Special Section Conference Abstracts 66 

 

 

 

 

 
 

 

Special Section Editor 

Bradley T. Conner, Ph.D. 

 

KEYNOTE ADDRESS 
 

How Cannabis Research and Cannabis Policy 

Can Inform Each Other 

 

Susan R. B. Weiss 

National Institute on Drug Abuse 

 

Cannabis policies in the United States (and 

globally) are changing rapidly, far outpacing the 

knowledge needed to determine and minimize the 

public health impacts of these changes. Cannabis 

is already the most commonly used federally illicit 

substance; and use has been increasing in young 

and older adult populations for the last 10 years; 

with at least 4 million individuals in the U.S. 

meeting diagnostic criteria for cannabis use 

disorder each year. Cannabis now refers to a 

diversity of products, with varying potencies (i.e., 

tetrahydrocannabinol (THC) concentrations), 

including some with little or no THC and in turn, 

minimal abuse liability (e.g., cannabidiol or CBD 

products). Federal and State laws don’t align, 

which creates multiple obstacles for researchers, 

patients, public health officials, and even those in 

the cannabis industry—who may be forced into 

unwieldy (e.g., cash only) business practices.  

 

This presentation highlights what we know and 

what we need to know about the adverse and 

potential therapeutic effects of cannabis, with the 

goal of identifying research needs and 

opportunities. While more evidence is needed on 

the long-term impact of cannabis, particularly in 

relation to causality and permanence, populations 

such as adolescents, pregnant women (and their 

offspring), and individuals with mental illness are 

likely to be the most at risk for adverse outcomes.  

The potential therapeutic uses of cannabis and its 

constituent cannabinoids are vastly understudied, 

despite their widespread use.  There are many 

reasons for this, including long-standing barriers 

to research, some of which are starting to be 

addressed.  As the evidence is being generated, it 

is crucial that we remain unbiased in our 

assessment of it, and forthright in our 

dissemination of information.  Only then can we 

advance the science and better inform policy and 

public health. 

 
Contact: Susan Weiss 

sweiss@nida.nih.gov 
 

KEYNOTE ADDRESS 
 

The Complex, Bi-Directional & Nuanced 

Associations between Cannabis and 

Posttraumatic Stress Disorder 

 

Marcel O. Bonn-Miller 

University of Pennsylvania Perelman School of 

Medicine 

 

Contact: Marcel Bonn-Miller 
mbonn@pennmedicine.upenn.edu 

 

Cannabis 
2018, 2 (1), 66-111 
© Author(s) 2018 
researchmj.org 
  

OPEN ACCESS 
 

Abstracts from the 2018 

Scientific Meeting of the 

Research Society on Marijuana 

July 28th-30th, 2018 

Fort Collins, Colorado 

 

 



Conference Abstracts  67 

 

POSTER PRESENTATIONS 
 

All poster presentations and symposia were peer-
reviewed by the 2018 Conference Program 
Committee of the Research Society on Marijuana 
(RSMj) (Chair: Bradley T. Conner, Colorado State 
University). All abstracts below were approved 
and voluntarily submitted for publication in 
Cannabis by the presenting or contact author. 

 

Association between Coping Strategies and 

Marijuana Use among Pre-Clinical Trauma-

Exposed Students 

 

Angeles Astorga, Nathan Kearns, Rachel 

Armour, Renee Cloutier, & Heidemarie 

Blumenthal 

University of North Texas 

 

Background: Marijuana is the most commonly 

used illicit substance in the United States (NIDA, 

2014) and has been linked to both short-term and 

long-term negative developmental, physical, and 

physiological outcomes (e.g., altered brain 

development, diminished life satisfaction; Volkow 

et al., 2014). Although epidemiological work has 

primarily emphasized sociodemographic 

characteristics (e.g., sex) associated with 

marijuana use (Pacek et al., 2015), a burgeoning 

literature also has focused on the impact of 

trauma exposure and posttraumatic stress. 

Indeed, extant work indicates that individuals 

exposed to trauma and/or that are evidencing 

posttraumatic stress disorder (PTSD) 

symptomatology are at increased risk for 

marijuana use and related problems (Boden et al., 

2013; Cougle et al., 2011). Expanding on this 

literature, recent work has focused on underlying 

mechanisms that may influence associations 

between trauma exposure and marijuana use, 

including specific coping motives and strategies 

that individuals use to manage posttraumatic 

stress (e.g., avoiding coping; Bonn-Miller et al., 

2007). However, much of this work has been 

conducted in clinical populations with PTSD 

and/or cannabis use disorder, providing little 

insight or guidance for prevention efforts in pre-

clinical populations following trauma. The 

current study begins to address this gap in the 

literature by evaluating 14 adaptive and 

maladaptive coping strategies that may be 

associated with increased or decreased likelihood 

of marijuana use in a heterogeneous sample of 

pre-clinical trauma survivors. Methods: The 

sample included 1,152 participants (M = 20.72, SD 

= 3.26; 73.7% female) drawn from a large 

university in a Southwestern state where 

marijuana use is not legal for medical or 

recreational purposes – that met the following 

eligibility criteria: (1) exposure to at least one 

DSM-5 Criteria A traumatic event (APA, 2013) 

according to the List of Events Checklist for DSM-

5 (LEC-5; Weather et al., 2013) and (2) their 

indexed trauma occurred more than 30 days prior 

to assessment. Participants completed the Brief 

COPE (Carver, 1997) to assess 14 coping 

strategies related to dealing with stress and a 

single-item, face-valid question to assess past-

month marijuana use. Responses were 

dichotomized to 0 (“No”) or 1 (“Yes”) based on any 

endorsement of past-month marijuana use. 

Forward stepwise binary logistic regression (Wald 

statistics) was conducted to isolate which coping 

strategies (or collection of variables) were most 

strongly associated with increased odds of 

marijuana use in the past-month. Results. 

Overall, five iterative steps identified the final 

model, which accounted for approximately 15% of 

the variance in past-month marijuana use. 

Expectedly, substance use coping has identified as 

the coping strategy most strongly associated with 

past-month marijuana use (OR = 1.65). Further, 

increased frequency of self-distraction (OR = 

0.92), denial (OR = 0.87), and religion (OR = 0.89) 

coping strategies were associated with decreased 

likelihood of using marijuana; conversely, 

frequency of active coping was positively 

associated with increased likelihood of using 

marijuana (OR = 1.13). Conclusion: The results of 

this study indicate that certain coping strategies 

are more strongly associated with marijuana use 

after trauma exposure, and may inform future 

prevention-oriented work. Coping strategy-

specific implications, study limitations, and 

future research will be discussed. 

 

Contact: Heidemarie Blumenthal 
Heidemarie.Blumenthal@unt.edu 

Supplemental Materials Here 

 



Conference Abstracts  68 

 
Does the Form of Marijuana Use Impact the 

Relation between Marijuana Use Motives and 

Daily Frequency? 

 

Chelsea L. Banks, Lillian J. Canfield, Alexa J. 

Pellegrino, Maryia M. Schneider, Megan H. 

Smith, Anastasia Robirds, Erica B. Peteja, Kerry 

D. Duck, Michael M. Phillips,  

& Kristina T. Phillips 

University of Northern Colorado 

 

Background: Past research has illustrated a 

relationship between a range of marijuana use 

motives (e.g., coping, social) and marijuana use 

frequency. With a growing market of new 

marijuana forms (e.g., concentrates, edibles), 

strains (hybrid, indica, sativa), and methods of 

ingestion (e.g., vaping, dabbing), it is important to 

gain a better understanding of how variations in 

marijuana might influence these established 

associations. Therefore, the goal of the current 

study is to examine whether marijuana form 

(flower vs. concentrates) moderates the relation 

between marijuana use motives and marijuana 

use frequency. Methods: Active college-student 

marijuana users (n = 79) were recruited and 

completed a baseline assessment that lasted 

approximately 90 minutes. Participants were 

trained on a signal-contingent ecological 

momentary assessment (EMA) protocol and 

responded to three random prompts each day for 

two weeks through a smartphone app. A baseline 

question examined the primary form of marijuana 

used (e.g., flower, concentrate, edibles). Five 

subscales from the Marijuana Motives Measure 

were used to assess specific reasons for using 

marijuana, including Coping, Social Anxiety, 

Enjoyment, Boredom, and Altered Perceptions. 

Marijuana use frequency was assessed using an 

EMA question assessing the average number of 

times marijuana was used each day over the two-

week period. Results: Participants were 60% 

female, 61% Caucasian, and averaged 20.35 (SD = 

3.65) years of age. At baseline, participants 

reported an average of 22.77 (SD = 7.3; 4 - 30) days 

of marijuana use in the last month and an average 

of 5.23 (SD = 1.81) days of use per week. EMA data 

showed that participants used, on average, once 

per day (M = 1.17; SD = .59). Over the two-week 

EMA period, participants reported using a mean 

of 16.32 (SD = 8.29) times over the two-week 

period. Most participants were using flower (n = 

60) or concentrates (n = 15) as their primary form 

of marijuana. Four participants reported 

primarily using edibles or other forms; these 

participants were excluded from further analyses. 

