7.6-33.4]* [9.3-56.5]* [14.1-49.3]* 29.4 33.7 43.4 51.6 high [6.8-77.9]* [8.8-83.5]* [9.7-124]* [14.1-132]* 0.42 4.8 0.4 4.3 t placebo last [0.15-≥8.3]* [0.17-≥8.3]* [0.15-≥8.3]* [0.17-≥8.3]* 3.5 3.5 4.8 6.3 h low [0.70-≥8.3]* [1.3-≥8.3]* [0.70-≥8.3]* [1.3-≥8.3]* 4.8 3.7 6.3 6.4 high [0.82-≥8.3]* [1.4-≥8.3]* [1.3-≥8.3]* [1.4-≥8.3]* 11-OH-THC Blood Plasma (LOQ 1 μg/L) No Alcohol Alcohol No Alcohol Alcohol 0 0 0 0 C placebo max [0-2.5]*#& [0-2.4]*#& [0-4.3]*# [0-3.2]*# 2.8 3.7 4.1 4.8 μg/L low [0-9.1]*#& [1.4-6.0]*#& [0-13.7]*# [1.3-8.0]*# 5.0 6.0 7.0 7.5 high #& #& # # [0-14.2]* [0-24.8]* [1.0-20.3]* [0-27.3]*

217

Table 25. (Continued from previous page) 0 0 0 0 C placebo max-BL [0-1.1]*#& [0-1.4]*#& [0-1.1]* [0-1.8]* 2.8 3.3 3.7 4.4 μg/L low [0-9.1]*#& [1.4-6.0]*#& [0-13.7]* [1.3-8.0]* 5.0 6.0 7.0 7.5 high #& #& [0-12.8]* [0-23.3]* [1.0-20.3]* [0-25.4]* 3.2 0.18 1.8 0.18 t placebo max [0.17-6.3] [0.17-2.3] [0.15-4.8] [0.17-4.8] 0.19 0.17 0.17 0.22 h low [0.15-0.58] [0.15-0.42] [0.15-0.4] [0.15-0.48] 0.18 0.18 0.18 0.18 high [0.15-0.43] [0.12-0.42] [0.15-0.4] [0.12-0.53] 0 0 0 0 AUC placebo 0-8.3h [0-18.2]* [0-11.6]* [0-28.3]* [0-21.2]* 3.4 4.4 5.8 6.4 h*μg/L low [0-25.9]* [1.1-15.0]* [0-39.0]* [1.1-28.3]* 6.8 7.2 9.8 11.8 high [0-29.8]* [0-42.0]* [0.4-48.3]* [0-51.3]* 0 0 0 0 AUC placebo >BL-8.3h [0-0.5]* [0-4.3]* [0-2.4]* [0-0.5]* 3.2 4.0 5.5 6.3 h*μg/L low [0-8.1]* [1.1-12.2]* [0-12.3]* [1.2-12.6]* 7.2 9.8 11.8 high 6.8 [0-28.8]* [0-29.4]* [0.42-41.5]* [0-33.4]* 6.3 4.8 3.4 [2.3-≥8.3] t placebo [0.42-≥8.3] [0.15-≥8.3] [0.17-≥8.3] last [n=2] [n=4] [n=4] [n=6] 1.4 1.5 2.3 2.3 h low [0.20-≥8.3]#& [0.42-≥8.3]#& [0.40-≥8.3] [0.42-≥8.3] 3.0 2.3 3.3 3.3 high #& #& [0.42-≥8.3] [0.42-≥8.3] [0.18-≥8.3] [0.42-≥8.3] THCCOOH Blood Plasma (LOQ 1 μg/L) No Alcohol Alcohol No Alcohol Alcohol 2.9 2.9 5.0 3.8 C placebo max [0-67.0]* [0-62.8]* [0-107]* [0-97.5]* 14.5 15 25.3 21.1 μg/L low [4.4-84.2]* [5.4-75.0]* [6.2-137]* [7.2-133]* 23.8 17.4 38.1 25.2 high [2.6-66.6]* [3.4-95.4]* [2.9-116]* [5.1-134]* 0 0.5 0.4 1.0 C placebo [-20.2- max-BL [-0.3-2.3]* [-1.1-43.3]* [-1.3-3.8]* 41.4]* 10.0 9.4 17.5 13.7 μg/L low [4.4-22.2]* [0-21.2]* [6.2-32.4]* [-0.8-47.3]* 18.8 17.5 11.9 26.0 high [-10.6- [2.6-36.9]* [0-53.2]* [2.9-61.1]* 82.9]*

218

Table 25. (Continued from previous page) 0.42 0.32 0.40 0.40 t placebo max [0.15-3.3] [0.17-≥8.3] [0.15-3.3] [0.07-3.4] 0.40 0.40 0.40 0.40 h low [0.17-1.6] [0.22-3.5] [0.17-1.3] [0.15-3.5] 0.40 0.42 0.40 0.42 high [0.17-0.82] [0.15-3.3] [0.17-1.3] [0.15-3.3] 17.1 13.5 25.2 20.7 AUC placebo 0-8.3h [0-437]* [0-358]* [0-682]* [0-568]* 56.8 56.8 97.3 84.2 h*μg/L low [13.4-579]* [11.8-424]* [18.2-883]* [24.8-659]* 88.4 62.5 134 100 high [9.6-361]* [8.3-572]* [14.9-665]* [16.6-816]* 0.4 0.1 1.1 0 AUC placebo >BL-8.3h [0-3.7]* [0-279]* [0-10.6]* [0-137]* 27.7 26.2 40.8 46.4 h*μg/L low [9.6-70.8]* [0-85.9]* [18.2-83.0]* [0-181]* 51.7 41.8 69.8 58.8 high [9.6-121]* [0-262]* [14.9-235]* [0-396]* ≥8.3 ≥8.3 ≥8.3 ≥8.3 t placebo last [0.18-≥8.3] [0.18-≥8.3] [0.43-≥8.3] [1.4-≥8.3] ≥8.3 ≥8.3 ≥8.3 ≥8.3 h low [8.2-≥8.3] [4.3-≥8.3] [≥8.3-≥8.3] [≥8.3-≥8.3] ≥8.3 ≥8.3 ≥8.3 ≥8.3 high [4.8-≥8.3] [3.3-≥8.3] [≥8.3-≥8.3] [6.2-≥8.3] THCCOOH-glucuronide Blood Plasma (LOQ 5 μg/L) No Alcohol Alcohol No Alcohol Alcohol 6.4 6.0 11.1 17.5 Cmax placebo [0-156]* [0-118]* [0-340]* [0-155]* 25.9 27.5 31.3 47.6 μg/L low [0-213]* [5.2-152]* [6.2-227]* [6.1-219]* 48.2 31.6 55.2 47.4 high [0-145]* [6.6-259]* [9.2-251]* [7.5-370]* 0 1.4 0.9 0.9 C max-BL placebo [-4.7-23]* [0-74.4]* [-6.0-93.0]* [-35.9-53.2]* 14.3 19.0 22.5 30.5 μg/L low [-7.0-31.1]* [5.2-39.0]* [-3.9-108]* [6.0-129]* 24.0 24.5 33.4 34.0 high [0-81.2]* [6.6-87.0]* [-4.7-107]* [-120-200]* 1.9 1.4 3.3 1.8 # # tmax placebo [0.17-6.3] [0.42-6.3] [0.15-≥8.3] [0.15-6.3] 2.3 1.7 1.7 2.4 h low [0.17-6.4]# [1.3-6.3]# [0.17-≥8.3] [0.42-≥8.3] 2.3 1.7 3.3 1.7 high # # [1.3-≥8.3] [1.3-≥8.3] [0.82-≥8.3] [0.18-≥8.3]

219

Table 25. (Continued from previous page) 30.0 22.0 55.5 63.2 AUC 0-8.3h placebo [0-1111]* [0-817]* [0-1796]* [0-855]* 173 177 181 301 h*μg/L low [0-1595]* [24.3-907]* [7.9-1425]* [21.9-1255]* 320 237 355 245 high [0-990]* [28.1-1796]* [8.0-1656]* [8.1-2656]* 0 1.8 0.36 1.5 AUC >BL-8.3h placebo [0-49.4]* [0-519]* [0-129]* [0-174]* 86.1 92.2 77.0 115 h*μg/L low [0-205]* [19.9-216]* [0-451]* [18.1-677]* 114 144 126 136 high [0-373]* [28.1-384]* [0-481]* [0-1171]* ≥8.3 ≥8.3 ≥8.3 ≥8.3 t last placebo [4.8-≥8.3]* [2.3-≥8.3]* [3.3-≥8.3] [1.4-≥8.3] ≥8.3 ≥8.3 ≥8.3 ≥8.3 h low [4.8-≥8.3]* [4.8-≥8.3]* [3.3-≥8.3] [4.8-≥8.3] ≥8.3 ≥8.3 ≥8.3 ≥8.3 high [≥8.3-≥8.3]* [4.8-≥8.3]* [0.82-≥8.3] [1.6-≥8.3] Data from 19 occasional-to-moderate cannabis smokers who participated in all dosing sessions (lower N reflects fewer participants with calculable ANOVA results due to negative placebo samples). See Table 29 (Supplemental) including Clast, and THC-glucuronide, cannabidiol, and cannabinol data. Statistical analysis performed by factorial repeated- measures analysis of variance. Boldface indicates statistical significance at p<0.05. Cannabis administered with Volcano® Medic vaporizer: 500 mg placebo (0.008±0.002% THC), low (2.9±0.14% THC) or high (6.7±0.05% THC) dose THC. Active alcohol dose was calculated to produce approximate peak 0.065% peak breath alcohol concentration.*Significant overall cannabis dose effect (p<0.05) by factorial repeated- measures analysis of variance [ANOVA]) ~Overall cannabis p<0.06 by factorial repeated-measures ANOVA. Post-hoc analysis revealed significant low- and high-vs.-placebo cannabis effect, but no significant low-vs.- high cannabis effect. #Significant overall alcohol dose effect by factorial repeated-measures ANOVA &Significant overall alcohol-cannabis effect by factorial repeated-measures ANOVA Abbreviations: THC, ∆9-tetrahydrocannabinol; LOQ, limit of quantification; 11-OH-THC, 11-hydroxy-THC; THCCOOH, 11-nor-9-carboxy-THC; Cmax, maximum concentration; Cmax- BL, Cmax accounting for baseline; tmax, time to maximum concentration; AUC0-8.3h, area under the 8.3h curve; AUC>BL-8.3h, AUC0-8.3h accounting for baseline; Clast, concentration at last detection.

220

Table 26. Effects of alcohol, cannabis, and alcohol*cannabis combination on blood cannabinoid pharmacokinetic parameters. Deg- Error Pairwise Overall rees of degrees Effect Analyte comparison by N F p-value effect free- of size, r parameter cannabis dose dom freedom THC Cmax Alcohol 19 8.03 1 18 0.56 0.011 Cannabis 42.84 1.21 21.73 <0.001a low vs. placebo 139.71 1 18 0.94 <0.001 high vs. placebo 57.23 1 18 0.87 <0.001 low vs. high 12.14 1 18 0.63 0.003 Alcohol* 1.91 1.15 20.74 0.182a Cannabis Cmax-BL Alcohol 18 8.03 1 17 0.57 0.011 Cannabis 42.00 1.21 20.62 <0.001a low vs. placebo 123.28 1 17 0.94 <0.001 high vs. placebo 55.74 1 17 0.88 <0.001 low vs. high 13.25 1 17 0.66 0.002 Alcohol* Cannabis 3.20 1.17 19.97 0.084a

tmax Alcohol 8 0.53 1 7 0.27 0.490 Cannabis 2.73 1.00 7.01 0.142a low vs. placebo 2.79 1 7 0.53 0.139 high vs. placebo 2.68 1 7 0.53 0.146 low vs. high 0.20 1 7 0.16 0.672 a Alcohol* 0.49 0.509 Cannabis 1.00 7.02 tlast Alcohol 8 1.46 1 7 0.42 0.266 Cannabis 9.18 1.15 8.04 0.014a low vs. placebo 10.11 1 7 0.77 0.016 high vs. placebo 9.34 1 7 0.76 0.018 low vs. high 0.61 1 7 0.28 0.461 Alcohol* 1.30 1.07 7.52 0.295a Cannabis AUC0-8.3h Alcohol 19 1.35 1 18 0.26 0.261 Cannabis 4.09 1.00 18.05 0.058a low vs. placebo 245.38 1 18 0.97 <0.001 high vs. placebo 5.13 1 18 0.47 0.036 low vs. high 2.53 1 18 0.35 0.129 Alcohol* 1.26 1.00 18.04 0.277a Cannabis AUC>BL-8.3h Alcohol 18 0.50 1 17 0.17 0.488 Cannabis 47.43 1.21 20.60 <0.001a low vs. placebo 119.56 1 17 0.94 <0.001 high vs. placebo 59.62 1 17 0.88 <0.001 low vs. high 17.18 1 17 0.71 0.001 Alcohol* Cannabis 0.63 1.27 21.55 0.473a 11-OH-THC Cmax Alcohol 19 9.95 1 18 0.60 0.005 Cannabis 28.88 1.16 20.81 <0.001a low vs. placebo 98.45 1 18 0.92 <0.001 high vs. placebo 38.44 1 18 0.83 <0.001 low vs. high 10.47 1 18 0.61 0.005

221

Table 26. (Continued from previous page) Deg- Error Pairwise Overall rees of degrees Effect Analyte comparison by N F p-value effect free- of size, r parameter cannabis dose dom freedom Alcohol* Cannabis 4.49 1.23 22.19 0.039a low vs. placebo 0.52 1 18 0.17 0.481 high vs. placebo 5.89 1 18 0.50 0.026 low vs. high 3.87 1 18 0.42 0.065 Cmax-BL Alcohol 18 8.50 1 17 0.58 0.010 Cannabis 29.61 1.16 19.74 <0.001a low vs. placebo 87.23 1 17 0.91 <0.001 high vs. placebo 39.09 1 17 0.83 <0.001 low vs. high 12.00 1 17 0.64 0.003 Alcohol* Cannabis 4.93 1.27 21.51 0.030a low vs. placebo 0.62 1 17 0.19 0.444 high vs. placebo 6.51 1 17 0.53 0.021 low vs. high 4.26 1 17 0.45 0.055 b tmax Alcohol low vs. high 16 1.63 1 15 0.31 0.221 Cannabis low vs. highb 0.09 1 15 0.08 0.769 Alcohol* 2.30 low vs. highb Cannabis 1 15 0.36 0.150 b tlast Alcohol low vs. high 16 0.01 1 15 0.03 0.910 Cannabis low vs. highb 16.35 1 15 0.72 0.001 Alcohol* low vs. highb Cannabis 4.81 1 15 0.50 0.043 AUC0-8.3h Alcohol 18 0.75 1 17 0.21 0.398 Cannabis 25.15 1.10 18.62 <0.001a low vs. placebo 53.57 1 17 0.87 <0.001 high vs. placebo 28.25 1 17 0.79 <0.001 low vs. high 14.08 1 17 0.67 0.002 Alcohol* Cannabis 0.60 1.20 20.37 0.475a AUC>BL-8.3h Alcohol 18 0.92 1 17 0.23 0.351 Cannabis 24.39 1.10 18.77 <0.001a low vs. placebo 63.20 1 17 0.89 <0.001 high vs. placebo 29.62 1 17 0.80 <0.001 low vs. high 13.60 1 17 0.67 0.002 Alcohol* Cannabis 0.10 1.29 21.99 0.823a THCCOOH Cmax Alcohol 19 0.03 1 18 0.04 0.871 Cannabis 27.35 1.39 25.02 <0.001a low vs. placebo 48.59 1 18 0.85 <0.001 high vs. placebo 46.38 1 18 0.85 <0.001 low vs. high 6.94 1 18 0.53 0.017 Alcohol* Cannabis 0.03 1.30 23.32 0.922a Cmax-BL Alcohol 18 0.00 1 17 0.00 0.995 Cannabis 26.34 1.44 24.43 <0.001a low vs. placebo 21.34 1 17 0.75 <0.001 high vs. placebo 32.78 1 17 0.81 <0.001 low vs. high 17.30 1 17 0.71 0.001

222

Table 26. (Continued from previous page) Deg- Error Pairwise Overall rees of degrees Effect Analyte comparison by N F p-value effect free- of size, r parameter cannabis dose dom freedom Alcohol* Cannabis 1.56 2 34 0.225 tmax Alcohol 13 0.56 1 12 0.21 0.470 Cannabis 1.46 1.03 12.40 0.250a low vs. placebo 1.33 1 12 0.32 0.271 high vs. placebo 1.61 1 12 0.34 0.229 low vs. high 0.82 1 12 0.25 0.383 Alcohol*

Cannabis 0.05 1.05 12.64 0.842a tlast Alcohol 13 0.25 1 12 0.14 0.628 Cannabis 4.10 1.04 12.43 0.064a low vs. placebo 4.50 1 12 0.52 0.055 high vs. placebo 3.81 1 12 0.49 0.075 low vs. high 0.60 1 12 0.22 0.455 Alcohol*

Cannabis 0.08 1.03 12.34 0.784a AUC0-8.3h Alcohol 19 0.18 1 18 0.10 0.675 Cannabis 17.94 1.49 26.87 <0.001a low vs. placebo 26.06 1 18 0.77 <0.001 high vs. placebo 36.45 1 18 0.82 <0.001 low vs. high 3.43 1 18 0.40 0.080 Alcohol* Cannabis 0.34 1.21 21.83 0.607a AUC>BL-8.3h Alcohol 18 0.12 1 17 0.08 0.731 Cannabis 10.30 1.42 24.21 0.002a low vs. placebo 4.18 1 17 0.44 0.057 high vs. placebo 13.07 1 17 0.66 0.002 low vs. high 13.56 1 17 0.67 0.002 Alcohol* Cannabis 1.32 2 34 0.282 THCCOOH-glucuronide Cmax Alcohol 19 0.50 1 18 0.16 0.490 Cannabis 16.46 1.46 26.31 <0.001a low vs. placebo 29.64 1 18 0.79 <0.001 high vs. placebo 31.94 1 18 0.80 <0.001 low vs. high 0.15 1 18 0.09 0.443 Alcohol* Cannabis 0.34 2 36 0.712 Cmax-BL Alcohol 18 1.03 1 17 0.24 0.325 Cannabis 17.98 2 34 <0.001 low vs. placebo 14.27 1 17 0.68 0.002 high vs. placebo 27.96 1 17 0.79 <0.001 low vs. high 8.52 1 17 0.58 0.010 Alcohol* Cannabis 1.18 2 34 0.318 tmax Alcohol 11 5.36 1 10 0.59 0.043 Cannabis 0.58 2 20 0.567 low vs. placebo 0.05 1 10 0.07 0.834 high vs. placebo 0.44 1 10 0.21 0.522 low vs. high 1.51 1 10 0.36 0.248

223

Table 26. (Continued from previous page) Deg- Error Pairwise Overall rees of degrees Effect Analyte comparison by N F p-value effect free- of size, r parameter cannabis dose dom freedom Alcohol*

Cannabis 0.25 2 20 0.780 tlast Alcohol 11 3.07 1 10 0.48 0.110 Cannabis 5.62 1.02 10.24 0.038a low vs. placebo 6.07 1 10 0.61 0.033 high vs. placebo 5.28 1 10 0.59 0.044 low vs. high 0.61 1 10 0.24 0.455 Alcohol*

Cannabis 1.74 1.06 10.63 0.216a AUC0-8.3h Alcohol 19 0.15 1 18 0.09 0.704 Cannabis 17.25 1.48 26.58 <0.001a low vs. placebo 37.23 1 18 0.82 <0.001 high vs. placebo 30.36 1 18 0.79 <0.001 low vs. high 1.67 1 18 0.29 0.212 Alcohol* Cannabis 0.66 1.52 27.36 0.487a AUC>BL-8.3h Alcohol 18 0.30 1 17 0.13 0.591 Cannabis 15.07 2 34 <0.001 low vs. placebo 8.93 1 17 0.59 0.008 high vs. placebo 20.68 1 17 0.74 <0.001 low vs. high 10.77 1 17 0.62 0.004 Alcohol* Cannabis 1.56 2 34 0.225 Data from 19 occasional-to-moderate cannabis smokers who participated in all dosing sessions (lower N reflects fewer participants with calculable ANOVA results due to negative placebo samples). See Table 30 (Supplemental) including Clast, and THC-glucuronide, cannabidiol, and cannabinol data. Statistical analysis performed by factorial repeated-measures analysis of variance. Boldface indicates statistical significance at p<0.05. Cannabis administered with Volcano® Medic vaporizer: 500 mg placebo (0.008±0.002% THC), low (2.9±0.14% THC) or high (6.7±0.05% THC) dose THC. Active alcohol dose was calculated to produce approximate peak 0.065% peak breath alcohol concentration. aMauchly’s test showed sphericity was violated on main effects, so Greenhouse-Geisser correction was utilized. bPlacebo doses not included in ANOVA due to too few positive specimens for comparison. Abbreviations: THC, ∆9-tetrahydrocannabinol; LOQ, limit of quantification; 11-OH-THC, 11-hydroxy- THC; THCCOOH, 11-nor-9-carboxy-THC; Cmax, maximum concentration; Cmax-BL, Cmax accounting for baseline; tmax, time to maximum concentration; Clast, concentration at last detection; tlast, time of last detection; AUC0-8.3h, area under the 8.3h curve; AUC>BL-8.3h, AUC0-8.3h accounting for baseline.

224

Table 27. Effects of alcohol, cannabis, and alcohol*cannabis combination on plasma cannabinoid pharmacokinetic parameters. Deg- Error Pairwise Analyte Overall rees of degrees Effect comparison by N F p-value parameter effect freedo of size, r cannabis dose m freedom THC Cmax Alcohol 19 5.20 1 18 0.47 0.035 Cannabis 40.28 1.17 20.99 <0.001a low vs. placebo 143.53 1 18 0.94 <0.001 high vs. placebo 53.52 1 18 0.87 <0.001 low vs. high 13.05 1 18 0.65 0.002 Alcohol* Cannabis 1.72 1.19 21.47 0.205a Cmax-BL Alcohol 18 5.32 1 17 0.49 0.034 Cannabis 37.64 1.19 20.25 <0.001a low vs. placebo 105.24 1 17 0.93 <0.001 high vs. placebo 50.14 1 17 0.86 <0.001 low vs. high 13.99 1 17 0.67 0.002 Alcohol* Cannabis 3.08 1.22 20.69 0.088a tmax Alcohol 11 4.53 1 10 0.56 0.059 Cannabis 4.75 1.00 10.01 0.054a low vs. placebo 4.85 1 10 0.57 0.052 high vs. placebo 4.66 1 10 0.56 0.056 low vs. high 0.24 1 10 0.15 0.636 Alcohol* Cannabis 4.43 1.00 10.01 0.062a tlast Alcohol 11 0.02 1 10 0.04 0.890 Cannabis 6.43 1.16 11.55 0.024a low vs. placebo 6.64 1 10 0.63 0.028 high vs. placebo 6.89 1 10 0.64 0.025 low vs. high 0.00 1 10 0.01 0.981 Alcohol* Cannabis 1.65 2 20 0.216 AUC0-8.3h Alcohol 19 1.35 1 18 0.26 0.261 Cannabis 4.09 1.00 18.05 0.058a low vs. placebo 245.38 1 18 0.97 <0.001 high vs. placebo 5.13 1 18 0.47 0.036 low vs. high 2.53 1 18 0.35 0.129 Alcohol* Cannabis 1.26 1.00 18.04 0.277a AUC>BL-8.3h Alcohol 18 1.39 1 17 0.27 0.255 Cannabis 42.73 1.15 19.57 <0.001a low vs. placebo 144.09 1 17 0.95 <0.001 high vs. placebo 54.40 1 17 0.87 <0.001 low vs. high 15.63 1 17 0.69 0.001 Alcohol* Cannabis 1.49 1.24 21.03 0.242a 11-OH-THC Cmax Alcohol 19 6.12 1 18 0.50 0.024 Cannabis 31.30 1.22 21.90 <0.001a low vs. placebo 73.17 1 18 0.90 <0.001 high vs. placebo 39.70 1 18 0.83 <0.001 low vs. high 12.24 1 18 0.64 0.003

225

Table 27. (Continued from previous page) Deg- Error Pairwise Analyte Overall rees of degrees Effect comparison by N F p-value parameter effect freedo of size, r cannabis dose m freedom Alcohol* Cannabis 2.77 1.34 24.15 0.100a Cmax-BL Alcohol 18 3.31 1 17 0.40 0.087 Cannabis 33.26 1.23 20.88 <0.001a low vs. placebo 60.95 1 17 0.88 <0.001 high vs. placebo 41.89 1 17 0.84 <0.001 low vs. high 15.74 1 17 0.69 0.001 Alcohol* Cannabis 3.57 1.49 25.30 0.055a b tmax Alcohol low vs. high 17 2.35 1 16 0.36 0.145 Cannabis low vs. highb 0.13 1 16 0.09 0.724 Alcohol* low vs. highb Cannabis 0.17 1 16 0.10 0.683 b tlast Alcohol low vs. high 17 3.37 1 16 0.42 0.085 Cannabis low vs. highb 4.04 1 16 0.45 0.062 Alcohol* low vs. highb Cannabis 0.65 1 16 0.20 0.432 AUC0-8.3h Alcohol 19 1.06 1 18 0.24 0.317 Cannabis 28.02 1.13 20.27 <0.001a low vs. placebo 75.29 1 18 0.90 <0.001 high vs. placebo 32.97 1 18 0.80 <0.001 low vs. high 12.54 1 18 0.64 0.002 Alcohol* Cannabis 1.92 1.21 21.73 0.179a AUC>BL-8.3h Alcohol 18 0.94 1 17 0.23 0.346 Cannabis 29.53 1.13 19.22 <0.001a low vs. placebo 82.54 1 17 0.91 <0.001 high vs. placebo 35.84 1 17 0.82 <0.001 low vs. high 13.51 1 17 0.67 0.002 Alcohol* Cannabis 1.10 1.40 23.85 0.327a THCCOOH Cmax Alcohol 19 0.01 1 18 0.03 0.910 Cannabis 26.04 1.52 27.30 <0.001a low vs. placebo 40.06 1 18 0.83 <0.001 high vs. placebo 49.99 1 18 0.86 <0.001 low vs. high 4.78 1 18 0.46 0.042 Alcohol* Cannabis 0.22 1.40 25.21 0.726a Cmax-BL Alcohol 18 0.65 1 17 0.19 0.431 Cannabis 44.15 1.15 19.50 <0.001a low vs. placebo 163.82 1 17 0.95 <0.001 high vs. placebo 55.51 1 17 0.87 <0.001 low vs. high 14.56 1 17 0.68 0.001 Alcohol* Cannabis 0.83 1.34 22.84 0.405a tmax Alcohol 13 0.56 1 12 0.21 0.470 Cannabis 1.46 1.03 12.40 0.250a low vs. placebo 1.33 1 12 0.32 0.271 high vs. placebo 1.61 1 12 0.34 0.229 low vs. high 0.82 1 12 0.25 0.383 226

Table 27. (Continued from previous page) Deg- Error Pairwise Analyte Overall rees of degrees Effect comparison by N F p-value parameter effect freedo of size, r cannabis dose m freedom Alcohol*

Cannabis 0.05 1.05 12.64 0.842a tlast Alcohol 14 0.03 1 13 0.05 0.858 Cannabis 2.51 1.03 13.41 0.136a low vs. placebo 2.73 1 13 0.42 0.123 high vs. placebo 2.33 1 13 0.39 0.151 low vs. high 0.56 1 13 0.20 0.467 Alcohol*

Cannabis 0.01 1.03 13.37 0.941a AUC0-8.3h Alcohol 19 0.17 1 18 0.10 0.689 Cannabis 19.47 2 36 <0.001 low vs. placebo 22.40 1 18 0.74 <0.001 high vs. placebo 48.87 1 18 0.85 <0.001 low vs. high 2.51 1 18 0.35 0.130 Alcohol* Cannabis 0.05 1.35 24.23 0.886a AUC>BL-8.3h Alcohol 18 0.02 1 17 0.03 0.888 Cannabis 29.55 1.22 20.71 <0.001a low vs. placebo 81.28 1 17 0.91 <0.001 high vs. placebo 39.24 1 17 0.84 <0.001 low vs. high 9.85 1 17 0.61 0.006 Alcohol* Cannabis 0.25 1.19 20.16 0.662a THCCOOH-glucuronide Cmax Alcohol 19 0.89 1 18 0.22 0.358 Cannabis 20.03 2 36 <0.001 low vs. placebo 19.77 1 18 0.72 0.001 high vs. placebo 28.55 1 18 0.78 <0.001 low vs. high 7.93 1 18 0.55 0.011 Alcohol* Cannabis 0.95 2 36 0.397 Cmax-BL Alcohol 18 0.10 1 17 0.08 0.759 Cannabis 18.67 2 34 <0.001 low vs. placebo 35.79 1 17 0.82 <0.001 high vs. placebo 33.41 1 17 0.81 <0.001 low vs. high 1.91 1 17 0.32 0.185 Alcohol* Cannabis 0.82 1.36 23.10 0.410a tmax Alcohol 12 1.07 1 11 0.30 0.323 Cannabis 0.35 2 22 0.712 low vs. placebo 0.11 1 11 0.10 0.751 high vs. placebo 0.55 1 11 0.22 0.474 low vs. high 0.27 1 11 0.16 0.612 Alcohol*

Cannabis 0.57 2 22 0.574 tlast Alcohol 12 0.16 1 11 0.12 0.693 Cannabis 1.04 2 22 0.371 low vs. placebo 2.44 1 11 0.43 0.147 high vs. placebo 0.00 1 11 0.01 0.975 low vs. high 2.12 1 11 0.40 0.173

227

Table 27. (Continued from previous page) Deg- Error Pairwise Analyte Overall rees of degrees Effect comparison by N F p-value parameter effect freedo of size, r cannabis dose m freedom Alcohol*

Cannabis 0.00 2 22 0.998 AUC0-8.3h Alcohol 19 0.88 1 18 0.22 0.362 Cannabis 11.87 1.23 22.16 0.001a low vs. placebo 22.55 1 18 0.75 <0.001 high vs. placebo 18.63 1 18 0.71 <0.001 low vs. high 4.59 1 18 0.45 0.046 Alcohol* Cannabis 1.21 1.38 24.87 0.299a AUC>BL-8.3h Alcohol 18 2.60 1 17 0.36 0.125 Cannabis 15.76 2 34 <0.001 low vs. placebo 26.93 1 17 0.78 <0.001 high vs. placebo 24.79 1 17 0.77 <0.001 low vs. high 3.23 1 17 0.40 0.090 Alcohol* Cannabis 0.40 1.45 24.62 0.609a Data from 19 occasional-to-moderate cannabis smokers who participated in all dosing sessions (lower N reflects fewer participants with calculable ANOVA results due to negative placebo samples). See Table 31 (Supplemental) including Clast and THC-glucuronide, cannabidiol, and cannabinol data. Statistical analysis performed by factorial repeated-measures analysis of variance. Boldface indicates statistical significance at p<0.05. Cannabis administered with Volcano® Medic vaporizer: 500 mg placebo (0.008±0.002% THC), low (2.9±0.14% THC) or high (6.7±0.05% THC) dose THC. Active alcohol dose was calculated to produce approximate peak 0.065% peak breath alcohol concentration. aMauchly’s test showed sphericity was violated on main effects, so Greenhouse-Geisser correction was utilized. aPlacebo doses not included in ANOVA due to too few positive specimens for comparison. Abbreviations: THC, ∆9-tetrahydrocannabinol; LOQ, limit of quantification; 11-OH-THC, 11- hydroxy-THC; THCCOOH, 11-nor-9-carboxy-THC; CBD, cannabidiol; CBN, cannabinol; Cmax, maximum concentration; Cmax-BL, Cmax accounting for baseline; tmax, time to maximum concentration; Clast, concentration at last detection; tlast, time of last detection; AUC0-8.3h, area under the 8.3 h curve; AUC>BL-8.3h, AUC0-8.3h accounting for baseline.

228

Figure 13. Median [interquartile range] blood and plasma cannabinoids after cannabis vaporization (N=19).THC content: placebo 0.008±0.002%, low 2.9±0.14%, high 6.7±0.05%. (*)Overall p<0.005 (Friedman’s ANOVA), p<0.005, N=19; (#) overall p<0.05; (a)p<0.006 (placebo-vs.- high, no alcohol); (b)p<0.006 (placebo-vs.-high, with alcohol); (c)p<0.006 (placebo-vs.-low, no alcohol); (d)p<0.006 (placebo-vs.-low, with alcohol); (e)difference<0.006 (low-vs.-high, no alcohol); (f)p<0.006 (low vs. high, alcohol).

229

time course for THCCOOH, and beginning at 1.4 h for THCCOOH-glucuronide. Only H- and H+ showed significant THCCOOH-glucuronide differences (post-hoc analysis) relative to P- and P+. Plasma observations were similar to blood (Figure 13, Figure 16

(Supplemental)). Figure 13 and Figure 15-Figure 16 (Supplemental) present post-hoc within-subjects dose differences at individual collection times. No significant blood or plasma L- vs. -H cannabinoid differences were observed at any discrete time point for any analyte, except plasma CBD immediately post-inhalation (Figure 13). Median THC- glucuronide, CBD and CBN tlast occurred within 0.5 h following inhalation. For all dosing conditions, ≥10.5% completers had blood THC ≥1 μg/L at baseline, and ≥16.7% throughout 8.3 h post-dose, even after P cannabis (Figure 14). With 2 μg/L blood THC cutoff, 5.3-10.5% were positive at baseline for all doses, and only 1 participant was positive after 0.42 h for P-. By 3.3 h, <50% were positive after any dose (Figure 14). In this occasional-to-moderate smoker cohort, 0-5.9% completers had blood THC ≥5 μg/L at baseline (all conditions), with ≤21.1% THC ≥5 μg/L by 2.3 h. Thus, 78.9% of occasional to moderate cannabis smokers were negative after only 2.3 h with a THC ≥5

μg/L cutoff.

Pharmacokinetic parameters for all participants are presented in Table 32

(Supplemental), Table 33 (Supplemental), Table 34 (Supplemental), Table 35

(Supplemental), Table 36 (Supplemental), Table 37 (Supplemental), and Table 38

(Supplemental). There were no significant differences (p>0.44, Mann-Whitney U [exact] test) between completers and non-completers in cannabis smoking history, age, weight, or body mass index. High inter-individual variability was observed in THC, 11-OH-THC,

THCCOOH, and THCCOOH-glucuronide concentrations. Observed THC, 11-OH-THC,

230

Figure 14. Percent completers (N=19) positive for ∆9- tetrahydrocannabinol (THC) in whole blood for various cutoffs after controlled administration of vaporized cannabis (placebo, low [2.9%], and high [6.7%] THC) and alcohol (placebo and active).

231

CBD and CBN tmax occurred immediately post-inhalation, whereas THCCOOH, THC- glucuronide and THCCOOH-glucuronide tmax reflected additional time needed for further metabolism. After active doses, full-study population median THC and 11-OH-THC observed tlast occurred 3.5-6.4 h and 1.4-3.3 h, respectively. Median THCCOOH and

THCCOOH-glucuronide tlast extended ≥8.3 h. CBD and CBN tlast occurred within 0.5 h.

Based on pharmacokinetic data, Participant 25 may have accessed active cannabis during his P+ session, despite being under observation throughout his stay (Figure 17

(Supplemental)). Blood and plasma THC Cmax were 18.5 and 25.6 μg/L. This participant’s oral fluid indicated he was negative on admission the night prior to dosing but positive just before dosing. It is possible these high concentrations resulted from dosing error; however, there was no indication from careful record review that an error occurred. These data were excluded from pharmacokinetic data analysis.

Blood/Plasma ratios

Median [range] blood/plasma ratios were 0.71 [0.13-1.5] (n=684) THC, 0.73

[0.42-1.4] (n=409) 11-OH-THC, 0.65 [0.39-1.5] (n=1,112) THCCOOH, 0.55 [0.40-1.3]

(n=12) THC-glucuronide, 0.80 [0.13-7.9] (n=926) THCCOOH-glucuronide, 0.73 [0.48-

1.0] (n=31) CBD, and 0.86 [0.49-1.3] (n=71) CBN. THC and metabolites’ blood/plasma ratios did not vary by time or dose (Figure 18 (Supplemental)).

THCCOOH-glucuronide/THCCOOH ratios

Blood and plasma THCCOOH-glucuronide/THCCOOH ratios decreased immediately (within the first half hour post-dose) after inhaling active cannabis,

232

subsequently rising, with substantial interindividual variability (Figure 19

(Supplemental)). Alcohol, cannabis and cannabis*time all significantly affected

THCCOOH-glucuronide/THCCOOH in blood [F(1,72)=8.173, p=0.006;

F(1.71,123.06)=24.17, p<0.001; and F(1.71,123.06)=15.12, p<0.001, respectively] and plasma [F(1,69)=10.51, p=0.002; F(2,138)=8.01, p=0.001; and F(2,138)=5.542, p=0.005, respectively]. Active alcohol (+) conditions produced higher THCCOOH- glucuronide/THCCOOH ratios than placebo alcohol (-).

Discussion

Here we obtained complete data for blood and plasma phase I and II cannabinoid concentrations following vaporized cannabis, with and without low-dose alcohol.

Inhaling vaporized bulk cannabis produced blood and plasma cannabinoid concentrations and pharmacokinetic curves similar to smoking (78, 93-94). Desrosiers et al (93) recently observed 34.4 [16.5-49.5] μg/L blood THC Cmax in 14 frequent smokers (≥4x/week) 0.5 h after smoking one 6.8% THC cigarette, similar to our occasional smokers’ L dose (500 mg, 2.9% THC) at tmax=0.17 h (32.7 [11.4-66.2] μg/L THC). However, inhaled THC concentrations peak prior to the last puff, rapidly decreasing as lipophilic THC is distributed to the tissues and rapidly metabolized (14). Thus, our 0.42 h post-dose time is comparable to Desrosiers’ 0.5 h, producing median [range] THC of 10.0 [1.6-17.9] μg/L

(L) and 13.2 [2.4-40.8] μg/L (H) and better illustrating occasional-vs.-frequent smoker differences. An occasional smoker cohort had smoked THC Cmax 12.1 [4.1-40.3] μg/L

(16), similar to our ~0.5 h findings. The only prior direct comparison of cannabis vaporization and smoking examined within-subjects plasma THC after 1.7%, 3.4%, and

233

6.8% THC (78). Cigarettes were halved; half were smoked, the other half vaporized. The two routes produced similar plasma THC concentrations and 6 h AUCs. Pulmonary THC intake after vaporization is similar to smoking (83), with ~54% of the THC dose delivered to the balloon for inhalation, and 30-40% exhaled. Smoking cannabis also has factors decreasing THC delivery relative to dose. Approximately 23-30% THC is lost by pyrolysis, and 40-50% as side-stream smoke (295). Our blood and plasma study corroborates evidence that vaporization is an effective alternative administration route

(mitigating health concerns from combustion byproduct inhalation due to the lower vaporization temperature), delivering THC in a similar manner to smoking and producing similar cannabinoid concentration profiles.

Participants inhaled ad libitum by controlling inhalation rate, depth, and hold time in the lungs (inhalation topography, allowing individual self-titration based on pharmacological response) (266), contributing to substantial inter-individual variability in cannabinoid concentration profiles. Significantly higher Cmax and AUC0-8.3h were observed for THC, THC-glucuronide, 11-OH-THC, THCCOOH, THCCOOH- glucuronide (only when accounting for baseline) and CBD, after H- vs. -L cannabis.

However, 52.6% of completers’ within-subject blood THC Cmax indicated self-titration:

21.0% had L and H Cmax within 20% of each other for ≥1 alcohol condition, and 31.6% had higher Cmax after L than H doses. For most compounds, noticeable median/range differences for the same THC potency with (+) and without (-) alcohol (Table 25 and

Table 29 (Supplemental)) generally occurred only after the H dose; L doses produced consistent Cmax and AUCs. This also supports self-titration: participants required less self-titration at the L dose to achieve intended results, likely consuming the full amount.

234

More variability after the H dose suggests greater self-titration. Apart from inhalation topography, factors affecting vaporized THC delivery include heating temperature, number of balloon fillings, cannabis amount and blend, and length of time between volatilization and inhalation (due to possible THC adherence to the balloon) (69, 83,

238).

Most previous cannabis and alcohol concentrations were reported from roadside drugged driving prevalence studies, providing no information about possible cannabinoid pharmacokinetic differences with alcohol (21). Some controlled-administration experiments provided limited cannabis and alcohol pharmacokinetic data in relation to performance impairment assessments (296-297). We showed significantly higher THC

Cmax with alcohol in blood and plasma, and additionally for other cannabinoids. Alcohol- cannabis interactions were statistically significant in blood 11-OH-THC Cmax, but not plasma, limiting conclusions from this observation. One study (169, 298) directly examined combined alcohol and cannabis pharmacokinetics in chronic smokers; but with only one cannabis dose (400 μg/kg THC) and three alcohol conditions (placebo, ~0.05%, and ~0.07% BAC). Alcohol prior to smoking did not significantly affect THC Cmax (169,

298). Similar results were reported in another study, which also found no significant differences in plasma THC Cmax or AUC after ingesting 420 and 850 mg/kg alcohol vs. placebo alcohol (cannabis smoked 0.25 h post-alcohol) (285). Plasma THC increased non-significantly but dose-dependently with increasing alcohol. Plasma THC 0.3 h post- smoking was “generally higher” 0.8 h after alcohol than without alcohol (296), but no statistics were provided. Moderate alcohol (0.35 g/kg) produced significantly higher plasma THC within 15 min after start of smoking, but significant differences were not

235

observed over a full 90-min THC curve (287). Contrastingly, 0.7 g/kg alcohol produced significantly lower serum THC 1 h post-smoking (298). Generally, these results corroborate ours, as observed THC Cmax occurred immediately post-inhalation (within 15 min).

Because alcohol increased THC and 11-OH-THC Cmax but not AUC0-8.3h (even accounting for baseline), possibly alcohol affected absorption (higher concentrations immediately post-inhalation). Possible alcohol-induced perfusion and distribution changes affect other drugs (298-299). Acute alcohol increases cardiac output within 30 min (284), possibly leading to more rapid THC absorption during inhalation due to increased pulmonary capillary flow. In contrast to prior studies with a time-delay [≥0.3 h] between alcohol and cannabis to allow for alcohol absorption (169, 287, 296-298), the present experiment administered cannabis and alcohol concurrently; the entire dosing process required ≤20 min. Our approach retains real-world validity for recreational intake. It also is possible that our higher blood cannabinoid Cmax reflects less careful cannabis self-titration after alcohol.

Overall, we observed minimal alcohol effects on THC metabolism. Higher blood and plasma 11-OH-THC Cmax (Table 25) could be due to increased metabolism, but probably results from higher THC Cmax. Blood THCCOOH-glucuronide Cmax occurred earlier with alcohol (+), but plasma THCCOOH-glucuronide did not. Non-glucuronidated

THCCOOH tmax was unaffected by alcohol or cannabis condition in either matrix.

Although the alcohol-cannabis interaction on metabolite 11-OH-THC tlast in blood was statistically significant, it was based only on L and H cannabis doses (too many negative samples after P) and no clear pattern emerged. Thus, it does not appear to be clinically

236

significant. No alcohol differences emerged at specific collection times, and there were no alcohol effects on THCCOOH concentrations. Limited other data are available on alcohol effects on cannabinoid metabolites (169, 298). Although lower THCCOOH was observed after alcohol than placebo alcohol over 4 h (298), the effect was not significant due to inter-individual variability from prior cannabis smoking history. Our observations were similar (Figure 15 (Supplemental)-Figure 16 (Supplemental)).

Participants 7, 13 and 22 had ≥10 μg/L blood THCCOOH and ≥40 μg/L blood

THCCOOH-glucuronide at baseline in ≥4 sessions, and at least one baseline blood THC

≥1.4 μg/L. Participant 7 additionally had one session with baseline 11-OH-THC 1.0 μg/L.

In all six sessions, participants 9 and 31 had ≥72.4 μg/L baseline THCCOOH- glucuronide, ≥17.9 μg/L THCCOOH, and ≥1.4 μg/L THC. These five participants were likely the most frequent smokers in our cohort. Fabritius et al (278) recently proposed that free blood THCCOOH thresholds differentiated occasional (≤3 μg/L) from frequent

(≥40 μg/L) cannabis smokers, although 38.7% of occasional smokers’ specimens had

THCCOOH >3 μg/L and 83.6% of frequent smokers’ specimens had THCCOOH ≤40

μg/L. By these criteria, 52.6% of our completers would be considered occasional smokers; the others fell between categories. Other factors in THCCOOH and

THCCOOH-glucuronide interpretation include metabolism time and residual cannabinoids (acute vs. chronic exposure). Observed THCCOOH-glucuronide tmax occurred later (median ≥1.4 h) than THCCOOH tmax (median <0.5 h) (Table 25), due to the additional phase II metabolic process. THCCOOH and THCCOOH-glucuronide median/range concentrations were considerably lower accounting for baseline, highlighting the effect of residual cannabinoid presence. THCCOOH-

237

glucuronide/THCCOOH ratios and variability (Figure 19 (Supplemental)) were similar to after smoking in occasional smokers (16).

This study has multiple strengths. With extensive vaporized cannabis pharmacokinetic data, we confirm the utility of vaporization as a viable and effective cannabis-smoking alternative. We also characterize cannabinoid blood and plasma pharmacokinetics with concurrent alcohol, and using grey-top Vacutainers, the collection device most commonly employed in forensic settings. Alcohol effects on cannabinoid pharmacokinetics are of interest due to the commonality of co-ingestion. Combining these drugs affects performance impairment (294), possibly in part due to higher cannabinoid concentrations. Our data provide a valuable pharmacokinetic reference for clinicians regarding future therapeutic use of vaporized cannabis. We also explicitly illustrate individual variability in inhalation behavior, documenting evidence of self- titration in half of participants. Increasing THC potency impacts people differently, depending on cannabis use history. An additional strength is inclusion of phase II THC- and THCCOOH-glucuronides, as well as minor cannabinoids CBD and CBN. Limited blood and plasma controlled-administration data exist for these compounds (93-94, 104,

126). Metabolites provide valuable information on smoking history and time since last intake (277-278). No study to date examined phase II metabolites following vaporization and alcohol; these data improve blood and plasma interpretation by toxicologists as medical and recreational cannabis prevalence expands. THC-glucuronide is detected at low concentrations, within 0.5 h post-smoking. CBD and CBN were not detected after

0.42 h in this study, so these compounds have utility as recent-use markers in blood. No known study to date detected CBD or CBN in blood or plasma after 2.1 h post-inhalation

238

(93-94), although controlled smoked administration studies usually contained low (≤1 mg) CBD and CBN doses. Blood collection may be delayed after an accident or traffic stop (142), making it unlikely to detect these compounds. Karschner et al (104) reported

CBD tmax 1.0-5.5 h after Sativex (1:1 CBD:THC oromucosal spray, 5 and 15 mg CBD).

This highlights CBD relevance in forensic cannabinoid testing, given increasing medical cannabis prevalence. We recommend controlled administration studies of smoked and vaporized high-CBD cannabis strains, utilized for antiepileptic, antiemetic, anti- inflammatory, and antipsychotic effects (99, 300).

Study limitations include blood and plasma collections for only 8.3 h.

Additionally, we did not directly compare vaporized cannabis to smoking to fully evaluate relative bioavailability. This investigation focused upon participants with self- reported occasional-to-moderate cannabis intake histories; additional research is needed to characterize vaporized cannabis and alcohol pharmacokinetics in chronic frequent smokers.

Different THC cutoffs yielded different positivity rates (Figure 14). At 1 μg/L,

THC was positive in ≥42.1% of participants 4.8 h after active (L and H) and ≥27.8% after

P, due to residual THC from previous self-administration. With THC ≥2 μg/L, 10.5-

15.8% were positive 3.3 h after L and 36.8-42.1% after H doses. THC ≥5 μg/L cutoffs resulted in only one THC-positive participant at 3.3 h. We expect positivity rates to be higher and for longer post-vaporization in frequent smokers (108), warranting investigation. These debated per se cutoffs yield different detection windows in these occasional-to-moderate smokers, with 2 μg/L limiting the window to ~4.8 h post-dose

(Figure 14), similar to the window of acute intoxication (41). A higher 5 μg/L cutoff

239

results in a short detection window for occasional-to-moderate smokers—shorter than impairment windows (41, 294)—emphasizing the challenge in establishing appropriate science-based per se cannabis drugged driving legislation.

Table 28 (Supplemental). Mean (standard deviation) approximate ∆9- tetrahydrocannabinol (THC), cannabidiol (CBD), and cannabinol (CBN) content in placebo, low-, and high-THC bulk cannabis doses administered by vaporizer balloon. THC CBD CBN

% mg % mg % mg 0.008 0.04 0.001 0.005 0.009 0.045 Placebo (0.002) (0.01) (0.001) (0.005) (0.003) (0.015) 2.9 14.5 0.05 0.25 0.22 1.1 Low (0.14) (0.7) (0.00) (0.0) (0.02) (0.1) 6.7 33.5 0.19 0.95 0.37 1.85 High (0.05) (0.25) (0.01) (0.05) (0.03) (0.15)

Table 29 (Supplemental). Median [range] blood and plasma pharmacokinetic parameters (all analytes; including last measured concentration) following controlled vaporized cannabis administration with and without oral alcohol. THC Blood Plasma (LOQ 1 μg/L) No Alcohol Alcohol No Alcohol Alcohol 2.1 0.6 3.2 1.4 # # # # Cmax placebo [0-7.6]* [0-5.2]* [0-9.8]* [0-9.6]* 32.7 35.3 46.5 48.6 μg/L low [11.4-66.2]*# [13.0-71.4]*# [16.6-114]*# [21.7-102]*# 42.2 67.5 62.1 97.8 high [15.2-137]*# [18.1-210]*# [23.6-196]*# [24.5-339]*# 1.7 0 2.4 0.7 C max-BL placebo [0-5.3]*# [0-3.2]*# [0-9.2]*# [-0.7-9.6]*# 32.7 35.3 45.7 48.6 μg/L low [11.4-66.2]*# [8.1-71.4]*# [16.6-113]*# [2.3-102]*# 42.2 67.5 62.1 96.1 high [15.2-137]*# [18.1-204]*# [23.6-196]*# [24.5-332]*# 0.17 0.18 0.17 0.22 t max placebo [0.15-6.3] [0.07-≥8.3] [0.15-6.3] [0.07-≥8.3] 0.17 0.17 0.17 0.17 h low [0.15-0.33] [0.15-0.25] [0.15-0.33] [0.15-0.25] 0.17 0.17 0.17 0.17 high [0.15-0.3] [0.12-0.37] [0.15-0.30] [0.12-0.37]

240

Table 29 (Supplemental). (Continued from previous page) THC Blood Plasma (LOQ 1 μg/L) No Alcohol Alcohol No Alcohol Alcohol 1.1 0.3 3.4 1.3 AUC 0-8.3h placebo [0-53.2]~ [0-36.4]~ [0-66.6]~ [0-103623]~ 31.9 36.2 44.6 49.4 h*μg/L low [10.6-84.2]~ [18.0-52.2]~ [14.1-124]~ [26.9-80]~ 43.1 62.2 56.2 93.2 high [10.6-113]~ [13.2-1445]~ [15.9-182]~ [19.4-2370]~ AUC>BL- 0.6 0 1.4 0.3 8.3h placebo [0-7.0]* [0-19.6]* [0-7.7]* [0-7.4]* 21.7 18.7 29.2 24.9 h*μg/L low [6.9-38.4]* [7.6-33.4]* [9.3-56.5]* [14.1-49.3]* 29.4 33.7 43.4 51.6 high [6.8-77.9]* [8.8-83.5]* [9.7-124]* [14.1-132]* 0.42 4.8 0.4 4.3 t last placebo [0.15-≥8.3]* [0.17-≥8.3]* [0.15-≥8.3]* [0.17-≥8.3]* 3.5 3.5 4.8 6.3 h low [0.70-≥8.3]* [1.3-≥8.3]* [0.70-≥8.3]* [1.3-≥8.3]* 4.8 3.7 6.3 6.4 high [0.82-≥8.3]* [1.4-≥8.3]* [1.3-≥8.3]* [1.4-≥8.3]* Clast placebo 1.4 [1.0-5.3] 1.8 [1.1-5.0] 1.5 [1.0-8.3] 1.6 [1.0-7.4] μg/L low 1.3 [1.0-7.0] 1.4 [1.0-5.3] 1.6 [1.0-12.0] 1.3 [1.0-8.6] high 1.5 [1.1-6.5] 1.6 [1.0-7.3] 1.8 [1.1-11.2] 1.6 [1.0-9.6]

THC-glucuronide Blood Plasma (LOQ 0.5 μg/L) No Alcohol Alcohol No Alcohol Alcohol

Cmax placebo 0 [0-0]* 0 [0-0]* 0 [0-0.8]* 0 [0-0]* μg/L low 0 [0-0]* 0 [0-0.6]* 0 [0-0.8]* 0 [0-0.8]* high 0 [0-0.8]* 0 [0-0.8]* 0.6 [0-1.7]* 0.6 [0-2.0]* Cmax-BL placebo 0 [0-0]* 0 [0-0]* 0 [0-0.2]* 0 [0-0]* μg/L low 0 [0-0]* 0 [0-0.6]* 0 [0-0.8]* 0 [0-0.8]* high 0 [0-0.8]* 0 [0-0.8]* 0.6 [0-1.7]* 0.6 [0-2.0]* tmax placebo n/a n/a 1.27 [n=1] n/a 0.40 0.43 0.42 h low n/a [n=1] [0.33-0.45] [0.22-0.48] 0.40 0.40 0.40 0.40 high [0.25-0.67] [0.38-0.53] [0.40-0.67] [0.37-0.53] AUC0-8.3h placebo 0 [0-0]* 0 [0-0]* 0 [0-1.7]* 0 [0-0]* h*μg/L low 0 [0-0]* 0 [0-0.4]* 0 [0-0.8]* 0 [0-0.6]* high 0 [0-0.5]* 0 [0-0.5]* 0.40 [0-1.5]* 0.3 [0-1.5]* AUC>BL- 8.3h placebo 0 [0-0]* 0 [0-0]* 0 [0-0.20]* 0 [0-0]* h*μg/L low 0 [0-0]* 0 [0-0.3]* 0 [0-0.8]* 0 [0-0.6]* high 0 [0-0.5]* 0 [0-0.5]* 0.40 [0-1.5]* 0.31 [0-1.5]*

241

Table 29 (Supplemental). (Continued from previous page) THC-glucuronide Blood Plasma (LOQ 0.5 μg/L) No Alcohol Alcohol No Alcohol Alcohol tlast placebo n/a n/a 3.3 [n=1] n/a 0.40 0.44 0.42 h low n/a [n=1] [0.42-0.58] [0.22-0.48] 0.40 0.40 0.41 0.42 high [0.25-0.67] [0.38-0.53] [0.40-1.3] [0.37-1.4] Clast placebo n/a n/a 0.5 [n=1] n/a μg/L low n/a 0.6 [n=1] 0.6 [0.5-0.8] 0.6 [0.5-0.8] high 0.6 [0.5-0.8] 0.6 [0.6-0.8] 0.8 [0.5-1.7] 0.7 [0-2.0]

11-OH-THC Blood Plasma (LOQ 1 μg/L) No Alcohol Alcohol No Alcohol Alcohol 0 0 0 0 #& #& # # Cmax placebo [0-2.5]* [0-2.4]* [0-4.3]* [0-3.2]* 2.8 3.7 4.1 4.8 μg/L low [0-9.1]*#& [1.4-6.0]*#& [0-13.7]*# [1.3-8.0]*# 5.0 6.0 7.0 7.5 high [0-14.2]*#& [0-24.8]*#& [1.0-20.3]*# [0-27.3]*# 0 0 0 0 C max-BL placebo [0-1.1]*#& [0-1.4]*#& [0-1.1]* [0-1.8]* 2.8 3.3 3.7 4.4 μg/L low [0-9.1]*#& [1.4-6.0]*#& [0-13.7]* [1.3-8.0]* 5.0 6.0 7.0 7.5 high [0-12.8]*#& [0-23.3]*#& [1.0-20.3]* [0-25.4]* 3.2 0.18 1.8 0.18 t max placebo [0.17-6.3] [0.17-2.3] [0.15-4.8] [0.17-4.8] 0.19 0.17 0.17 0.22 h low [0.15-0.58] [0.15-0.42] [0.15-0.4] [0.15-0.48] 0.18 0.18 0.18 0.18 high [0.15-0.43] [0.12-0.42] [0.15-0.4] [0.12-0.53] 0 0 0 0 AUC 0-8.3h placebo [0-18.2]* [0-11.6]* [0-28.3]* [0-21.2]* 3.4 4.4 5.8 6.4 h*μg/L low [0-25.9]* [1.1-15.0]* [0-39.0]* [1.1-28.3]* 6.8 7.2 9.8 11.8 high [0-29.8]* [0-42.0]* [0.4-48.3]* [0-51.3]* AUC>BL- 0 0 0 0 8.3h placebo [0-0.5]* [0-4.3]* [0-2.4]* [0-0.5]* 3.2 4.0 5.5 6.3 h*μg/L low [0-8.1]* [1.1-12.2]* [0-12.3]* [1.2-12.6]* 6.8 7.2 9.8 11.8 high [0-28.8]* [0-29.4]* [0.42-41.5]* [0-33.4]*

242

Table 29 (Supplemental). (Continued from previous page) 11-OH-THC Blood Plasma (LOQ 1 μg/L) No Alcohol Alcohol No Alcohol Alcohol 6.3 4.8 3.4 [2.3-≥8.3] t placebo [0.42-≥8.3] [0.15-≥8.3] [0.17-≥8.3] last [n=2] [n=4] [n=4] [n=6] 1.4 1.5 2.3 2.3 h low [0.20-≥8.3]#& [0.42-≥8.3]#& [0.40-≥8.3] [0.42-≥8.3] 3.0 2.3 3.3 3.3 high [0.42-≥8.3] #& [0.42-≥8.3]#& [0.18-≥8.3] [0.42-≥8.3] Clast placebo 1.5 [1.1-1.9] 1.2 [1.1-1.7] 1.3 [1.0-3.0] 1.3 [1.0-2.2] 1.2 [1.1- 1.3 [1.0- μg/L low 1.3 [1.0-2.5]& 2.0]& 1.2 [1.0-4.6]& 2.7]& 1.4 [1.0- 1.3 [1.0- high 1.2 [1.0-2.4]& 2.4]& 1.4 [1.0-4.9]& 3.7]&

THCCOOH Blood Plasma (LOQ 1 μg/L) No Alcohol Alcohol No Alcohol Alcohol 2.9 2.9 5.0 3.8 Cmax placebo [0-67.0]* [0-62.8]* [0-107]* [0-97.5]* 14.5 15 25.3 21.1 μg/L low [4.4-84.2]* [5.4-75.0]* [6.2-137]* [7.2-133]* 23.8 17.4 38.1 25.2 high [2.6-66.6]* [3.4-95.4]* [2.9-116]* [5.1-134]* 0.5 0.4 1.0 0 C max-BL placebo [-0.3-2.3]* [-1.1-43.3]* [-1.3-3.8]* [-20.2-41.4]* 10.0 9.4 17.5 13.7 μg/L low [4.4-22.2]* [0-21.2]* [6.2-32.4]* [-0.8-47.3]* 17.5 11.9 26.0 18.8 high [2.6-36.9]* [0-53.2]* [2.9-61.1]* [-10.6-82.9]* 0.42 0.32 0.40 0.40 t max placebo [0.15-3.3] [0.17-≥8.3] [0.15-3.3] [0.07-3.4] 0.40 0.40 0.40 0.40 h low [0.17-1.6] [0.22-3.5] [0.17-1.3] [0.15-3.5] 0.40 0.42 0.40 0.42 high [0.17-0.82] [0.15-3.3] [0.17-1.3] [0.15-3.3] 17.1 13.5 25.2 20.7 AUC 0-8.3h placebo [0-437]* [0-358]* [0-682]* [0-568]* 56.8 56.8 97.3 84.2 h*μg/L low [13.4-579]* [11.8-424]* [18.2-883]* [24.8-659]* 88.4 62.5 134 100 high [9.6-361]* [8.3-572]* [14.9-665]* [16.6-816]*

243

Table 29 (Supplemental). (Continued from previous page) THCCOOH Blood Plasma (LOQ 1 μg/L) No Alcohol Alcohol No Alcohol Alcohol AUC>BL- 0.4 0.1 1.1 0 8.3h placebo [0-3.7]* [0-279]* [0-10.6]* [0-137]* 27.7 26.2 40.8 46.4 h*μg/L low [9.6-70.8]* [0-85.9]* [18.2-83.0]* [0-181]* 51.7 41.8 69.8 58.8 high [9.6-121]* [0-262]* [14.9-235]* [0-396]* ≥8.3 ≥8.3 ≥8.3 ≥8.3 t last placebo [0.18-≥8.3] [0.18-≥8.3] [0.43-≥8.3] [1.4-≥8.3] ≥8.3 ≥8.3 ≥8.3 ≥8.3 h low [8.2-≥8.3] [4.3-≥8.3] [≥8.3-≥8.3] [≥8.3-≥8.3] ≥8.3 ≥8.3 ≥8.3 ≥8.3 high [4.8-≥8.3] [3.3-≥8.3] [≥8.3-≥8.3] [6.2-≥8.3] 3.5 4.7 5.5 8.5 C last placebo [1.1-33.7]* [1.0-32.5]* [1.0-63.5]* [1.1-45.0]* 3.5 4.2 5.7 6.7 μg/L low [1.0-53.1]* [1.4-46.4]* [1.6-92.8]* [2.1-71.5]* 5.9 5.2 7.9 7.9 high [1.1-36.7]* [1.2-49.8]* [1.2-62.1]* [1.2-63.4]*

THCCOOH- Blood Plasma glucuronide (LOQ 5 μg/L) No Alcohol Alcohol No Alcohol Alcohol 6.4 6.0 11.1 17.5 Cmax placebo [0-156]* [0-118]* [0-340]* [0-155]* 25.9 27.5 31.3 47.6 μg/L low [0-213]* [5.2-152]* [6.2-227]* [6.1-219]* 48.2 31.6 55.2 47.4 high [0-145]* [6.6-259]* [9.2-251]* [7.5-370]* 0 1.4 0.9 0.9 C max-BL placebo [-4.7-23]* [0-74.4]* [-6.0-93.0]* [-35.9-53.2]* 14.3 19.0 22.5 30.5 μg/L low [-7.0-31.1]* [5.2-39.0]* [-3.9-108]* [6.0-129]* 24.0 24.5 33.4 34.0 high [0-81.2]* [6.6-87.0]* [-4.7-107]* [-120-200]* 1.9 1.4 3.3 1.8 t max placebo [0.17-6.3]# [0.42-6.3]# [0.15-≥8.3] [0.15-6.3] 2.3 1.7 1.7 2.4 h low [0.17-6.4]# [1.3-6.3]# [0.17-≥8.3] [0.42-≥8.3] 2.3 1.7 3.3 1.7 high [1.3-≥8.3]# [1.3-≥8.3]# [0.82-≥8.3] [0.18-≥8.3]

244

Table 29 (Supplemental). (Continued from previous page) THCCOOH- Blood Plasma glucuronide (LOQ 5 μg/L) No Alcohol Alcohol No Alcohol Alcohol 30.0 22.0 55.5 63.2 AUC 0-8.3h placebo [0-1111]* [0-817]* [0-1796]* [0-855]* 173 177 181 301 h*μg/L low [0-1595]* [24.3-907]* [7.9-1425]* [21.9-1255]* 320 237 355 245 high [0-990]* [28.1-1796]* [8.0-1656]* [8.1-2656]* AUC>BL- 0 1.8 0.36 1.5 8.3h placebo [0-49.4]* [0-519]* [0-129]* [0-174]* 86.1 92.2 77.0 115 h*μg/L low [0-205]* [19.9-216]* [0-451]* [18.1-677]* 114 144 126 136 high [0-373]* [28.1-384]* [0-481]* [0-1171]* ≥8.3 ≥8.3 ≥8.3 ≥8.3 t last placebo [4.8-≥8.3]* [2.3-≥8.3]* [3.3-≥8.3] [1.4-≥8.3] ≥8.3 ≥8.3 ≥8.3 ≥8.3 h low [4.8-≥8.3]* [4.8-≥8.3]* [3.3-≥8.3] [4.8-≥8.3] ≥8.3 ≥8.3 ≥8.3 ≥8.3 high [≥8.3-≥8.3]* [4.8-≥8.3]* [0.82-≥8.3] [1.6-≥8.3] 13.8 18.7 20.8 15.2 C last placebo [5.0-94.8]* [5.2-82.1]* [5.3-161]* [6.2-117]* 18.0 16.1 19.2 31.4 μg/L low [5.1-155]* [5.1-95.5]* [5.8-221]* [6.1-134]* 25.8 21.6 35.5 28.3 high [7.0-134]* [5.8-177]* [7.1-248]* [5.1-250]*

CBD Blood Plasma (LOQ 1 μg/L) No Alcohol Alcohol No Alcohol Alcohol & & Cmax placebo 0 [0-0]* 0 [0-0]* 0 [0-0]* 0 [0-0]* μg/L low 0 [0-0]* 0 [0-0]* 0 [0-0]*& 0 [0-0]*& high 1.0 [0-3.6]* 1.1 [0-3.8]* 1.2 [0-4.5]*& 1.8 [0-5.2]*& #& #& Cmax-BL placebo 0 [0-0]* 0 [0-0]* 0 [0-0]* 0 [0-0]* μg/L low 0 [0-0]* 0 [0-0]* 0 [0-0]*#& 0 [0-0]*#& 1.8 [0- high 1.0 [0-3.6]* 1.1 [0-3.8]* 1.2 [0-4.5]*#& 5.2]*#& tmax placebo n/a n/a n/a n/a h low n/a n/a n/a n/a 0.17 0.17 0.17 0.17 high [0.15-0.25] [0.12-0.32] [0.15-0.30] [0.12-0.37] AUC0-8.3h placebo 0 [0-0]* 0 [0-0]* 0 [0-0]* 0 [0-0]* h*μg/L low 0 [0-0]* 0 [0-0]* 0 [0-0]* 0 [0-0]* high 0.3 [0-1.1]* 0.4 [0-1.1*] 0.4 [0-2.5]* 0.6 [0-3.2]*

245

Table 29 (Supplemental). (Continued from previous page) CBD Blood Plasma (LOQ 1 μg/L) No Alcohol Alcohol No Alcohol Alcohol AUC>BL- 8.3h placebo 0 [0-0]* 0 [0-0]* 0 [0-0]* 0 [0-0]* h*μg/L low 0 [0-0]* 0 [0-0]* 0 [0-0]* 0 [0-0]* high 0.2 [0-0.8]* 0.2 [0-0.8]* 0.3 [0-1.7]* 0.4 [0-2.3]* tlast placebo n/a n/a n/a n/a h low n/a n/a n/a n/a 0.17 0.17 0.18 0.18 high [0.15-0.25] [0.12-0.32] [0.15-0.52] [0.12-0.47] Clast placebo n/a n/a n/a n/a μg/L low n/a n/a n/a n/a high 1.7 [1.0-3.6] 1.9 [1.0-3.8] 1.5 [1.1-4.5] 1.5 [1.1-4.4]

CBN Blood Plasma (LOQ 1 μg/L) No Alcohol Alcohol No Alcohol Alcohol # # # # Cmax placebo 0 [0-1.3]* 0 [0-1.4]* 0 [0-1.4]* 0 [0-1.3]* μg/L low 1.3 [0-3.4]*# 1.8 [0-4.2]*# 1.9 [0-5.2]*# 1.8 [0-5.8]*# high 1.1 [0-3.4]*# 1.6 [0-5.3]*# 1.5 [0-4.0]*# 1.7 [0-7.3]*# # # Cmax-BL placebo 0 [0-1.3]* 0 [0-1.2]* 0 [0-1.4]* 0 [-1.2-0]* μg/L low 1.3 [0-3.4]*# 1.8 [0-4.2]*# 1.9 [0-5.2]* 1.8 [0-5.8]* high 1.1 [0-3.4]*# 1.6 [0-5.3]*# 1.5 [0-4.0]* 1.7 [0-7.3]* 0.17 0.30 0.17 0.17 t max placebo [n=1] [0.17-0.42] [0.17-0.17] [0.17-0.17] 0.17 0.17 0.17 0.17 h low [0.15-0.20] [0.15-0.25] [0.15-0.33] [0.15-0.25] 0.17 0.17 0.17 0.17 high [0.15-0.25] [0.12-0.32] [0.15-0.30] [0.12-0.32] # # AUC0-8.3h placebo 0 [0-0.4]* 0 [0-0.7]* 0 [0-0.4]* 0 [0-0.4]* h*μg/L low 0.5 [0-1.1]*# 0.5 [0-2.3]*# 0.6 [0-1.6]* 0.5 [0-3.1]* high 0.3 [0-1.2]*# 0.5 [0-2.4]*# 0.6 [0-2.5]* 0.5 [0-3.3]* AUC>BL- # # 8.3h placebo 0 [0-0.3]* 0 [0-0.8]* 0 [0-0.3]* 0 [0-0]* h*μg/L low 0.3 [0-0.8]*# 0.4 [0-1.5]*# 0.4 [0-1.2]* 0.4 [0-1.8]* high 0.2 [0-1.1]*# 0.3 [0-1.6]*# 0.4 [0-1.7]* 0.4 [0-2.1]* 0.17 0.30 0.17 0.17 t last placebo [n=1] [0.17-0.42] [n=1] [n=1] 0.17 0.18 0.17 0.18 h low [0.15-0.20] [0.15-0.40] [0.15-0.40] [0.15-0.40] 0.17 0.18 0.17 0.18 high [0.15-0.40] [0.12-0.47] [0.15-0.52] [0.12-0.47]

246

Table 29 (Supplemental. (Continued from previous page) CBN Blood Plasma (LOQ 1 μg/L) No Alcohol Alcohol No Alcohol Alcohol Clast placebo 1.3 [n=1] 1.3 [1.2-1.4] 1.4 [n=1] 1.3 [n=1] μg/L low 1.9 [1.1-3.4] 1.8 [1.0-4.1] 2.1 [1.0-5.2] 1.8 [1.0-5.4] high 1.9 [1.0-3.4] 1.7 [1.0-3.4] 1.9 [1.0-3.0] 1.5 [1.0-2.9] Data from 19 occasional-to-moderate cannabis smokers who participated in all dosing sessions (lower N reflects fewer participants with calculable ANOVA results due to negative placebo samples). Statistical analysis performed by factorial repeated-measures analysis of variance. Boldface indicates statistical significance at p<0.05. Cannabis administered with Volcano® Medic vaporizer: 500 mg placebo (0.008±0.002% THC), low (2.9±0.14% THC) or high (6.7±0.05% THC) dose THC. Active alcohol dose was calculated to produce approximate peak 0.065% peak breath alcohol concentration.*Significant overall cannabis dose effect (p<0.05) by factorial repeated-measures analysis of variance [ANOVA]) ~Overall cannabis p<0.06 by factorial repeated-measures ANOVA. Post-hoc analysis revealed significant low- and high-vs.-placebo cannabis effect, but no significant low-vs.-high cannabis effect. #Significant overall alcohol dose effect by factorial repeated-measures ANOVA &Significant overall alcohol*cannabis effect by factorial repeated-measures ANOVA Abbreviations: THC, ∆9-tetrahydrocannabinol; LOQ, limit of quantification; 11-OH-THC, 11-hydroxy- THC; THCCOOH, 11-nor-9-carboxy-THC; CBD, cannabidiol; CBN, cannabinol; Cmax, maximum concentration; Cmax-BL, Cmax accounting for baseline; tmax, time to maximum concentration; AUC0-8.3h, area under the 8.3h curve; AUC>BL-8.3h, AUC0-8.3h accounting for baseline; Clast, concentration at last detection.