We found that form of marijuana was related to 

the EMA daily average of marijuana use. When 

examining the interactions between form and 

motives in five separate models, we found a 

moderated effect of form on daily EMA average 

only for enjoyment motives (p = .04). More 

specifically, as enjoyment motives increased, 

concentrate users reported greater marijuana use 

daily instances via EMA. In the relationship 

between altered perceptions and marijuana use 

frequency, marijuana form approached 

significance as a moderator (p = .08). Conclusions: 

Our findings suggest that the relationship 

between enjoyment motives and marijuana use 

frequency was particularly relevant for marijuana 

concentrate users. As more forms of marijuana 

become available, a more nuanced approach to 

understanding the relation between motives and 

frequency is needed. 

 

Contact: Kristina Phillips 
kristina.phillips@unco.edu 

 

Cannabis and Schizotypy: Revisiting an Old 

Problem 

 

Brittany E. Blanchard, Angela K. Stevens, & 

Andrew K. Littlefield 

Texas Tech University 

 

The link between cannabis and schizotypy (i.e., 

patterns of behavior which typify a genetic 

predisposition schizophrenia-spectrum pathology) 

is well documented, with some research indicating 

individuals with schizotypy are two-to-four more 

likely to report cannabis use (e.g., Cohen et al., 

2010). However, as previous work has indicated 

(e.g., Earleywine, 2006), several markers of 

schizotypy assessed on self-report measures (e.g., 

speech-related problems, such as forgetting what 

one is saying) may also represent common effects 

of cannabis use (e.g., working memory deficits). 

Given this, it is imperative to determine whether 

these items intended to assess schizotypal 

personality traits function the same across users 

and nonusers of cannabis. If these items exhibit 

differential item functioning, cannabis users may 

be erroneously classified as exhibiting schizotypy, 



Conference Abstracts  69 

 
which can yield deleterious consequences in 

clinical work and impede research progress. The 

current study aimed to determine whether 

speech-related items on the Schizotypy 

Personality Questionnaire-Brief Revised (SPQ-

BR) exhibit measurement invariance as a function 

of cannabis use status. 

Participants consisted of college students from a 

southwestern university (N = 602, 63% female, 

66% White, 27% LatinX) who completed a battery 

of self-report measures, including the SPQ-BR 

and a modified Daily Drinking Questionnaire to 

assess cannabis use. To ensure adequate power, 

lifetime use (n = 353) versus non-use (n = 249) was 

used for the current analyses. To assess 

differential item functioning, measurement 

invariance was tested using structural equation 

models in Mplus version 7.4 with SPQ-BR items 

modeled as ordinal indicators using a WLSMV 

estimator. The DIFFTEST was used to assess 

statistically significant changes in model fit. 

Preliminary results indicated the four-item Odd 

Speech subscale was not invariant across 

cannabis use status when comparing the 

configural (i.e., freely estimated) to scalar model 

(i.e., factor loadings and thresholds constrained 

across groups; DIFFTEST χ(15) = 36.96, p = .001). 

Next, partial scalar invariance was tested by 

freeing one item at a time. Partial measurement 

invariance could not be achieved when freeing any 

one item, suggesting all four items of the Odd 

Speech subscale of the SPQ-BR function 

differently across cannabis users versus nonusers 

(DIFFTEST χ(14) = 25.61- 44.48, p < .05). 

Although it is difficult to interpret differential 

item functioning, speculatively, it may be that 

cannabis users are reporting on cannabis-induced 

outcomes on items of the Odd Speech subscale 

(e.g., “I sometimes forget what I am trying to say”; 

“Do you tend to wander off the topic when having 

a conversation?”), rather than endorsing 

indications of schizotypy. Although directions on 

the SPQ-BR specifically ask individuals not to 

consider substance-induced experiences when 

answering, this may not be sufficient. Cognitive 

interviewing with cannabis users while taking the 

SPQ-BR may be beneficial in understanding why 

these items function differently across users and 

nonusers. Given these preliminary findings, 

establishing measurement invariance across 

cannabis use status for measures assessing 

schizotypy and similar constructs will be 

necessary going forward before any relation, 

causal or correlational, between cannabis use and 

schizophrenia-spectrum pathology can be 

elucidated. 

 

Contact: Brittany Blanchard 
Brittany.Blanchard@ttu.edu 

 

The Intersectionality of Race, Biological Sex, and 

Past 30-Day Marijuana Use among Trauma 

Exposed Emerging Adults 

 

Caitlyn N. Carey, Nathan T. Kearns, Quadreon 

K. Miller, Alisa C. Payne, Melissa N. Whitted, & 

Heidemarie Blumenthal  

University of North Texas 

 

Background: Marijuana is the most widely used 

illicit drug in the United States (SAMHSA, 2013), 

and is linked to various health, emotional, social, 

and legal consequences (Degenhardt & Hall, 

2006). An extensive literature has consistently 

identified trauma exposure as a risk factor for 

problematic marijuana use (Vlahov et al.,2002; 

Kilpatrick et al., 2000; Bremner et al., 1996). In 

an attempt to better understand substance use 

after trauma exposure, related research (e.g., 

alcohol) has identified both biological sex and 

race/ethnicity as risk markers (Danielson et al., 

2009). However, limited work has focused on these 

factors in relation to marijuana use in trauma-

exposed populations, and no work has evaluated 

the intersection of these sociodemographic 

markers (e.g., black males, white females). The 

current project begins to address this gap by 

assessing marijuana use frequency and the 

intersectionality of race/ethnicity and biological 

sex among trauma-exposed college students. 

Method: A sample of 805 trauma-exposed 

undergraduates (Mage= 20.68; 70.7% female) were 

recruited through an online participant pool. 

Participants were included in the study if they 

reported (a) a Diagnostic and Statistical Manual 

for Mental Disorders–Fifth Edition (DSM-5) 

Criteria A traumatic event (APA 2013), and (b) 

some level of PTSD symptom severity in the past 

30 days (i.e., not asymptomatic). Due to limited 

sample sizes in other reported racial/ethnic 

categories (e.g., Asian, Pacific Islander), only 

those primarily identifying as Black/African 

American, White/Caucasian, or Hispanic/Latinx 

were examined in the current analyses. A single-



Conference Abstracts  70 

 
item, face-valid question was used to assess past 

30-day marijuana use, with response options 

ranging from 1 (no use) to 7 (20 or more times). A 

3 X 2 factorial ANOVA was conducted to examine 

the differences in marijuana use frequency 

between groups based on race/ethnicity and 

biological sex (i.e., male and female). Results: 

Results indicated no main effect for race/ethnicity 

[F(2, 799) = 0.59, p = .592, η2= .00], but a 

significant main effect was found for biological sex 

[F(1, 799) = 11.46, p < .001, η2= .01], indicating 

that individuals identifying as biologically male 

endorsed greater past 30-day marijuana use 

frequency (M = 2.03, SD = 2.58) than female peers 

(M = 1.52, SD = 2.18). Further, the interaction was 

significant [F(2, 799) = 3.05, p = .048, ηp
2= .01]. 

Specifically, men who identified as Black/African 

American endorsed the most frequent marijuana 

use (M = 2.63, SD = 3.03), and females who 

identified as Black/African American endorsed 

the least frequent marijuana use (M = 1.09, SD = 

1.85). Discussion: These results indicate that, 

across all race/ethnicity groups, trauma-exposed 

males endorsed greater frequency of past-month 

marijuana use than females. However, for 

individuals who identified as Black/African 

American, that disparity appears to be more 

extreme, with Black/African American males 

endorsing the highest frequency of marijuana use 

and Black/African American females endorsing 

the lowest frequency of marijuana use. This 

finding highlights the importance of evaluating 

and meaningfully addressing the intersectionality 

of sociodemographic variables. Further, these 

findings emphasize the need for prevention 

intervention methods that are tailored towards 

identifying the at-risk intersecting identities of 

minority groups. 