Table 30 (Supplemental). Effects of alcohol, cannabis, and alcohol*cannabis combination on blood cannabinoid pharmacokinetic parameters (all analytes; including last measured concentration). Deg- Error Pairwise rees Analyte Overall degrees Effect comparison by N F of p-value parameter effect of size, r cannabis dose free- freedom dom THC Cmax Alcohol 19 8.03 1 18 0.56 0.011 Cannabis 42.84 1.21 21.73 <0.001a low vs. placebo 139.71 1 18 0.94 <0.001 high vs. placebo 57.23 1 18 0.87 <0.001 low vs. high 12.14 1 18 0.63 0.003 Alcohol* 1.91 1.15 20.74 0.182a Cannabis Cmax-BL Alcohol 18 8.03 1 17 0.57 0.011 Cannabis 42.00 1.21 20.62 <0.001a low vs. placebo 123.28 1 17 0.94 <0.001 high vs. placebo 55.74 1 17 0.88 <0.001 low vs. high 13.25 1 17 0.66 0.002 Alcohol* Cannabis 3.20 1.17 19.97 0.084a

247

tmax Alcohol 8 0.53 1 7 0.27 0.490 Cannabis 2.73 1.00 7.01 0.142a low vs. placebo 2.79 1 7 0.53 0.139 high vs. placebo 2.68 1 7 0.53 0.146 low vs. high 0.20 1 7 0.16 0.672 a Alcohol* 0.49 0.509 Cannabis 1.00 7.02 Clast Alcohol 8 0.51 1 7 0.26 0.498 Cannabis 1.03 7.23 0.56 0.483a low vs. placebo 0.85 1 7 0.33 0.387 high vs. placebo 0.00 1 7 0.00 0.983 low vs. high 5.29 1 7 0.66 0.055 Alcohol* 0.61 0.558 Cannabis 2 14 tlast Alcohol 8 1.46 1 7 0.42 0.266 Cannabis 9.18 1.15 8.04 0.014a low vs. placebo 10.11 1 7 0.77 0.016 high vs. placebo 9.34 1 7 0.76 0.018 low vs. high 0.61 1 7 0.28 0.461 Alcohol* 1.30 1.07 7.52 0.295a Cannabis AUC0-8.3h Alcohol 19 1.35 1 18 0.26 0.261 Cannabis 4.09 1.00 18.05 0.058a low vs. placebo 245.38 1 18 0.97 <0.001 high vs. placebo 5.13 1 18 0.47 0.036 low vs. high 2.53 1 18 0.35 0.129 Alcohol* 1.26 1.00 18.04 0.277a Cannabis AUC>BL-8.3h Alcohol 18 0.50 1 17 0.17 0.488 Cannabis 47.43 1.21 20.60 <0.001a low vs. placebo 119.56 1 17 0.94 <0.001 high vs. placebo 59.62 1 17 0.88 <0.001 low vs. high 17.18 1 17 0.71 0.001 Alcohol* Cannabis 0.63 1.27 21.55 0.473a

11-OH-THC Cmax Alcohol 19 9.95 1 18 0.60 0.005 Cannabis 28.88 1.16 20.81 <0.001a low vs. placebo 98.45 1 18 0.92 <0.001 high vs. placebo 38.44 1 18 0.83 <0.001 low vs. high 10.47 1 18 0.61 0.005 Alcohol* Cannabis 4.49 1.23 22.19 0.039a low vs. placebo 0.52 1 18 0.17 0.481 high vs. placebo 5.89 1 18 0.50 0.026 low vs. high 3.87 1 18 0.42 0.065

248

THCCOOH Cmax Alcohol 19 0.03 1 18 0.04 0.871 Cannabis 27.35 1.39 25.02 <0.001a low vs. placebo 48.59 1 18 0.85 <0.001 high vs. placebo 46.38 1 18 0.85 <0.001 low vs. high 6.94 1 18 0.53 0.017 Alcohol* Cannabis 0.03 1.30 23.32 0.922a

249

Cannabis 0.05 1.05 12.64 0.842a Clast Alcohol 13 0.12 1 12 0.10 0.740 Cannabis 6.15 2 24 0.007 low vs. placebo 7.56 1 12 0.62 0.018 high vs. placebo 13.63 1 12 0.73 0.003 low vs. high 0.97 1 12 0.27 0.344 Alcohol*

Cannabis 0.56 2 24 0.580 tlast Alcohol 13 0.25 1 12 0.14 0.628 Cannabis 4.10 1.04 12.43 0.064a low vs. placebo 4.50 1 12 0.52 0.055 high vs. placebo 3.81 1 12 0.49 0.075 low vs. high 0.60 1 12 0.22 0.455 Alcohol*

250

Cannabis Clast Alcohol -- Cannabis Alcohol*

Cannabis tlast Alcohol -- Cannabis Alcohol*

Cannabis AUC0-8.3h Alcohol 19 0.28 1 18 0.12 0.601 Cannabis 8.13 1.16 20.90 0.007a low vs. placebo 1.00 1 18 0.23 0.331 high vs. placebo 9.47 1 18 0.59 0.006 low vs. high 7.55 1 18 0.54 0.013 Alcohol* Cannabis 1.47 1.19 21.44 0.244a AUC>BL-8.3h Alcohol 18 2.08 1 17 0.33 0.167 Cannabis 7.82 2 34 0.002 low vs. placebo n/a 1 17 n/a high vs. placebo 7.82 1 17 0.56 0.012 low vs. high 7.82 1 17 0.56 0.012 Alcohol* Cannabis 2.08 2 34 0.140

251

Cannabis 0.25 2 20 0.780 Clast Alcohol 11 0.16 1 10 0.12 0.699 Cannabis 8.89 2 20 0.002 low vs. placebo 20.29 1 10 0.82 0.001 high vs. placebo 12.15 1 10 0.74 0.006 low vs. high 1.45 1 10 0.36 0.256 Alcohol*

Cannabis 0.66 2 20 0.529 tlast Alcohol 11 3.07 1 10 0.48 0.110 Cannabis 5.62 1.02 10.24 0.038a low vs. placebo 6.07 1 10 0.61 0.033 high vs. placebo 5.28 1 10 0.59 0.044 low vs. high 0.61 1 10 0.24 0.455 Alcohol*

Cannabis Clast Alcohol -- Cannabis Alcohol*

Cannabis tlast Alcohol -- Cannabis Alcohol*

Cannabis AUC0-8.3h Alcohol 19 1.33 1 18 0.26 0.264 Cannabis 33.16 2 36 <0.001 low vs. placebo n/a 1 18 n/a high vs. placebo 33.16 1 18 0.81 <0.001 low vs. high 33.16 1 18 0.81 <0.001 Alcohol* Cannabis 1.33 2 36 0.278 AUC>BL-8.3h Alcohol 18 2.73 1 17 0.37 0.117 Cannabis 28.58 2 34 <0.001 low vs. placebo n/a 1 17 n/a high vs. placebo 28.58 1 17 0.79 <0.001 low vs. high 28.58 1 17 0.79 <0.001 Alcohol* Cannabis 2.73 2 34 0.080

253

Table 30 (Supplemental). (Continued from previous page) Deg- Error Pairwise rees Analyte Overall degrees Effect comparison by N F of p-value parameter effect of size, r cannabis dose free- freedom dom CBN Cmax Alcohol 19 10.22 1 18 0.60 0.005 Cannabis 20.52 2 36 <0.001 low vs. placebo 37.76 1 18 0.82 <0.001 high vs. placebo 25.81 1 18 0.77 <0.001 low vs. high 0.40 1 18 0.15 0.535 Alcohol* Cannabis 0.96 1.39 25.00 0.367a Cmax-BL Alcohol 18 8.87 1 17 0.59 0.008 Cannabis 20.60 2 34 <0.001 low vs. placebo 33.15 1 17 0.81 <0.001 high vs. placebo 28.32 1 17 0.79 <0.001 low vs. high 0.07 1 17 0.07 0.789 Alcohol* Cannabis 1.32 1.33 22.63 0.274a b tmax Alcohol low vs. high 4 0.22 1 3 0.26 0.674 Cannabis low vs. highb 1.19 1 3 0.53 0.356 Alcohol* low vs. highb Cannabis 0.00 1 3 0.03 0.966 b Clast Alcohol low vs. high 4 0.34 1 3 0.32 0.602 Cannabis low vs. highb 0.36 1 3 0.33 0.589 Alcohol* low vs. highb Cannabis 0.39 1 3 0.34 0.577 b tlast Alcohol low vs. high 4 0.71 1 3 0.44 0.460 Cannabis low vs. highb 1.08 1 3 0.51 0.375 Alcohol* low vs. highb Cannabis 0.30 1 3 0.30 0.624 AUC0-8.3h Alcohol 17 4.72 1 16 0.48 0.045 Cannabis 17.58 2 32 <0.001 low vs. placebo 40.29 1 16 0.85 <0.001 high vs. placebo 21.95 1 16 0.76 <0.001 low vs. high 0.09 1 16 0.08 0.764 Alcohol* Cannabis 0.83 1.42 22.76 0.411a AUC>BL-8.3h Alcohol 18 5.03 1 17 0.48 0.039 Cannabis 13.82 2 34 <0.001 low vs. placebo 21.67 1 17 0.75 <0.001 high vs. placebo 18.58 1 17 0.72 <0.001 low vs. high 0.11 1 17 0.08 0.747 Alcohol* Cannabis 0.56 2 34 0.579 Data from 19 occasional-to-moderate cannabis smokers who participated in all dosing sessions (lower N reflects fewer participants with calculable ANOVA results due to negative placebo samples). Statistical analysis performed by factorial repeated-measures analysis of variance. Boldface indicates statistical significance at p<0.05. Cannabis administered with Volcano® Medic vaporizer: 500mg placebo (0.008±0.002% THC), low (2.9±0.14% THC) or high (6.7±0.05% THC) dose THC. Active alcohol dose was calculated to produce approximate peak 0.065% peak breath alcohol concentration. aMauchly’s test showed sphericity was violated on main effects, so Greenhouse-Geisser correction was utilized. bPlacebo doses not included in ANOVA due to too few positive specimens for comparison. Abbreviations: THC, ∆9-tetrahydrocannabinol; LOQ, limit of quantification; 11-OH-THC, 11-hydroxy-THC; THCCOOH, 11-nor-9- carboxy-THC; CBD, cannabidiol; CBN, cannabinol; Cmax, maximum concentration; Cmax-BL, Cmax accounting for baseline; tmax, time to maximum concentration; Clast, concentration at last detection; tlast, time of last detection; AUC0-8.3h, area under the 8.3 h curve; AUC>BL- 8.3h, AUC0-8.3h accounting for baseline. 254

Table 31 (Supplemental). Effects of alcohol, cannabis, and alcohol*cannabis combination on plasma cannabinoid pharmacokinetic parameters (all analytes; including last measured concentration). Error Degr- Pairwise deg- Analyte Overall ees of Effect comparison by N F rees of p-value parameter effect free- size, r cannabis dose freedom dom THC Cmax Alcohol 19 5.20 1 18 0.47 0.035 Cannabis 40.28 1.17 20.99 <0.001a low vs. placebo 143.53 1 18 0.94 <0.001 high vs. placebo 53.52 1 18 0.87 <0.001 low vs. high 13.05 1 18 0.65 0.002 Alcohol* Cannabis 1.72 1.19 21.47 0.205a Cmax-BL Alcohol 18 5.32 1 17 0.49 0.034 Cannabis 37.64 1.19 20.25 <0.001a low vs. placebo 105.24 1 17 0.93 <0.001 high vs. placebo 50.14 1 17 0.86 <0.001 low vs. high 13.99 1 17 0.67 0.002 Alcohol* Cannabis 3.08 1.22 20.69 0.088a tmax Alcohol 11 4.53 1 10 0.56 0.059 Cannabis 4.75 1.00 10.01 0.054a low vs. placebo 4.85 1 10 0.57 0.052 high vs. placebo 4.66 1 10 0.56 0.056 low vs. high 0.24 1 10 0.15 0.636 Alcohol* Cannabis 4.43 1.00 10.01 0.062a Clast Alcohol 11 0.01 1 10 0.04 0.911 Cannabis 1.74 1.28 12.81 0.214a low vs. placebo 0.98 1 10 0.30 0.347 high vs. placebo 2.31 1 10 0.43 0.159 low vs. high 2.12 1 10 0.42 0.176 Alcohol* Cannabis 1.47 2 20 0.254 tlast Alcohol 11 0.02 1 10 0.04 0.890 Cannabis 6.43 1.16 11.55 0.024a low vs. placebo 6.64 1 10 0.63 0.028 high vs. placebo 6.89 1 10 0.64 0.025 low vs. high 0.00 1 10 0.01 0.981 Alcohol* Cannabis 1.65 2 20 0.216 AUC0-8.3h Alcohol 19 1.35 1 18 0.26 0.261 Cannabis 4.09 1.00 18.05 0.058a low vs. placebo 245.38 1 18 0.97 <0.001 high vs. placebo 5.13 1 18 0.47 0.036 low vs. high 2.53 1 18 0.35 0.129 Alcohol* Cannabis 1.26 1.00 18.04 0.277a

255

11-OH-THC Cmax Alcohol 19 6.12 1 18 0.50 0.024 Cannabis 31.30 1.22 21.90 <0.001a low vs. placebo 73.17 1 18 0.90 <0.001 high vs. placebo 39.70 1 18 0.83 <0.001 low vs. high 12.24 1 18 0.64 0.003 Alcohol* Cannabis 2.77 1.34 24.15 0.100a Cmax-BL Alcohol 18 3.31 1 17 0.40 0.087 Cannabis 33.26 1.23 20.88 <0.001a low vs. placebo 60.95 1 17 0.88 <0.001 high vs. placebo 41.89 1 17 0.84 <0.001 low vs. high 15.74 1 17 0.69 0.001 Alcohol* Cannabis 3.57 1.49 25.30 0.055a b tmax Alcohol low vs. high 17 2.35 1 16 0.36 0.145 Cannabis low vs. highb 0.13 1 16 0.09 0.724 Alcohol* low vs. highb Cannabis 0.17 1 16 0.10 0.683 b Clast Alcohol low vs. high 16 0.00 1 15 0.00 0.984 Cannabis low vs. highb 0.01 1 15 0.02 0.939 Alcohol* low vs. highb Cannabis 10.27 1 15 0.64 0.006 b tlast Alcohol low vs. high 17 3.37 1 16 0.42 0.085 Cannabis low vs. highb 4.04 1 16 0.45 0.062 Alcohol* low vs. highb Cannabis 0.65 1 16 0.20 0.432 AUC0-8.3h Alcohol 19 1.06 1 18 0.24 0.317 Cannabis 28.02 1.13 20.27 <0.001a low vs. placebo 75.29 1 18 0.90 <0.001 high vs. placebo 32.97 1 18 0.80 <0.001 low vs. high 12.54 1 18 0.64 0.002 Alcohol* Cannabis 1.92 1.21 21.73 0.179a AUC>BL-8.3h Alcohol 18 0.94 1 17 0.23 0.346 Cannabis 29.53 1.13 19.22 <0.001a low vs. placebo 82.54 1 17 0.91 <0.001 high vs. placebo 35.84 1 17 0.82 <0.001 low vs. high 13.51 1 17 0.67 0.002 Alcohol* Cannabis 1.10 1.40 23.85 0.327a

256

Cannabis 0.05 1.05 12.64 0.842a Clast Alcohol 14 0.01 1 13 0.02 0.946 Cannabis 7.37 2 26 0.003 low vs. placebo 8.86 1 13 0.64 0.011 high vs. placebo 21.37 1 13 0.79 <0.001 low vs. high 0.51 1 13 0.19 0.487 Alcohol*

Cannabis 0.35 2 26 0.707 tlast Alcohol 14 0.03 1 13 0.05 0.858 Cannabis 2.51 1.03 13.41 0.136a low vs. placebo 2.73 1 13 0.42 0.123 high vs. placebo 2.33 1 13 0.39 0.151 low vs. high 0.56 1 13 0.20 0.467 Alcohol*

THC-glucuronide Cmax Alcohol 19 0.01 1 18 0.02 0.924 Cannabis 21.39 1.16 20.95 <0.001a low vs. placebo 21.62 1 18 0.74 <0.001 high vs. placebo 26.11 1 18 0.77 <0.001 low vs. high 15.61 1 18 0.68 0.001 Alcohol* Cannabis 1.08 2 36 0.350 Cmax-BL Alcohol 18 0.03 1 17 0.04 0.860 Cannabis 18.88 1.17 19.88 <0.001a low vs. placebo 19.53 1 17 0.73 <0.001 high vs. placebo 22.86 1 17 0.76 <0.001 low vs. high 13.73 1 17 0.67 0.002 Alcohol* Cannabis 0.87 2 34 0.428 tmax Alcohol -- Cannabis Alcohol*

Cannabis Clast Alcohol -- Cannabis Alcohol*

Cannabis tlast Alcohol -- Cannabis Alcohol*

Cannabis AUC0-8.3h Alcohol 19 0.55 1 18 0.17 0.469 Cannabis 17.42 1.53 27.47 <0.001a low vs. placebo 2.70 1 18 0.36 0.118 high vs. placebo 23.51 1 18 0.75 <0.001 low vs. high 16.22 1 18 0.69 0.001 Alcohol* Cannabis 0.58 1.51 27.21 0.522a AUC>BL-8.3h Alcohol 18 0.17 1 17 0.10 0.682 Cannabis 18.99 1.15 19.54 <0.001a low vs. placebo 16.41 1 17 0.70 0.001 high vs. placebo 22.54 1 17 0.76 <0.001 low vs. high 14.90 1 17 0.68 0.001 Alcohol* Cannabis 0.46 1.45 24.66 0.574a

258

Cannabis 0.57 2 22 0.574 Clast Alcohol 12 0.35 1 11 0.17 0.568 Cannabis 8.07 2 22 0.002 low vs. placebo 2.65 1 11 0.44 0.132 high vs. placebo 13.75 1 11 0.75 0.003 low vs. high 5.39 1 11 0.57 0.040 Alcohol*

Cannabis 0.14 2 22 0.868 tlast Alcohol 12 0.16 1 11 0.12 0.693 Cannabis 1.04 2 22 0.371 low vs. placebo 2.44 1 11 0.43 0.147 high vs. placebo 0.00 1 11 0.01 0.975 low vs. high 2.12 1 11 0.40 0.173 Alcohol*

Cannabis Clast Alcohol -- Cannabis Alcohol*

Cannabis tlast Alcohol -- Cannabis Alcohol*

Cannabis AUC0-8.3h Alcohol 19 1.31 1 18 0.26 0.267 Cannabis 22.85 2 36 <0.001 low vs. placebo n/a 1 18 n/a high vs. placebo 21.85 1 18 0.74 <0.001 low vs. high 22.85 1 18 0.75 <0.001 Alcohol* Cannabis 1.31 2 36 0.282 AUC>BL-8.3h Alcohol 18 0.89 1 17 0.22 0.359 Cannabis 21.61 2 34 <0.001 low vs. placebo n/a 1 17 n/a high vs. placebo 21.61 1 17 0.75 <0.001 low vs. high 21.61 1 17 0.75 <0.001 Alcohol* Cannabis 0.89 2 34 0.420

260

Table 31 (Supplemental). (Continued from previous page) Error Degr- Pairwise deg- Analyte Overall ees of Effect comparison by N F rees of p-value parameter effect free- size, r cannabis dose freedom dom CBN Cmax Alcohol 19 5.40 1 18 0.48 0.032 Cannabis 23.98 2 36 <0.001 low vs. placebo 45.41 1 18 0.85 <0.001 high vs. placebo 31.73 1 18 0.80 <0.001 low vs. high 0.61 1 18 0.18 0.445 Alcohol* Cannabis 0.83 1.26 22.75 0.399a Cmax-BL Alcohol 18 3.24 1 17 0.40 0.090 Cannabis 22.31 2 34 <0.001 low vs. placebo 34.68 1 17 0.82 <0.001 high vs. placebo 32.81 1 17 0.81 <0.001 low vs. high 0.10 1 17 0.08 0.759 Alcohol* Cannabis 1.57 1.19 20.27 0.229a b tmax Alcohol low vs. high 7 0.46 1 6 0.27 0.524 Cannabis low vs. highb 0.37 1 6 0.24 0.564 Alcohol* low vs. highb Cannabis 0.69 1 6 0.32 0.438 b Clast Alcohol low vs. high 7 0.56 1 6 0.29 0.482 Cannabis low vs. highb 4.49 1 6 0.65 0.078 Alcohol* low vs. highb Cannabis 1.00 1 6 0.38 0.356 b tlast Alcohol low vs. high 7 0.84 1 6 0.35 0.395 Cannabis low vs. highb 0.67 1 6 0.32 0.445 Alcohol* low vs. highb Cannabis 0.35 1 6 0.23 0.577 AUC0-8.3h Alcohol 14 0.73 1 13 0.23 0.408 Cannabis 12.56 2 26 <0.001 low vs. placebo 29.15 1 13 0.83 <0.001 high vs. placebo 19.00 1 13 0.77 0.001 low vs. high 0.45 1 13 0.18 0.513 Alcohol* Cannabis 0.14 1.33 17.34 0.874a AUC>BL-8.3h Alcohol 18 1.02 1 17 0.24 0.327 Cannabis 15.32 2 34 <0.001 low vs. placebo 31.22 1 17 0.80 <0.001 high vs. placebo 21.24 1 17 0.75 <0.001 low vs. high 0.27 1 17 0.13 0.609 Alcohol* Cannabis 0.32 1.39 23.63 0.651a Data from 19 occasional-to-moderate cannabis smokers who participated in all dosing sessions (lower N reflects fewer participants with calculable ANOVA results due to negative placebo samples). Statistical analysis performed by factorial repeated-measures analysis of variance. Boldface indicates statistical significance at p<0.05. Cannabis administered with Volcano® Medic vaporizer: 500 mg placebo (0.008±0.002% THC), low (2.9±0.14% THC) or high (6.7±0.05% THC) dose THC. Active alcohol dose was calculated to produce approximate peak 0.065% peak breath alcohol concentration. aMauchly’s test showed sphericity was violated on main effects, so Greenhouse-Geisser correction was utilized. bPlacebo doses not included in ANOVA due to too few positive specimens for comparison. Abbreviations: THC, ∆9-tetrahydrocannabinol; LOQ, limit of quantification; 11-OH-THC, 11-hydroxy-THC; THCCOOH, 11-nor-9- carboxy-THC; CBD, cannabidiol; CBN, cannabinol; Cmax, maximum concentration; Cmax-BL, Cmax accounting for baseline; tmax, time to maximum concentration; Clast, concentration at last detection; tlast, time of last detection; AUC0-8.3h, area under the 8.3 h curve; AUC>BL- 8.3h, AUC0-8.3h accounting for baseline. 261

Table 32 (Supplemental). Median [range] blood and plasma ∆9-THC pharmacokinetic parameters for all 32 participants (full study population, including noncompleters) after controlled cannabis inhalation by vaporizer. THC Blood Plasma (LOQ 1 μg/L) No Alcohol n Alcohol n No Alcohol n Alcohol n 1.9 1.2 3.1 2.0 Cmax placebo [0-7.6] 26 [0-7.9] 26 [0-10.2] 26 [0-12.0] 26 32.7 37.2 46.5 57.6 μg/L low [11.4-75.7] 23 [6.7-72.0] 25 [16.6-114] 23 [11.2-102] 25 60.0 67.5 87.5 97.8 high [15.2-137] 28 [18.1-210] 19 [23.6-210] 28 [24.5-339] 19 1.5 0.7 2.2 1.1 Cmax-BL placebo [0-6.5] 26 [0-7.9] 26 [-0.2-10.2] 26 [-0.7-12.0] 26 32.7 37.2 45.7 57.6 μg/L low [11.4-75.7] 23 [4.1-72.0] 25 [16.6-113] 23 [2.3-102] 25 60.0 67.5 87.5 96.1 high [15.2-137] 28 [18.1-204] 19 [23.6-208] 28 [24.5-332] 19 0.17 0.18 0.17 0.18 tmax placebo [0.15-6.3] 22 [0.07-8.3] 15 [0.15-6.3] 24 [0.07-8.3] 17 0.17 0.17 0.17 0.17 h low [0.13-0.33] 23 [0.13-0.25] 25 [0.13-0.33] 23 [0.13-0.25] 25 0.17 0.17 0.17 0.17 high [0.13-0.30] 28 [0.12-0.37] 19 [0.13-0.30] 28 [0.12-0.37] 19 1.0 0.4 3.3 1.1 AUC0-8.3h placebo [0-53.2] 26 [0-36.4] 26 [0-66.6] 26 [0-103,623] 26 31.9 37.0 44.6 51.7 h*μg/L low [10.6-84.2] 23 [18.0-63.4] 25 [14.1-124] 23 [26.9-86.7] 25 48.2 62.2 70.2 93.2 high [10.6-113] 28 [13.2-1445] 19 [15.9-182] 28 [19.4-2370] 19 0.5 0.3 1.0 0.5

AUC>BL-8.3h placebo [0-7.1] 26 [0-19.6] 26 [0-12.7] 26 [0-7.4] 26 21.7 20.8 29.2 27.6 h*μg/L low [6.9-38.3] 23 [2.5-33.3] 25 [9.3-56.5] 23 [6.0-49.3] 25 31.2 33.7 49.9 51.6 high [6.8-77.9] 28 [8.8-83.5] 19 [9.7-124] 28 [14.1-132] 19 0.41 1.3 1.4 2.4 tlast placebo [0.15-≥8.3] 22 [0.13-≥8.3] 15 [0.15-≥8.3] 24 [0.13-≥8.3] 17 3.5 3.5 4.8 6.3 h low [0.70-≥8.3] 23 [1.3-≥8.3] 25 [0.70-≥8.3] 23 [1.3-≥8.3] 25 3.4 3.7 6.3 6.4 high [0.82-≥8.3] 28 [1.4-≥8.3] 19 [1.3-≥8.3] 28 [1.4-≥8.3] 19 1.4 1.7 1.4 1.9 Clast placebo [1.0-5.3] 22 [1.1-5.0] 15 [1.0-8.3] 24 [1.0-7.4] 17 1.3 1.4 1.6 1.3 μg/L low [1.0-7.0] 23 [1.0-5.3] 25 [1.0-12.0] 23 [1.0-8.6] 25 1.6 1.6 1.8 1.6 high [1.1-6.5] 28 [1.0-7.3] 19 [1.0-11.2] 28 [1.0-9.6] 19 Cannabis administered with Volcano® Medic vaporizer: 500 mg placebo (0.008±0.002% THC), low (2.9±0.14% THC) or high (6.7±0.05% THC) dose THC. Active alcohol dose was calculated to produce approximate peak 0.065% peak breath alcohol concentration. THC limit of quantification (LOQ) was 1 µg/L. For completer pharmacokinetic data only, refer to Table 25. 9 Abbreviations: THC, ∆ -tetrahydrocannabinol; LOQ, limit of quantification; Cmax, maximum concentration; Cmax-BL, Cmax accounting for baseline; tmax, time to maximum concentration; AUC0-8.3h, area under the 8.3 h curve; AUC>BL-8.3h, AUC0-8.3h accounting for baseline; Clast, concentration at last detection.

262

Table 33 (Supplemental). Median [range] blood and plasma 11-OH-THC pharmacokinetic parameters for all 32 participants (full study population, including noncompleters) after controlled cannabis inhalation by vaporizer. 11-OH- Blood Plasma THC (LOQ 1 n n n n No Alcohol Alcohol No Alcohol Alcohol μg/L) 0 0 0 0 Cmax placebo [0-2.5] 26 [0-2.4] 26 [0-4.3] 26 [0-3.0] 26 2.6 3.7 3.8 4.8 μg/L low [0-9.1] 23 [1.4-8.5] 25 [0-13.7] 23 [1.3-12.1] 25 4.4 6.0 5.8 7.5 high [0-14.2] 28 [0-24.8] 19 [1.0-20.3] 28 [0-27.3] 19 0 0 0 0 Cmax-BL placebo [0-1.1] 26 [0-2.1] 26 [0-1.5] 26 [0-1.8] 26 2.6 3.4 3.7 4.4 μg/L low [0-9.1] 23 [0.71-8.5] 25 [0-13.7] 23 [0.87-12.1] 25 4.4 6.0 5.8 7.5 high [0-12.8] 28 [0-23.3] 19 [1.0-20.3] 28 [0-25.4] 19 3.2 0.18 0.40 0.18 tmax placebo [0.17-6.3] 2 [0.15-2.3] 6 [0.15-4.8] 5 [0.17-8.3] 9 0.17 0.17 0.17 0.20 h low [0.13-0.58] 22 [0.13-0.42] 25 [0.13-0.40] 22 [0.13-0.48] 25 0.17 0.18 0.17 0.18 high [0.13-0.43] 27 [0.12-0.42] 18 [0.13-0.43] 28 [0.12-0.53] 18 0 0 0 AUC0-8.3h placebo [0-18.2] 26 [0-11.6] 26 0 [0-28.3] 26 [0-21.2] 26 3.4 4.9 5.5 8.1 h*μg/L low [0-25.9] 22 [1.1-15.0] 25 [0-39.0] 23 [1.1-28.3] 24 7.1 7.2 10.2 11.8 high [0-29.8] 28 [0-42.0] 19 [0.43-48.3] 28 [0-51.3] 19 0 0 0 0

AUC>BL-8.3h placebo [0-0.54] 26 [0-4.3] 26 [0-2.4] 26 [0-1.5] 26 4.1 4.7 6.4 h*μg/L low 3.2 [0-8.1] 23 [0.83-13.7] 25 [0-12.3] 23 [0.81-17.4] 25 6.9 7.2 10.2 11.8 high [0-28.8] 28 [0-29.4] 19 [0.42-41.5] 28 [0-33.4] 19 5.3 6.3 3.3 6.3 tlast placebo [2.3-≥8.3] 2 [0.18-≥8.3] 6 [0.15-≥8.3] 5 [0.17-≥8.3] 9 1.4 1.5 2.3 2.3 h low [0.20-≥8.3] 22 [0.42-≥8.3] 25 [0.40-≥8.3] 22 [0.42-≥8.3] 25 2.8 2.3 3.3 3.3 high [0.40-≥8.3] 27 [0.42-≥8.3] 18 [0.18-≥8.3] 28 [0.42-≥8.3] 18 1.5 1.3 1.5 1.5 Clast placebo [1.1-1.9] 2 [1.1-1.7] 6 [1.0-3.0] 5 [1.0-2.2] 9 1.3 1.2 1.2 1.4 μg/L low [1.0-2.5] 22 [1.1-2.5] 25 [1.0-4.6] 22 [1.0-2.7] 25 1.2 1.4 1.4 1.3 high [1.0-2.4] 27 [1.0-2.4] 18 [1.0-4.9] 28 [1.0-3.7] 18 Cannabis administered with Volcano® Medic vaporizer: 500 mg placebo (0.008±0.002% THC), low (2.9±0.14% THC) or high (6.7±0.05% THC) dose THC. Active alcohol dose was calculated to produce approximate peak 0.065% peak breath alcohol concentration. 11-OH-THC limit of quantification (LOQ) was 1 µg/L. For completer pharmacokinetic data only, refer to Table 25. Abbreviations: 11-OH-THC, 11-hydroxy-tetrahydrocannabinol; THC, ∆9-tetrahydrocannabinol ; LOQ, limit of quantification; Cmax, maximum concentration; Cmax-BL, Cmax accounting for baseline; tmax, time to maximum concentration; AUC0-8.3h, area under the 8.3 h curve; AUC>BL-8.3h, AUC0-8.3h accounting for baseline; Clast, concentration at last detection.

263

Table 34 (Supplemental). Median [range] blood and plasma THCCOOH pharmacokinetic parameters for all 32 participants (full study population, including noncompleters) after controlled cannabis inhalation by vaporizer. THCCOOH Blood Plasma (LOQ 1 μg/L) No Alcohol n Alcohol n No Alcohol n Alcohol n 2.9 2.9 5.6 3.7 Cmax placebo [0-67.0] 27 [0-62.8] 25 [0-107] 26 [0-97.5] 26 15.5 23 15.6 25 25.3 23 21.2 25 μg/L low [4.0-84.2] [5.4-75.0] [6.2-137] [7.2-133] 24.2 28 17.4 19 39.0 28 25.2 19 high [2.6-66.6] [3.4-95.4] [2.9-116] [5.1-134] 0.57 26 0 26 1.0 26 0 26 C max-BL placebo [-0.32-2.3] [-1.1-43.3] [-1.3-4.8] [-20.2-41.4] 10.0 23 10.7 25 17.5 23 14 25 μg/L low [4.0-22.2] [0-21.2] [6.2-32.4] [-7.0-47.3] 16.9 28 11.9 19 26.9 28 18.8 19 high [2.6-36.9] [0-53.2] [2.9-61.1] [-10.6-82.9] 0.42 20 0.27 20 0.40 22 0.31 20 t max placebo [0.15-3.3] [0.15-8.3] [0.15-3.3] [0.07-3.4] 0.40 23 0.40 25 0.40 23 0.40 25 h low [0.17-1.6] [0.17-3.5] [0.17-1.3] [0.15-3.5] 0.40 28 0.42 19 0.40 28 0.42 19 high [0.17-0.82] [0.15-3.3] [0.17-1.8] [0.15-3.3] 17.1 27 13.5 25 32.3 26 20.7 26 AUC 0-8.3h placebo [0-437] [0-358] [0-682] [0-568] 68.0 23 62.1 25 99.3 23 90.5 25 h*μg/L low [8.9-579] [11.8-424] [16.9-883] [24.8-659] 88.3 28 62.5 19 135 28 100 19 high [9.6-361] [8.3-572] [14.9-665] [16.6-816] AUC>BL- 0.46 26 0 26 0.58 26 0 26 8.3h placebo [0-4.4] [0-279] [0-10.6] [0-137] 23.9 23 26.2 25 35.4 23 46.4 25 h*μg/L low [8.9-70.8] [0-85.9] [17.3-83.0] [0-181] 50.4 28 41.8 19 67.1 28 58.8 19 high [8.4-150] [0-262] [5.3-235] [0-396] ≥8.3 20 ≥8.3 20 ≥8.3 22 ≥8.3 20 t last placebo [0.18-≥8.3] [0.18-≥8.3] [0.18-≥8.3] [0.50-≥8.3] ≥8.3 23 ≥8.3 25 ≥8.3 23 ≥8.3 25 h low [3.3-≥8.3] [4.3-≥8.3] [6.3-≥8.3] [8.0-≥8.3] ≥8.3 28 ≥8.3 19 ≥8.3 28 ≥8.3 19 high [4.8-≥8.3] [3.3-≥8.3] [6.6-≥8.3] [6.2-≥8.3] 4.1 20 3.5 20 5.6 22 6.1 20 C last placebo [1.1-33.7] [1.0-34.5] [1.0-63.5] [1.1-49.8] 4.2 23 4.7 25 6.7 23 7.1 25 μg/L low [1.0-53.1] [1.4-46.4] [1.1-92.8] [2.1-71.5] 6.2 28 5.2 19 9.1 28 7.9 19 high [1.1-36.7] [1.2-49.8] [1.2-62.1] [1.2-63.4] Cannabis administered with Volcano® Medic vaporizer: 500mg placebo (0.008±0.002% THC), low (2.9±0.14% THC) or high (6.7±0.05% THC) dose THC. Active alcohol dose was calculated to produce approximate peak 0.065% peak breath alcohol concentration. THCCOOH limit of quantification (LOQ) was 1 µg/L. For completer pharmacokinetic data only, refer to Table 25. Abbreviations: THCCOOH, 9 11-nor-9-carboxy-THC; THC, ∆ -tetrahydrocannabinol ; LOQ, limit of quantification; Cmax, maximum concentration; Cmax-BL, Cmax accounting for baseline; tmax, time to maximum concentration; AUC0-8.3h, area under the 8.3 h curve; AUC>BL-8.3h, AUC0-8.3h accounting for baseline; Clast, concentration at last detection.

264

Table 35 (Supplemental). Median [range] blood and plasma THC-glucuronide pharmacokinetic parameters for all 32 participants (full study population, including noncompleters) after controlled cannabis inhalation by vaporizer. THC-glucuronide Blood Plasma (LOQ 0.5 μg/L) No Alcohol n Alcohol n No Alcohol n Alcohol n 0 0 0 0 C placebo max [0-0] 26 [0-0] 26 [0-0.8] 26 [0-0] 26 0 0 0 0 μg/L low [0-0] 23 [0-0.6] 25 [0-0.8] 23 [0-0.8] 25 0 0 0.7 0.6 high [0-1.4] 28 [0-0.8] 19 [0-2.0] 28 [0-2.0] 19 0 0 0 0 C placebo max-BL [0-0] 26 [0-0] 26 [0-0.2] 26 [0-0] 26 0 0 0 0 μg/L low [0-0] 23 [0-0.6] 25 [0-0.8] 23 [0-0.8] 25 0 0 0.7 0.6 high [0-1.4] 28 [0-0.8] 19 [0-2.0] 28 [0-2.0] 19 t placebo n/a n/a 1.3 n/a max 0 0 1 0 0.42 0.42 h low n/a 0.40 0 1 [0.33-0.45] 5 [0.17-0.48] 11 0.40 0.40 0.40 0.40 high [0.25-0.67] 9 [0.38-0.53] 3 [0.13-0.67] 19 [0.37-0.53] 11 0 0 0 0 AUC placebo 0-8.3h [0-0] 26 [0-0] 26 [0-1.7] 26 [0-0] 26 0 0 0 0 h*μg/L low [0-0] 23 [0-0.35] 25 [0-0.75] 23 [0-0.60] 25 0 0 0.40 0.31 high [0-1] 28 [0-0.49] 19 [0-1.5] 28 [0-1.5] 19 0 0 0 0 AUC placebo >BL-8.3h [0-0] 26 [0-0] 26 [0-0.20] 26 [0-0] 26 0 0 0 0 h*μg/L low [0-0] 23 [0-0.34] 25 [0-0.75] 23 [0-0.59] 25 0 0 0.40 0.31 high [0-1.0] 28 [0-0.49] 19 [0-1.5] 28 [0-1.5] 19

tlast placebo n/a 0 n/a 0 3.3 1 n/a 0 0.43 0.42 h low n/a 0.40 0 1 [0.38-0.58] 5 [0.17-0.48] 11 0.40 0.40 0.40 0.42 high [0.25-0.67] 9 [0.38-0.53] 3 [0.13-1.6] 19 [0.37-1.4] 11 0.5 C placebo n/a n/a n/a last 0 0 [0.5-0.5] 1 0 0.6 0.6 μg/L low n/a 0.6 0 1 [0.5-0.8] 5 [0.5-0.8] 11 0.6 0.6 0.7 0.7 high [0.5-1.4] 9 [0.6-0.8] 3 [0.5-2.0] 19 [0-2.0] 12 Cannabis administered with Volcano® Medic vaporizer: 500 mg placebo (0.008±0.002% THC), low (2.9±0.14% THC) or high (6.7±0.05% THC) dose THC. Active alcohol dose was calculated to produce approximate peak 0.065% peak breath alcohol concentration. THC-glucuronide limit of quantification (LOQ) was 0.5 µg/L. For completer pharmacokinetic data only, refer to Table 25. Abbreviations: THC-glucuronide, ∆9-tetrahydrocannabinol-glucuronide; THC, ∆9-tetrahydrocannabinol; LOQ, limit of quantification; Cmax, maximum concentration; Cmax-BL, Cmax accounting for baseline; tmax, time to maximum concentration; AUC0-8.3h, area under the 8.3 h curve; AUC>BL-8.3h, AUC0-8.3h accounting for baseline; Clast, concentration at last detection.

265

Table 36 (Supplemental). Median [range] blood and plasma THCCOOH-glucuronide pharmacokinetic parameters for all 32 participants (full study population, including noncompleters) after controlled cannabis inhalation by vaporizer THCCOOH- Blood Plasma glucuronide (LOQ 5 μg/L) No Alcohol n Alcohol n No Alcohol n Alcohol n 10.5 5.9 15.7 11.7 Cmax placebo [0-156] 26 [0-272] 26 [0-340] 26 [0-370] 26 25.9 30.4 31.3 47.6 μg/L low [0-214] 23 [5.2-276] 25 [6.2-257] 23 [6.1-229] 25 48.8 31.6 62.0 47.4 high [0-276] 28 [6.6-259] 19 [9.2-378] 28 [7.5-370] 19 0.7 0.9 0.4 1.2 Cmax-BL placebo [-4.7-24.8] 26 [0-74.4] 26 [-11.6-93.0] 26 [-35.9-104] 26 14.3 19.4 21.4 29.0 μg/L low [-7.0-31.1] 23 [5.2-92.0] 25 [-3.9-108] 23 [6.0-129] 25 25.5 24.5 30.0 34.0 high [0-81.2] 28 [6.6-87.0] 19 [-4.7-107] 28 [-120-200] 19 2.8 1.4 3.3 1.3 tmax placebo [0.17-≥8.3] 16 [0.20-6.3] 15 [0.15-≥8.3] 17 [0.15-6.3] 18 2.3 2.3 2.3 3.3 h low [0.17-6.4] 22 [0.13-≥8.3] 25 [0.13-≥8.3] 23 [0.42-≥8.3] 25 2.3 1.7 3.3 1.7 high [0.13-≥8.3] 27 [1.3-≥8.3] 19 [0.43-≥8.3] 28 [0.18-≥8.3] 19 AUC0- 61.8 19.8 63.3 53.0 8.3h placebo [0-1111] 26 [0-1998] 26 [0-1796] 26 [0-1235] 26 173 193 178 298 h*μg/L low [0-1595] 23 [24.3-2114] 25 [7.9-1579] 23 [21.9-1255] 25 303 237 280 245 high [0-2114] 28 [28.1-1796] 19 [8.0-1693] 28 [8.1-2656] 19 AUC>BL 1.1 0.9 0.1 1.5 -8.3h placebo [0-49.4] 26 [0-519] 26 [0-129] 26 [0-198] 26 74.7 93.3 72.5 103 h*μg/L low [0-205] 23 [7.5-331] 25 [0-451] 23 [18.1-677] 25 135 144 120 136 high [0-382] 28 [28.1-384] 19 [0-481] 28 [0-1171] 19 ≥8.3 ≥8.3 ≥8.3 ≥8.3 tlast placebo [4.8-≥8.3] 16 [2.3-≥8.3] 15 [3.3-≥8.3] 17 [1.4-≥8.3] 18 ≥8.3 ≥8.3 ≥8.3 ≥8.3 h low [3.3-≥8.3] 22 [4.8-≥8.3] 25 [3.3-≥8.3] 23 [4.8-≥8.3] 25 ≥8.3 ≥8.3 ≥8.3 ≥8.3 high [6.6-≥8.3] 27 [4.8-≥8.3] 19 [0.82-≥8.3] 28 [1.6-≥8.3] 19 19.9 18.7 21.7 15.2 Clast placebo [5.0-94.8] 16 [5.2-212] 15 [5.3-161] 17 [5.6-126] 18 17.2 17.8 19.2 31.4 μg/L low [5.1-155] 22 [5.1-254] 25 [5.8-221] 23 [6.1-136] 25 25.6 21.6 32.5 28.3 high [7.0-254] 27 [5.8-177] 19 [5.7-248] 28 [5.1-250] 19 Cannabis administered with Volcano® Medic vaporizer: 500 mg placebo (0.008±0.002% THC), low (2.9±0.14% THC) or high (6.7±0.05% THC) dose THC. Active alcohol dose was calculated to produce approximate peak 0.065% peak breath alcohol concentration. THCCOOH-

glucuronide limit of quantification (LOQ) was 5 µg/L. For completer pharmacokinetic data only, refer to Table 25. Abbreviations: THCCOOH-glucuronide, 11-nor-9-carboxy-tetrahydrocannabinol-glucuronide; THC, ∆9-tetrahydrocannabinol; LOQ, limit of quantification; Cmax, maximum concentration; Cmax-BL, Cmax accounting for baseline; tmax, time to maximum concentration; AUC0-8.3h, area under the 8.3 h curve; AUC>BL-8.3h, AUC0-8.3h accounting for baseline; Clast, concentration at last detection.

266

Table 37 (Supplemental). Median [range] blood and plasma CBD pharmacokinetic parameters for all 32 participants (full study population, including noncompleters) after controlled cannabis inhalation by vaporizer. CBD Blood Plasma

(LOQ 1 μg/L) No Alcohol n Alcohol n No Alcohol n Alcohol n 0 0 0 0 Cmax placebo [0-0] 26 [0-0] 26 [0-0] 26 [0-0] 26 0 0 0 0 μg/L low [0-0] 23 [0-0] 25 [0-1.1] 23 [0-0] 25 1.1 1.1 1.7 1.8 high [0-4.4] 28 [0-3.8] 19 [0-6.4] 28 [0-5.2] 19 0 0 0 0 Cmax-BL placebo [0-0] 26 [0-0] 26 [0-0] 26 [0-0] 26 0 0 0 0 μg/L low [0-0] 23 [0-0] 25 [0-1.1] 23 [0-0] 25 1.1 1.1 1.7 1.8 high [0-4.4] 28 [0-3.8] 19 [0-6.4] 28 [0-5.2] 19

tmax placebo n/a 0 n/a 0 n/a 0 n/a 0 h low n/a 0 n/a 0 0.17 1 n/a 0 0.17 0.17 0.17 0.17 high [0.13-0.25] 18 [0.12-0.32] 13 [0.13-0.30] 20 [0.12-0.37] 17 0 0 0 0 AUC0-8.3h placebo [0-0] 26 [0-0] 26 [0-0] 26 [0-0] 26 0 0 0 0 h*μg/L low [0-0] 23 [0-0] 25 [0-0.3] 23 [0-0] 25 0.3 0.4 0.5 0.6 high [0-1.3] 28 [0-1.1] 19 [0-2.5] 28 [0-3.2] 19 AUC>BL- 0 0 0 0 8.3h placebo [0-0] 26 [0-0] 26 [0-0] 26 [0-0] 26 0 0 0 0 h*μg/L low [0-0] 23 [0-0] 25 [0-0.2] 23 [0-0] 25 0.2 0.2 0.4 0.4 high [0-1.0] 28 [0-0.8] 19 [0-1.7] 28 [0-2.3] 19

tlast placebo n/a 0 n/a 0 n/a 0 n/a 0 h low n/a 0 n/a 0 0.17 1 n/a 0 0.17 0.17 0.17 0.18 high [0.13-0.25] 18 [0.12-0.32] 13 [0.13-0.52] 20 [0.12-0.47] 17

Clast placebo n/a 0 n/a 0 n/a 0 n/a 0 μg/L low n/a 0 n/a 0 1.1 1 n/a 0 1.5 1.9 1.7 1.5 high [1.0-4.4] 18 [1.0-3.8] 13 [1.1-6.4] 20 [1.1-4.4] 17 Cannabis administered with Volcano® Medic vaporizer: 500 mg. CBD content in placebo, low, and high doses was 0.001±0.001%, 0.05±0.00%, and 0.19±0.01%, respectively. Active alcohol dose was calculated to produce approximate peak 0.065% peak breath alcohol concentration. CBD limit of quantification (LOQ) was 1 µg/L. For completer pharmacokinetic data only, refer to Table 25. Abbreviations: CBD, cannabidiol; LOQ, limit of quantification; Cmax, maximum concentration; Cmax-BL, Cmax accounting for baseline; tmax, time to maximum concentration; AUC0-8.3h, area under the 8.3 h curve; AUC>BL-8.3h, AUC0-8.3h accounting for baseline; Clast, concentration at last detection.

267

Table 38 (Supplemental). Median [range] blood and plasma cannabinol CBN pharmacokinetic parameters for all 32 participants (full study population, including noncompleters) after controlled cannabis inhalation by vaporizer. CBN Blood Plasma (LOQ 1 μg/L) No Alcohol n Alcohol n No Alcohol n Alcohol n 0 0 0 0 Cmax placebo [0-1.3] 26 [0-1.2] 26 [0-1.4] 26 [0-0] 26 1.3 1.9 1.9 1.8 μg/L low [0-3.9] 23 [0-4.2] 25 [0-5.2] 23 [0-5.8] 25 1.2 1.6 1.8 1.7 high [0-3.7] 28 [0-5.3] 19 [0-4.0] 28 [0-7.3] 19 0 0 0 0 Cmax-BL placebo [0-1.3] 26 [0-1.2] 26 [0-1.4] 26 [-1.2-0] 26 1.3 1.9 1.9 1.8 μg/L low [0-3.9] 23 [0-4.2] 25 [0-5.2] 23 [0-5.8] 25 1.2 1.6 1.8 1.7 high [0-3.7] 28 [0-5.3] 19 [0-4.0] 28 [0-7.3] 19 0.30 tmax placebo 0.17 1 [0.17-0.42] 2 0.17 1 0.17 1 0.17 0.17 0.17 0.17 h low [0.13-0.20] 14 [0.13-0.25] 21 [0.13-0.33] 17 [0.13-0.25] 20 0.17 0.17 0.17 0.17 high [0.13-0.25] 17 [0.12-0.32] 14 [0.13-0.30] 20 [0.12-0.32] 16 0 0 0 0 AUC0-8.3h placebo [0-0.4] 26 [0-0.7] 26 [0-0.4] 26 [0-0] 26 0.5 0.6 0.6 0.5 h*μg/L low [0-1.1] 21 [0-2.5] 25 [0-2.1] 21 [0-3.1] 25 0.4 0.5 0.6 0.5 high [0-1.2] 28 [0-2.4] 19 [0-2.5] 23 [0-3.3] 18 AUC>BL- 0 0 0 0 8.3h placebo [0-0.3] 26 [0-0.8] 26 [0-0.3] 26 [0-0] 26 0.3 0.4 0.4 0.4 h*μg/L low [0-0.8] 23 [0-1.6] 25 [0-1.5] 23 [0-1.8] 25 0.3 0.3 0.4 0.4 high [0-1.1] 28 [0-1.6] 19 [0-1.7] 28 [0-2.1] 19 0.30 tlast placebo 0.17 1 [0.17-0.42] 2 0.17 1 0.17 1 0.17 0.17 0.17 0.18 h low [0.13-0.20] 14 [0.13-0.40] 21 [0.15-0.40] 17 [0.13-0.40] 20 0.17 0.18 0.17 0.18 high [0.13-0.40] 17 [0.12-0.47] 14 [0.13-0.52] 20 [0.12-0.47] 16 1.3 Clast placebo 1.3 1 [1.2-1.4] 2 1.4 1 1.3 1 2.0 1.9 2.1 2.0 μg/L low [1.0-3.9] 14 [1.0-4.1] 21 [1.0-5.2] 17 [1.0-5.4] 20 1.8 1.7 2.0 1.5 high [1.0-3.7] 17 [1.0-3.4] 14 [1.0-3.7] 20 [1.0-2.9] 16 Cannabis administered with Volcano® Medic vaporizer: 500mg. CBN content in placebo, low, and high doses was 0.009±0.003%, 0.22±0.02%, and 0.37±0.03%. Active alcohol dose was calculated to produce approximate peak 0.065% peak breath alcohol concentration. CBN limit of quantification (LOQ) was 1 µg/L. For completer pharmacokinetic data only, refer to Table 25. Abbreviations: CBN, cannabinol; LOQ, limit of quantification; Cmax, maximum concentration; Cmax-BL, Cmax accounting for baseline; tmax, time to maximum concentration; AUC0-8.3h, area under the 8.3 h curve; AUC>BL-8.3h, AUC0-8.3h accounting for baseline; Clast, concentration at last detection.

268

Figure 15 (Supplemental). Median [interquartile range] blood cannabinoids in 19 study completers after controlled cannabis vaporization and placebo and active alcohol. THC content: placebo 0.008±0.002%, low 2.9±0.14%, high 6.7±0.05%. (*)Overall p<0.005 (Friedman's ANOVA); (#)overall p<0.05; (a)p<0.006 (placebo-vs.-high, no alcohol); (b)p<0.006 (placebo-vs.-high, with alcohol); (c)p<0.006 (placebo-vs.-low, no alcohol); (d)p<0.006 (placebo-vs.-low, with alcohol); (e)p<0.006 (low-vs.-high, no alcohol); (f)p<0.006 (low-vs.- high, alcohol). No significant differences were observed (alcohol vs. no alcohol) for any analyte at any individual time point. Abbreviations: THC, D9-tetrahydrocannabinol; 11-OH-THC, 11-hydroxy-THC; THCCOOH, 11-nor-9- carboxy-THC; CBD, cannabidiol; CBN, cannabinol; ANOVA, analysis of variance.

269

Figure 16 (Supplemental). Median [interquartile range] plasma cannabinoids in 19 study completers after controlled cannabis vaporization and placebo and active alcohol. THC content: placebo 0.008±0.002%, low 2.9±0.14%, high 6.7±0.05%. (*)Overall p<0.005 (Friedman's ANOVA); (#)overall p<0.05; (a)p<0.006 (placebo-vs.-high, no alcohol); (b)p<0.006 (placebo-vs.-high, with alcohol); (c)p<0.006 (placebo-vs.-low, no alcohol); (d)p<0.006 (placebo-vs.-low, with alcohol); (e)p<0.006 (low-vs.-high, no alcohol); (f)p<0.006 (low-vs.-high, alcohol). No significant differences were observed (alcohol vs. no alcohol) for any analyte at any individual time point. Abbreviations: THC, ∆9-tetrahydrocannabinol; 11-OH-THC, 11-hydroxy-THC; THCCOOH, 11-nor-9- carboxy-THC; CBD, cannabidiol; CBN, cannabinol; ANOVA, analysis of variance.

270

Figure 17 (Supplemental). Participant 25 blood and plasma ∆9-tetrahydrocannabinol (THC) concentrations following the placebo THC with active alcohol dose. Time course suggests this individual consumed active THC, possibly clandestinely (prior to inhalation). Data are consistent with oral fluid.

271

272

Figure 18 (Supplemental). Median [interquartile range] blood/plasma cannabinoid ratios after controlled cannabis vaporization and placebo and active alcohol. THC content: placebo 0.008±0.002%, low 2.9±0.14%, high 6.7±0.05%. A) Blood/plasma D9-tetrahydrocannabinol (THC) ratios. B) Blood/plasma 11-hydroxy-THC (11-OH-THC) ratios. C) Blood/plasma 11-nor-9-carboxy-THC (THCCOOH) ratios. D) THCCOOH-glucuronide ratios.

Figure 19 (Supplemental). Median [interquartile range] blood and plasma 11-nor-9-carboxy- tetrahydrocannabinol [THCCOOH]-glucuronide to THCCOOH ratios after controlled vaporized cannabis and placebo and active alcohol. THC content: placebo 0.008±0.002%, low 2.9±0.14%, high 6.7±0.05%. (*) Cannabis produced significantly lower ratio (p<0.006). (@)Alcohol produced significantly higher ratio (p<0.006).

273

Chapter 7 – Cannabinoid disposition in oral fluid after controlled vaporizer administration with and without alcohol

As published in Forensic Toxicology, March 2015)3

Abstract

Oral fluid (OF) is an advantageous matrix for cannabis detection, with on-site tests available for roadside drug-impaired driving screening. Limited data exist for device performance following vaporized cannabis, which reduces exposure to harmful combustion byproducts. We assessed cannabinoid OF disposition, with and without alcohol, and evaluated on-site Dräger DrugTest® 5000 performance (Dräger) following controlled vaporized cannabis. Forty-three cannabis smokers (≥1x/3 months, ≤3 days/week) reported 10-16 h prior to dosing, and drank placebo or low (target ~0.065% peak breath-alcohol concentration [BrAC]) dose alcohol 10 min prior to inhaling 500 mg placebo, low (2.9% THC)-, or high (6.7% THC)-dose vaporized cannabis (within- subjects; 6 possible alcohol-cannabis combinations; 19 completers). BrAC and OF

(QuantisalTM, Dräger) were collected before and up to 8.3 h post-dose. Median [range] maximum OF concentrations (Cmax) for low and high doses (no alcohol, N=19) were 848

[32.1-18230] and 764 [25.1-23680] µg/L THC; 6.0 [0-100] and 26.8 [1.0-1106] µg/L cannabidiol; 54.4 [1.8-941] and 29.7 [0-766] µg/L cannabinol; and 24.1 [0-686] and 18.0

[0-414] ng/L 11-nor-9-carboxy-THC (THCCOOH). Lack of significant low-vs.-high- dose THC concentration differences indicated participants may have titrated doses. THC, cannabidiol and cannabinol Cmax occurred immediately post-inhalation, but metabolite

THCCOOH tmax showed interindividual variability. Concurrent alcohol did not affect OF

3 Hartman RL et al. Forensic Toxicology, Epub 10 March 2015. doi: 10.1007/s11419-015-0269-6. 274

cannabinoid concentrations or on-site test sensitivity. At 5 µg/L THC confirmation cutoff, Dräger sensitivity, specificity, and efficiency were 60.8%, 98.2% and 82.5%.

Dräger had lower sensitivity after 6.7% THC vaporization (53.8%, THC ≥2 µg/L confirmation cutoff) than reported following smoking a 6.8% THC cigarette, but high specificity (99.3%) and comparable efficiency (65.0%). Vaporized THC bioavailability may be lower than smoked. Confirmation cutoff, time course, intake histories, and additional cannabinoid analytes affect OF interpretation.

Introduction

Cannabis is the most prevalent illicit drug identified in drivers (122, 136).

Cannabis is frequently consumed together with alcohol, the most common licit drug, and driving under the influence of drugs (DUID) cases often show this combination (274,

301). Both drugs are associated with impairment, alone and combined (34, 41, 261, 274,

297). States that decriminalized medical or recreational cannabis observed increased cannabis-driving cases (275-276), presenting challenges for traffic safety enforcement.

Oral fluid (OF) is an advantageous sampling matrix for drug screening due to ease of collection, non-invasiveness, and facility for on-site testing (110). Observed collection is a deterrent to adulteration, and drugs in OF are frequently associated with recent intake

(74, 109-110). OF is often collected in roadside surveys and case-control studies, wherein participants might elect not to undergo blood collection (122, 136, 302-303). With better knowledge of OF cannabinoid disposition, new workplace and DUID OF drug testing cutoffs (∆9-tetrahydrocannabinol [THC, the primary psychoactive phytocannabinoid] ≥2

µg/L, THC ≥2 µg/L and/or 11-nor-9-carboxy-THC [THCCOOH] ≥20 ng/L) were

275

proposed by the Substance Abuse and Mental Health Services Administration

(SAMHSA) (304). The European Driving under the Influence of Drugs, Alcohol and

Medicines (DRUID) project used THC ≥1 µg/L to ensure identical analytical cutoffs in all laboratories participating in the program (305).

To date, most OF cannabinoid disposition research focused on smoking as the route of administration, because it remains the most prevalent route of intake (63). By controlling inhalation topography (the manner in which the cannabis joint or blunt is smoked), individuals can titrate doses to their desired level, achieving maximum THC concentrations prior to the end of smoking (107, 115, 280). Cannabis vaporization is increasing as a smoking alternative, as it produces lower combustion byproduct-to-THC ratios (66, 69). Vaporizers reduce exposure to harmful polycyclic aromatic hydrocarbons and other respiratory-hazardous combustion products (71, 78-79). A survey querying

6,883 individuals who consumed cannabis at least once in the previous month indicated those who utilized vaporizers were significantly less likely (OR 0.40 controlling for age, sex, cigarette smoking, amount of cannabis consumed) to report respiratory problems than those who smoked or employed other inhalation techniques (70). Subjective effects and plasma THC concentrations are similar for vaporization and smoking, and studies indicated participant preference for vaporization (78). Increasingly, anti-smoking legislation in public facilities causes smokers to search for alternatives; popular e- cigarettes or “vape pens” can conceal cannabis consumption in public settings. As states continue to decriminalize medical or recreational cannabis, vaporization may become more common among health-conscious or discreet smokers. Quantifying OF cannabinoid

276

disposition after vaporization is critical for guiding further development of OF as a test matrix for workplace and DUID investigation.

On-site cannabis screening tests are common tools in DUID cases in the last decade (116, 122-123, 130-131). These technologies’ goals include assisting law enforcement officers to evaluate drug-impaired driving roadside—before drug effects recede during lengthy arrest and booking procedures—and deterring DUID (122-123).

The Dräger DrugTest® 5000 is considered among the most reliable devices for smoked cannabis testing (111, 123-124, 297), but limited data exist for on-site OF devices following vaporized cannabis (74).

We aimed to address these knowledge gaps by evaluating OF cannabinoids and an on-site screening device after vaporization, hypothesizing cannabis vaporization OF results similar to smoking. We quantified and assessed cannabinoid OF disposition, with and without alcohol, and evaluated on-site Dräger DrugTest® 5000 performance (Dräger) following controlled vaporized cannabis administration.

Materials and Methods

Participants

Healthy adult volunteers provided written informed consent for this University of

Iowa Institutional Review Board-approved controlled cannabis administration study.

Participants received comprehensive medical and psychological evaluations to ensure eligibility. Inclusion criteria included: ages 21-55; self-reported average cannabis consumption ≥1x/3 months but ≤3 days/week over the past 3 months; self-reported

“light” or “moderate” alcohol consumption according to quantity-frequency-variability

277

scale; or if “heavy”, not more than 3-4 servings in a typical drinking occasion. Exclusion criteria included: current clinically significant medical history or illness; history of clinically significant adverse event associated with cannabis or alcohol intoxication; ≥450 mL blood donation in 2 weeks preceding drug administration; pregnant or nursing

(pregnancy tests conducted at screening and each dosing visit); interest in drug abuse treatment within 60 days preceding enrollment; and currently taking drugs contraindicated with cannabis or alcohol.

Study design

We utilized a 3x2 factorial design with three cannabis levels (placebo, low, high) and two alcohol levels (placebo, active). Participants entered the research unit approximately 10-16 h prior to drug administration, to preclude intoxication at the time of dosing. Over 10 min ad libitum, participants drank low-dose 90% grain alcohol

(calculated to produce approximately 0.065% peak BrAC) mixed with juice or placebo

(same volume of juice with alcohol-swabbed rim and topped with 1 ml alcohol for taste and odor). After drinking, participants inhaled 500 mg vaporized cannabis plant material over 10 min (Volcano® Medic vaporizer, Storz & Bickel, Tuttlingen, Germany).

Participants received placebo (0.008±0.002% THC, 0.001±0.001% cannabidiol [CBD],

0.009±0.003% cannabinol [CBN]), low (2.9±0.14% THC, 0.05±0.00% CBD,

0.22±0.02% CBN), or high (6.7±0.05% THC, 0.19±0.01% CBD, 0.37±0.03% CBN) cannabis [obtained through NIDA Chemistry and Physiological Systems Research

Branch (Research Triangle Institute, Oxford, MS)] doses. In this within-subjects design, completing participants received each alcohol/cannabis combination, for 6 total sessions.

278

Sessions were separated by ≥1 week to prevent carryover from study drug administration in randomized order.

Breath alcohol concentration (BrAC) was measured with an on-site breath-testing device (Alco-Sensor® IV, Intoximeters, St. Louis, MO) on admission, -0.8 and 0.17, 0.42,

1.4, 2.3, 3.3, 4.3, 5.3, 6.3, 7.3, and 8.3 h after start of cannabis dosing. This measurement device reports results in g/210L breath (limit of quantification [LOQ] 0.006 g/210L), equivalent to approximate BAC. OF specimens were collected immediately following each BrAC measurement (except 0.42 h), with the QuantisalTM collection device

(Immunalysis, Pomona, CA) and the Dräger DrugTest® 5000 (Dräger Safety Diagnostics,

Lübeck, Germany) on-site test, in that sequence.

The QuantisalTM consists of an absorbent pad on a plastic stick, placed under the tongue to collect 1.0±0.1 mL OF. The device comes with a tube containing a standard amount of stabilizing buffer, into which the pad is deposited post-collection. The Dräger cassette contains a polymeric non-compressible pad which is swiped throughout the mouth, tongue, and cheeks to collect 270 µL±15%. Both devices contain a volume adequacy indicator, which changes color when sufficient sample is collected. OF for each device was collected until the indicator turned blue, or after a maximum of 10 min. Low- volume specimens were noted and no weight correction was performed. Oral intake

(eating, drinking, inhaling/smoking) was prohibited 10 min prior to OF collection.

Specimen analysis

Dräger specimens were analyzed in real time on the analyzer, producing a qualitative “Positive”/“Negative” or “Invalid” (if improper lateral flow was detected)

279

response at 5 µg/L THC manufacturer screening cutoff. Confirmatory (QuantisalTM) specimens were diluted in the elution/stabilization buffer at 4°C ≥12 h prior to pad removal and transferred to cryotubes for storage at 4°C. Specimens were analyzed within a month±1 week of collection based on our previous stability study (279). Specimens were quantified for THC, CBD, CBN and the THCCOOH metabolite by a published validated two-dimensional gas chromatography-mass spectrometry method (257), with minor modifications as follows. Before loading the initial elution solvent, 0.4 mL hexane was added to solid-phase extraction columns. THC, THCCOOH, CBD and CBN respective linear ranges were 0.5-50 µg/L, 15-500 ng/L, 1-50 and 1-50 µg/L. Inter- and intra-assay imprecision were <12.3%, and inaccuracy was ≤14.4% (n=21). If concentrations exceeded the upper LOQ, OF specimens were diluted with drug-free

QuantisalTM buffer to achieve concentrations within the method’s linear range.

Data Analysis

Maximum concentration (Cmax), time to Cmax (tmax), and time of last detection

(tlast) were calculated with concentrations observed post-dose. Because some individuals were cannabinoid-positive at baseline, an additional parameter was calculated (Cmax as difference from baseline, Cmax-C0) to account for previously self-administered cannabis.

Area under the curve from baseline to 8.3 h (AUC0-8.3h) was calculated by linear trapezoidal method. If sessions were terminated early (voluntary participant withdrawal), specimens provided were analyzed and included in Dräger calculations. Other measures

(Cmax/tmax/tlast) were assessed only if ≥2 successive subsequent samples were negative or

<20% of maximum. AUC0-8.3h was not evaluated for early terminations.