 

Contact: Heidemarie Blumenthal 
Heidemarie.Blumenthal@unt.edu 

A Meta-Analytic Investigation of the 

Associations between Cannabis Use and 

Negative Consequences 

 

Gabriel J. Carrion-Gonzalez, Sarah L. Simons, & 

Matthew R. Pearson 

University of New Mexico 

 

In the current climate on policy change regarding 

cannabis (i.e., decriminalization, medicalization, 

and legalization), various stakeholders have 

strong interest in determining the associations 

between cannabis use and important outcomes, 

ranging from positive outcomes (e.g., medicinal 

benefits like decreased pain) to negative outcomes 

(e.g., psychosis). The present study sought to 

quantify the association between indicators of 

cannabis use and the experience of negative 

cannabis-related consequences. We are 

conducting an exhaustive search of the cannabis 

literature to identify studies that examine the 

use-consequence association, and are using meta-

analytic methods to quantify this association and 

examine possible moderators of this association 

(e.g., percentage of females, measure of 

consequences, indicator of cannabis use, 

population). To date, we have coded 19 articles 

and present these findings here. Given our 

expectation that we would find significant 

heterogeneity in effect sizes across studies, we 

conducted a random-effects meta-analysis. We 

found that cannabis use had a medium-sized 

effect on consequences, rw = .293, indicating that 

about 8.6% of the variance in consequences was 

attributed to the frequency or amount of cannabis 

use. In addition, we found significant 

heterogeneity of this effect, Q = 217.832, p < .001, 

I2 = 93.114. In fact, over 90% of various across 

studies was due to true heterogeneity rather than 

chance/sampling error. When we have coded all 

articles, we will examine moderators of this 

association to further probe this heterogeneity. 

This study is important in that it describes the 

average effect size of cannabis use on a broad 

measure of negative consequences. Given that 

91.4% of the variance in negative cannabis-

related consequences was not explained by any 

single indicator of cannabis use, pointing to the 

fact that additional factors need to be examined to 

explain the experience of negative consequences 



Conference Abstracts  71 

 
from cannabis use and that additional indicators 

of cannabis use may be needed. 

 
Contact: Matthew R. Pearson 

mateo.pearson@gmail.com 
Supplemental Materials Here 

 

Prevalence Rates of Marijuana, Alcohol, and 

Simultaneous Use and Associations with Alcohol-

Related Consequences in At-Risk Youth 

 

Daniel J. Delaney, Sara G. Balestrieri, Shayna S. 

Bassett, & L.A.R Stein 

University of Rhode Island 

 

Objective: Studies suggest that concurrent and 

simultaneous use of alcohol and marijuana (CAM 

and SAM, respectively) are associated with 

experiencing more alcohol-related consequences 

and higher rates of binge drinking, than alcohol 

use alone. However, rates of CAM and SAM, and 

possible associations with alcohol-related 

consequences have been little studied in samples 

of youth (e.g., ages 9-18) and at-risk samples.  This 

study aimed to assess prevalence rates of 

marijuana and alcohol use (including CAM and 

SAM use) and associations with alcohol-related 

consequences in a sample of at-risk youth. 

Methods: Youth ages 9-18 (N=505, M=14.36, 

SD=2.57) were recruited from community mental 

health agencies and a juvenile correctional facility 

in the Northeast United States (53% male; 25% 

Hispanic; 67% White, 18% Black, 6% American 

Indian/Alaskan Native; 4.7% Pacific Islander; 

2.2% Asian).  Past 30-day prevalence rates of 

alcohol and marijuana use were measured using 

Timeline Follow-Back (TLFB). Five distinct use 

groups were identified: Youth with (1) no 

substance use, (2) alcohol use only, (3) marijuana 

use only, (4) simultaneous use (operationalized as 

reporting same day use of alcohol and marijuana), 

and (5) concurrent use (reported using both 

substances but not same-day use). Number of 

binge drinking occasions as defined by NIAAA for 

youth was also measured using TLFB. Alcohol-

related consequences were measured by the Risks 

and Consequences Questionnaire.  Negative 

binomial regression and zero-inflated negative 

binomial regression (ZINB) analyses were 

conducted to compare alcohol-only versus SAM 

users on rates of binge drinking and alcohol-

related consequences. Results:  Overall, 64% of 

youth reported no substance use in the past 30 

days, 12.5% reported marijuana use only, 5.5% 

reported alcohol use only, 15% reported SAM, and 

3% reported CAM. Past 30-day prevalence across 

age was also calculated, and all youth ages 9-11 

reported no substance use. As expected, 

prevalence of substance use increased from ages 

12-18 (e.g., 0% of 12 year olds versus 51% of 18 

year olds reported SAM use). An equal proportion 

of males and females reported no substance use 

(50% vs. 50%), but a larger percentage of males as 

compared to females endorsed SAM use (22% vs. 

8%). Negative binomial regression and ZINB 

analyses indicated that youth reporting SAM use 

had significantly higher rates of binge drinking 

compared to those who endorsed alcohol use only 

(IRR = 2.74, p < .001), and significantly higher 

rates of drinking and driving (IRR = 10.94, p < 

.001). No other consequences (e.g., alcohol related 

injuries, unprotected sex) were found to be 

significant. Conclusions: Among the substance 

using youth, SAM and marijuana use only were 

most prevalent, especially among the older youth. 

These findings are alarming given the strong 

associations found with SAM use and drinking 

and driving and higher rates of binge drinking. 

More research is needed to determine causal 

relationships between SAM use and harmful 

drinking consequences.  

 

Contact: Daniel J. Delaney 
dannyjdelaney@uri.edu 

 

Impulsivity-like Traits, Use of Protective 

Behavioral Strategies, and Marijuana-Related 

Outcomes 

 

Chloe J. Espinosa, Adrian J. Bravo, Matthew R. 

Pearson, & Marijuana Outcomes Study Team 

University of New Mexico 

 

The UPPS-P model of impulsivity (Cyders et al., 

2007) posits that there are five unique 

impulsivity-like traits that result in impulsive 

behaviors: Negative Urgency, Premeditation (lack 

of), Perseverance (lack of), Sensation Seeking, 

Positive Urgency. Similar to findings in the 

alcohol field (Coskunpinar et al., 2013), negative 

urgency has been found to be robustly associated 

with marijuana-related consequences (Bravo et 

al., 2017). Although a 5-factor model of 

impulsivity-like traits has much empirical 



Conference Abstracts  72 

 
support, alternative factor structures have been 

examined, including a 3-factor model with 

positive and negative urgency loading onto a 

higher-order Urgency factor, and perseverance 

and premeditation loading onto a higher-order 

Deliberation factor. The present study examined 

whether use of marijuana protective behavioral 

strategies (PBS) mediates the effects of 

impulsivity on marijuana-related outcomes. 

Further, we examine whether the mediating role 

of PBS use is best understood in the context of 1-, 

3-, and 5-factor models of impulsivity-like traits. 

Using a sample of 6,584 students recruited from 

one of eight universities across the U.S., we 

compared a 5-factor, a higher-order 3-factor 

model, and a 1-factor model. The 5-factor 

structure of impulsivity-like traits fit best and 

unique effects support examining these facets 

independently in the prediction of marijuana-

related outcomes. Consistent with previous 

research with alcohol (Coskunpinar et al., 2013) 

and marijuana (Bravo et al., 2017), negative 

urgency was directly related to negative 

marijuana-related consequences. However, the 

effect of negative urgency on both typical 

marijuana quantity and negative marijuana 

consequences was partially accounted for by lower 

use of marijuana PBS. Personality-targeted 

interventions have demonstrated promise among 

adolescents in particular (Conrod, 2016), but 

perhaps PBS use could be one promising 

intervention target for these interventions 

individuals with elevated levels of negative 

urgency and/or sensation seeking. The fact that 

negative urgency demonstrated a direct effect 

even after controlling for PBS use suggests that 

additional factors need to be examined to explain 

why individuals with higher negative urgency 

experience more harms from their marijuana use. 

 

Contact: Matthew R. Pearson 
mateo.pearson@gmail.com 

Supplemental Materials Here 

 

Individual Difference Factors Influencing 

Cannabis Non-Use Motives 

 

Reagan E. Fitzke, Samuel R. Davis, Mark A. 

Prince, Randall C. Swaim, & Linda Stanley 

Colorado State University 

 

Introduction: Research on cannabis use motives is 

well established. However, there is limited 

research exploring cannabis non-use motives (i.e., 

motivations to not use cannabis). Previous 

research on non-drinking motives shows evidence 

that greater non-use motives predict lower 

lifetime rates of alcohol use. Additionally, 

research on American Indian (AI) populations 

specifically shows factors such as ethnic identity 

can protect against substance use. AI populations 

have a higher prevalence of substance use 

compared to non-AI populations. These studies 

suggest that exploring non-use motives in relation 

to cannabis can assist in understanding cannabis 

use decisions. Methods: This study investigates 

cannabis non-use motives in a national sample of 

individuals that report no lifetime cannabis use 

and live on or near reservations in the United 

States (N = 1867). Approximately 22% of the 

present sample identified as White and 73% 

identified as AI. Additionally, the relation 

between non-use motives and ethnic identity, 

ethnic pride, perceived discrimination due to 

ethnicity, parental awareness, and parental 

care/concern was explored. Twelve non-use 

motives were included in the analysis. All 

analyses were conducted in MPlus version 8 and 

were specified as complex to account for clustering 

by school. First, an exploratory factor analysis 

(EFA) was run to assess the factor structure of the 

non-use motives scale. Second, structural 

equation modeling (SEM) was used to assess the 

relation between cannabis non-use motives and 

several demographic and other factors. Results: 

Results of the EFA indicated that a one-factor 

solution best fit the data, with all goodness-of-fit 

indices falling within acceptable ranges. 