280

Statistical evaluation was performed with IBM SPSS Statistics Version 19 for

Windows. For statistical purposes, concentrations

Analysis of Variance (ANOVA, factors: cannabis, alcohol) with Bonferroni correction for individuals who completed all 6 sessions. Friedman’s ANOVA was utilized to confirm BAC did not vary significantly by cannabis dose at any time. For alcohol- positive sessions, THC Cmax vs. BAC was compared for placebo, low, and high doses via linear regression on GraphPad Prism 5 (La Jolla, CA). Dräger sensitivity (100*true positives [TP]/(TP+false negatives [FN])), specificity (100*true negatives

[TN]/(TN+false positives [FP])), and efficiency (100*(TN+TP)/(TN+TP+FN+FP)) were calculated for different confirmation cutoffs. Low- vs. high-dose times of last detection

(tlast) were compared for different Dräger screening/confirmation cutoffs via Mann-

Whitney U Test. Fisher’s Exact Test was utilized to compare Dräger performance in the presence and absence of alcohol, at baseline and up to 4.3 h post-inhalation (median alcohol tlast, to ensure comparison of the same time course and prevent over- representation from alcohol-negative sessions). Figures were created on GraphPad Prism

Results and Discussion

Participants

Forty-three healthy adults (26 men, 17 women), ages 21-42 years, provided OF for this study (Table 39). Self-reported cannabis history varied considerably between

281

Table 39. Self-reported demographic characteristics and recent cannabis and alcohol consumption history of 43 healthy adult occasional cannabis smokers. Hours Doses Race Typical Time since Amount last Alcohol Cannabis “stoned” received Parti- Age and BMI drinks last cannabis consumedb Sex intake intake on typical (Reason for cipant (years) ethni- (kg/m2) per consumed (joint or joint frequency frequency cannabis withdrawal) city occasion (days) equivalent) occasiona 1 F 30.6 W 21.4 2-4x/m 2-4 2-3x/wk 1-2 1 2 2 (P) 2c M 23.7 W 24.3 2-3x/wk 2-4 2-4x/m 1-2 1 1 6 3c F 28.4 AA 23.8 ≥4x/wk 2-4 2-4x/m 3-4 14 1 6 4 M 27.8 W 33.2 2-3x/wk 2-4 2-3x/wk 1-2 1 1 3 (P) 5c M 21.9 W 24.7 2-3x/wk 5-6 2-4x/m 1-2 6 1 6 6c M 37.8 W 26.1 2-3x/wk 2-4 2-3x/wk 1-2 3 2.5 6 7c M 26.6 W 21.6 ≤1x/m 2-4 ≤1x/m 1-2 11 3.5 6 8 F 34.9 W 24.5 2-3x/wk 2-4 2-3x/wk 1-2 2 0.25 1 (AE) 9c F 26.3 W 20.0 2-3x/wk 2-4 2-3x/wk 3-4 1 0.25 6 10c M 25.8 W 40.6 2-4x/m 2-4 2-3x/wk 1-2 0.3 0.5 6 282 11c M 26.1 H 31.5 2-4x/m 1-2 2-3x/wk 1-2 3 1 6

12 M 29.5 W 32.6 2-3x/wk 1-2 ≤1x/m 5-6 21 1 2 (AE) 13 M 26.9 W 22.9 2-3x/wk 1-2 ≤1x/m 3-4 2 1 3 (P) 14c M 23.2 W 19.5 2-3x/wk 2-4 2-3x/wk 3-4 2 1 6 15 F 24.0 As 19.6 2-3x/wk 2-4 2-4x/m <1 3 1 1 (AE) 16c M 23.1 W 23.9 2-4x/m 2-4 ≤1x/m 1-2 2 0.25 6 17 M 22.7 W, H 23.4 2-3x/wk 2-4 2-4x/m 1-2 3 2 1 (P) 18 M 21.1 W 20.6 2-3x/wk 5-6 2-3x/wk 1-2 2 2 3 (P) 19c M 32.3 O, H 28.9 2-3x/wk 2-4 2-3x/wk 1-2 4 1 6 20c F 23.4 W 23.3 2-3x/wk 2-4 2-4x/m 3-4 4 1 6 21c F 30.3 AA 24.1 2-3x/wk 2-4 ≤1x/m <1 120 1 6 22c M 24.6 W 23.3 2-3x/wk 2-4 2-4x/m 1-2 7 0.8 6 23 F 34.8 W 21.2 2-3x/wk 2-4 2-4x/m 3-4 2 1 1 (AE) 24 M 40.8 W 31.7 2-3x/wk 2-4 2-4x/m 3-4 5 3 2 (P) 25 F 21.8 W 30.8 2-4x/m 2-4 2-3x/wk 1-2 183 0.5 4 (P) 26 M 42.1 W 24.2 2-4x/m 1-2 ≤1x/m 1-2 45 2 2 (P) 27 M 39.4 W,As 34.6 2-4x/m 2-4 2-4x/m 3-4 1 4.5 4 (P) AI, As, 28 M 21.1 24.0 2-4x/m 2-4 2-3x/wk 5-6 2 1 2 (P) AA, W

Table 39. (Continued from previous page) Hours Doses Race Typical Time since Amount last Alcohol Cannabis “stoned” received Parti- Age and BMI drinks last cannabis consumedb Sex intake intake on typical (Reason for cipant (years) ethni- (kg/m2) per consumed (joint or joint frequency frequency cannabis withdrawal) city occasion (days) equivalent) occasiona 29 F 24.6 W, H 19.1 2-3x/wk 2-4 2-4x/m 3-4 28 0.5 3 (AE) 30c,d M 21.8 W 32.7 ≤1x/m 1-2 2-4x/m 1-2 7 0.13 6 31 F 24.8 W, H 26.7 2-3x/wk 1-2 2-4x/m 3-4 21 4 1 (AE) 32 M 29.0 O 28.0 2-3x/wk 2-4 ≤1x/m <1 30 0.2 2 (P) 33 F 23.0 W 21.0 2-3x/wk 2-4 2-4x/m 5-6 7 0.3 2 (P) 34c F 21.7 AA, W 23.0 2-4x/m 1-2 2-3x/wk 1-2 1.1 1.5 6 35c M 28.7 W 18.3 2-3x/wk 2-4 ≤1x/m 3-4 45 0.5 6 36 F 24.4 W 21.6 2-3x/wk 2-4 2-3x/wk 3-4 2 2 1 (P) 37c M 28.1 W 48.3 2-4x/m 2-4 2-4x/m 3-4 5 1 6 38c F 22.9 W 21.6 2-4x/m 5-6 2-3x/wk 3-4 1 1 6 39 F 37.3 W 24.8 2-4x/m 1-2 2-4x/m 1-2 4 1 1 (P)

283 40 F 22.5 W 19.7 2-3x/wk 2-4 2-3x/wk 1-2 1 1 1 (P) 41 M 25.8 AA 28.8 2-3x/wk 2-4 2-4x/m <1 14 1 1 (AE) 42 M 22.7 W 26.1 2-4x/m 1-2 2-4x/m 1-2 8 1 3 (P) 43 M 26.7 W 23.5 2-3x/wk 2-4 ≤1x/m 1-2 11 2 1 (AE) Median (all) 25.8 24.0 4.0 1.0 Mean (all) 27.3 25.7 14.8 1.3 StDev (all) 5.7 6.0 33.1 1.0 Median 25.8 23.9 4.0 1.0 (completers) Mean 26.1 26.3 12.5 1.0 (completers) StDev 4.1 7.5 27.9 0.8 (completers) a‘Hours “stoned” ’ wording originates from Cannabis Use Disorders Identification Test, source of self-reported cannabis frequency data bCannabis amount last consumed is based on empirically-normalized joint consumption, to account for various administration routes and self-reported “sharing” between multiple individuals cParticipant completed all 6 study sessions dMay have consumed active cannabis during placebo-cannabis sessions Abbreviations: W, White; AA, African American; H, Hispanic or Latino; As, Asian; O, Other; AI, American Indian/Native American; P, withdrew for personal reasons (job obligations/scheduling/choice); AE, withdrew due to adverse event (nausea/emesis or dizziness, related to study drugs or other study procedures); StDev, standard deviation

individuals. Two participants (21 and 25) reported most recent cannabis intake 4 and 6 months prior to admission, despite indicating overall average intake at least once/3 months. However, most consumed cannabis within the last week. Nineteen participants completed all 6 dosing sessions. The 24 other participants withdrew for personal reasons

(e.g., job obligations, scheduling, elected to withdraw) or adverse events (e.g., nausea/emesis or dizziness related to study drugs or other study procedures) (Table 39) lists doses received and reasons for withdrawal). There were no significant differences between completers and noncompleters in age, weight, BMI, or self-reported cannabis history (p>0.21, Mann-Whitney U [exact] test).

Alcohol

Completers’ breath alcohol Cmax, tmax, tlast, and AUC0-8.3h are summarized in Table

40. For informational purposes, pharmacokinetic data from all participants (including noncompleters) is provided in Table 44 (Supplemental). Within-subjects alcohol doses produced similar AUC0-8.3h. Alcohol concentration did not significantly differ between alcohol-positive doses at any time point, nor did overall alcohol Cmax and AUC0-8.3h

(Figure 20). Active cannabis (relative to placebo) resulted in significantly later alcohol

2 tmax (ANOVA χ (2)=6.621, p=0.037), but alcohol tmax did not significantly differ between active (low vs. high) cannabis doses. Alcohol did not significantly affect THC Cmax

(Figure 21, no slopes differed significantly from 0) or THC tmax. Alcohol displayed a typical zero-order elimination profile (197, 306), and was not detected after 5.3 h.

284

Table 40. Median [range] QuantisalTM oral fluid pharmacokinetic parameters following controlled vaporized placebo, low (2.9%), or high (6.7%) THC cannabis with or without low-dose alcohol for 19 occasional to moderate smokers who completed all six dosing conditions. BrAC Breath Alcohol Concentration (LOQ 0.006 g/210L) (Active Alcohol Sessions)

Cmax placebo THC dose 0.063 [0.034-0.135] μg/L low THC dose 0.062 [0.035-0.097] high THC dose 0.053 [0.036-0.087]

tmax placebo THC dose 0.42 [0.17-1.4] h low THC dose 0.42 [0.17-2.3] high THC dose 1.4 [0.17-2.3] tlast placebo THC dose 4.3 [3.3-5.3] h low THC dose 4.3 [2.3-5.3] high THC dose 4.3 [2.3-5.3] AUC0-8.3h placebo THC dose 0.166 [0.103-0.234] h*μg/L low THC dose 0.171 [0.074-0.257]

high THC dose 0.151 [0.104-0.226]

THC Oral Fluid Concentration (LOQ 0.5 μg/l) No Alcohol Alcohol

Cmax placebo 5.0 [0-25.9]* 3.9 [0-27.2]* μg/L low 848 [32.1-18230]* 735 [72.9-7494]* high 764 [25.1-23680]* 952 [22.7-66200]* C0 placebo 0.62 [0-14.2] 0 [0-11.3] μg/L low 0.54 [0-30.7] 0 [0-72.9] high 0 [0-11.7] 0.55 [0-34.2] Cmax-C0 placebo 4.2 [-3.0-24] 2.1 [-2.2-22.6] μg/L low 847 [32.1-18206] 735 [71-7494] high 762 [25.1-23671] 952 [22.7-66192] tmax placebo 0.17 [0.17-1.4] 0.17 [0.17-2.3] h low 0.17 [0.17-0.17] 0.17 [0.17-0.17] high 0.17 [0.17-3.3] 0.17 [0.17-0.17] tlast placebo 5.8 [1.4-8.3]* 8.3 [1.4-8.3]* h low 8.3 [3.3-8.3]* 8.3 [8.3-8.3]* high 8.3 [7.3-8.3]* 8.3 [4.3-8.3]* AUC0-8.3h placebo 7.1 [0-56.1] 8.8 [0-39.4] h*μg/L low 723 [29.8-3865] 625 [88.8-8146]

high 880 [38.4-19090] 917 [25.2-53984]

THCCOOH Oral Fluid Concentration (LOQ 15 ng/l) No Alcohol Alcohol

Cmax placebo 0 [0-361] 0 [0-370] ngLl low 24.1 [0-686] 37.7 [0-992] high 18.0 [0-414] 24.0 [0-909] C0 placebo 0 [0-249] 0 [0-243] ng/L low 0 [0-505] 0 [0-911] high 0 [0-223] 0 [0-468] Cmax-C0 placebo 0 [-18.6-113]* 0 [-17.3-193]* ng/L low 22.8 [0-182]* 32.5 [0-219]* high 18.0 [0-192]* 24.0 [0-441]*

285

Table 40. (Continued from previous page) THCCOOH Oral Fluid Concentration (LOQ 15 ng/l) No Alcohol Alcohol tmax placebo 2.3 [1.4-8.3] 2.3 [1.4-8.3] h low 2.3 [0.17-8.3] 1.4 [0.17-7.3] high 2.3 [0.17-5.3] 1.4 [0.17-3.3] tlast placebo 8.3 [7.3-8.3] 8.3 [1.4-8.3] h low 8.3 [0.17-8.3] 8.3 [1.4-8.3] high 8.3 [0.17-8.3] 8.3 [2.3-8.3] AUC0-8.3h placebo 0 [0-1941] 0 [0-1904] h*ng/L low 42.9 [0-2935] 185 [0-5153] high 14.2 [0-1827] 70.0 [0-3536]

CBD Oral Fluid Concentration (LOQ 1 μg/l) No Alcohol Alcohol

Cmax placebo 0 [0-0]* 0 [0-1.1]* μg/L low 6.0 [0-100]* 2.4 [0-46.5]* high 26.8 [1.0-1106]* 37.1 [0-2331]* C0 placebo 0 [0-0] 0 [0-0] μg/L low 0 [0-0] 0 [0-0] high 0 [0-0] 0 [0-0] Cmax-C0 placebo 0 [0-0]* 0 [0-1.1]* μg/L low 6.0 [0-100]* 2.4 [0-46.5]* high 26.8 [1-1106]* 37.1 [0-2331]* tmax placebo -- 0.17 [0.17-0.17] h low 0.17 [0.17-0.17] 0.17 [0.17-0.17] high 0.17 [0.17-3.3] 0.17 [0.17-0.17] tlast placebo -- 0.8 [0.17-1.4] h low 0.17 [0.17-2.3]*#lh 0.17 [0.17-3.3]*#lh high 2.3 [0.17-8.3]*#lh 3.3 [0.17-8.3]*#lh AUC0-8.3h placebo 0 [0-0] 0 [0-0.82] h*μg/L low 3.1 [0-79] 1.7 [0-53.2] high 30.3 [0.72-912] 38.8 [0-1911]

CBN Oral Fluid Concentration (LOQ 1 μg/l) No Alcohol Alcohol

Cmax placebo 0 [0-1.8] 0 [0-2.1] μg/L low 54.4 [1.8-941] 49.4 [3.2-312] high 29.7 [0-766] 31.7 [0-2650] C0 placebo 0 [0-0] 0 [0-0] μg/L low 0 [0-1.1] 0 [0-3.6] high 0 [0-0] 0 [0-0] Cmax-C0 placebo 0 [0-1.8] 0 [0-2.1] μg/L low 54.4 [1.8-941] 49.4 [3.2-312] high 29.7 [0-766] 31.7 [0-2650] tmax placebo 0.17 [0.17-0.17] 0.17 [0.17-0.17] h low 0.17 [0.17-0.17] 0.17 [0.17-0.17] high 0.17 [0.17-3.3] 0.17 [0.17-0.17] tlast placebo 0.17 [0.17-0.17] 0.17 [0.17-1.4] h low 2.3 [0.17-7.3]#lh 2.3 [0.17-8.3] #lh high 2.3 [0.17-8.3] #lh 3.3 [0.17-8.3] #lh

286

Table 40. (Continued from previous page) CBN Oral Fluid Concentration (LOQ 1 μg/l) No Alcohol Alcohol AUC0-8.3h placebo 0 [0-1.3] 0 [0-1.5] h*μg/L low 44.1 [1.3-246] 39.7 [2.3-405] high 25.9 [0-617] 29.6 [0-2226] *Significant overall cannabis dose effect by factorial repeated-measures analysis of variance ANOVA) #Significant overall alcohol dose effect by factorial repeated-measures ANOVA lhSignificant overall effects based on low vs. high dose THC ANOVA only, due to insufficient positive placebo Abbreviations: THC, ∆9-tetrahydrocannabinol; LOQ, limit of quantification; THCCOOH, 11-nor-9- carboxy-THC; CBD, cannabidiol; CBN, cannabinol

Figure 20. Median [interquartile range] breath alcohol concentration (BrAC) in 19 completers following drinking placebo and three equivalent Everclear grain alcohol doses at separate sessions, with controlled inhalation of placebo, low (2.9%) or high (6.7%) ∆9- tetrahydrocannabinol (THC) vaporized cannabis. In total, there were three sessions (placebo, low, high cannabis) with no alcohol; alcohol was never detected in any of these sessions. Vertical dotted line represents start of cannabis administration. *Overall cannabis p≤0.004 by repeated-measures analysis of variance (ANOVA) with Bonferroni correction for repeated measures. Significance level set to p<0.05/12 measurements = p<0.004

287

Figure 21. ∆9-Tetrahydrocannabinol (THC) maximum oral fluid concentration vs. breath alcohol concentration (BrAC) for placebo, low (2.9%), and high (6.7%) THC doses (all participant data) following drinking alcohol and inhaling controlled cannabis by vaporizer. Line correlations were not significantly non-zero (THC concentrations did not vary with BrAC). Except for Participant 30, THC concentrations did not exceed 42.6 µg/L after the placebo dose, and did not exceed 21.0 µg/L when baseline OF THC was negative

QuantisalTM OF cannabinoids

Completers’ OF THC, CBD, CBN, and THCCOOH pharmacokinetic data and statistical analysis (Factorial ANOVA) are presented in Table 40 and Table 41. No significant alcohol*cannabis interactions were observed. All participants’ data and pairwise comparisons (Table 45 (Supplemental), Table 46 (Supplemental), Table 47

(Supplemental), and Table 48 (Supplemental)) corroborated results from completers.

THC Cmax was significantly higher after low and high doses (with and without alcohol) than placebo, and AUC0-8.3h was significantly higher after the low dose than placebo

(high vs. placebo had a trend for completers, p=0.056) (Table 40, Table 45

(Supplemental)). No dose difference was observed in THC tmax (immediately after

288

Table 41. Overall effects of alcohol, cannabis, and alcohol*cannabis interaction on oral fluid maximum concentration (Cmax), time to maximum concentration (tmax), time of last 9 detection (tlast), and area under the curve (AUC0-8.3h) for cannabinoids ∆ - tetrahydrocannabinol (THC), cannabidiol (CBD), cannabinol (CBN), and 11-nor-9- carboxy-THC (THCCOOH) after inhalation of vaporized cannabis Error Pairwise Degrees Analyte Overall degrees Effect comparison by N F of p-value parameter effect of size, r cannabis dose freedom freedom THC Cmax Alcohol 19 1.403 1 18 0.27 0.252 Cannabis 4.957 1.10 19.75 0.035a low vs. placebo 10.097 1 18 0.60 0.005 high vs. placebo 6.027 1 18 0.50 0.024 low vs. high 3.227 1 18 0.39 0.089 Alcohol* 1.963 1.12 20.22 0.176a Cannabis Cmax-C0 Alcohol 18 0.917 1 17 0.23 0.352 Cannabis 4.234 1.10 18.62 0.051a low vs. placebo 8.503 1 17 0.58 0.010 high vs. placebo 5.141 1 17 0.48 0.037 low vs. high 2.786 1 17 0.38 0.113 Alcohol* 1.347 1.13 19.20 0.266a Cannabis tmax Alcohol 17 1.250 1 16 0.27 0.280 Cannabis 2.292 1.25 23.61 0.134a Alcohol* 0.278 1.21 19.28 0.647a Cannabis tlast Alcohol 14 0.019 1 13 0.04 0.894 Cannabis 11.798 1.01 13.11 0.004a low vs. placebo 11.729 1 13 0.69 0.005 high vs. placebo 11.939 1 13 0.69 0.004 low vs. high 0 1 13 0 1.00 Alcohol* 0.065 1.05 13.71 0.815a Cannabis AUC0-8.3h Alcohol 17 1.643 1 16 0.31 0.218 Cannabis 3.283 1.04 16.64 0.087a low vs. placebo 15.605 1 16 0.70 0.001 high vs. placebo 4.231 1 16 0.46 0.056 low vs. high 2.008 1 16 0.33 0.176 Alcohol* 1.136 1.05 16.85 0.305a Cannabis

CBD Cmax Alcohol 19 0.970 1 18 0.23 0.338 Cannabis 5.829 1.00 18.05 0.027a low vs. placebo 12.461 1 18 0.64 0.002 high vs. placebo 6.158 1 18 0.50 0.023 low vs. high 5.487 1 18 0.48 0.031 Alcohol* 1.098 1.01 18.10 0.309a Cannabis

289

Table 41. (Continued from previous page) Error Pairwise Degrees Analyte Overall degrees Effect comparison by N F of p-value parameter effect of size, r cannabis dose freedom freedom CBD

Cmax-C0 Alcohol 18 0.626 1 17 0.19 0.440 Cannabis 5.142 1.00 17.04 0.037a low vs. placebo 11.188 1 17 0.63 0.004 high vs. placebo 5.435 1 17 0.49 0.032 low vs. high 4.838 1 17 0.47 0.042 Alcohol* 0.721 1.01 17.09 0.408a Cannabis 0.19 b tmax Alcohol low vs. high 11 1 1 10 0.30 0.341 Cannabis low vs. highb 1 1 10 0.30 0.341 Alcohol* low vs. highb 1 1 10 0.30 0.341 Cannabis b tlast Alcohol low vs. high 11 7.784 1 10 0.66 0.019 Cannabis low vs. highb 25.339 1 10 0.84 0.001 Alcohol* low vs. highb 3.272 1 10 0.50 0.101 Cannabis AUC0-8.3h Alcohol 17 1.284 1 16 0.27 0.274 Cannabis 4.245 1.00 16.05 0.056a low vs. placebo 9.186 1 16 0.60 0.008 high vs. placebo 4.564 1 16 0.47 0.048 low vs. high 3.919 1 16 0.44 0.065 Alcohol* 1.404 1.01 16.11 0.254a Cannabis

CBN

Cmax Alcohol 19 0.982 1 18 0.23 0.335 Cannabis 3.921 1.23 22.16 0.053a low vs. placebo 11.606 1 18 0.63 0.003 high vs. placebo 5.179 1 18 0.47 0.035 low vs. high 1.110 1 18 0.24 0.306 Alcohol* 1.494 1.20 21.58 0.240a Cannabis Cmax-C0 Alcohol 18 0.775 1 17 0.21 0.391 Cannabis 3.573 1.22 20.80 0.066a low vs. placebo 9.707 1 17 0.60 0.006 high vs. placebo 4.718 1 17 0.47 0.044 low vs. high 1.178 1 17 0.25 0.293 Alcohol* 1.138 1.20 20.34 0.332 Cannabis b tmax Alcohol low vs. high 15 1.775 1 14 0.34 0.204 Cannabis low vs. highb 1.775 1 14 0.34 0.204 Alcohol* low vs. highb 1.775 1 14 0.34 0.204 Cannabis b tlast Alcohol low vs. high 16 8.477 1 15 0.60 0.011 Cannabis low vs. highb 0.008 1 15 0.02 0.929 Alcohol* low vs. highb 2.583 1 15 0.38 0.129 Cannabis

290

Table 41. (Continued from previous page) Error Pairwise Degrees Analyte Overall degrees Effect comparison by N F of p-value parameter effect of size, r cannabis dose freedom freedom CBN

AUC0-8.3h Alcohol 17 1.871 1 16 0.32 0.190 Cannabis 2.666 1.08 17.23 0.119a low vs. placebo 17.478 1 16 0.72 0.001 high vs. placebo 3.634 1 16 0.43 0.075 low vs. high 0.942 1 16 0.23 0.346 Alcohol* 1.018 1.09 17.40 0.334a Cannabis

THCCOOH Cmax Alcohol 19 1.340 1 18 0.26 0.262 Cannabis 3.740 1.43 25.68 0.051a low vs. placebo 3.873 1 18 0.42 0.065 high vs. placebo 5.301 1 18 0.48 0.033 low vs. high 0.087 1 18 0.07 0.772 Alcohol* 1.273 1.26 22.63 0.282a Cannabis Cmax-C0 Alcohol 19 4.314 1 18 0.44 0.052 Cannabis 9.427 1.10 19.66 0.005a low vs. placebo 14.611 1 18 0.67 0.001 high vs. placebo 9.828 1 18 0.59 0.006 low vs. high 4.920 1 18 0.46 0.040 Alcohol* 1.828 1.19 21.68 0.191a Cannabis b tmax Alcohol low vs. high 10 2.018 1 9 0.43 0.189 Cannabis low vs. highb 1.755 1 9 0.40 0.218 Alcohol* low vs. highb 1.932 1 9 0.42 0.198 Cannabis b tlast Alcohol low vs. high 10 0.670 1 9 0.26 0.434 Cannabis low vs. highb 0.522 1 9 0.23 0.488 Alcohol* low vs. highb 0.264 1 9 0.17 0.619 Cannabis b AUC0-8.3h Alcohol low vs. high 17 1.124 1 16 0.26 0.305 Cannabis low vs. highb 2.041 1.52 24.33 0.160a Alcohol* low vs. highb 0.213 1.32 21.15 0.716a Cannabis Data are from 19 individuals who participated in all dosing sessions. Statistical analysis performed by factorial repeated-measures analysis of variance (ANOVA). aMauchly’s test showed sphericity was violated on main effects, so Greenhouse-Geisser correction was utilized. bPlacebo doses not included in ANOVA due to too few positive specimens for comparison. Boldface represents statistical significance at p<0.05. Where “low vs. high” is the only overall effect compared, there were too few positive specimens after the placebo dose for a statistical comparison

291

dosing). After active cannabis, the median tlast was ≥8.3 h, not significantly different between low and high doses. No significant low- vs. high-dose differences were observed for OF THC at any time post-dose (Figure 22); however, the high dose showed greater interindividual variability, particularly after alcohol (Table 40). Placebo cannabis contained 0.008±0.002% THC, and low THC concentrations were detected in OF after this dose, even after accounting for baseline. However, OF THC never exceeded 42.6

µg/L after placebo, except for Participant 30 (described below). When baseline OF THC was 0, placebo THC tmax did not exceed 21.0 µg/L. Figure 24 (Supplemental) depicts

THC and THCCOOH before dosing and over 8.3 h for placebo sessions.

CBD Cmax was significantly greater and had substantial variability after the high compared to low dose cannabis (Table 40, Figure 22). CBD tmax occurred immediately after inhalation; after placebo and low doses, tlast was typically 0.17 h. After the high dose, median tlast shifted significantly (p=0.033) to 2.3 and 3.3 [0.17-≥8.3] h for non- alcohol and alcohol conditions, respectively (Table 40). At individual sampling times over the first 7.3 h, there was a significant overall dose difference (p<0.05, Figure 22).

Specific differences by post-dose time are provided in Figure 22. CBD was only detected in OF after placebo (0.05% potency) in one and two sessions without and with alcohol, respectively. In the placebo-without-alcohol session (Participant 24), Cmax-C0=-0.4 μg/L, indicating CBD detected was carryover from a previous self-administration. In the active- alcohol sessions all participants were negative for CBD at baseline (C0=0 μg/L), indicating detected CBD (Participants 6 and 39) came from the placebo cannabis dose

(Table 40). CBD low vs. high dose differences can be explained by the 4-fold difference

292

Figure 22. Median (interquartile range) oral fluid a) ∆9-tetrahydrocannabinol (THC), b) cannabidiol (CBD), c) cannabinol (CBN) and d) 11-nor-9-carboxy-THC (THCCOOH) vs. time after controlled vaporized cannabis inhalation in 19 completers. Horizontal dotted line represents analyte limit of quantification (LOQ); vertical dotted line represents start of cannabis administration. (*)Doses significantly different overall by Friedman’s ANOVA, p≤0.001. (#)Overall dose effect p<0.05 by Friedman’s ANOVA, for informational purposes. (Bonferroni correction sets significance level at p<0.05/11 measurements = p<0.005). (+)All placebo doses significantly different to all active THC doses (p<0.005), with no significant differences between any active doses. (‡)All placebo doses different to all active THC doses (p<0.05), with no significant differences between any active doses. (α)p<0.05 for placebo vs. low (no alcohol), placebo vs. high (with and without alcohol), and low vs. high (with alcohol). (β)p<0.05 for placebo vs. high (with and without alcohol), and for low vs. high (with and without alcohol). (γ)p<0.05 for placebo vs. high (with and without alcohol), and for low vs. high (with alcohol). (δ)p<0.05 for placebo vs. low (with alcohol, and for placebo vs. high (with alcohol). (ε)p<0.05 for placebo vs. high (with alcohol)

in CBD potency of the cannabis. Participants titrate their dose based on psychoactive

THC concentrations (only 2-fold low-high dose THC potency difference); titration is not based on CBD, as CBD is non-psychoactive.

Low- vs. high-dose CBN Cmax and AUC0-8.3h did not significantly differ (Table 40 and Table 41). CBN tmax occurred within 3.3 h post-dose, but was 0.17 h in 98% of

293

sessions. Participants 3 and 37 had 1.4 h and 3.3 h CBN tmax, respectively in their high dose/no alcohol sessions; all other CBN tmax were 0.17 h. The only significant pairwise alcohol difference was tlast with high cannabis (Table 47 (Supplemental)). As with CBD, alcohol produced significantly later CBN tlast. CBN concentrations and specific differences by post-dose time are provided in Figure 22. CBN was only detected in 5 participants’ OF after placebo cannabis, in both alcohol conditions.

THCCOOH displayed substantial interindividual OF concentration variability at all doses, reflecting participants’ smoking history (Table 40, Figure 22). There were no significant low- vs. high-dose differences in Cmax or AUC0-8.3h, but THCCOOH Cmax accounting for baseline was significantly higher after the high than low dose in completers (Table 41). Alcohol had no effect on any THCCOOH results. Low- and high- dose Cmax-C0 were significantly higher than placebo, and Cmax-C0 ranges demonstrated notable differences relative to Cmax ranges (Table 40). Median THCCOOH tmax occurred

1.4-2.8 h post-dose for every condition; however, substantial variability was noted due to smoking history/body burden and individual metabolic rates. When detected, THCCOOH tlast was typically ≥8.3 h.

On admission the night prior to dosing, 51% of QuantisalTM specimens were positive for THC (0.52-440.8 µg/L); 5% for CBD (1.1-41.7 µg/L), 16% for CBN (1.0-

33.3 µg/L), and 38% for THCCOOH (15.1-887 ng/L). The following morning at baseline, 47% of all specimens remained positive for THC (0.54-72.9 µg/L), 0.6% for

CBD (2.1 µg/L), 2% for CBN (1.1-3.6 µg/L), and 34% for THCCOOH (15.1-911 ng/L).

Participants 6, 7, 10, 27, and 38 were THCCOOH-positive at baseline (after overnight) for all doses received, and each had at least one baseline QuantisalTM OF with

294

THCCOOH>100 ng/L. OF THC was ≥1 µg/L and THCCOOH≥20 ng/L at baseline across all their sessions.

Based on pharmacokinetic data, Participant 30 may have accessed active cannabis during his placebo sessions, despite being under observation throughout his stay (Figure

25 (Supplemental)). For his placebo with alcohol session, THC was negative on admission to the unit, but positive prior to dosing; THC, CBD, and CBN Cmax were 569,

17.8, and 54.8 µg/L at 0.17 h. It is possible these high concentrations resulted from dosing error; however, records were carefully reviewed and there was no indication that an error occurred. Because he was negative on admission and positive at baseline, we cannot rule out clandestine intake prior to dosing. For his placebo without alcohol session, THC and CBN Cmax were 22.7 and 2.2 µg/L at 5.3 h, despite being lower/negative earlier post-dose. His active doses did not contain anomalous findings.

Data from these placebo-cannabis sessions were excluded from median [range] calculations for Cmax, tmax, tlast, and AUC0-8.3h and for matched-pairs analyses.

Dräger DrugTest® 5000 performance and confirmation comparison

In total, 1,723 OF Dräger-QuantisalTM specimen pairs were obtained. Thirteen

Dräger specimens (0.8%) produced “invalid” results, leaving 1,710 for comparison.

Dräger performance at various quantitative cutoffs examined previously for smoking (due to proposed SAMHSA guidelines or utilized in the DRUID program) (117, 124, 307) is summarized in Table 49 (Supplemental). Alcohol presence did not affect Dräger performance. Overall sensitivity at manufacturer-specified 5 µg/l THC confirmation cutoff was 60.8% over 8.3 h. Specificity was high at 98.2%, yielding 82.5% overall

295

efficiency. At lower QuantisalTM THC confirmation cutoffs, sensitivity decreased.

Including CBD and CBN as required confirmatory analytes produced higher sensitivity

(89.2% [CBD] and 86.4% [CBN]) for THC ≥2 µg/L and for THC ≥1 µg/L. These numbers were identical for both THC cutoffs because CBD and CBN were not present when THC <2 µg/L. This also explains the higher sensitivity, since fewer confirmatory specimens were considered positive when CBD or CBN were required. Detection rates

(from the 19 completers) vs. post-dose time for several possible confirmation criteria are presented in Figure 23. Data were identical for THC ≥2 or 1 µg/L for completers. Dräger tlast for the various cutoffs are presented in Table 42 (low vs. high, completers) and Table

49 (Supplemental) (full study population). Overall, the DrugTest® 5000 was positive for

THC 3.3 [0.17-≥8.3] h (median [range]) after dosing. The only significant high vs. low tlast difference among the various confirmation cutoffs was when CBD and THC were required. This corresponds to the finding that CBD had significantly later low vs. high tlast. To compare to smoking a 6.8% THC cigarette (111), the 546 tests (549, 3 “invalid”) from high-dose sessions also were evaluated (Table 43). Sensitivity for the high-dose only increased relative to overall results (for all confirmation cutoffs except those requiring CBD), but sensitivity was lower after vaporization than smoking (111).

Discussion

We present, for the first time, THC, CBD, CBN and THCCOOH disposition in

OF following controlled vaporized cannabis administration. Prior clinical data following cannabis vaporization are limited. One other study examined OF after vaporization, but specimens were only collected 0.08 and 1.3 h post-inhalation (74).

296

297

Figure 23. Dräger DrugTest® 5000 oral fluid cannabis detection rates over time in 19 completers with different confirmation cutoff criteria.

Table 42. Median [range] low (2.9% THC) and high (6.7% THC) dose time of last Dräger DrugTest® 5000 on-site test positive detection in 19 completers only (5 µg/L ∆9- tetrahydrocannabinol (THC) oral fluid screening cutoff) with different oral fluid confirmation cutoffs, following oral inhalation of cannabis by Volcano Medic vaporizer. Quantitative Confirmation Median [Range] p-value Cutoffs t (h) (Low vs. µg/l (THC, CBD, CBN) last Lowa, Highb High) ng/l (THCCOOH) THC ≥5 3.3a,b [0.17-8.3] 0.189 THC ≥2 (SAMHSA) 3.3a,b [0.17-8.3] 0.330 THC ≥1 (DRUID) 3.3a,b [0.17-8.3] 0.330 3.3a [0.17-8.3] THC ≥2 and THCCOOH ≥20 0.171 5.3b [1.4-8.3] 3.3a [0.17-8.3] THC ≥1 and THCCOOH ≥20 0.171 5.3b [1.4-8.3] 0.17a [0.17-3.3] THC ≥2 and CBD ≥1 <0.001 3.3b [1.4-8.3] THC ≥2 and CBN ≥1 2.3a,b [0.17-8.3] 0.579 3.3a [0.17-8.3] THCCOOH ≥20 0.171 5.3b [1.4-8.3] 0.17a [0.17-3.3] THC ≥1 and CBD ≥1 <0.001 3.3b [0.17-8.3] THC ≥1 and CBN ≥1 2.3a,b [0.17-8.3] 0.579 THC ≥2 or THCCOOH ≥20 3.3a,b [0.17-8.3] 0.330 THC ≥1 or THCCOOH ≥20 3.3a,b [0.17-8.3] 0.330 THC ≥2 or CBD ≥1 3.3a,b [0.17-8.3] 0.330 THC ≥2 or CBN ≥1 3.3a,b [0.17-8.3] 0.330 THC ≥1 or CBD ≥1 3.3a,b [0.17-8.3] 0.330 THC ≥1 or CBN ≥1 3.3a,b [0.17-8.3] 0.330 Abbreviations: THC, ∆9-tetrahydrocannabinol; CBD, cannabidiol; CBN, cannabinol; THCCOOH, 11-nor-9-carboxy-THC; SAMHSA, Substance Abuse and Mental Health Services Administration; DRUID, Driving Under the Influence of Drugs, Alcohol and Medicines

298

Table 43. Performance characteristics for the Dräger DrugTest® 5000 on-site test (5 µg/L ∆9-tetrahydrocannabinol (THC) oral fluid screening cutoff) with different oral fluid confirmation cutoffs, following inhalation of high-dose (6.7% THC) cannabis by Volcano Medic vaporizer, for comparison to smoking a similar-potency cigarette (111). Quantitative Confirmation Cutoffs Sensitivity Specificity Efficiency TP TN FP FN µg/l (THC, CBD, CBN) % % % ng/l (THCCOOH) THC ≥5 216 207 6 117 64.9 97.2 77.5 THC ≥2 (SAMHSA) 221 134 1 190 53.8 99.3 65.0 THC ≥1 (DRUID) 222 90 0 234 48.7 100 57.1 THC ≥2 and 108 244 114 80 57.4 68.2 64.5 THCCOOH ≥20 THC ≥1 and 108 239 114 85 56.0 67.7 63.6 THCCOOH ≥20 THC ≥2 and 151 303 71 21 87.8 81.0 83.2 CBD ≥1 THC ≥2 and 150 312 72 12 92.6 81.3 84.6 CBN ≥1 THCCOOH ≥20 108 238 114 86 55.7 67.6 63.4 THC ≥1 and 151 303 71 21 87.8 81.0 83.2 CBD ≥1 THC ≥1 and CBN ≥1 150 312 72 12 92.6 81.3 84.6 THC ≥2 or 221 128 1 196 53.0 99.2 63.9 THCCOOH ≥20 THC ≥1 or 222 89 0 235 48.6 100 57.0 THCCOOH ≥20 THC ≥2 or CBD ≥1 221 134 1 190 53.8 99.3 65.0 THC ≥2 or CBN ≥1 221 134 1 190 53.8 99.3 65.0 THC ≥1 or CBD ≥1 222 90 0 234 48.7 100 57.1 THC ≥1 or CBN ≥1 222 90 0 234 48.7 100 57.1 Abbreviations: THC, ∆9-tetrahydrocannabinol; CBD, cannabidiol; CBN, cannabinol; THCCOOH, 11-nor-9-carboxy-THC; TP, true positives; TN, true negatives; FP, false positives; FN, false negatives; SAMHSA, Substance Abuse and Mental Health Services Administration; DRUID, Driving Under the Influence of Drugs, Alcohol and Medicines

Our data represent a broad cannabis history spectrum, suggested by the highly variable self-report data, residual cannabinoids present in some participants at baseline, and large intersubject variability (discussed below). We only recruited individuals who self-reported cannabis intake ≤3x/week, less than our cutoff for chronic frequent smoking

(≥4x/week) (117, 124, 307). Although some current study participants were occasional smokers (≤2x/week), several fit into an intermediate category (2-3x/week), including 8

299

completers. After a night on the research unit, previous frequent smokers were still positive for THC and THCCOOH in 79 and 100% of participants (117). Contrastingly, current participants’ OF baseline (10-16 h after admission) was still positive in half

(THC) and one-third (THCCOOH) of study sessions. Previous frequent smokers were

100% negative for CBD and CBN after a night on the research unit (117); but we detected CBD and CBN in 1/163 and 4/163 baseline specimens. This study was conducted in a different geographic region (Iowa City, IA) than our previous work

(Baltimore, MD). Possibly, typical cannabinoid potencies are different in the two areas.

CBD and CBN often are identified as markers of recent intake (115, 117, 307); it may be prudent to consider potency in consumed cannabis when evaluating time since exposure.

CBD potency in particular may become more variable as medical and recreational cannabis decriminalization increase, due to its other pharmacological properties

(antiemetic, antipsychotic, anti-inflammatory, antiepileptic) (99, 101-102).

Three high-dose THC OF Cmax, exceeding 20,000 µg/L, were among the highest ever reported (74, 115-117, 307-308). After vaporizing two successive THC doses 80 min apart, Wille et al found median (range) OF THC concentrations 1,952 (77.7-12,360) ng/g (74). Another controlled cannabis (smoking) study utilizing the QuantisalTM device had lower median (range) THC Cmax (644 [68.0-10,284] µg/L) (115). Our ranges were considerably wider, but medians at any dose were <1,000 µg/L (Table 40). These data resemble Toennes et al (116, 308) after controlled smoked cannabis of similar potency.

Both administration routes showed large intersubject variability. In that study, median

Cmax was 4,800 ng/g. One frequent smoker displayed 71,747 ng/g maximum OF THC

(308), and 5/17 frequent smokers attained OF THC >20,000 ng/g with 0.5-0.7g/L blood

300

alcohol (116). The authors assert concentrations in ng/g are comparable to µg/L because

OF specific gravity is only 0.2-1.2% different from 1 kg/L. After our low and high doses without alcohol, 75% of THC Cmax were <1,589 and <3,933 µg/L, respectively, compared to 75% of THC Cmax ≤6,236 ng/g in the Toennes study (116). With alcohol, 75% were

<2,811 and <5,288 µg/L in our study, compared to 74% with alcohol ≤9,210 ng/g (116).

As in that study, our findings indicated alcohol did not produce significant OF THC effects. Collectively, these data indicate vaporization produces similar but slightly lower

OF THC concentrations relative to smoking.

The relative lack of significant dose effects on cannabinoids’ Cmax and AUC0-8.3h after high vs. low doses suggests several participants titrated their cannabis dose to individual subjective and cardiovascular comfort levels. Despite similar median THC

Cmax across all active doses, ranges varied >1000-fold (Table 40). However, in only 3/151 complete sessions was the cannabis balloon not fully inhaled. This occurred twice for high dose without alcohol (participants 19 and 20); and once, for low dose without alcohol (also participant 20). The balloons were left approximately ¼ full. Participant 19 had 477 µg/L OF THC Cmax with this unfinished balloon. For comparison, his Cmax at low dose/no alcohol and high dose/alcohol were 209 and 2,348 µg/L, respectively. This dose- dependent intra-subject variability markedly contrasts with participant 20, who had similar Cmax for high dose/no alcohol, low dose/no alcohol, and low dose/alcohol (746,

707, and 735 µg/L, respectively). This could indicate titration, particularly given she did not finish the balloon in two sessions (those without alcohol). Her high dose/alcohol session produced 951 µg/L Cmax. Apart from these three instances, participants consumed the entire bag, except in three sessions that were terminated for drug-related adverse

301

events (panic attack [Participant 23] and emesis/dizziness [Participants 29 and 31]). A recent survey of 96 people who vaporize cannabis (open-ended questions on “best” and

“worst” characteristics of vaporizing) identified >10% claiming it provided more effect for the same cannabis quantity; one respondent indicated it was “easy to consume too much” (80). It is unclear from that survey whether anyone thought it had less effect.

Since our participants usually finished the entire balloon, perhaps titration occurred instead by controlling inhalation rate and depth, and hold time in the lungs. These factors may affect absorption and true tmax. Participants were allowed to inhale ad libitum over

10 min; the first post-dose specimen was not collected until after the full time elapsed.

Cannabinoid concentrations may have peaked earlier. Individual vaporizer experiences vary considerably with cannabis history and inhalation topography, contributing to the substantial variability observed. Experienced smokers often achieve higher THC concentrations with more practiced inhalation technique and some tolerance to its effects.

The within-subjects design of this study was advantageous, providing a framework for examining participants’ data relative to their own unique smoking patterns.

High initial OF THC, CBD, and CBN concentrations arise mainly from contamination of the oral mucosa during inhalation due to minimal transfer from blood to

OF (109-110, 115-118). However, this effect is strongest within the first 0.75 h of exposure, dissipating thereafter such that OF cannabinoids better correlate with plasma, possibly due to transmucosal absorption (74, 110, 309). Vaporized THC, CBD, and CBN tmax immediately followed inhalation (active doses), which is consistent with smoking data (116-117, 307), except in two instances after the high dose without alcohol.

Participant 3 had high dose tmax 1.4 h for THC, CBD, and CBN, but results were within

302

±10% at 0.17 and 1.4 h (159 vs. 165, 7.1 vs. 7.7, and 6.1 vs. 6.6 µg/L, respectively). For both specimens the QuantisalTM adequacy indicator did not turn blue at 0.17 h, indicating insufficient sample volume. Dry mouth is a well-documented phenomenon following cannabis exposure (310-311), possibly explaining these inconsistent results. These specimens’ concentrations were likely underestimated due to analysis without weight correction (117). Even under these unusual circumstances, initial THC concentrations

>100 µg/L greatly exceeded proposed SAMHSA (304) and DRUID analytical (305) cutoffs (2 and 1 µg/L THC, respectively). Although the results may not be quantitatively accurate, short samples generally contain sufficient cannabinoid concentrations to document recent exposure.

Because THCCOOH is not present in smoke but passively diffuses into OF from the bloodstream, it can help rule out acute passive from active cannabis exposure (110,

293, 310, 312). When present, THCCOOH is detected at low ng/l concentrations. Even after the high cannabis dose, THCCOOH was not detected in OF in some participants.

Median Cmax-C0 was similar to Cmax, but accounting for baseline concentrations produced much lower maximum THCCOOH Cmax values for all doses. This demonstrates new vs. residual (built up with more frequent intake) cannabinoids concentrations. THCCOOH tmax varied throughout the session, reflecting differential metabolic rates and residual cannabinoid concentrations.

Although our inclusion criteria targeted occasional to moderate cannabis intake

(≥1x/3 months but ≤3 days/week over the past 3 months), participants 6, 7, 10, 27, and

38’s baseline cannabinoids (THCCOOH ≥20 ng/L and THC ≥1 µg/L after an overnight stay, at least one baseline THCCOOH >100 ng/L) suggested these five individuals were

303

frequent smokers. Occasional smokers did not meet these conditions after 13 h post- smoking in a recent study (307). Fabritius et al (288) found mean THCCOOH 100 ng/L and max 500 ng/L at baseline in frequent smokers, but it is unclear how long participants were admitted prior to baseline. Other participants in the current study had THC ≥1 µg/L and THCCOOH ≥20 ng/L at baseline during some, but not all, of their sessions, and baseline THCCOOH never exceeded 100 ng/L.

Residual THC content in placebo cannabis was only 0.008±0.002%, but this low vaporized quantity still produced observable OF THC, shown by Cmax-C0>0. This effect was not limited to participants with residual THC at baseline. Concentrations were always <21.0 µg/L (for baseline-negative participants) in these cases and decreased more rapidly than active doses. OF THC and THCCOOH following placebo sessions are presented in Figure 24 (Supplemental). In participants positive for THCCOOH on admission, concentrations usually decreased by baseline, but some remained consistent or increased. In baseline-positive participants, THCCOOH concentrations increased and decreased without pattern throughout the time course after placebo dosing. Participants negative for THCCOOH at baseline remained THCCOOH-negative throughout placebo sessions, except in two instances. In Participant 11’s placebo with alcohol session, residual THC decreased throughout the session from 96.3 µg/L on admission and 7.5

µg/L at baseline to 0.65 µg/L at 8.3 h; THCCOOH was 16.9 ng/L on admission, 0 ng/L at baseline, and 0 ng/L at all times post-dose except 1.4 h (15.7 ng/L). Since both positive

THCCOOH specimens were near the 15 ng/L LOQ, it is likely residual THCCOOH was just below this limit during that time. In Participant 13’s placebo with alcohol session,

THC and THCCOOH were negative prior to dosing and at all times except 0.17 h (3.7

304

µg/L and 72.6 ng/L, respectively). THCCOOH detected in both of these placebo sessions was consistent with residual cannabinoids from previous self-administration. Toennes et al (308) observed similar THC-positive OF specimens following controlled placebo- cannabis smoking, and Wille et al (74) found OF THC concentrations up to 746 ng/g

(median 8 ng/g, no reported baseline) after vaporized placebo cannabis. In those studies,

THCCOOH was not quantified in OF. This observation will not likely confound OF THC interpretation in forensic cases, because outside the laboratory setting there is little cause to consume placebo cannabis.

OF THC and THCCOOH were detectable in QuantisalTM specimens ≥8.3 h post- dose after active cannabis, consistent with smoking administration. Further study is required to adequately assess extended detection times following vaporization. Previous studies after smoking one similar-potency cannabis cigarette documented THC and

THCCOOH in some individuals’ OF ≥22 h (115, 310) or ≥30 h (117, 307) post-smoking, especially for frequent smokers. During sustained monitored abstinence in chronic frequent cannabis smokers, THC was often present 48 h after admission, and THCCOOH for many days (313). Participants in the present study were screened as occasional or moderate smokers; but because some were more frequent smokers based on cannabinoid concentrations, we hypothesize OF detection times would be similar to or higher than those for occasional smokers. Such data (THC 13.5-≥30 h, CBD 1-6 h, CBN 2-13.5 h,

THCCOOH 0-28 h) exist for other collection devices (StatSure, Oral-Eze®) (117, 307); our CBD and CBN data appear similar. QuantisalTM was only characterized for frequent smokers (THC and THCCOOH 6-≥22 h, CBD and CBN 2-≥6 h) (115), with no collection times between 6 and 22 h.

305

Although no overall alcohol*cannabis interactive effects were statistically significant (Table 41), high- vs. low-dose AUC0-8.3h was only significantly different with co-administration of alcohol (Table 45 (Supplemental)). Additionally, alcohol produced later tlast for CBD and CBN after high doses (Table 46 (Supplemental) and Table 47

(Supplemental)). AUC and tlast both rely upon longer-term analyte measurements, extending beyond primary absorption and distribution phases. If only AUC0-8.3h (THC) and tlast (CBD, CBN) were affected without impacting Cmax or tmax, this may imply alcohol slightly slowed excretion. Limited other data exist on OF cannabis combined with alcohol. An early controlled-administration study noted lower THC concentrations

(58.3, 73.5 µg/L) 1 h post-dose in two participants who drank 200 mL beer immediately after smoking 10 mg THC, relative to two (250, 96.0 µg/L) who did not drink (314). The authors concluded the differences resulted from a “washing” effect from the drink. This is possible, but given the low N, it may be difficult to draw such a conclusion. Equally likely, their observations reflected normal interindividual variability unrelated to the beverage. No OF was collected prior to 1 h and all participants were cannabis-naïve.

Another study examining OF THC in concert with alcohol found no significant differences between alcohol conditions (116). The authors further noted that drinking a

300 mL alcoholic beverage would not affect roadside THC detectability; our results concur. Despite similar blood alcohol AUC0-8.3h, THC appeared to slightly alter alcohol’s absorption phase (Figure 20, Table 40), producing significantly lower and later alcohol

Cmax. This corroborates previous findings (286). It is possible this resulted from cannabinoids’ slowing effects on gastrointestinal motility and decreased gastric emptying

(315-316), since alcohol is absorbed via passive diffusion along concentration gradients

306

in the stomach and small intestine (317). It is important to consider that the 1.4 h median alcohol tmax after the high cannabis dose (rather than 0.42 h from the low dose) reflected the immediate next alcohol measurement time, so median tmax data should be interpreted with caution.

Factors affecting apparent on-site performance include chosen confirmation cutoff, frequency of cannabis intake, time course, and administration route. The

DrugTest® 5000 demonstrated good specificity and efficiency for OF obtained over 8.3 h after cannabis vaporization in these occasional smokers, but sensitivity was lower than observed in frequent smokers after smoking a cannabis cigarette with the same THC potency (sensitivity 90.7% at THC ≥2 µg/L) (111). At this confirmation cutoff, we observed 47.0% sensitivity, but 99.6% specificity (due to few FP) for overall 70.1% efficiency. Low vaporized sensitivity arose from high FN rates, even within the first 4.3 h post-dose. Figure 23 demonstrates the effect of different confirmation cutoffs when evaluating on-site Dräger screening performance. After active THC, 70.7% of tests over the first 3.3 h were positive by Dräger and confirmed at THC ≥2 µg/L (SAMHSA proposed cutoff). At 5.3 h and 8.3 h, detection rates were 28.9% and 14.9%, respectively.

Confirming with THC ≥2µg/L or another analyte (CBD, CBN, or THCCOOH) produced the same results, showing that in this occasional/moderate smoker cohort, when the

Dräger was positive and CBD, CBN, or THCCOOH was ≥1 µg/L, ≥1 µg/L, or ≥20 ng/L respectively, THC was always ≥2 µg/L. THCCOOH was proposed as a potential additional confirmatory criterion because it helps rule out passive environmental exposure, detects oral cannabis use, and can extend detection windows in chronic frequent cannabis smokers (293, 313). In this population, THCCOOH was not always

307

detected; so including THCCOOH as a requirement for confirmation decreased sensitivity. Additional CBD or CBN ≥1 µg/L confirmation requirements increased apparent sensitivity relative to THC ≥1 µg/L or THC ≥2 µg/L only. However, this finding should be interpreted carefully, because it reflects CBD and CBN as recent-use cannabinoid markers. FN were reduced by requiring minor cannabinoid detection to be considered “positive”. This created an on-site detection window similar to the performance-impairment window (Table 42, Figure 23). Although CBD and CBN may be markers of recent intake, their absence does not preclude it. CBD and CBN decreased confirmed detection rates especially after 4 h. In a study with a longer time course, requiring these markers for confirmation would decrease apparent sensitivity relative to our results (111). Using the manufacturer-specified 5 µg/L THC screening cutoff as the confirmation cutoff showed 60.8% sensitivity, 98.2% specificity, and 82.5% overall efficiency, higher than all other evaluated THC cutoffs except those additionally requiring CBD or CBN.

Our results are similar to an early roadside Dräger study whose authors also noted high numbers of FN (318). In contrast, recent smoked and roadside studies demonstrated higher sensitivity (58.3-94.4%), but lower specificity (15.4-75%) sometimes resulting from few TN (111, 116, 123, 131, 319). Some of these studies only quantified plasma rather than OF confirmations during Dräger performance evaluation. THC cutoffs in plasma were 1-2 µg/L (123, 131), and in OF 2-5 µg/L (111, 116). Roadside studies may have inherently fewer true positives than controlled-administration studies, decreasing apparent sensitivity and efficiency (which depend upon total detected TP). Alcohol produced no impact on Dräger performance post-smoking (116), agreeing with our

308

findings post-vaporization. To date, the only other on-site testing device evaluated with vaporized cannabis, the DrugWipe-5S, produced remarkably similar results to the Dräger at 1 µg/L OF THC cutoff. Wille et al (74) observed 43%, 100%, and 57% DrugWipe-5S sensitivity, specificity, and efficiency; here, Dräger performance was 40.4%, 99.8%, and

60.7%.

Volatilization by hot air is a different heating mechanism than combustion, altering the properties of inhaled vapor versus smoke (67, 69, 125). As far as we are aware, pH and other chemical properties of cannabis smoke and vapor are not yet elucidated; but tobacco smoke can vary even during the process of smoking a cigar (320).

Cannabis vapor may interact with oral mucosa differently to smoke, altering Dräger performance. Lower volatilization heating temperature (210°C) releases less THC than smoking (≥230°C) (69, 238), and some THC could adhere to the balloon (83).

Vaporization causes less exposure to combustion byproducts, cannabinoids, and other chemicals (66, 69). It is possible that lower THC contamination of oral mucosa contributed to the lower vaporized sensitivity. Another possible explanation is OF collection with the Dräger collection device involves moving it throughout the entire mouth, mildly stimulating salivary production, whereas the QuantisalTM device is held sublingually. Dräger also recommends collecting the confirmation OF specimen first, which may help stimulate OF production. We followed these guidelines in specimen collection. Stimulation can decrease OF drug concentrations due to further dilution (321).

These and other factors may contribute to observed sensitivity differences relative to smoking. Finally, the time course of the current experiment was shorter than our previous studies, and doses included placebo, low (2.9%) and a comparable (6.7%) THC dose.

309

Including only high-dose results (Table 43) increased sensitivity overall but still resulted in lower sensitivity relative to smoking (111). Another possible consideration is that the

THC cigarettes contained more total cannabis (0.79 g) than the amount vaporized (0.5 g).

Median Dräger tlast was 3-4 h for evaluated cutoffs, but for all cutoffs some specimens were positive ≥8.3 h. This coincides with previous smoking findings, showing that some Dräger OF specimens were positive ≥4 h (116) and 6-≥22 h (111). More recently, significant differences in Dräger tlast were observed between occasional and frequent smokers when OF confirmation results also considered the presence of

THCCOOH (124). Dräger tlast varied considerably overall and by chosen confirmation cutoff criteria, highlighting the importance of careful interpretation. Further study is required to determine extended detection windows following vaporization.

Conclusions

For the first time following controlled cannabis vaporization, we document cannabinoid disposition in OF over 8.3 h with and without low-dose alcohol, and performance of an on-site screening device. The Dräger on-site device best reflected the cannabis impairment window when combined with recent use markers CBD and CBN, because these analytes shortened detection windows to approximately 2-4 h. However, possible increased variability in CBD potency may result in different or extended CBD detection; future research with cannabis containing higher CBD is recommended. Chosen confirmation cutoff, time since dosing, length of monitoring, frequency of use, and additional detected analytes all affect interpretation. The Dräger DrugTest® 5000 displayed lower sensitivity after vaporization than smoking, but high specificity and

310

comparable efficiency. Concurrent alcohol (albeit at least 10 min prior to vaporization) did not affect cannabinoid OF concentrations or on-site test sensitivity. Future studies should directly compare cannabis vaporization to smoking over extended periods.

311

Table 44 (Supplemental). Median [range] maximum (Cmax) breath alcohol concentration (BrAC, reported in g/210L or approximate BAC g/dL), time to maximum BrAC (tmax), time of last alcohol detection (tlast), and area under the curve (AUC0-8.3h), following drinking Everclear grain alcohol over 10 min and followed immediately by controlled inhalation of placebo, low (2.9%) and high (6.7%) ∆9-tetrahydrocannabinol (THC) over 10 min. Alcohol (LOQ 0.006 g/210L) Median p-value (N)a N [Range] vs. placebo vs. low vs. high C , 0.065 max 30 ------BrAC g/210L [0.034-0.135] during t , 0.42 max 30 ------Placebo h [0.17-1.4] THC Dose t , 4.3 last 29 ------h [2.3-5.3] AUC 0.1618 0-8.3h 29 ------h*g/210L [0.1025-0.2620] C , 0.062 0.056 max 25 -- -- g/dL [0.035-0.097] (24) BrAC t , 0.42 0.007 max 25 -- -- during h [0.17-2.3] (24) Low THC t , 4.3 1.000 last 25 -- -- Dose h [2.3-5.3] (24) AUC 0.1705 0.317 0-8.3h 25 -- -- h*g/210L [0.0592-0.2572] (24) C , 0.053 0.006 0.380 max 21 -- g/dL [0.036-0.087] (20) (20) BrAC t , 1.4 0.005 0.283 max 21 -- during h [0.17-2.3] (20) (20) High THC t , 4.3 0.257 0.366 last 19 -- Dose h [2.3-5.3] (19) (19) AUC 0.1513 0.295 0.260 0-8.3h 19 -- h*g/210L [0.1040-0.2256] (19) (19) Alcohol dose was calculated to produce 0.065 g/210L approximate BrAC based on sex, height, weight, and age. BrAC limit of quantification (LOQ) was 0.006 g/210L. aMatched pairs for within-subjects Wilcoxon Matched Pairs statistical comparison. Boldface indicates statistical significance at p<0.05. Note the 1.4 h median alcohol tmax after the high cannabis dose (rather than 0.42 h median after the low dose) reflects the next alcohol measurement time. For alcohol dose calculations: The Sahlgrenska formula for body water was utilized to estimate participant body water. Body water (L) if male = -10.759 + 0.192 × height (cm) + 0.312 × weight (kg) – 0.078 × age (years). Body water (L) if female = -29.994 + 0.294 × height (cm) + 0.214 × weight (kg) – 0.0004 × age (years). Alcohol volume for dosing was calculated based on estimated body water, according to the following formula: Alcohol (mL) = (((desired peak BAC + ((total absorption time [10min])/(60 × standard clearance rate)))/(H2O in blood)) × ((Sahlgrenska body water × 10)/(specific gravity of alcohol)))/0.90

312

TM 9 Table 45 (Supplemental). Median [range] Quantisal oral fluid ∆ -tetrahydrocannabinol (THC) maximum concentration (Cmax), baseline concentration (C0), time to maximum concentration (tmax), time of last detection (tlast), and area under the curve (AUC0-8.3h) after controlled cannabis inhalation by vaporizer. THC (LOQ No Alcohol With Alcohol 0.5 µg/L)

a a p-value (N) p-value (N) p-value (N)a, Median Median N N alcohol vs. [Range] vs. vs. vs. [Range] vs. vs. vs. no alcohol placebo low high placebo low high C 4.7 <0.001 <0.001 3.7 <0.001 <0.001 0.520 max, 25 -- 29 -- µg/L [0-25.9] (18) (25) [0-42.6] (23) (18) (19) C 0.62 0 0, 26 n/a n/a n/a 30 n/a n/a n/a n/a Placebo THC µg/L [0-14.2] [0-11.3] C -C , 4.2 <0.001 <0.001 1.9 <0.001 <0.001 0.936 max 0 25 -- 29 -- µg/L [-3.0-24.7] (18) (25) [-2.2-41.9] (23) (18) (19) t , 0.17 0.317 0.414 0.17 0.063 0.102 0.180 max 23 -- 27 -- h [0.17-1.4] (17) (23) [0.17-2.3] (22) (17) (17)

313 t , 6.3 0.005 0.002 6.8 0.011 0.017 0.611 last 19 -- 18 -- h [1.4-≥8.3] (15) (19) [1.4-≥8.3] (17) (14) (14)

AUC 6.6 <0.001 <0.001 4.5 <0.001 <0.001 0.687 0-8.3h 25 -- 28 -- h*µg/L [0-56.1] (17) (25) [0-39.4] (23) (18) (19) C 809 <0.001 0.217 882 <0.001 0.053 0.476 max, 25 -- 25 -- µg/L [16.3-18,230] (18) (21) [29.3-7,494] (23) (19) (21) C 0.54 0 0, 25 n/a n/a n/a 27 n/a n/a n/a n/a µg/L [0-30.7] [0-72.9] C -C , 809 <0.001 0.217 880 <0.001 0.053 0.476 max 0 25 -- 25 -- Low (2.9%) µg/L [16.3-18,206] (18) (21) [3.3-7,494] (23) (19) (21) THC t , 0.17 0.317 0.180 0.17 0.063 0.317 1.000 max 25 -- 25 -- h [0.17-0.17] (17) (21) [0.17-0.17] (22) (19) (21) t , ≥8.3 0.005 0.104 ≥8.3 0.011 0.317 0.109 last 24 -- 25 -- h [1.4-≥8.3] (15) (21) [≥8.3-≥8.3] (17) (19) (21) AUC 647 <0.001 0.159 779 <0.001 0.040 0.881 0-8.3h 24 -- 25 -- h*µg/L [13.9-3,865] (17) (19) [88.8-8,146] (23) (19) (20)

Table 45 (Supplemental). (Continued from previous page) C 862 <0.001 0.217 932 <0.001 0.053 0.247 max, 28 -- 20 -- µg/L [25.1-23,680] (25) (21) [22.7-66,200] (18) (19) (20) C 0 0 0, 31 n/a n/a n/a 24 n/a n/a n/a n/a µg/L [0-11.7] [0-34.2] C -C , 861 <0.001 0.217 930 <0.001 0.053 0.218 max 0 28 -- 20 -- High (6.7%) µg/L [25.1-23,671] (25) (21) [22.7-66,192] (18) (19) (20) THC t , 0.17 0.414 0.180 0.17 0.102 0.317 0.414 max 28 -- 20 -- h [0.17-3.3] (23) (21) [0.17-0.17] (17) (19) (20) t , ≥8.3 0.002 0.104 ≥8.3 0.017 0.317 0.655 last 27 -- 20 -- h [6.3-≥8.3] (19) (21) [4.3-≥8.3] (14) (19) (20) AUC 934 <0.001 0.159 877 <0.001 0.040 0.376 0-8.3h 27 -- 20 -- h*µg/L [38.4-19,090] (25) (19) [25.2-53984] (18) (19) (19) THC content of placebo, low, and high doses was 0.008±0.002%, 2.9±0.14%, and 6.7±0.05%, respectively. a Matched pairs for statistical comparison. N<19 for Wilcoxon Matched Pairs within-subjects analysis of tlast, tmax, or AUC0-8.3h indicates THC never detected (tmax) or missing data for t=8.3 h (tlast, AUC). Boldface indicates statistical significance at p<0.05. THC limit of quantification (LOQ) was 0.5 µg/L. For completer pharmacokinetic data only, refer to Table 40.

314 Table 46 (Supplemental). Median [range] oral fluid cannabidiol (CBD) maximum concentration (Cmax), baseline concentration (C0),

time to maximum concentration (tmax), time of last detection (tlast), and area under the curve (AUC0-8.3h) after controlled cannabis inhalation by vaporizer. CBD (LOQ No Alcohol With Alcohol 1 µg/L) p-value (N)a p-value (N)a p-value (N)a, Median Median [Range] N N alcohol vs. no vs. vs. vs. [Range] vs. vs. vs. placebo low high placebo low high alcohol C 0 0.002 <0.001 0 <0.001 <0.001 0.317 max, 25 -- 29 -- µg/L [0-1.7] (18) (25) [0-1.6] (23) (18) (19) C 0 0 0, 26 n/a n/a n/a 30 n/a n/a n/a n/a Placebo (THC) µg/L [0-2.1] [0-0] C -C , 0 0.002 <0.001 0 <0.001 <0.001 0.317 Dose max 0 25 -- 29 -- µg/L [-0.4-0] (18) (25) [0-1.6] (23) (18) (19) t , 0.17 -- -- max 0.17 1 ------2 -- -- h [0.17-0.17] (2) (2) t , 0.17 -- -- last 0.17 1 ------2 -- -- h [0.17-0.17] (2) (2) AUC 0 0.003 <0.001 0 <0.001 <0.001 0.317 0-8.3h 25 -- 28 -- h*µg/L [0-1.4] (17) (25) [0-0.8] (23) (18) (19)

Table 46 (Supplemental). (Continued from previous page) C 2.6 0.002 0.006 3.6 <0.001 <0.001 0.586 max, 25 -- 25 -- µg/L [0-100] (18) (21) [0-75.7] (23) (19) (21) C 0 0 0, 25 n/a n/a n/a 27 n/a n/a n/a n/a µg/L [0-0] [0-0] C -C , 2.6 0.002 0.006 3.6 <0.001 <0.001 0.586 max 0 25 -- 25 -- Low (THC) µg/L [0-100] (18) (21) [0-75.7] (23) (19) (21) Dose t , 0.17 0.180 0.17 -- 0.317 1.000 max 17 n/a -- 18 -- h [0.17-0.17] (14) [0.17-0.17] (2) (13) (12) t , 0.17 0.004 0.17 -- 0.001 0.194 last 17 n/a -- 18 -- h [0.17-0.17] (14) [0.17-3.3] (2) (13) (12) AUC 1.8 0.003 0.005 2.6 <0.001 <0.001 0.723 0-8.3h 24 -- 25 -- h*µg/L [0-79.0] (17) (19) [0-57.3] (23) (19) (20) C 34.7 <0.001 0.006 35.8 <0.001 <0.001 0.654 max, 28 -- 20 -- µg/L [1.0-1106] (25) (21) [0-2331] (18) (19) (20) C 0 0 0, 31 n/a n/a n/a 24 n/a n/a n/a n/a µg/L [0-0] [0-0] C -C , 34.7 <0.001 0.006 35.8 <0.001 <0.001 0.627 max 0 28 -- 20 -- High (THC) µg/L [1.0-1106] (25) (21) [0-2331] (18) (19) (20) Dose t , 0.17 0.180 0.17 -- 0.317 0.180 max 28 n/a -- 19 -- h [0.17-3.3] (14) [0.17-1.4] (2) (13) (19)

315 t , 2.3 0.004 3.3 -- 0.001 0.033 last 28 n/a -- 19 -- h [0.17-≥8.3] (14) [0.17-≥8.3] (2) (13) (19)

AUC 32.0 <0.001 0.005 35.6 <0.001 <0.001 0.872 0-8.3h 27 -- 20 -- h*µg/L [0.7-912] (25) (19) [0-1910] (18) (19) (19) CBD content in placebo, low, and high doses was 0.001±0.001%, 0.05±0.00%, and 0.19±0.01% CBD. aMatched pairs for within-subjects Wilcoxon Matched Pairs statistical comparison. Boldface indicates statistical significance at p<0.05. CBD limit of quantification (LOQ) was 1 µg/L. For completer pharmacokinetic data only, refer to Table 40.

Table 47 (Supplemental). Median [range] oral fluid cannabinol (CBN) maximum concentration (Cmax), baseline concentration (C0), time to maximum concentration (tmax), time of last detection (tlast), and area under the curve (AUC0-8.3h) after controlled cannabis inhalation by vaporizer. CBN (LOQ No Alcohol With Alcohol 1 µg/L) p-value (N)a p-value (N)a p-value (N)a, Median Median [Range] N N alcohol vs. no vs. vs. vs. [Range] vs. vs. vs. placebo low high placebo low high alcohol C 0 <0.001 <0.001 0 <0.001 <0.001 0.893 max, 25 -- 29 -- µg/L [0-1.9] (18) (25) [0-2.4] (23) (18) (19) C 0 0 0, 26 n/a n/a n/a 30 n/a n/a n/a n/a Placebo µg/L [0-1.3] [0-0] C -C , 0 <0.001 <0.001 0 <0.001 <0.001 0.893 (THC) Dose max 0 25 -- 29 -- µg/L [0-1.9] (18) (25) [0-2.4] (23) (18) (19) t , 0.17 1.000 1.000 0.17 1.000 1.000 -- max 5 -- 5 -- h [0.17-0.17] (3) (5) [0.17-0.17] (4) (3) (1) t , 0.17 0.180 0.066 0.17 0.066 0.109 -- last 5 -- 5 -- h [0.17-0.17] (3) (5) [0.17-0.17] (4) (3) (1)

316 AUC 0 <0.001 <0.001 0 <0.001 <0.001 0.893 0-8.3h 25 -- 28 -- h*µg/L [0-1.4] (17) (25) [0-1.5] (23) (18) (19)

C 54.0 <0.001 0.614 93.0 <0.001 0.398 0.931 max, 25 -- 25 -- µg/L [0-941] (18) (21) [0-484] (23) (19) (21) C 0 0 0, 25 n/a n/a n/a 27 n/a n/a n/a n/a µg/L [0-1.3] [0-3.6] C -C , 54.0 <0.001 0.614 93.0 <0.001 0.398 0.931 max 0 25 -- 25 -- Low (THC) µg/L [0-941] (18) (21) [0-484] (23) (19) (21) Dose t , 0.17 1.000 0.180 0.17 1.000 1.000 1.000 max 24 -- 24 -- h [0.17-0.17] (3) (18) [0.17-0.17] (4) (18) (19) t , 2.3 0.180 0.321 2.8 0.066 0.583 0.381 last 24 -- 24 -- h [0.17-7.3] (3) (19) [0.17-≥8.3] (4) (18) (20) AUC 41.4 <0.001 0.872 72.8 <0.001 0.355 0.654 0-8.3h 24 -- 25 -- h*µg/L [0-246] (17) (19) [0-405] (23) (19) (20)

Table 47 (Supplemental). (Continued from previous page) C 32.0 <0.001 0.614 31.5 <0.001 0.398 0.296 max, 28 -- 20 -- µg/L [0-766] (25) (21) [0-2650] (18) (19) (20) C 0 0 0, 31 n/a n/a n/a 24 n/a n/a n/a n/a µg/L [0-0] [0-0] C -C , 32.0 <0.001 0.614 31.5 <0.001 0.398 0.296 max 0 28 -- 20 -- High (THC) µg/L [0-766] (25) (21) [0-2650] (18) (19) (20) Dose t , 0.17 1.000 0.180 0.17 1.000 1.000 0.180 max 26 -- 19 -- h [0.17-3.3] (5) (18) [0.17-0.17] (3) (18) (17) t , 2.3 0.066 0.321 3.3 0.109 0.583 0.021 last 26 -- 19 -- h [0.17-≥8.3] (5) (19) [0.17-≥8.3] (3) (18) (17) AUC 28.9 <0.001 0.872 27.8 <0.001 .355 0.445 0-8.3h 27 -- 20 -- h*µg/L [0-617] (25) (19) [0-2226] (18) (19) (19) CBN content in placebo, low, and high doses was 0.009±0.003%, 0.22±0.02%, and 0.37±0.03%. aMatched pairs for within-subjects Wilcoxon Matched Pairs statistical comparison. Boldface indicates statistical significance at p<0.05. CBN limit of quantification (LOQ) was 1 µg/L. For completer pharmacokinetic data only, refer to Table 40.