Structural equation modeling results showed AI 

ethnic identity, ethnic pride, parental awareness, 

and parental care/concern were positively related 

to non-use motives. In contrast, perceived 

discrimination and White ethnic identity were not 

related to non-use motives. Conclusion: AI ethnic 

identity was positively associated with non-use 

motives, but White ethnic identity had no 



Conference Abstracts  73 

 
association with non-use motives. AI populations 

have previously been regarded to be particularly 

at-risk for substance use and misuse. The positive 

relation between AI ethnic identity and non-use 

motives suggests that AI cultural identity may be 

protective against marijuana use. Parental care, 

awareness of cannabis use, and concern about 

cannabis use were also associated with a greater 

number of reported non-use motives. These 

findings indicate potential protective factors that 

all contribute to an individual’s choice to 

ultimately not engage in cannabis use. 

Furthermore, these analyses suggest non-use 

motives are strongly related to protective factors 

in predicting use patterns. Incorporation of non-

use motives has the potential to direct future 

prevention strategies, primarily for at-risk and 

underrepresented populations. 

 

Contact: Reagan Fitzke 

refitzke@rams.colostate.edu 
 

The Influence of Mentor Relationship Quality on 

Marijuana Use during the Campus Connections 

Mentoring Program 

 

Gereon Fredrickson, Shelley Haddock, & 

Kimberly Henry  

Colorado State University 

 

Background: Adolescent substance abuse is a 

serious public health concern. Adolescents who 

abuse substances, particularly those who begin 

during early adolescence, are more likely to 

engage in risky sexual behavior, commit crimes, 

drop out of school, and develop a substance use 

disorder. Additionally, adolescent marijuana use 

is associated with adverse consequences in later 

adulthood, including short-term memory 

impairments, altered brain development, and 

addiction. Identification of programs, practices, 

and policies that have the capacity to prevent 

adolescent substance abuse are needed. 

Mentoring has been recommended as one such 

initiative; however, not all mentoring programs 

are successful. Nearly 50% of all mentoring 

relationships end prematurely and many do not 

produce improvements in the adolescent’s 

developmental trajectory. We propose that 

positive outcomes of mentoring will be heightened 

when the adolescent and mentor develop a high-

quality relationship. Methods: The data for this 

study comes from a mentoring program called 

Campus Connections (CC). We hypothesized that 

adolescents who develop a high-quality 

relationship with their mentor during the CC 

program will be less likely to engage in marijuana 

use compared to youth who develop a low-quality 

relationship with their mentor. Mentees self-

reported their perception of the Mentor 

Relationship Quality (MRQ) using the Cavell et 

al. (2009) Mentor Alliance Scale (α = 0.86) at the 

end of the mentoring program. Mentees also self-

reported marijuana use at baseline and at the end 

of the program. Marijuana use appeared to be 

over-dispersed and was specified as a count 

variable. A negative binomial regression model 

was used to estimate the association between 

MRQ and marijuana use at the end of the CC 

program, controlling for baseline use and a set of 

confounders (gender, age, ethnicity, socioeconomic 

status, and baseline marijuana use prior to CC 

program start). Results: A negative binomial 

regression model was used to calculate the 

results. Adjusting for baseline marijuana use and 

relevant confounders, the expected days of 

marijuana use is 67% lower for each one unit 

increase in MRQ. Conclusions: Consistent with 

prior research, we find that mentoring 

relationship quality is associated with better 

outcomes for adolescents. Although having a 

mentor may be beneficial, having a strong alliance 

and relationship quality with a mentor appears to 

be more favorable. Methods to enhance 

relationship quality are needed, and effective 

strategies may substantially enhance the 

effectiveness of mentoring interventions, 

including reductions in substance use for 

participating adolescents. 

 

Contact: Gereon Fredrickson 
Gereon.Fredrickson@colostate.edu 

 

Meaning in Life and Sensation Seeking on 

Marijuana Use 

 

Cara L. Fresquez, Alex Tyskiewicz, Theodore J. 

Fetterling, & Mark A. Prince 

Colorado State University 

 

Presence of meaning in life has been identified as 

a protective factor decreasing the likelihood of 

marijuana use. Meaning in life is defined as the 

sense and significance regarding the nature of 



Conference Abstracts  74 

 
one’s being and existence. Meaning in life is 

composed of two constructs, search for meaning 

and presence of meaning in life. Sensation seeking 

is defined as the desire for novel experiences and 

the willingness to take risks for these experiences. 

Past studies identify experience seeking as a 

predictor of initial drug use, but not of drug 

maintenance. However, risk seeking has been 

shown to predict drug maintenance and not 

initiation. Previous research on meaning in life, 

sensation seeking, and substance use have 

demonstrated that religious meaning in life is a 

protective factor for marijuana use regardless of 

an individual’s level of risk seeking. The relations 

among existential (meaning in life not associated 

with religion) and personality factors (sensation 

seeking) and marijuana use has not yet been 

explored. The current study examined the relation 

among meaning in life, sensation seeking, 

marijuana use, using the Meaning of Life, 

Sensation Seeking Personality Type, and Risky 

Behavior Inventory scales. We hypothesized 

meaning of life would moderate the relation 

between sensation seeking and past 30-day 

marijuana use. Further, we hypothesized that 

individuals who had low search for meaning in life 

and high experience seeking, would use 

marijuana more frequently than individuals with 

high search for meaning in life and low experience 

seeking. The current study is a secondary data 

analysis of data that was collected in fall 2016 as 

a part of a larger study on health risk behaviors. 

Study hypotheses were tested using negative 

binomial regression due to past 30-day use being 

a highly skewed count variable. Marijuana use 

was determined by an individual’s reported 

frequency of use over the previous 30 days. We 

found a significant interaction between search for 

meaning in life and experience seeking on past 30-

day marijuana use. At low levels of the moderator, 

experience seeking negatively predicted past 30-

day marijuana use, whereas high levels of the 

moderator were associated with experience 

seeking predicting increased use. These results 

suggest that search for meaning in life moderates 

the relation between experience seeking and 

marijuana use. Contrary to previous studies, 

which have identified experience seeking being a 

protective factor against drug use, the present 

study implicates it as a risk factor when 

moderated by search for meaning in life. Thus, if 

a client present high in experience seeking and is 

searching for meaning in life, then he or she may 

be at risk for marijuana use. 

 

Contact: Cara L. Fresquez 
Cara.fresquez@colostate.edu 

Supplemental Materials Here 

 

Repetitive Transcranial Magnetic Stimulation of 

the Medial Prefrontal Cortex for Reduction of 

Polysubstance Use Craving: A Pilot Study 

 

Jordan A. Gette, Gavin K. Ueland, Andrew K. 

Littlefield, & Yi-Yuan Tang  

Texas Tech University 

 

Polysubstance use has been associated with 

myriad deleterious consequences in college 

students including increased risk for substance 

use disorders, poor academic performance, and 

risky sex behaviors. Recent studies among college 

students suggest that over 50% of past-month 

substance users have engaged in use of two or 

more of the following substance: alcohol, tobacco, 

and cannabis (ATC). Repetitive transcranial 

magnetic stimulation (rTMS) has been 

implemented as an alternative treatment to 

psychotherapy and has shown promise as a brief, 

noninvasive intervention for substance craving. 

The medial prefrontal cortex (mPFC) has been 

implicated in cognitive processes such as 

impulsivity and drug related cue responses 

making the mPFC a brain region of interest for 

the application of rTMS. Thus, the aim of the 

current study was to determine whether rTMS on 

the mPFC could be utilized to reduce 

polysubstance craving over a brief four day 

intervention. Participants (N = 14, 5 females, Mage 

= 21.5, 50% white) completed demographics, the 

Penn Alcohol Craving Scale (PACS), the PACS 

adjusted for cannabis, and the PACS adjusted for 

tobacco. Participants completed TMS stimulation 

sessions of either two or three cycles on four 

consecutive days. In following with previous 

literature, rTMS was set to 50Hz with 1800 pulses 

per cycle at 110% of the individual’s resting motor 

threshold. Dependent means t-tests were 

conducted to assess for decreases in ATC craving 

from pre- to post-rTMS sessions.  Results indicate 

a significant decrease in cannabis craving (p < .01) 

amongst cannabis users (N = 12). Decreases in 

alcohol craving amongst users (N = 12) 

approached significance (p < .10). Removal of an 



Conference Abstracts  75 

 
outlier resulted in nominally significant decreases 

in alcohol craving (p = .01). Results did not 

indicate significant decreases in tobacco craving 

(p = .43) amongst tobacco users (N = 8). Although 

replication is necessary, these results show 

promise in using rTMS as a means of reducing 

alcohol and cannabis craving. Lack of significant 

decreases in tobacco craving may suggest that 

tobacco craving evidences differential patterns of 

cognitive control and impulsivity as compared to 

alcohol or cannabis. Future studies should 

compare findings to a control group of ATC users. 

Additionally, comparison of decreases in ATC 

craving between those receiving active rTMS and 

those in a sham rTMS condition would assist in 

dismantling if it is rTMS or other factors (e.g. 

expectancy effects) driving the decreases in 

craving. Future studies would also benefit from 

assessing how long decreases in craving are 

maintained and how decreases in craving relate to 

decrease in ATC use. 

Contact: Jordan A. Gette 
Jordan.Gette@ttu.edu 

Supplemental Materials Here 

 
Problems Associated with Adolescents’ and 

Young Adults’ Marijuana and Alcohol Co-Use 

 

Joel W.  Grube & Sharon Lipperman-Kreda 

Prevention Research Center, Pacific Institute for 

Research and Evaluation 

 

Although the adverse effects of marijuana use are 

debated, its use during adolescence and young 

adulthood has been associated with a variety of 

negative consequences, including increased risk of 

accidental injury, fatal motor vehicle crashes, 

respiratory illness, psychotic disorders, 

impairment of cognitive functioning and brain 

development, low educational attainment, drug 

dependence, and involvement in crime.  These 

adverse outcomes may be exacerbated by the co-

use or simultaneous use of marijuana and other 

substances, including alcohol.  We investigated 

the associations of co-use of marijuana and alcohol 

with experiences of negative outcomes using 

survey data obtained from 706 adolescents and 

young adults aged 16-21 years (mean age 18.3 

years; 49% female) who lived in 24 midsized 

California cities.  Based on past year self-reported 

alcohol and marijuana use, we identified three 

groups: (a) non-users, (b) alcohol only users, and 

(c) co-users of alcohol and marijuana.  

Interestingly, there were few marijuana only 

users in the sample (3%). Marijuana only users 

were thus excluded from the analyses.  Of the 

study sample, 30% reported alcohol use only and 

40% reported past year marijuana and alcohol co-

use.  Multi-level mixed effects logistic regression 

analyses were used to predict whether 

participants who reported alcohol only use or co-

use of alcohol and marijuana had experienced 

each of eight problems in the past year (got into 

argument/fight; got hurt/injured; had unprotected 

sex, had sex with a stranger; had an unwanted 

sexual experience, rode with an intoxicated 

driver, and drove while intoxicated). We 

controlled for individual-level demographics (age, 

sex, race/ethnicity) and city-level variables (% 

minors, % White, % Hispanic, median household 

income).  Compared with non-users, young people 

who used alcohol only were significantly more 

likely to report unprotected sex (OR = 1.85; 95% 

CI = 1.14, 3.02).  They did not differ significantly 

from non-users in the likelihood of reporting any 

of the other problems.  In contrast, those who 

reported alcohol and marijuana co-use were more 

likely than non-users to report getting into a fight 

(OR = 1.53; 95% CI = 1.04, 2.26), unprotected sex 

(OR = 4.98; 95% CI = 3.16, 7.85), sex with a 

stranger (OR = 8.25; 95% CI = 3.83, 17.73), 

unwanted sex (OR = 3.05; 95% CI =1.23, 7.52), 

riding with a drinking driver (OR = 3.57; 95% CI 

= 2.04, 6.28), and driving after drinking (OR = 

10.44; 95% CI = 3.68, 29.62).  Overall, the findings 

suggest that young people who use both alcohol 

and marijuana are at greater risk for problem 

outcomes than are those who use alcohol only.  It 

is unclear whether the associations are causal 

(i.e., co-use leads to greater problems) or the result 

of other common factors (i.e., impulsivity) that 

predispose young people to engage in risky 

behaviors more generally.  Nonetheless, 

prevention efforts should target the co-use of 

marijuana and alcohol. 

 

Contact: Joel W. Grube 
grube@prev.org 

Supplemental Materials Here 

 



Conference Abstracts  76 

 
Are We Asking the Right Questions: Results of a 

Pilot Study on Marijuana Use Patterns and 

Terminology? 

 

Kerri Hayes, Helene R. White, Sheila Vandal, & 

Kristina M. Jackson 

Brown University & Rutgers University 

 

With the quickly changing landscape of marijuana 

products, it is difficult to accurately capture 

marijuana use. Frequency can be relatively easily 

ascertained, but quantity is much more 

challenging; a standard in the field is yet to be 

widely accepted. Without a common terminology 

and standard of measurement between user and 

researcher, either scant data is collected or 

tremendous detail is gathered that is difficult to 

distill into usable measures. In an effort to 

balance participant burden and reporting 

accuracy, it is important to better understand 

common terminology and patterns of use. An 

online screening survey was sent to a nationwide 

Qualtrics panel. Inclusion criteria for the full 

survey were age 18-24, weekly alcohol and 

marijuana use in the past-month. 811 individuals 

completed the screening survey, 207 met inclusion 

criteria with an average age of 21.9 years. 123 

completed an anonymous online survey. The 

sample used marijuana an average of 20 days of 

the last 30 (range 2-30). To understand if patterns 

of marijuana use exist “When you use 

marijuana… how often would you say that your 

use follows a pattern or routine” was asked; 97.6% 

reported that their use at least sometimes follows 

a pattern; 30.1% reported their use was always 

patterned. Participants were queried on several 

factors that contribute to patterns of use; they 

reported always using with the same people 

(39%), using the same mode of delivery (35.8%), in 

the same place (33.3%) and the same time of day 

(25.2%). Participants also described their pattern 

in their own words. A content analysis of 

responses revealed mention of the following 

factors in descending order of prevalence: time of 

day, method, who with, quantity, associated 

activities, and motivations. Participants were also 

asked to report in detail on their use in three 

different formats: hourly, block of day 

(day/evening/night), and with an open-ended 

item. In terms of reported quantity here 

(measured in grams), the hourly report generated 

the highest daily quantity with an average of 3.2 

grams (range 0.1g-32g). The block format 

produced a daily average of 2.0g (range 0.1-12). 

When comparing the daily report to the block 

report 24.3% reported an equal quantity (defined 

by less than 0.1g disparity) with the 2 methods, 

while 45.5% reported greater quantity with the 

hourly, and 30.0% reported greater quantity with 

the block. Those that had an equal report across 

the two methods reported using lower quantities 

(range 0.1-4, mean=0.9g) compared to those that 

had disparities between methods (mean range 

1.6-5.4). The open-ended responses were 

examined for content–terms bowls and joints were 

most frequently used, followed by grams, hits and 

blunts. Findings indicated that the terminology 

used around an important factor of use such as 

quantity needs to be thoroughly considered. 

Additionally, the measurement terminology 

researchers utilize may not be how users conceive 

of their use, and different timeframes appear to 

elicit different responses, especially for heavier 

users. In this frequently-using sample, patterned 

use is common. With this in mind, future research 

may be able to reduce participant burden by 

continuing to understand these nuances. 

 
Contact: Kerri Hayes 

kerri_hayes@brown.edu 
 

Development of an mHealth Diary that 

Examines the Effects of Cannabis Use on 

Chronic Neuropathic Pain in Individuals Living 

with HIV  

 

Brook L. Henry, Sarah Henry, Elizabeth 

Quintana, & David J. Moore 

University of California, San Diego 

 

Background: HIV sensory neuropathy (HIV-SN), 

typically characterized by sensory deficits and 

pain in the feet or hands, affects more than 50% 

of people living with HIV (PLWH). HIV-related 

neuropathic pain (NP) is often described as a 

chronic “stabbing”, “burning” or “aching” 

sensation that is a frequent source of disability in 

this population. Traditional analgesic treatments 

are ineffective for many PLWH with NP, but 

recent data indicate that administration of 

smoked or vaporized cannabis can provide pain 

relief. However, previous clinical trials are limited 

by several factors, including short duration 

(typically 1-2 weeks), a focus on low doses of delta-



Conference Abstracts  77 

 
9-tetrahydrocannabinol (THC), but not 

cannabidiol (CBD), and a lack of data about 

various methods of cannabis administration, such 

as edible products or dabbing. There is thus a 

significant need to close the gap between existing 

clinical data and assessing the effects of “real-life” 

cannabis use on NP. This abstract describes 

qualitative data from focus groups designed to 

obtain feedback about the effects of self-

administered cannabis on pain and the utility of 

mHealth methods (collection of health data via 

cell phones or other mobile devices) to track this 

information for subsequent intervention studies. 