317 Table 48 (Supplemental). Median [range] oral fluid 11-nor-9-carboxy-THC (THCCOOH) maximum concentration (Cmax), baseline concentration (C0), time to maximum concentration (tmax), time of last detection (tlast), and area under the curve (AUC0-8.3h) after

controlled cannabis inhalation by vaporizer. THCCOOH (LOQ No Alcohol With Alcohol 15 ng/L) p-value (N)a p-value (N)a p-value (N)a, Median Median [Range] N N alcohol vs. no vs. vs. vs. [Range] vs. vs. vs. placebo low high placebo low high alcohol C 0 0.028 0.001 0 0.030 0.016 0.214 max, 25 -- 29 -- ng/L [0-361] (18) (25) [0-392] (23) (18) (19) C 0 0 0, 26 n/a n/a n/a 30 n/a n/a n/a n/a Placebo (THC) ng/L [0-249] [0-243] C -C , 0 0.005 <0.001 0 0.016 0.016 0.110 Dose max 0 25 -- 29 -- ng/L [-18.6-113] (18) (25) [-17.3-193] (23) (18) (19) t , 2.3 0.684 0.441 2.3 0.352 0.072 0.785 max 10 -- 11 -- h [0.17-≥8.3] (7) (10) [0.17-≥8.3] (10) (8) (6) t , ≥8.3 1.000 0.336 ≥8.3 0.416 0.180 0.157 last 10 -- 11 -- h [5.3-≥8.3] (7) (10) [0.17-≥8.3] (10) (8) (6) AUC 0 0.021 0.019 0 0.140 0.041 0.314 0-8.3h 25 -- 28 -- h*ng/L [0-1941] (17) (25) [0-2018] (23) (18) (19)

Table 48 (Supplemental). (Continued from previous page) C 24.1 0.028 0.975 37.7 0.030 0.507 0.158 max, 25 -- 25 -- ng/L [0-686] (18) (21) [0-992] (23) (19) (21) C 0 0 0, 25 n/a n/a n/a 27 n/a n/a n/a n/a ng/L [0-505] [0-911] C -C , 20.3 0.005 0.245 33.0 0.016 0.087 0.551 max 0 25 -- 25 -- Low (THC) ng/L [0-182] (18) (21) [0-219] (23) (19) (21) Dose t , 2.3 0.684 0.058 1.4 0.352 0.865 0.154 max 17 -- 16 -- h [0.17-≥8.3] (7) (12) [0.17-7.3] (10) (11) (14) t , ≥8.3 1.000 0.891 ≥8.3 0.416 0.854 0.201 last 16 -- 16 -- h [0.17-≥8.3] (7) (12) [1.4-≥8.3] (10) (11) (13) AUC 60.6 0.021 0.754 185 0.140 0.650 0.463 0-8.3h 24 -- 25 -- h*ng/L [0-2,935] (17) (19) [0-5153] (23) (19) (20) C 18.6 0.001 0.975 22.9 0.016 0.507 0.140 max, 28 -- 20 -- ng/L [0-464] (25) (21) [0-909] (18) (19) (20) C 0 0 0, 31 n/a n/a n/a 24 n/a n/a n/a n/a ng/L [0-223] [0-468] C -C , 18.6 <0.001 0.245 22.0 0.016 0.087 0.140 max 0 28 -- 20 -- High (THC) ng/L [0-250] (25) (21) [0-441] (18) (19) (20) Dose t , 2.8 0.441 0.058 1.4 0.072 0.865 0.733 max 20 -- 12 -- h [0.17-6.3] (10) (12) [0.17-3.3] (8) (11) (10)

318 t , ≥8.3 0.336 0.891 ≥8.3 0.180 0.854 0.285 last 20 -- 12 -- h [0.17-≥8.3] (10) (12) [2.3-≥8.3] (8) (11) (10)

AUC 66.2 0.019 0.754 57.4 0.041 0.650 0.249 0-8.3h 27 -- 20 -- h*ng/L [0-2,181] (25) (19) [0-3536] (18) (19) (19) aMatched pairs for within-subjects Wilcoxon Matched Pairs statistical comparison. Boldface indicates statistical significance at p<0.05. THCCOOH limit of quantification (LOQ) was 15 ng/L. For completer pharmacokinetic data only, refer to Table 40.

Table 49 (Supplemental). Performance characteristics for the Dräger DrugTest® 5000 on- site test (5 µg/L ∆9-tetrahydrocannabinol (THC) oral fluid screening cutoff) with different oral fluid confirmation cutoffs, following inhalation of cannabis by Volcano Medic vaporizer Quantitative Median p- Confirmation Sensi- Speci- Effi- [Range] value Cutoffs TP TN FP FN tivity ficity ciency tlast (h) (Low a µg/L (THC, CBD, CBN) % % % Low , vs. ng/L (THCCOOH) Highb High) 3.3a,b THC ≥5 437 973 18 282 60.8 98.2 82.5 0.391 [0.17-≥8.3] THC ≥2 3.3a,b 452 746 3 509 47.0 99.6 70.1 0.674 (SAMHSA) [0.17-≥8.3] 3.3a,b THC ≥1 (DRUID) 454 584 1 671 40.4 99.8 60.7 0.530 [0.17-≥8.3] 3.3a THC ≥2 and [0.17-≥8.3] 232 1,003 223 252 47.9 81.8 72.2 0.393 THCCOOH ≥20 3.8b [1.43-≥8.3] 3.3a THC ≥1 and [0.17-≥8.3] 232 957 223 298 43.8 81.1 69.5 0.393 THCCOOH ≥20 3.8b [1.43-≥8.3] 0.17a THC ≥2 and [0.17-3.3] 214 1,229 241 26 89.2 83.6 84.4 0.001 CBD ≥1 2.3b [0.17-≥8.3] THC ≥2 and 2.3 306 1,207 149 48 86.4 89.0 88.5 0.872 CBN ≥1 [0.17-≥8.3] 3.3a [0.17-≥8.3] THCCOOH ≥20 232 899 223 356 39.5 80.1 66.1 0.393 3.8b [1.43-≥8.3] 0.17a THC ≥1 and [0.17-3.3] 214 1,229 241 26 89.2 83.6 84.4 0.001 CBD ≥1 2.3b [0.17-≥8.3] THC ≥1 and 2.3a,b 306 1,207 149 48 86.4 89.0 88.5 0.872 CBN ≥1 [0.17-≥8.3] THC ≥2 or 3.3a,b 452 642 3 613 42.4 99.5 64.0 0.674 THCCOOH ≥20 [0.17-≥8.3] THC ≥1 or 3.3a,b 454 526 1 729 38.4 99.8 57.3 0.530 THCCOOH ≥20 [0.17-≥8.3] 3.3a,b THC ≥2 or CBD ≥1 452 746 3 509 47.0 99.6 70.1 0.674 [0.17-≥8.3] 3.3a,b THC ≥2 or CBN ≥1 452 746 3 509 47.0 99.6 70.1 0.674 [0.17-≥8.3] 3.3a,b THC ≥1 or CBD ≥1 454 584 1 671 40.4 99.8 60.7 0.530 [0.17-≥8.3] 3.3a,b THC ≥1 or CBN ≥1 454 584 1 671 40.4 99.8 60.7 0.530 [0.17-≥8.3] Boldface indicates statistically significant difference in low (2.9% THC) vs. high (6.7% THC) dose time of last detection (tlast) based on screening/cutoff criteria. Abbreviations: THC, ∆9-tetrahydrocannabinol; CBD, cannabidiol; CBN, cannabinol; THCCOOH, 11-nor-9-carboxy- THC; TP, true positives; TN, true negatives; FP, false positives; FN, false negatives; SAMHSA, Substance Abuse and Mental Health Services Administration; DRUID, Driving Under the Influence of Drugs, Alcohol and Medicines 319

Figure 24 (Supplemental). Median (interquartile range) oral fluid a) Δ9- tetrahydrocannabinol (THC) and b) 11-nor-9-carboxy-THC (THCCOOH) vs. time in 19 completers after controlled placebo cannabis inhalation by vaporizer. Horizontal dotted line represents analyte limit of quantification (LOQ); vertical dotted line represents start of cannabis administration.

320

321

Figure 25 (Supplemental). Oral fluid cannabinoids a) ∆9-tetrahydrocannabinol (THC), b) cannabidiol (CBD), c) cannabinol (CBN) and d) 11- nor-9-carboxy-THC (THCCOOH) in participant 30 after controlled inhalation of vaporized placebo-THC cannabis. Horizontal dotted line represents analyte limit of quantification (LOQ); vertical dotted line represents start of cannabis administration. Data suggest participant consumed active drug after 4.3 h post-dose (no-alcohol condition) and prior to or near dose time (alcohol condition).

Chapter 8 – Conclusions and Future Research

Research Summary

The primary aims of this research were to 1) evaluate current knowledge on cannabis’ effects on driving and related skills, with and without alcohol; 2) directly assess cannabis’ effects on driving skills via a driving simulator, with and without low-dose alcohol, by blood THC concentration; 3) establish cannabis effects on subjective feelings after vaporization by blood and oral fluid THC concentrations, with and without alcohol;

4) assess cannabis vaporization’s similarity to smoking as an administration route; 5) determine blood, plasma, and OF cannabinoid pharmacokinetics after vaporization, and possible cannabis-alcohol interactive effects; 6) determine whether cannabis effects correlate to OF THC concentrations. This project involved interagency agreements with

ONDCP and NHTSA to perform the first illicit drug administration study at the NADS, in order to investigate our primary driving hypotheses. The main research findings are summarized in Table 50.

322

Table 50. Primary research findings Research Question Findings Chapter(s) Cannabis affected SDLP in a blood THC concentration-dependent manner, by 0.26 cm per μg/L THC. SDLP increased 8.2% (relative to median) at 10 μg/L THC and 16% at 20 μg/L. These SDLP increases were greater Does cannabis affect lateral control than those produced by 0.05 and 0.08 g/210L BrAC, illegal alcohol concentrations for driving in most during driving, and at what blood THC 4 countries. Lower concentrations produced smaller SDLP increases. THC concentrations during driving are concentrations? generally much higher than at the time of blood draw in driving under the influence of cannabis (DUIC) cases. There were no significant blood THC effects on lane departures/min and maximum lateral acceleration. How does concurrent alcohol interact Alcohol dose-dependently impaired lateral control, increasing lane departures/min and maximum lateral with cannabis’ effects on driving lateral acceleration in addition to SDLP. Cannabis-alcohol interactive effects on SDLP were additive rather than 4 control? synergistic. Is cannabis vaporization an effective Yes, cannabis vaporization produced subjective effects similar to those previously described for smoking. method of inhaled THC delivery, Effects vs. blood THC concentration curves produced counterclockwise hysteresis, with maximum effects 5 producing expected subjective effects? occurring after THC Cmax. How does alcohol interact with Concurrent alcohol administered with cannabis increased the duration of subjective effects experienced 5 cannabis’ subjective effects? relative to either drug alone. Vaporized cannabis produced similar blood, plasma, and OF pharmacokinetic profiles as observed following Are vaporized cannabis’ blood, plasma, smoking, with observed Cmax immediately post-inhalation and rapidly decreasing concentrations soon 323 and OF cannabinoid dispositions and thereafter, followed by a more gradual secondary decrease after ~2 h post-dose. All matrices show high 6, 7

pharmacokinetics similar to those intersubject variability, as observed after smoking, and OF THC concentrations’ large variability reflected observed after smoking? oromucosal contamination. Within-subjects data indicated self-titration of dose in about half of participants, achieved by controlling inhalation topography.

Blood and plasma THC and 11-OH-THC Cmax were significantly higher with alcohol than without, Do cannabis and alcohol interactions particularly after the high (6.7% THC) dose, but alcohol did not affect THC tmax or AUC0-8.3h. Concurrent alter cannabinoid or alcohol alcohol did not affect OF cannabinoid concentrations, although THC AUC0-8.3h increased and CBD and CBN 6, 7 pharmacokinetic dispositions? had later tlast with alcohol. The high cannabis dose produced significantly lower and later alcohol Cmax, but no effect on AUC0-8.3h. The Dräger DrugTest® 5000 on-site OF screening device had lower sensitivity after 6.7% THC vaporization Does cannabis vaporization or alcohol than smoking a similar potency (6.8% THC) cigarette (111), but high specificity. Factors affecting test significantly alter on-site OF test 7 interpretation included amount of cannabis, confirmation cutoff, cannabis history, and time since last intake. performance or detection windows? Concurrent alcohol did not alter on-site OF test performance or detection windows. Can OF directly predict blood or plasma No; although OF and blood or plasma are significantly correlated after inhalation, OF variability is 4, 5 cannabinoid concentrations? substantially greater (2-5 fold higher) than paired blood specimens, producing variable OF/blood ratios. Does alcohol alter OF/blood and Alcohol did not affect OF/blood or OF/plasma ratios. 5 OF/plasma cannabinoid relationships? Do OF concentrations correlate to OF identifies cannabis intake, but poorly predicts effects due to >1000 fold variability in OF THC 5 cannabis’ pharmacodynamic effects? concentrations.

Research Findings and Conclusions

Does cannabis affect lateral control during driving, and at what blood THC concentrations?

Prior to this research, cannabis’ effects on lateral control were poorly understood, due to equivocal results in previous studies (25, 28-29, 44-46, 322). Road tracking is a crucial skill for successful driving, and lane weaving is a common observable sign of

DUI. Cannabis increased SDLP, confirming our hypothesis. THC concentrations ≥8.2

μg/L produced SDLP impairment to at least the same degree as 0.05 g/210L BrAC, with

10 and 20 μg/L blood THC increasing SDLP 8.2% and 16%, respectively, relative to median (THC-negative). (For comparison, 0.05, 0.08, and 0.10 g/210L BrAC increased

SDLP 6.7%, 11%, and 13%.) Blood collection occurs about 90 min after arrest (141) and

3-4 h after an accident (142)—long enough that many specimens are cannabinoid- negative or contain substantially lower THC concentrations than those present during driving.

Unlike alcohol—which displays comparatively slow and consistent zero-order elimination (197, 306) and can be assessed more readily in real time by breath-testing instruments (323-324)—blood THC concentrations decrease rapidly and time-delayed blood collections may not reflect concentrations present during driving. Models were established for determining time since last intake in less than daily smokers by plasma cannabinoids THC (Model 1) and THC/THCCOOH (Model 2) (325). Combining 95% CI from Models 1 and 2 produced the best results (326), by creating the widest possible time range. Studies (277, 326-328) assessing their combined predictive validity for plasma

324

after smoked and oral THC showed ≥94% accuracy within 8 h (follow-up duration) post- dose (smoked) (327-328); over the 24 h follow-up (single-dose oral) or up to 16 h

(multiple oral doses) (326); and 90% accuracy up to 10.5 h but only 70% and 20% for

12.5 h and 22.5 h after last oral dose, respectively (chronic frequent smokers) (277). The present study (occasional-to-moderate smokers) found 93.7% overall accuracy up to 8.3 h

(follow-up duration) after vaporization from the combined models (N=618 calculable plasma specimens), with no significant differences with vs. without alcohol or low vs. high dose. The remaining 6.3% of cases whose collection times were outside the combined models’ estimated CI comprised 28 (4.5%) underestimations (models’ upper confidence limit less than actual time) and 11 (1.8%) overestimations (models’ lower confidence limit more than actual time). The underestimations, which occurred in 13 participants (10 completers), typically occurred after 6.3 or 8.3 h. Upper confidence limits from the combined models in these cases were 1.5 [0.1-3.6] h (median [range]) less than the actual collection time. Five of the 10 completers (and one noncompleter) with underestimations were identified as heavier smokers by their blood or OF cannabinoids

(256, 260). Models were less accurate for chronic frequent smokers (328), because neither model nor the model combination accurately predicted the late excretion phase.

These individuals have long cannabinoid detection windows due to the large body burden of lipophilic THC built up in adipose tissue (108, 163, 329), and argue that they could still have blood THC greater than debated per se cutoffs despite being outside the window of acute impairment. THC was detectable in blood (LOQ 0.3 μg/L) for up to 30 days sustained abstinence in chronic, frequent cannabis smokers (108), with residual psychomotor impairment on critical tracking and divided attention through 3 weeks

325

relative to a control group of occasional cannabis smokers and/or 3,4- methylenedioxymethamphetamine (MDMA) users (330). Although they showed improvement from baseline over 3 weeks, performance never reached that of the control group (however, the control group was located on a different site and statistics did not account for possible between-group IQ differences). Chronic frequent smokers also exhibited significant downregulation of CB1 receptors (positron emission tomography

[PET] imaging) relative to a control group with <10 lifetime cannabis exposures, reversible after ~4 weeks abstinence (331). Thus residual effects of long-term chronic exposure to THC also bear consideration, even at low THC concentrations. Accurately estimating time since last intake and associated impact on driving at the time of the incident remains a challenge, and large interindividual pharmacokinetic variability adds complexity. Lower THC concentrations (such as 1, 2, and 5 μg/L, debated for cannabis per se laws) obtained forensically do not preclude substantially impairing higher concentrations during driving. For all these reasons, accurate science-based per se THC cutoffs for driving remain elusive.

Other lateral control measures examined in this study were not affected by blood

THC concentration according to our descriptive models. Lane departures and maximum lateral acceleration are less sensitive measures, because increases represent substantially more erratic driving than weaving within the confines of a lane. Previous on-road research showed neither 100 or 200 μg/kg (only ~7 or 14 mg) THC alone nor 0.04%

(target BAC) alcohol alone significantly increased time out of lane (34, 47). Conversely,

SD steering angle is not sensitive because minute changes that do not affect steering angle SD can still produce effects on SDLP. Although we hypothesized SD steering angle

326

would increase based on results from one previous study (25), the lack of observed THC effect on this parameter is not surprising. Small steering alterations can lead to wider

SDLP due to forward motion: a minor course adjustment immediately followed by wheel straightening would still put a car on a sideways trajectory. The ability to maintain consistent and stable lane position (low SDLP) requires continuous monitoring and minor iterative adjustments.

How does concurrent alcohol interact with cannabis’ effects on driving lateral control?

Concurrent alcohol produced an additive (rather than synergistic) effect with THC on SDLP. This was indicated by the lack of interaction term entering the GLM Select model. Combining low 2 or 5 μg/L THC with 0.05 g/210L BrAC produced additive

SDLP impairment similar to 10 μg/L THC or 0.08 g/210L BrAC. Although the SDLP impairment effect was not synergistic, adding the relatively lower impairment from minor concentrations of both drugs can increase overall SDLP impairment to more substantial levels.

Prior to this study, the effect of combining THC and alcohol on lateral control was not well characterized because analyses were mostly empirical and conducted by dose rather than drug concentrations. Ronen et al (44) noted 13 mg THC and alcohol

(~0.05% BAC) significantly increased SDLP relative to THC-only and alcohol-only as well as to placebo. In that study, neither drug alone significantly increased SDLP.

Similarly, THC and alcohol alone produced “minor” to “moderate” SDLP increases (2.2-

3.5 cm higher than baseline), whereas combinations produced “severe” increases (5.3 or

8.5 cm for 100 or 200 μg/kg THC with ~0.04% BAC) (34, 47). To our knowledge, only

327

Lenné et al (25) directly investigated possible cannabis-alcohol lateral control interactions via robust linear model analyses. In a driving simulator, cannabis and alcohol significantly increased SDLP, with no significant cannabis*alcohol interaction, consistent with the results of this research.

In our model only alcohol increased lane departures/min and maximum lateral acceleration (precluding evaluation of additive effects), and no interaction term was detected. Although neither THC nor alcohol independently increased time out of lane in previous on-road research (34, 47), combining THC and alcohol did. This is not inconsistent with our findings, because the dose-wise effect size was low (~1.1% increase after 200 μg/kg THC with ~0.04% BAC). With concentration-dependent analysis, it is reasonable that this difference would not be detectable.

Is cannabis vaporization an effective method of inhaled THC delivery, producing expected subjective effects?

Cannabis vaporization is an effective administration route, significantly and dose- dependently increasing subjective “high”, “good drug effect”, “stimulated”, and “stoned” over 3-4 h post-dose, as expected. Observed effects and patterns were consistent with cannabis smoking. As with smoked cannabis (81, 332), subjective effects vs. THC concentration curves displayed counterclockwise hysteresis, reflecting the lag time for equilibration in the brain.

Because smoking is an inappropriate route for pharmacotherapy (64) and vaporization produces similar effects while reducing harmful PAH exposure (69, 71, 78-

79, 83), it is a preferable alternative for medical and recreational marijuana. The

328

vaporizer device employed in this research was the first to be approved for medical use

(333), and is currently approved for this purpose in Canada and various European Union countries.

How does alcohol interact with cannabis’ subjective effects?

Alcohol potentiates cannabis-mediated effects, extending their duration. Overall, alcohol*cannabis interaction terms produced approximately-additive effects initially and synergistic effects over time. This finding is important because it indicates that combining these drugs could produce effects lasting beyond the consumer’s expectations or intent.

Prior to this investigation, limited data were available for combined alcohol and cannabis subjective effects (169, 285, 287). Alcohol decreased the latency to experiencing cannabis’ effects and increased duration of euphoria (287), consistent with the present research. Our investigation presented the most thorough evaluation of cannabis-alcohol interaction on THC subjective effects to date, examining several potential subjective effects rather than just variations of “high”, and quantitatively assessing the interaction between THC concentrations and BrAC.

Are vaporized cannabis’ blood, plasma, and OF cannabinoid dispositions and pharmacokinetics similar to those observed after smoking?

Vaporized cannabis pharmacokinetics in blood, plasma, and OF were similar to those observed after smoking. Maximum concentrations occurred immediately post-dose, with substantial interindividual variability. After the initial rapid THC decrease, blood

329

and plasma THC decreased more gradually after 1.4 h post-dose. THCCOOH and its glucuronide were present for extended periods, displaying large variability consistent with variation in smoking history, whereas THC-glucuronide was only detected within a half hour after inhalation. OF cannabinoids were characterized by THC, CBD, and CBN initial oromucosal contamination, producing high initial concentrations with 2-5 fold more THC variability than in matched blood. If present, OF THCCOOH concentrations were low in these occasional-to-moderate smokers. Although THC Cmax was significantly higher after the high than low dose, blood and plasma THC concentrations showed evidence of self-titration in about half of participants. Prior research indicated similar cannabinoid time courses in OF, blood, and plasma after smoking similar-potency THC

(93-94, 115, 117, 307), and even after widely varying (1.75-13%) potencies (107, 285,

328). Lack of significant high vs. low THC Cmax dose differences reflected self-titration and large interindividual variability.

Consistent with prior findings (93-94), CBD and CBN were only detected in blood and plasma within an hour after inhalation. These compounds identified recent intake, although detection windows in blood and plasma were short compared to normal delays in collection after an arrest or accident (142). OF CBD (0.19%) and CBN (0.37%) were detectable ~2.3 h (median) post-dose, meaning they are more likely to be detected in forensic OF than blood specimens. Karschner et al (104) reported CBD tmax 1.0-5.5h after Sativex (1:1 CBD:THC oromucosal spray, 5 and 15 mg CBD) administration. This highlights CBD relevance in forensic cannabinoid testing, given increasing medical marijuana prevalence. The fourfold difference in CBD potency in low vs. high doses

(0.05% and 0.19%, respectively) created significantly later OF tlast after the high than low

330

doses, an important consideration as medical strains with high-CBD potency become more prevalent.

Although the properties of cannabis vapor are different than those of smoke

(higher cannabinoids:byproducts ratios, many additional compounds present in smoke relative to vapor) (69, 125, 334), this research indicated that the rising popularity of vaporizers for medical and recreational cannabis need not alter interpretation of cannabinoid concentrations. In occasional-to-moderate smokers, a 5 μg/L blood THC cutoff produced detection windows shorter than those for acute performance impairment

(41); whereas 1 μg/L produced extended windows ≥8.3 h. These new data will facilitate forensic cannabinoid blood, plasma and OF interpretation, possibly informing the debate on drugged driving legislation.

Do cannabis and alcohol interactions alter cannabinoid or alcohol pharmacokinetic dispositions?

Significant alcohol*cannabis interactions were noted for blood and plasma THC and 11-OH-THC Cmax (higher with concurrent alcohol) but not AUC0-8.3h, possibly indicating higher absorption rates. Blood and plasma alcohol*cannabis interactions did not consistently impact metabolism. These data are consistent with the limited previous observations. During the THC ascending phase (within 15 min of smoking initiation), alcohol significantly increased plasma THC concentrations and Cmax occurred 5 min sooner (287), whereas the descending phase (20-120 min) showed no significant differences. The current research more robustly demonstrated this absorption effect, examining metabolites as well as THC, and blood in addition to plasma. Alcohol effects

331

on OF cannabinoids were limited, and no additional alcohol*cannabis interactions were observed. Only AUC0-8.3h (THC) and tlast (CBD and CBN) were significantly increased with alcohol co-administration, possibly indicating slightly slowed excretion.

Concurrent alcohol and cannabis dosing in this study also enabled evaluation of cannabis’ effects on alcohol pharmacokinetics. Alcohol had significantly lower and later

Cmax after high-dose cannabis, consistent with previous findings (286). This indicates that cannabis slows alcohol absorption. However, cannabis did not affect alcohol tlast or AUC, suggesting that only absorption was affected. Because no alcohol measurements were made between 0.42 and 1.4 h, the degree to which cannabis delayed alcohol tmax could not be accurately measured in this study.

Does cannabis vaporization or alcohol significantly alter on-site OF test performance or detection windows?

The Dräger on-site test performance is similar after vaporization compared to previously-reported results after smoking (111), albeit with decreased sensitivity (53.8% vs. 90.7% at 2 µg/L THC confirmation cutoff). However, this decrease may be related to several factors including lower administered dose (0.5 g 6.7% THC bulk cannabis vs.

0.79 g 6.8% THC cigarette) (111), release of less THC due to lower volatilization temperature (210°C) than combustion (≥230°C) (69, 238), and possible THC adherence to the balloon (83). Dräger specificity after vaporized cannabis was high (99.3%, similar to smoking (111)) and efficiency was 65.0%. OF tests were performed on two different sites—the clinical research unit and NADS—by different research staff. It is possible that tests were performed differently by the two groups, because the NADS staff had

332

additional training. However, no significant device performance differences were observed.

Alcohol did not significantly affect on-site OF test performance, an important new finding due to the frequency of co-administration.

Can OF directly predict blood or plasma cannabinoid concentrations?

OF cannot directly predict blood or plasma cannabinoid concentrations. Although

OF THC concentrations strongly correlated to blood and plasma after both active cannabis doses, OF/blood and OF/plasma ratios had high variability and were not consistent throughout the time course. Similar wide ranges were recently reported for

OF/serum ratios (116, 291-292), corroborating that OF variability is too great to predict blood/plasma or serum concentrations. Despite this challenge, OF THC concentrations

>600 µg/L likely indicate intake within the last 2-3 h. OF/blood and OF/plasma

THCCOOH were more stable (albeit still with large variability), but THCCOOH was not always detected in OF.

Does alcohol alter OF/blood and OF/plasma cannabinoid relationships?

Alcohol did not significantly affect OF/blood and OF/plasma ratios. This is consistent with prior research (116), and is important given the frequency of co- administration (274).

333

Do OF cannabinoid concentrations correlate to cannabis’ pharmacodynamic effects?

Most subjective effects lacked significant main effects of OF THC that were observed with blood concentrations. Due to the 2-5 fold increased variability relative to matched blood, this finding is unsurprising. Because OF contamination is strongest immediately post-inhalation and decreases thereafter, significant positive time*THC interaction findings in our statistical models may introduce a mechanism to normalize some of the OF THC concentration variability. Meanwhile, OF retains value in identifying recent cannabis exposure (289), but is limited in predicting cannabis effects.

Future Directions

Cannabis’ effects upon other critical driving skills also need examination.

Particularly, longitudinal control, RT, and divided attention are important areas still requiring investigation. Data analyses for longitudinal control, distracted-driving divided attention tasks, and executive function and decision-making divided attention tasks are currently underway through continued collaboration with ONDCP, NHTSA and the

University of Iowa team. Manuscripts for these driving results will be prepared over the next four months, and submitted by autumn 2015. Additional planned driving articles include performance on the monotonous rural straightaway and eye tracking results, longitudinal control, divided attention performance, and an overall summary of cannabis’ effects on driving as measured by this study.

This research focused on occasional smokers, but driving performance among frequent smokers also needs further study. Some evidence indicates tolerance toward

THC effects on predictors of driving performance in chronic frequent smokers (41, 169),

334

but this tolerance may not compensate for all effects. A follow-up study of frequent smokers’ driving performance in this sophisticated simulator would be valuable, particularly as medical and recreational cannabis expand.

Additionally, although these data indicated vaporized cannabinoid disposition and effects similar to smoking, they lacked a direct comparison. As vaporization and new edible cannabis formulations increase in prevalence, within-subjects equivalent dose comparisons among multiple administration routes are necessary. I helped write a protocol (NIDA protocol 14-DA-N135) to investigate this, comparing cannabis subjective, physiological and psychomotor effects and pharmacokinetic disposition after controlled smoked, vaporized, and brownie administration. This protocol, currently being conducted, also will compare frequent and occasional smokers.

Finis

Cannabis and alcohol remain the most commonly encountered drugs in DUI cases. DUIC is a growing current concern capturing national and international attention as medicalized and legalized cannabis becomes increasingly commonplace and per se laws continue to be debated. These new data demonstrate cannabis’ concentration- dependent impairing effects on SDLP, an important measure of lateral control, indicating impairment similar to a 0.05 g/210L BrAC at blood THC concentrations as low as 7-10

μg/L. As changing legislation and public attitudes shift toward greater acceptance for cannabis, alternative (less-harmful) administration routes such as vaporization will become more ubiquitous. Inhaled cannabis has similar pharmacodynamic and pharmacokinetic dispositions after vaporization and smoking, information that will aid

335

forensic interpretation. Blood is preferable to OF for evaluating cannabis’ effects, but OF is convenient and beneficial for real-time drug exposure screening and establishing recent intake.

Additional Research

Over the course of my doctoral work at NIDA, I participated in several additional projects along with my primary dissertation research. These included investigation of

MDMA and metabolite disposition after controlled oral administration, a large-scale review of synthetic cannabinoids knowledge as part of a Department of Defense contract with NIDA, and participation in a military-prevalence study of novel psychoactive substances (NPS) involving immunoassay screening of >20,000 urine specimens.

Appendix A presents my first-author MDMA blood method and data manuscript. Based upon my work with cannabinoids and experience with cannabis-driving research, I also composed an editorial response [Appendix B] to a recently published article (335) with critical errors. Appendix C presents abstracts from the additional published manuscripts that I coauthored.

336

Appendix A

Reprinted with permission from Springer Science+Business Media, 13 March 2015.

337

338

339

340

341

342

343

344

345

346

347

348

349

Appendix B

Reprinted with permission from Oxford University Press 13 March 2015.

350

351

Appendix C (MDMA) and Metabolite Correlation after Controlled Oral MDMA Administration4

Oral fluid (OF) offers a non-invasive sample collection for drug testing. However,

3,4-methylenedioxymethamphetamine (MDMA, ecstasy) in OF has not been adequately characterized in comparison to plasma. We administered oral low (1.0 mg/kg) and high

(1.6 mg/kg) dose MDMA to 26 participants and collected simultaneous OF and plasma specimens for up to 143 h after dosing. We compared OF/plasma (OF/P) ratios, time of initial detection (tfirst), maximal concentrations (Cmax), time of peak concentrations (tmax), time of last detection (tlast), clearance, and 3,4-methylenedioxyamphetamine (MDA) to

MDMA ratios over time.

For OF MDMA and MDA, Cmax was higher, tlast was later, and clearance was slower compared to plasma. For OF MDA only, tfirst was later compared to plasma.

Median (range) OF/P ratios were 5.6 (0.1-52.3) for MDMA and 3.7 (0.7-24.3) for MDA.

OF and plasma concentrations were weakly but significantly correlated (MDMA R2=

0.438, MDA R2= 0.197, p<0.0001). Median OF/P ratios were significantly higher following high dose: MDMA low 5.2 (0.1-40.4) and high 6.0 (0.4-52.3) (p<0.05); MDA low 3.3 (0.7-17.1) and high 4.1 (0.9-24.3) (p<0.001). There was large inter-subject variation in OF/P ratios. MDA/MDMA ratios in plasma were higher than those in OF

(p<0.001), and MDA/MDMA ratios significantly increased over time in OF and plasma.

MDMA and MDA concentrations were higher in OF than in plasma. OF and plasma concentrations were correlated, but large inter-subject variability precludes estimation of plasma concentrations from OF.

4 Abstract from Desrosiers NA et al. Anal Bioanal Chem 2013 May;405(12):4067-76 352

Synthetic Cannabinoids: Epidemiology, Pharmacodynamics and Clinical Implications5

Synthetic cannabinoids (SC) are a heterogeneous group of compounds developed to probe the endogenous cannabinoid system or as potential therapeutics. Clandestine laboratories subsequently utilized published data to develop SC variations marketed as abusable “designer drugs.” In the early 2000’s, SC became popular as “legal highs” under brand names such as “Spice” and “K2,” in part due to their ability to escape detection by standard cannabinoid screening tests.