Methods: We recruited outpatients (n = 20) 

experiencing chronic pain (neuropathy, arthritis, 

or back pain) and self-reported cannabis use 

during the prior three months, including both 

PLWH and HIV-uninfected participants, to obtain 

diverse feedback about cannabis use in the 

context of chronic pain. Two focus groups, with 8 

and 12 individuals, respectively, were conducted 

at the HIV Neurobehavioral Research Program in 

San Diego to discuss how to report the effects of 

cannabis exposure on daily pain using a mobile 

interface on their phones. Results: Two 

investigators independently coded focus group 

transcripts using the MAXQDA software. The 

kappa value for interrater reliability was 0.85, 

indicating a high degree of rater agreement. Data 

analysis revealed three primary themes: 1) 

effective methods for reporting pain; 2) effective 

methods for reporting cannabis use; 3) the impact 

of cannabis on pain. Participants expressed 

preferences for both qualitative (impact on 

everyday functioning) and quantitative (0-10 pain 

scale with anchors at 0 and 5) measures of pain, 

endorsed the feasibility of reporting cannabis 

content (THC and CBD), dose, and route of 

administration via a once-a-day text-message link 

to an online survey, reported that cannabis use 

allowed them to reduce their exposure to narcotic 

analgesics (opioids), and indicated that cannabis-

reduced pain was associated with improvement in 

affect and everyday function. Conclusions: Focus 

group feedback enabled the development of a 

mobile smartphone platform designed to record 

daily pain and cannabis effects on pain, including 

technical design (text links to survey), pain 

question content (inclusive of affect/everyday 

functioning), and a detailed assessment of 

cannabis use. These data were used to support an 

ongoing study that evaluates the consequences of 

THC and CBD on HIV-related NP, including a 6-

month observational electronic diary (Individual 

Monitoring of Pain and Cannabis Taken or 

IMPACT). 

 

Contact: Brooks L. Henry 
blhenry@mail.ucsd.edu 

 

Marijuana Policy Research Tool 

 

Mike Hilton 

National Institute on Alcohol Abuse and 

Alcoholism 

 

The purpose of the poster is to spread the word 

about a research tool that can be used to study the 

effects of marijuana legalization policy.  NIAAA's 

Alcohol Policy Information System (APIS) now 

contains an authoritative and systematic coding 

of policies to legalize the recreational use of 

cannabis.  The original APIS coverage is being 

expanded to cover about double the number of 

policy features covered by its earlier coverage.  

Members of the Research Society on Marijuana 

are positioned to take the lead in studying the 

effects of legalized cannabis, and hence need to 

know about the availability of this tool.  Also 

depicted in the poster is an example of the fact 

that policy has been changing quite rapidly in the 

recreational cannabis field compared to policies 

affecting alcohol use.  Finally, calls for grant 

applications are highlighted.  These NIH Funding 

Opportunity Announcements are sponsored by 

both NIAAA and NIDA. 

 

Contact: Mike Hilton 
mhilton@willco.niaaa.nih.gov 

 

Incidental Effects of Naltrexone Maintenance on 

Cannabis Use in the COMBINE Study 

 

Jon M. Houck 

University of New Mexico 

 

Reciprocal functional interactions between the 

endogenous cannabinoid and opioid systems are 

well known. THC can reduce pain perception 

(Vivian et al., 1998), and rhesus monkeys 

administered THC reduced their self-

administration of heroin (Li et al., 2015). 

Consistent with this, recent work in humans 

found that joint administration of cannabis and 



Conference Abstracts  78 

 
low doses of oxycodone produce analgesia similar 

to that of larger doses of oxycodone alone (Cooper 

et al., 2018). Naltrexone is an opioid antagonist 

that also acts on cannabinoid receptors. In 

nonhuman primates, acute naltrexone reduces 

THC self-administration (Justinova et al., 2004). 

In humans, the effects of naltrexone on cannabis 

use appear to vary with the experience of the 

individual. In non-cannabis-smokers, acute 

naltrexone potentiates the effects of low cannabis 

doses (Haney, 2007), while in daily cannabis 

smokers, maintenance doses of naltrexone are 

associated with decreased cannabis self-

administration (Haney et al., 2015). The 

COMBINE Study was a large multi-site 

medication (naltrexone, acamprosate) clinical 

trial for alcohol use disorder that did not exclude 

participants on the basis of cannabis dependence 

(Anton et al., 2006). COMBINE participants who 

used cannabis generally had higher alcohol use at 

the end of treatment did than those who did not 

use cannabis (Subbaraman et al., 2016). However, 

cannabis users randomized to receive naltrexone 

and the Combined Behavioral Intervention (CBI) 

reported end-of-treatment drinking equivalent to 

those randomized to receive naltrexone who did 

not use cannabis. The goal of the present study 

was to text the incidental effects of naltrexone 

maintenance for alcohol use disorder on self-

reported cannabis use in the COMBINE Study. 

Data on cannabis use days were obtained from the 

Structured Clinical Interview for DSM-IV (SCID-

I). Naltrexone dose estimates in COMBINE were 

generated by inspection of used blister packs. 206 

participants (19.9% female) reported cannabis use 

during the 16-week treatment period. Cannabis 

use was significantly lower in the naltrexone-

taking groups than in other groups (X2(1)=3.865, 

p=.049). In a follow-up analysis comparing 85 

participants assigned to receive either naltrexone 

or placebo, negative binomial regression 

controlling for baseline use indicated that 

participants randomized to receive naltrexone 

had significantly fewer cannabis use days during 

treatment than did those assigned to receive 

placebo (b=-1.969, p=.043). There was not a 

significant effect of CBI (b=-.563, p=.078). At the 

12-month follow-up, there were no significant 

effects on cannabis use days. Results of the 

present analysis suggest that in the COMBINE 

study, naltrexone had incidental effects on 

participant cannabis use. That is, although 

COMBINE was not designed to influence 

cannabis use, participants who used cannabis and 

took naltrexone had fewer cannabis use days 

during the treatment period than did those who 

did not take naltrexone, and participants who 

took the prescribed dose of naltrexone had fewer 

cannabis use days than those who did not. After 

participants stopped taking naltrexone, cannabis 

use did not differ between groups. Although 

limited data on cannabis use reduce confidence in 

the study findings, these results suggest that 

naltrexone should be examined further as a 

treatment for cannabis use disorder, particularly 

in populations that also use alcohol. 

 
Contact: Jon M. Houck 

jhouck@unm.edu 
 

Greater Past Month and Lifetime Marijuana Use 

Are Associated with Poorer Cognitive Flexibility 

in Young Adult College Students 

 

Sarah Lahanas & Anita Cservenka 

Oregon State University 

 

Objective: Marijuana (MJ) is the most commonly 

used illicit substance in the United States. The 

use of MJ is particularly common among college 

and university students with 4.9% of them 

reporting daily or near-daily MJ use. Substantial 

evidence suggests frequent MJ users exhibit 

deficits in memory and attention, beyond the 

acutely intoxicated state. However, less is known 

about how recent and lifetime MJ use affect 

components of executive functioning, such as 

cognitive flexibility, which is critical to set-

shifting and overcoming habitual actions. Due to 

the continued maturation of the prefrontal cortex 

during young adulthood, and the importance of 

this region for executive functioning, it is 

necessary to examine frequent MJ users in this 

age range for aberrations in cognitive flexibility 

related to MJ use. This is particularly important 

to investigate in young adult college students, as 

MJ use has been associated with adverse 

academic outcomes. The current study examined 

the association between quantity of MJ use and 

cognitive flexibility in frequent MJ users, ages 18-

22, who were recruited as part of a larger study on 

MJ use and executive functioning. Method: 

Seventeen heavy MJ users (mean age: 20.12 ± .27; 

9 male, 8 female) reporting 5 or more episodes of 



Conference Abstracts  79 

 
MJ use per week over the past year were included 

in the current study. Participants were instructed 

to remain abstinent from MJ use for 12 hours 

prior to their study visit to insure they were not 

acutely intoxicated during testing. The 30-day 

Timeline Followback (TLFB) was used to assess 

recent MJ use and participants were asked to 

estimate lifetime MJ use occasions. Participants 

completed the Modified Wisconsin Card Sorting 

Test (M-WCST), a measure of cognitive flexibility. 

Results: Greater past 30-day MJ use was 

associated with poorer performance on the M-

WCST. Past 30-day use was negatively correlated 

with T-scores for the following variables: 

categories correct (r = -.59, p = .01), perseverative 

errors (r = -.67, p = .003), percent perseverative 

errors (r = -.62, p = .008), and executive function 

composite (r = -.67, p = .003). Greater lifetime use 

was negatively correlated with perseverative 

errors (r = -.59, p = .01), percent perseverative 

errors (r = -.65, p = .005), and executive function 

composite (r = -.49, p = 0.046) T-scores. When 

controlling for recent use, lifetime use 

significantly predicted percent perseverative 

errors (F(2,14) = 8.60, p = .004, ∆R2 = .17, β = -.46, 

t = -2.31, p = .04), above and beyond recent use. 

Conclusion: These findings suggest a dose-

dependent relationship between MJ use and 

cognitive flexibility in young adult college 

students who are frequent users of MJ. Cognitive 

inflexibility may result in the maintenance of 

craving and preoccupation with MJ use due to an 

impaired ability to consider alternatives to and 

consequences of MJ use. Further research is 

needed to understand whether these deficits are 

the result of neurotoxic effects of MJ or pre-

existing impairments in cognitive flexibility prior 

to initiation of use, and whether these deficits 

persist after longer periods of abstinence. 

 

Contact: Anita Cservenka 
anita.cservenka@oregonstate.edu 

Supplemental Materials Here 

 

Mental Health Symptoms and Marijuana 

Consequences: Protective Behavioral Strategies 

as a Mediator 

 

Jenae B. Lynch, Andrew P. Weinstein, Matthew 

R. Pearson, & Marijuana Outcomes Study Team 

University of New Mexico 

 

Individuals that have elevated mental health 

symptoms are more likely to suffer from 

consequences of marijuana use than individuals 

without such mental health symptoms. Several 

mechanisms could be responsible for this 

relationship, including the individuals’ motives 

for use, their expectancies, or their use of 

protective behavioral strategies (PBS). PBS are 

behavioral strategies that can be used to reduce 

the negative consequences from substance abuse, 

such as only using during a certain time or 

limiting the amount used. PBS use was examined 

as a possible mediator for the effect of mental 

health symptoms on marijuana consequences. 

From a larger sample of 6,584 college students 

recruited from 8 universities in the U.S., we 

examined a subsample of marijuana users who 

completed the DSM-5 Level 1 Cross-Cutting 

Symptom Measure (ns range from 909 – 922 

across models), which assesses 13 psychiatric 

domains. In separate models (i.e., each mental 

health symptom), PBS use significantly mediated 

the association between each mental health 

domain and marijuana consequences (14.29% - 

43.90% of the total effect) except for anger. In a 

combined model with all 13 mental health 

symptoms, there were significant indirect effects 

through PBS use for memory, dissociation, and 

anger, though the effect through anger was 

opposite of its total effect. Lastly, we found PBS 

use partially mediated the effect of a latent factor 

of all 13 mental health symptoms on marijuana 

consequences. PBS use was found to be a partial 

mediator for the amount of marijuana 

consequences experienced by individuals that 

have elevated mental health symptoms, 

suggesting that PBS use may be a promising 

intervention target for these individuals. 

Research is needed to determine why these 

individuals use fewer PBS (e.g., lower perceived 

harm of marijuana use, lower capability of 

planning their use, etc.) in order to determine 

optimal intervention strategies. Importantly, PBS 

use only accounted for some of the effects of 



Conference Abstracts  80 

 
mental health symptoms on consequences, 

indicating that other possible mediators should be 

examined 

 

Contact: Matthew R. Pearson 
mateo.pearson@gmail.com 

Supplemental Materials Here 

 

Trauma Cues Elicit Greater Conditioned Craving 

Responses than Do Cannabis Cues in Cannabis 

Users with Trauma Histories 

 

Ioan T. Mahu, Pablo Romero-Sanchiz, Tristan 

Park, Jennifer Swansburg, Joshua Salmon, 

Mohammed Al-Hamdani, Daniel Rasic, Sean 

Barrett, & Sherry H. Stewart 

Dalhousie University 

 

Conditioned craving is said to develop in drug 

users in response to drug cue exposure. Through 

repeated pairing of drug cues with desired 

outcomes of drug use, strong associations are 

formed in memory, such that presentation of the 

drug cue alone is sufficient to elicit a conditioned 

craving response. Through a similar learning 

process, exposure to trauma cues is theorized to 

elicit conditioned drug craving in individuals with 

trauma histories, following the repeated use of the 

drug for the relief of trauma-related symptoms. 

Little is known regarding whether this type of 

trauma cue-induced conditioned craving occurs in 

traumatized cannabis users. We examined the 

link between trauma cue exposure and cannabis 

cravings using a cue-reactivity paradigm in 

cannabis users with trauma histories. 

Participants are exposed to three types of 

personalized audio and visual scripts (trauma, 

cannabis, neutral) in random order. Following 

exposure to each cue, participants rate their 

cannabis cravings (Marijuana Craving 

Questionnaire – Short Form (MCQ-SF); 

Heishman et al., 2009) and mood (Positive and 

Negative Affect Schedule (PANAS); Watson, 

Clark, & Tellegen, 1988; Mood-VAS scale; 

Lundahl & Greenwald, 2016). Although data 

collection is ongoing, we present preliminary 

results on our first eight participants (mean age = 

42.63, 87.5% male, 50% military, 37.5% met 

criteria for PTSD). We conducted one-way (cue 

type) repeated measures ANOVAs (with 

Greenhouse-Geisser adjustment) separately for 

each mood and cannabis craving scale, followed by 

LSD post-hoc tests. On the PANAS, the trauma 

cue induced significantly more negative mood 

than both cannabis and neutral cues, F(1.684, 

11.787 ) = 7.963, p = .008, although there was no 

difference between cue types for positive mood, 

F(1.106, 7.744) = .457, p = .538. Participants also 

reported feeling significantly more “anxious” 

(F(1.382, 9.674) = 11.580, p = .005) and “down” 

(F(1.905, 13.334) = 10.848, p = .002) on the Mood-

VAS scale after trauma vs. both cannabis and 

neutral cue exposure, supporting the validity of 

the trauma cue exposure in inducing negative 

mood. On the MCQ-SF subscales, trauma cues 

were significantly more likely to elicit cannabis 

cravings for relief (i.e. “emotionality”; F(1.878, 

13.145) = 6.522, p = .012), induce an inability to 

control cannabis use (i.e. “compulsivity”; F(1.538, 

10.768) = 7.808, p = .011), create an anticipation 

of positive outcomes from using cannabis (i.e. 

“expectancy”; F(1.448, 10.135) = 6.752, p = .019), 

and induce cravings reflecting an intention and 

planning of using cannabis (i.e. “purposefulness”; 

F(1.431, 10.017) = 17.516, p = .001), when 

compared to both cannabis and neutral cues. 

Cannabis cues in turn elicited more craving in the 

purposefulness domain than neutral cues. 

Overall, these preliminary results indicate that 

trauma cues elicit more craving, as well as more 

negative affect (anxiety, sadness), than both 

cannabis and neutral cues in this cannabis-using 

population with significant trauma histories. This 

suggests that exposure to trauma cues (i.e., 

external trauma reminders and/or internal 

trauma memories) may be a strong predictor and 

motivator for cannabis use in this population. 

Thus, exposure to trauma cues may promote 

cannabis use for regulating negative affect via 

conditioned craving processes. 

 

Contact: Ioan T. Mahu 
Ioan.Tiberiu.Mahu@Dal.Ca 

 

Campus Climate Matters: The Impact of Racial 

Microaggressions on Marijuana Use among 

Undergraduate Students of Color 

 

Miesha Marzell, Woojae Han, Elora Orazio, Tess 

K. Drazdowski, & Ada Robinson-Perez 

Binghamton University 

 

Background: Microaggressions are verbal, 

behavioral, or environmental indignities, 



Conference Abstracts  81 

 
intentional or not, that communicate hostile, 

derogatory, or other negative racial assertions to 

people of color. Recent research focused on racial 

microaggressions indicates that experiencing 

them is associated with a higher risk for stress, 

low self-esteem, and anxiety among college 

students of color. Also, students of color who 

perceive an unwelcoming campus climate can 

experience academic difficulties and increased 

risky substance use. At the same time, marijuana 

use by all college students has seen a steady 

increase over the last decade. In 2016, more than 

20% reported use in the past 30 days, up 5% from 

2006. Many students use marijuana to cope with 

stress and social anxiety, with related negative 

health and academic outcomes. We sought to 

identify associations between perceived 

microaggressions, marijuana use, and 

motivations for use among undergraduate 

students of color, hypothesizing that the more a 

student experienced racial microaggressions, the 

more likelihood of marijuana use. Methods: A 

sample of 251 undergraduate students of color 

from a medium-size university in the 

northeastern US responded to a racial 

microaggressions and substance use survey. 

Original data were collected using the Racial 

Microaggressions Scale (RMAS), which is factored 

into six subscales: invisibility, criminality, low-

achieving/undesirable culture, sexualization, 

foreigner/not belonging, and environmental. We 

explored relationships among racial 

microaggressions, motives for marijuana use, and 

marijuana use, conducting a path analysis to test 

the significance of hypothesized connections 

between variables. Results: The majority of 

students were aged 18-20 (62.5%) and female 

(69.9%). Approximately 42% of the sample 

reported using marijuana during the past 30 days, 

22% reported use 1-3 times per month, and 17% 

multiple times per week. Path analysis indicated 

that motives for marijuana use mediated the 

relationship between racial microaggressions and 

actual marijuana use. Discussion: and 

Implications. Although our analyses showed no 

direct relationship between perception of racial 

microaggressions and marijuana use, taking into 

account motives for use revealed that racial 

microaggressions do indirectly influence 

marijuana use. Specifically, our study suggests 

that when undergraduates of color perceive 

negative characteristics of their environment 

related to race (e.g., absence of people from one’s 

racial background in school or being in contexts 

where an individual is the “only person of color”), 

their motives to use marijuana, especially those 

involving social reasons (e.g., to fit in with peers), 

increase the likelihood of their using marijuana. 