The majority of SC detected in herbal products have greater binding affinity to the

9 cannabinoid CB1 receptor than does Δ -tetrahydrocannabinol (THC), the primary psychoactive compound in the cannabis plant, and greater affinity at the CB1 than the CB2 receptor. In-vitro and animal in-vivo studies show SC pharmacological effects 2-100 times more potent than THC, including analgesic, anti-seizure, weight-loss, anti- inflammatory, and anti-cancer growth effects.

SC produce physiological and psychoactive effects similar to THC, but with greater intensity, resulting in medical and psychiatric emergencies. Human adverse effects include nausea and vomiting, shortness of breath or depressed breathing, hypertension, tachycardia, chest pain, muscle twitches, acute renal failure, anxiety, agitation, psychosis, suicidal ideation, and cognitive impairment. Long-term or residual effects are unknown. Due to these public health consequences, many SC are classified as controlled substances. However, frequent structural modification by clandestine

5 Abstract from Castaneto MS et al. Drug Alcohol Depend 2014 Nov 1;144:12-41 353

laboratories results in a stream of novel SC that may not be legally controlled or detectable by routine laboratory tests.

We present here a comprehensive review, based on a systematic electronic literature search, of SC epidemiology and pharmacology and their clinical implications.

Synthetic cannabinoids pharmacokinetics and detection methods in biological matrices6

Synthetic cannabinoids (SC), originally developed as research tools, are now highly abused novel psychoactive substances. We present a comprehensive systematic review covering in vivo and in vitro animal and human pharmacokinetics and analytical methods for identifying SC and their metabolites in biological matrices. Of two main phases of SC research, the first investigated therapeutic applications, and the second abuse-related issues. Administration studies showed high lipophilicity and distribution into brain and fat tissue. Metabolite profiling studies, mostly with human liver microsomes and human hepatocytes, structurally elucidated metabolites and identified suitable SC markers. In general, SC underwent hydroxylation at various molecular sites, defluorination of fluorinated analogs, and phase II metabolites were almost exclusively glucuronides.

Analytical methods are critical for documenting intake, with different strategies applied to adequately address the continuous emergence of new compounds.

Immunoassays have different cross-reactivities for different SC classes, but cannot keep pace with changing analyte targets. Gas chromatography and liquid chromatography

6 Abstract from Castaneto MS et al. Drug Metab Rev 2015; in press. doi: 10.3109/03602532.2015.1029635 354

mass spectrometry assays—first for a few, then numerous analytes—are available but constrained by reference standard availability, and must be continuously updated and revalidated. In blood and oral fluid, parent compounds are frequently present, albeit in low concentrations; for urinary detection, metabolites must be identified and interpretation is complex due to shared metabolic pathways. A new approach is non- targeted HRMS screening that is more flexible, and permits retrospective data analysis.

We suggest that streamlined assessment of new SC’s pharmacokinetics and advanced

HRMS screening provide a promising strategy to maintain relevant assays.

355

Protocol Appendices

Protocol Appendix 1. Telephone screening questions

Inclusion Criteria ~ General Questions

Before the list of questions is administered, please communicate the following:

There are several criteria that must be met for participation in this study. I will need to ask you several questions to determine your eligibility.

If a subject fails to meet one of the following criteria, (answers must be YES unless otherwise specified), proceed to Closing.

1) Do you use cannabis at least once every three months? 2) Are you between the ages of 21-55 years of age? 3) Do you currently have a valid US Drivers’ License? 4) Is your driver’s license free of restrictions (vision restriction not considered if corrected to 20/20 with corrected lenses)? 5) Do you drive at least 1300 miles per year? 6) Do you consider yourself in good physical and mental health? 7) Do you live within a 80 miles of the National Advanced Driving Simulator, located at University Research Park? 8) Are you able to come to 6 study visits the night before and stay until the following night? 9) Are you able to drive without the use of special equipment such as pedal extensions, hand brake or throttle, spinner wheel knobs or other nonstandard equipment? 10) Would this be the first time you have you participated in a driving simulator study at NADS? (If NO to above) Was the study about impairment? (Must answer NO) Was the study about distraction? (Must answer NO)

General Inclusion Criteria are met. Because we are conducting a study to determine how various blood alcohol concentrations and cannabis impact driving performance, the following questions ask you about the quantity, frequency, and regularity of alcohol (and cannabis) you consume. Your answers will determine if you continue to meet the study qualifications.

Administer Phone Screening Quantity, Frequency, Variability (Phone Screening_QFV) to determine if potential subject meets study criteria for light or moderate drinker of alcoholic beverages or, , if a heavy drinker, not drink more than 1-2 times a week and not have a modal quantity of 5-6 drinks .

356

Administer AUDIT to determine if potential subject meets study criteria. Administer CUDIT to determine if potential subject meets study criteria.

Administer Phone Screening Quantity-Frequency-Variability (QFV) Questionnaire (Appendix 12) . If all Inclusion Criteria are met, proceed to General Health Exclusion . If subject doesn’t meet criteria, proceed to Closing If potential subject meets study criteria, proceed to Exclusion Criteria

Exclusion Criteria ~ General Health Questions Overview

 Before administering this list of questions, please communicate the following:  Because of pre-existing health conditions, some people are not eligible for participation in this study.  I need to ask you several health-related questions before you can be scheduled for a study session.  Your responses are voluntary and all answers are confidential.  You can refuse to answer any questions and only a record of your motion sickness susceptibility will be kept as part of this study.  No other responses will be kept. . If a participant fails to meet one of the following criteria, proceed to the Closing (If unsure about exclusion criteria, consult Principal Investigator )

1) If the subject is female:  Are you, or is there any possibility that you are pregnant? Exclusion criteria:  If there is ANY possibility of pregnancy

2) Have you been diagnosed with a serious illness?  If YES, is the condition still active?  Are there any lingering effects?  If YES, do you care to describe? Exclusion criteria:  Cancer (receiving any radiation and/or chemotherapy treatment within last 6 months)  Crohn’s disease  Hodgkin’s disease  Currently receiving any radiation and/or chemotherapy treatment

3) Do you have Diabetes?  Have you been diagnosed with hypoglycemia?  If yes, do you take insulin or any other medication for blood sugar? NOTE: Type II Diabetes accepted if controlled (medicated and under the supervision of

357

physician) Exclusion criteria:  Type I Diabetes - insulin dependent  Type II – Uncontrolled (see above)

4) Do you suffer from a heart condition such as disturbance of the heart rhythm or have you had a heart attack or a pacemaker implanted within the last 6 months?  If YES, please describe? Exclusion criteria:  History of ventricular flutter or fibrillation  Systole requiring cardio version (atrial fibrillation may be acceptable if heart rhythm is stable following medical treatment or pacemaker implants)

5) Have you ever suffered brain damage from a stroke, tumor, head injury, or infection?  If YES, what are the resulting effects?  Do you have an active tumor?  Any visual loss, blurring or double vision?  Any weakness, numbness, or funny feelings in the arms, legs or face?  Any trouble swallowing or slurred speech?  Any uncoordination or loss of control?  Any trouble walking, thinking, remembering, talking, or understanding? Exclusion criteria:  A stroke within the past 6 months  An active tumor  Any symptoms still exist

6) Have you ever been diagnosed with seizures or epilepsy?  If YES, how frequently and what type? Exclusion criteria:  A seizure within the past 12 months

7) Do you have Ménière's Disease or any inner ear, dizziness, vertigo, hearing, or balance problems?  Wear hearing aides - full correction with hearing aides acceptable  If YES, please describe.  Ménière's Disease is a problem in the inner ear that affects hearing and balance. Symptoms can be low- pitched roaring in the ear (tinnitus), hearing loss, which may be permanent or temporary, and vertigo.  Vertigo is a feeling that you or your surroundings are moving when there is no actual movement, described as a feeling of spinning or whirling and can be sensations of falling or tilting. It may be difficult to walk or stand and you may lose your balance and fall.

Exclusion criteria:  Meniere’s Disease  Any recent history of inner ear, dizziness, vertigo, or balance problems

358

8) Do you currently have a sleep disorder such as sleep apnea, narcolepsy, Chronic Fatigue Syndrome, etc.?  If YES, please describe.  Sleep apnea: how long under treatment and was treatment successful Exclusion criteria:  Untreated sleep apnea  Narcolepsy  Chronic Fatigue Syndrome

9) Do you have migraine or tension headaches that require you to take medication daily?  If YES, please describe. Exclusion criteria:  Any narcotic medications

10) Do you currently have untreated depression, anxiety disorder, ADHD or claustrophia?  If YES, please describe. Exclusion criteria:  Untreated depression  Agoraphobia, hyperventilation, or anxiety attacks  ADHD (treated and untreated)

11) Are you currently taking any prescription or over the counter medications?  If YES, what is the medication?  Check PDR for possible interaction with alcohol or cannabis  Are there any warning labels on your medications? Warning about using medication with alcohol or cannabis or warning about drowsiness  Ask potential subject to check with his/her physician that use of their medication is acceptable with drinking alcohol beverages  Over the counter medications: ask potential subject to not use medication for 48 hours prior to visit if able to discontinued and does not compromise them medically and is acceptable to them to not use prior to visit. Exclusion criteria:  Any sedating medications or drowsiness label on medication  Drugs that interact with alcohol  Subject’s physician objects to use of medication while drinking alcoholic beverages  Warning on label about use of medication with alcohol  Unable to discontinue use of over the counter medication

12) Do you experience any kind of motion sickness?  If YES, what were the conditions you experienced: when occurred (age), what mode of transportation, (boat, plane, train, car), and what was the intensity of your motion sickness? 359

 On a scale of 0 to 10, how often do you experience motion sickness with 0 = Never and 10 = Always  On a scale of 0 to 10, how severe are the symptoms when you experience motion sickness with 0 = Minimal and 10 = Incapacitated Exclusion criteria:  One single mode of transportation where intensity is high and present  More than 2 to 3 episodes for mode of transportation where intensity is moderate or above  Severity and susceptibility scores rank high

13) Have you donated any blood recently?  If YES, when did you donate?  Schedule screening appointment for 14 days post blood donation if possible Exclusion criteria:  Unable to schedule screening appointment at least 14 days post donation

13) Are you able to avoid donating blood from two weeks prior to the study through to two weeks after the study  If YES, when did you last donate?

Exclusion criteria:  Unable to avoid donating blood

15) Have you ever had a negative reaction from using cannabis/marijuana/hashish

Exclusion criteria:  Negative reaction

16) Have you ever participated in drug abuse treatment?  If YES, when did you participate in treatment? Exclusion criteria:  If participated in treatment in the last 60 days  Currently participating in treatment

17) Are you currently interested in participating in drug abuse treatment? Exclusion criteria:  If interested in receiving treatment

360

Protocol Appendix 2: Driving Survey

Study: Effects of Inhaled Cannabis on Driving Performance Participant:______Date: ______

Driving Survey

As part of this study, it is useful to collect information describing each participant. The following questions ask about you and your health, your driving patterns, and your alcohol consumption. Please read each question carefully. If something is unclear, ask the researcher for help. Your participation is voluntary and you have the right to omit questions if you choose. Please remember that all of your answers will be kept confidential.

Background Information

1) What is your birth date? ______/ ______/ ______Month Day Year

2) What age are you today? ______

3) What is your gender?  Male  Female

4) What is your marital status? (Check only one)  Single, never married  Married  Domestic Partnership  Separated or Divorced  Widowed

5) What was your total household income last year? (Check only one)  $0- $24,999  $25,000- $29,999  $30,000 - $34,999  $35,000 - $39,999  $40,000 - $49,999  $50,000 - $59,999  $60,000 - $69,999  $70,000 - $79,999  $80,000 - $89,999  $90,000 - $99,999  $100,000 or more

6) What is your present employment status? (Check only one)  Unemployed  Retired  Work part-time  Work full-time  None of the above

361

7) What type of work do you do (e.g., teacher, homemaker)? ______

10) Of which ethnic origin(s) do you consider yourself? (Check all that apply)  American Indian/Alaska Native  Asian  Black/African American  Hispanic/Latino  Native Hawaiian/Other Pacific Islander  White/Caucasian  Other

11) What is the highest level of education that you have completed? (Check only one)  Primary School  High School Diploma or equivalent  Technical School or equivalent  Some College or University  Associate’s Degree  Bachelor’s Degree  Some Graduate or Professional School  Graduate or Professional Degree

Driving Experience

12) How old were you when you started to drive? ______years of age

13) For which of the following do you currently hold a valid driver’s license within the United States? (Check all that apply)

Vehicle Type Year When FIRST Licensed (May be Approximate)  Passenger Vehicle License ______ Commercial Truck License ______ Motorcycle License ______ Other: ______ Other: ______

14) How often do you drive? (Check the most appropriate category)  Less than once weekly  At least once weekly  At least once daily

15) Approximately how many miles do you drive per year in each vehicle type, excluding miles driven for work-related activities? (Check only one for each vehicle) Other: Car Motorcycle Truck ______ Do not drive  Do not drive  Do not drive  Do not drive  Under 2,000  Under 2,000  Under 2,000  Under 2,000  2,000 - 7,999  2,000 - 7,999  2,000 - 7,999  2,000 - 7,999  8,000 - 12,999  8,000 - 12,999  8,000 - 12,999  8,000 - 12,999  13,000 - 19,999  13,000 - 19,999  13,000 - 19,999  13,000 - 19,999  20,000 or more  20,000 or more  20,000 or more  20,000 or more 362

16) How frequently do you drive in the following environments? (Check only one for each environment) Never Yearly Monthly Weekly Daily Residential      Business District      Rural Highway (e.g., Route 6)      Interstate (e.g., Interstate 80)      Gravel Roads     

17) What speed do you typically drive in a residential area when the speed limit is 25? ______mph

18) What speed do you typically drive in a business district when the speed limit is 35? ______mph

19) What speed do you typically drive on a rural highway when the speed limit is 55? ______mph

20) What speed do you typically drive on the Interstate when the speed limit is 65? ______mph

21) What speed do you typically drive on a gravel road? ______mph

22) Have you ever had to participate in any driver improvement courses due to moving violations?  No  Yes (Please describe) ______23) When driving, how frequently do you perform each of the following tasks/maneuvers? (Check the most appropriate answer for each task/maneuver) Not Occas- Fre- Applic- Never Rarely ionally quently Always able Change lanes on Interstate or       freeway Keep up with traffic in town       Keep up with traffic on two-lane       highway Keep up with traffic on       Interstate or freeway Pass other cars on Interstate or       freeway Exceed speed limit       Wear a safety belt       Make left turns at uncontrolled       intersections 24) How comfortable do you feel when you drive in the following conditions or perform the following maneuvers? (Check the most appropriate answer for each condition) Very Slightly Slightly Very Not

363

Uncomfort- Uncomfort- Comfort- Comfort- Applicable able able able able Highway/freeway      After drinking alcohol      After smoking      marijuana With children      High-density traffic      Passing other cars      Changing lanes      Making left turns at uncontrolled      intersections

Violations

25) Within the past five years, how many tickets have you received for the following? (Please check a response for each ticket) 0 1 2 3+ Speeding     Going too slowly     Failure to yield right of way     Disobeying traffic lights     Disobeying traffic signs     Improper passing     Improper turning     Reckless driving     Following another car too closely     Operating While Intoxicated (OWI) or Driving Under     the influence (DUI) Other (please specify type and frequency of violation) ______

Accidents 26) In the past five years, how many times have you been the driver of a car involved in an accident?  0 (Go to question # 27 on page 7)  1  2  3  4 or more Please provide the following information for each accident on the next page. Accident 1 Was another vehicle involved?  No  Yes

364

Was a pedestrian involved?  No  Yes Were you largely responsible for this  No  Yes accident? Did you go to driver’s rehabilitation?  No  Yes

Weather Condition: ______Month/Year: ______

Description:______

Accident 2 Was another vehicle involved?  No  Yes

Was a pedestrian involved?  No  Yes

Were you largely responsible for this accident?  No  Yes

Did you go to driver’s rehabilitation?  No  Yes

Weather Condition: ______Month/Year: ______

Description:______

______

Accident 3 Was another vehicle involved?  No  Yes

Was a pedestrian involved?  No  Yes

Were you largely responsible for this accident?  No  Yes

Did you go to driver’s rehabilitation?  No  Yes

Weather Condition: ______Month/Year: ______

Description:______

Health Status

365

27) How often do you experience motion sickness? (Circle only one)

0 1 2 3 4 5 6 7 8 9 10 Never Always

28) How severe are your symptoms when you experience motion sickness (Circle only one)

0 1 2 3 4 5 6 7 8 9 10 None Severe

29) Have you taken any medication in the past 48 hours? (Check only one)

 No  Yes (Please list all) ______

30) What is your normal bedtime (hour of the day)? ______

Alcohol Consumption History

31) When you drink alcoholic beverages, where do you usually drink?  At home  At a friend’s home  Public place (restaurant, bar, sports arena, etc.)

32) In a typical month, how many hours do you wait to drive after consuming what number of drinks?  Not Applicable, proceed to next question  Applicable, complete table below (Mark all that apply. For example, if every week you have one glass of wine with dinner at your favorite restaurant and drive home less than one hour before you finish the glass and on three Sundays each month you have four beers while watching the game and wait 1 hour after finishing your last drink to drive, you would mark the following:

EXAMPLE <1 hour 1 hour 2 hours TABLE 1 drink 4 2 drinks 3 drinks 4 drinks 3

< 1 hour 1 hour 2 hours 3 hours 4 hours 5 hours 6 or more 1 drink 2 drinks 3 drinks 4 drinks 5 drinks 6 or more 33) Compare your driving to other people’s driving:

366

a) After one alcoholic beverage, I rate my driving as compared with others who have consumed as much alcohol as I have as: (make a slash mark anywhere along the line)

Far worse about the same considerably better

b) After two alcoholic beverages, I rate my driving as compared with others who have consumed as much alcohol as I have as: (make a slash mark anywhere along the line)

Far worse about the same considerably better

c) After five alcoholic beverages, I rate my driving as compared with others who have consumed as much alcohol as I have as: (make a slash mark anywhere along the line)

Far worse about the same considerably better

34) How many times per month do you operate a motor vehicle after consuming too much alcohol?  Never  1-2  3-5  6-10  11-15  16 or more

35) How many times per month do you drive after someone has encouraged you not to because of your level of alcohol intoxication?  Never  1-2  3-5  6-10  11-15  16 or more

36) Do you have a maximum number of alcoholic beverages you can consume after which you do not operate a motor vehicle?  No  Yes, how many_____

37) Have you ever been pulled over by the police and been given a breathalyzer or standard field sobriety test?  No  Yes, how many times? ______

38) Have you

367

a) ever been charged for operating while intoxicated (OWI) or for driving under the influence of alcohol (DUI)?  No  Yes, how many times? ______

b) been charged for operating while intoxicated (OWI) or for driving under the influence of alcohol (DUI) in the past 3 years?  No  Yes, how many times? ______

39) Have you ever caused an accident while under the influence of alcohol?  No  Yes

40) Have you ever been in an accident where someone else was under the influence of alcohol?  No  Yes

41) Have your driving privileges ever been suspended or revoked for operating while intoxicated (OWI) or for driving under the influence of alcohol (DUI)?  No  Yes

Cannabis Consumption History

42) When you use marijuana, where do you usually use it?  At home  At a friend’s home  Public place (restaurant, bar, sports arena, etc.)

43) What is your preferred way to take marijuana/hash/weed/cannabis?

 Smoking  Inhaling from a vaporizer  Hookah/Bong  Oral (brownies, etc)

44) In a typical month, how many hours do you normally wait to drive after consuming cannabis?  Not Applicable, proceed to next question  < 1 hour  1 hour  2 hours  3 hours  4 hours  5 hours  6 or more

368

45) Compare your driving to other people’s driving:

a) After one joint, I rate my driving as compared with others who have consumed as much alcohol as I have as: (make a slash mark anywhere along the line)

Far worse about the same considerably better  Not Applicable

b) After two joints, I rate my driving as compared with others who have consumed as much alcohol as I have as: (make a slash mark anywhere along the line)

Far worse about the same considerably better  Not Applicable

c) After five joints, I rate my driving as compared with others who have consumed as much alcohol as I have as: (make a slash mark anywhere along the line)

Far worse about the same considerably better  Not Applicable

46) How many times per month do you operate a motor vehicle after consuming too much marijuana/hash/weed/cannabis?  Never  1-2  3-5  6-10  11-15  16 or more

47) How many times per month do you drive after someone has encouraged you not to because of your level of cannabis intoxication?  Never  1-2  3-5  6-10  11-15  16 or more

48) Do you have a maximum amount of marijuana/hash/weed/cannabis you can consume after which you do not operate a motor vehicle?  No  Yes, how much_____

49) Have you ever been pulled over by the police for driving while stoned?  No  Yes, how many times? ______

50) Have you ever caused an accident while under the influence of marijuana/hash/weed/cannabis?  No  Yes

369

51) Have you ever been in an accident where someone else was under the influence of marijuana?  No  Yes

Other Studies

52) Have you participated in other driving studies?  No (End of questionnaire)  Yes (please provide details for each study you have participated in below)

Study 1 What vehicle was used for this study? (Check only one)  Actual car - only  Another simulator - only  National Advanced Driving Simulator (Motion Simulator)  National Advanced Driving Simulator (Static Simulator)  Both - actual car and another simulator  Both - actual car and the National Advanced Driving Simulator (Motion Simulator) Brief Description: ______

Study 2 What vehicle was used for this study? (Check only one)  Actual car - only  Another simulator - only  National Advanced Driving Simulator (Motion Simulator)  National Advanced Driving Simulator (Static Simulator)  Both - actual car and another simulator  Both - actual car and the National Advanced Driving Simulator (Motion Simulator) Brief Description: ______

Study 3 What vehicle was used for this study? (Check only one)

 Actual car - only  Another simulator - only  National Advanced Driving Simulator (Motion Simulator)  National Advanced Driving Simulator (Static Simulator)  Both - actual car and another simulator  Both - actual car and the National Advanced Driving Simulator (Motion Simulator)

Brief Description: ______

370

Protocol Appendix 3: Realism Survey

Study: Effects of Inhaled Cannabis on Driving Performance Participant: ______Visit: Date: ______REALISM SURVEY For each of the following items, circle the number that best indicates how closely the simulator resembles an actual car in terms of appearance, sound, and response. If an item is not applicable, circle NA. Not at Com- all pletely General Driving realis- Realis tic -tic 1 Response of the seat adjustment levers 0 1 2 3 4 5 6 NA 2 Response of the mirror adjustment levers 0 1 2 3 4 5 6 NA 3 Response of the door locks and handles 0 1 2 3 4 5 6 NA 4 Response of the fans 0 1 2 3 4 5 6 NA 5 Response of the gear shift 0 1 2 3 4 5 6 NA 6 Response of the brake pedal 0 1 2 3 4 5 6 NA 7 Response of accelerator pedal 0 1 2 3 4 5 6 NA 8 Response of the speedometer 0 1 2 3 4 5 6 NA Response of the steering wheel while 9 0 1 2 3 4 5 6 NA driving straight Response of the steering wheel while 10 0 1 2 3 4 5 6 NA driving on curves 11 Feel when accelerating 0 1 2 3 4 5 6 NA 12 Feel when braking 0 1 2 3 4 5 6 NA 13 Ability to read road and warning signs 0 1 2 3 4 5 6 NA 14 Appearance of car interior 0 1 2 3 4 5 6 NA 15 Appearance of signs 0 1 2 3 4 5 6 NA 16 Appearance of roads and road markings 0 1 2 3 4 5 6 NA 17 Appearance of urban scenery 0 1 2 3 4 5 6 NA 18 Appearance of rural scenery 0 1 2 3 4 5 6 NA 19 Appearance of freeway scenery 0 1 2 3 4 5 6 NA 20 Appearance of intersections 0 1 2 3 4 5 6 NA 21 Appearance of headlights 0 1 2 3 4 5 6 NA 22 Appearance of gravel road 0 1 2 3 4 5 6 NA 23 Appearance of other vehicles 0 1 2 3 4 5 6 NA 24 Appearance of rear-view mirror image 0 1 2 3 4 5 6 NA 25 Sound of the car 0 1 2 3 4 5 6 NA 26 Sound of other vehicles 0 1 2 3 4 5 6 NA 27 Overall feel of the car when driving 0 1 2 3 4 5 6 NA 28 Overall similarity to real driving 0 1 2 3 4 5 6 NA 29 Overall Appearance of driving scenes 0 1 2 3 4 5 6 NA

371

Not at Compl

all etely Situational Driving realis- Realis

tic -tic 30 Feel of driving straight while going 25 mph 0 1 2 3 4 5 6 NA 31 Feel of driving straight while going 35 mph 0 1 2 3 4 5 6 NA 32 Feel of driving straight while going 55 mph 0 1 2 3 4 5 6 NA 33 Feel of driving straight while going 65 mph 0 1 2 3 4 5 6 NA Feel of driving on a curved road while 34 0 1 2 3 4 5 6 NA going 25 mph Feel of driving on a curved road while 35 0 1 2 3 4 5 6 NA going 55 mph Feel of driving on a curved road while 36 0 1 2 3 4 5 6 NA going 65 mph Feel of accelerating from a stopped 37 0 1 2 3 4 5 6 NA position 38 Feel of braking to a stop 0 1 2 3 4 5 6 NA Performing a 90 degree turn to the left 39 0 1 2 3 4 5 6 NA while going 25 mph Performing a 90 degree turn to the right 40 0 1 2 3 4 5 6 NA from a stopped position 41 Feel of driving on the freeway 0 1 2 3 4 5 6 NA 42 Feel of changing lanes on the freeway 0 1 2 3 4 5 6 NA 43 Feel of driving on a freeway on/exit ramp 0 1 2 3 4 5 6 NA 44 Feel of driving on gravel road 0 1 2 3 4 5 6 NA 45 Ability to stop the vehicle 0 1 2 3 4 5 6 NA 46 Ability to respond to other vehicles 0 1 2 3 4 5 6 NA 47 Ability to keep straight in your lane 0 1 2 3 4 5 6 NA 48 Ability to respond at intersections 0 1 2 3 4 5 6 NA

372

Protocol Appendix 4: Neuromotor Examination

ARC #:______Date:______

Mental Status: Pre: Post:

Alert Y N Alert Y N Drowsy Y N Drowsy Y N Oriented to: Oriented to: Person Y N Person Y N Place Y N Place Y N Time Y N Time Y N Situation Y N Situation Y N

Neuromotor Skills

Pre: Post:

Walk and Turn: (circle one) Walk and Turn: (circle one)

Can’t keep balance Normal Can’t keep balance Normal Abnormal Abnormal Stops walking Normal Stops walking Normal Abnormal Abnormal Steps off line Normal Steps off line Normal Abnormal Abnormal Misses heel /toe Normal Misses heel /toe Normal Abnormal Abnormal Raises arms Normal Raises arms Normal Abnormal Abnormal

Pre-Dosing Evaluator’s Signature______Date:______Time______Comments

Post-Dosing Evaluator’s Signature______Date:______Time______Comments:______

373

Protocol Appendix 5: Risk Perception Questionnaire with Randomization Instructions

ARC#: ______Date of Rating: __ __/__ __/______Subject ID: ______Subject Initials: ______Rater ID: ______Rater Initials: ______

NOTES: ______

Was Form Completed? ____ (1=Yes,2=No) If Not, Please Specify: ______

Much A little A little Much Missing Less Equally More less less more more data likely likely likely likely likely likely likely -3 -2 -1 0 1 2 3 9 I. Form: A II. “Compared to other adults of your age and gender, how likely is it that you will at some point in your lifetime?” 1 ______Develop gum disease 2 ______Suffer from frostbite 3 ______Develop diabetes 4 ______Contract lung cancer 5 ______Be involved in an automobile crash as a passenger 6 ______Experience obesity 7 ______Get sick from exposure to chemical fertilizer 8 ______Experience asbestos poisoning 9 ______Experience a non-fatal case of appendicitis 10 ______Experience non-fatal heat stroke 11 ______Accidentally drown 12 ______Develop glaucoma 13 ______Contract a non-fatal STD (Sexually Transmitted Disease) 14 ______Be the victim of a non-fatal violent crime 15 ______Get a cut, while preparing food, severe enough to need treatment 16 ______Become anemic 17 ______Be pressured by friends or family to commit a crime 18 ______Develop asthma 19 ______Get robbed on the street

374

20 ______Get sick due to toxic gas in your home 21 ______Get a dog bite that requires medical treatment 22 ______Experience a fatal overdose of a prescription painkiller 23 ______Contract a non-fatal case of influenza 24 ______Be injured in a car as a result of not wearing a seatbelt 25 ______Lose a large amount of money while gambling 26 ______Contract a case of the common cold 27 ______Get sick from using someone else's toothbrush 28 ______Get sick as a result of drinking water contamination 29 ______Witness a crime being committed 30 ______Get sick as the result of a vaccination 31 ______Experience a fatal fall 32 ______Be injured by debris falling from space 33 ______Be sent to jail 34 ______Contract a non-fatal case of pneumonia 35 ______Get sick from exposure to hazardous waste 36 ______Experience a fatal overdose of aspirin or Tylenol 37 ______Develop arthritis 38 ______Develop cataracts 39 ______Undergo an organ transplant operation 40 ______Experience an earthquake

Randomization instructions:

1.) Highlight the first column in the table above (which contains the question numbers). 2.) Copy the column to your clipboard. 3.) Go to http://www.random.org/lists/ and Paste the number column into the field provided. 4.) Click Randomize. 5.) Highlight randomized number list, and Copy to clipboard. 6.) Paste into the first column of the table in this document, replacing the original numbers. 7.) Turn OFF the automatic numbering for the first column. This will eliminate the ordinal numbers produced as an artifact of copying the randomized list. 8.) Select entire table. Under Table Tools  Layout, choose Sort. 9.) Sort by Column 1; Type: Number; Using: Paragraphs; Ascending.

375

Protocol Appendix 6: Self-Assessments of Risk Perception, Risk- Taking/Impulsivity and Sensation-Seeking

RISK PERCEPTION SCALE How likely is it that something bad would happen to you if you: Score* 1. drove over the speed limit? ______2. drove while drunk? ______3. drove without a seat belt? ______4. drank a lot? ______5. had sex with someone you just met? ______6. got drunk and had sex with someone you just met? ______

*1 – 5 scale: 1=Very unlikely; 5=Very likely.

RISK -TAKING AND IMPULSIVITY SCALE How much does each of the following statements escribed you? Score* 1. I often act on the spur of the moment without stopping to think. ______2. I get a real kick out of doing things that are a little dangerous. ______3. You might say I act impulsively. ______4. I like to test myself every now and then by doing something a little chancy. ______5. Many of my actions seem to be hasty. ______

*1 – 4 scale: 1=Not at all; 4=Quite a lot.

SENSATION -SEEKING SCALE How much does each of the following statements describe you? Score* 1. I'm always up for a new experience. ______2. I like to try new things for the excitement. ______3. I go for the thrills in life when I get a chance. ______4. I like to experience new and different sensations. ______

*1 – 4 scale: 1=Not at all; 4=Quite a lot.

376

Protocol Appendix 7: Impulsive Sensation-Seeking Subscale from the Zuckerman-Kuhlman Personality Questionnaire

377

Protocol Appendix 8: Barratt Impulsiveness Scale (Version 11)

Instructions: Please tick the response that best indicates your usual style (Rarely/Never, Occasionally, Often, Almost Always/Always). 1. I plan tasks carefully. 2. I do things without thinking. 3. I make-up my mind quickly. 4. I am happy-go-lucky. 5. I don’t “pay attention.” 6. I have “racing” thoughts. 7. I plan trips well ahead of time. 8. I am self-controlled. 9. I concentrate easily. 10. I save regularly. 11. I “squirm” at plays or lectures. 12. I am a careful thinker. 13. I plan for job security. 14. I say things without thinking. 15. I like to think about complex problems.’ 16. I change .jobs. 17. I act “on impulse.” 18. I get easily bored when solving thought problems. 19. I act on the spur of the moment. 20. I am a steady thinker. 21. I change residences. 22. I buy things on impulse. 23. I can only think about one problem at a time. 24. I change hobbies 25. I spend or charge more than I earn. 26. I often have extraneous thoughts when thinking. 27. I am more interested in the present than the future.

378

28. I am restless at the theater or lectures. 29. I like puzzles. 30. I am future oriented.

Protocol Appendix 9: Subjective Effects

Visual-analogue scales (100 mm anchored with “not at all” and “most ever”)

Good drug effect High Stoned Stimulated Sedated Anxious Restless

5-Point Likert scales (none, slight, mild, moderate, severe)

Difficulty concentrating Altered sense of time Slowed or slurred speech Body feels sluggish or heavy Feel hungry Feel thirsty Shakiness/tremulousness Nausea Headache Palpitations Upset stomach Dizzy Dry mouth or throat

379

Protocol Appendix 10: Diagram of the Balloon Analogue Risk Task (BART)

380

Protocol Appendix 11. Quantity-Frequency-Variability (QFV) scale I am going to ask you some questions about your recent use of alcohol (past 60-90 days). We need participants with a variety of levels and patterns of alcohol use, so there are no right or wrong answers. Please respond as honestly and accurately as you can. Keep in mind that a glass of wine is 5 ounces, a can of beer is 12 ounces, and a drink of liquor or spirits is 1.5 ounces. Quantity Think of all the times you have had wine recently, How often do you have as many as... Nearly every More than half Less than half Once in a Never time the time the time while 5 – 6 glasses 3 - 4 glasses 1 – 2 glasses

Think of all the times you have had beer recently, How often do you have as many as... Nearly More than half the Less than half Once in a Never every time time the time while 5 – 6 cans 3 – 4 cans 1 – 2 cans

Think of all the times you have had whiskey/liquor recently, How often do you have as many as... Nearly More than half the Less than half Once in a Never every time time the time while 5 – 6 drinks 3 – 4 drinks 1 – 2 drinks Frequency Please tell us how often you have the following kinds of drinks…. Liquor Any kind of Wine Beer (Spirits) alcoholic drink 3 or more times a day 2 times a day Nearly every day 3-4 times per week Once or twice per week 2-3 times a month About once a month Less than once a month but at least

once a year Never

381

Protocol Appendix 12. Phone Screening QFV

382

383

AUDIT Pre Post

40777 Today's Date Id No. Gender 1 / Protocol/ Appendix 13. AUDITMale Female Survey

Using a black pen, please fill in the bubble that is correct for you.

1. How often do you have a drink containing alcohol?

NEVER MONTHLY TWO TO FOUR TWO OR THREE FOUR OR MORE OR LESS TIMES A MONTH TIMES A WEEK TIMES A WEEK

NOTE: For answering these questions, one "drink" is equal to 12 ounces of beer, or 5 ounces of wine, or 1 ounce of liquor.

2. How many drinks containing alcohol do you have on a typical day when you are drinking?

1 OR 2 2 TO 4 5 OR 6 7 TO 9 10 OR MORE

3. How often do you have six or more drinks on one occasion?