Our study demonstrates that students of color 

may have unique reasons for marijuana use and 

reinforces the importance of college 

administrators’ efforts to make the campus 

climate more inclusive and welcoming for all 

students. Implications for practitioners include 

tailoring substance use prevention programs and 

counseling services to better address the health 

and well-being of students of color. 

 

Contact: Miesha Marzell 
mmarzell@binghamton.edu 

 
Who Volunteers for Cannabis Research? 

Examining Potential Participation in Cannabis 

Research among Emerging Adults 

 

Maha N. Mian, Brianna Altman,  

& Mitch Earleywine  

University at Albany, SUNY 

 

While many contemporary empirical studies draw 

inferences from the undergraduate student 

population, little is known about the patterns of 

research participation in this group. The sensitive 

nature of substance use may play an influential 

role in studies investigating illicit use, 

particularly in the emerging adult population. 

The present study examined cannabis use and 

willingness to participate in cannabis-related 

research in emerging adults. Undergraduate 

students enrolled in a psychology course (n=262) 

reported frequency of cannabis use and if they 

would be willing to participate in hypothetical 

research studies on cannabis, varying in study 

type and compensation. Chi-square tests for 

independence revealed significant relations 

between gender and cannabis use per week, as 

well as use per year and lifetime use. Notable 

significant associations also appeared between 

willingness to ingest marijuana for pay and 

lifetime use (Cramer’s V=.433, p<.0001), as well 

as willingness to ingest marijuana for scientific 

contribution with both yearly and lifetime use 

(Cramer’s V=.434, p<.001; Cramer’s V=.580, 

p<.001). Additionally, significant relations with 



Conference Abstracts  82 

 
gender suggested that males were less willing to 

complete a survey for a prize, but more willing to 

come to a lab study for scientific contribution, and 

to ingest cannabis for scientific contribution; odds 

ratios (respectively) were, OR=.372 [95% CI, .171-

.811], OR=2.336 [95% CI, 1.318-4.138] and 

OR=2.182 [95% CI, 1.154-4.123]. These results 

demonstrate important trends exist in potential 

participation for cannabis-related research in the 

emerging adult population, specifically around 

gender, and suggest cautious interpretation for 

the generalizability of studies on cannabis in 

undergraduate students. 

 

Contact: Maha N. Mian 
mmian@albany.edu 

 

Marijuana Expectancies as a Predictor of 

Substance Use Initiation among High School 

Students 

 

Kevin S. Montes, Katie Witkiewitz, Matthew R. 

Pearson, & Adam M. Leventhal 

University of New Mexico & University of 

Southern California 

 

Adolescence is a high-risk period for substance 

use onset. Expectancies have been identified as 

proximal predictors of adolescent onset of 

substance use. However, little is known about 

whether change in marijuana expectancies is a 

risk factor for marijuana use onset. The present 

study was conducted to examine trajectories of 

marijuana expectancies and to investigate 

whether change in marijuana expectancies was 

predictive of marijuana use onset. Data from 

3,396 ethnically diverse high school students were 

collected across eight waves of assessment (follow-

up assessments were 6-months apart with 

baseline assessment commencing when 

participants were in 9th grade and concluding 

when participants were in 12th grade) and 

analyzed within a latent growth modeling 

framework. Parallel process latent growth models 

with robust weighted least squares estimation 

were conducted to derive the slopes and intercepts 

from the first four waves of positive and negative 

marijuana expectancy data to examine whether 

these parameters were predictive of marijuana 

use onset that occurred during the last four waves 

of assessment. The results from the study indicate 

that the slopes of positive marijuana use 

expectancies were prospectively predictive of 

marijuana use onset. Specifically, the slope of 

positive marijuana expectancies was positively 

associated with the likelihood of initiating 

marijuana use for the first time after controlling 

for gender, age, and baseline level of positive 

marijuana expectancies (B=1.26, p<.001; β=.32). 

Moreover, the slope of positive marijuana 

expectancies accounted for unique variance that 

was not accounted for by baseline levels of positive 

marijuana expectancies. That is, regardless of the 

level of positive marijuana expectancies endorsed 

by participants at baseline, change in positive 

marijuana expectancies was still predictive of 

marijuana use onset. Baseline level of positive 

(B=.63, p<.001; β=.54) and negative (B=-.19, 

p=.03, β=-.15) marijuana expectancies were also 

found to be prospective predictors of marijuana 

use onset. Results from the present study indicate 

that change in marijuana expectancies – as well 

as level of positive and negative marijuana 

expectancy endorsement at 9th grade –  may be 

markers of risk propensity for prospective 

marijuana use onset that occurs in high school.  

 
Contact: Kevin S. Montes 

kmontes@csudh.edu 
 

Retail Availability of Cannabis in Oregon 

Counties, Perceived Availability, and Cannabis 

Use among Adolescents 

 

Mallie J. Paschall, Joel W. Grube,  

& Anthony Biglan 

Prevention Research Center &  

Oregon Research Institute 

 

Research on local cannabis policies, retail 

cannabis availability and cannabis use among 

adolescents is limited.  This study examines 

relationships between cannabis sales policy in 

Oregon counties, retail availability of cannabis, 

perceived availability of cannabis, normative 

beliefs, and cannabis use among adolescents.  We 

hypothesized that a higher density of cannabis 

retail outlets would be present in counties that 

allowed cannabis sales for medical or recreational 

use, and that greater retail availability would be 

positively associated with perceived cannabis 

availability, normative beliefs favorable to 

cannabis use, and cannabis use among 

adolescents. We also hypothesized that perceived 



Conference Abstracts  83 

 
availability of cannabis and normative beliefs 

more favorable to cannabis use would act as 

mediators of the association between retail 

cannabis availability and cannabis use among 

youth.  Data on Oregon county cannabis sales 

policy and licensed retail outlets were obtained 

from the Oregon Liquor Control Commission, 

indicating which cities and counties prohibited 

sales of cannabis for medical and recreational use, 

and the locations of licensed retail outlets, 

respectively.  We computed licensed retail outlet 

density per 10,000 county population.  We 

obtained county-level prevalence rates for past-

30-day cannabis use among 11th graders from the 

2015-16 Student Wellness Survey (N=57,742) 

conducted by the Oregon Health Authority.  The 

SWS also included questions about perceived 

availability of cannabis and perceived disapproval 

of cannabis use by parents and friends.  Bivariate 

correlations for 32 Oregon counties indicated 

significant positive associations between the 

prevalence of past-30-day cannabis use among 

11th graders and counties allowing sales of 

cannabis for medical or recreational use (r = .58, 

p<.01), density of licensed outlets (r = .48, p<.01), 

and the perception that cannabis would be very 

easy to get (r = .70, p<.01).  Parents’ disapproval 

of cannabis use was inversely related to the 

prevalence of past-30-day cannabis use among 

11th graders (r = -.51, p<.01), as was friends’ 

disapproval (r = -.48, p<.01).  Perceived 

availability of cannabis was positively associated 

with counties allowing cannabis sales (r = .55, 

p<.01) and a higher density of licensed cannabis 

outlets (r = .40, p<.05).  Parents’ and friends’ 

disapproval of cannabis use were inversely related 

to counties allowing cannabis sales and density of 

licensed cannabis outlets.  Linear regression 

analyses indicated possible indirect effects of 

county cannabis sales policies and cannabis outlet 

density on past-30-day cannabis use among 11th 

graders, as the beta coefficients for these 

predictors were no longer significant in models 

with perceived cannabis availability.  However, 

the normative beliefs variables did not appear to 

act as mediators.  These results suggest that 

legalization of cannabis for medical and 

recreational use is associated with greater retail 

availability of cannabis and greater perceived 

availability of cannabis among adolescents, which 

may increase their risk for cannabis use.  Further 

research is needed to better understand effects of 

local cannabis policies and availability on 

cannabis use and related problems among youth. 

 
Contact: Mallie J. Paschall 

paschall@prev.org 
 

PTSD Symptom Severity and Cannabis Use: The 

Moderating Role of Race/Ethnicity 

 

Alisa Payne, Nathan Kearns, Renee Cloutier, 

Quadreon Miller, Anabel Potts,  

& Heidemarie Blumenthal 

University of North Texas 

 

Elevated posttraumatic stress disorder (PTSD) 

symptoms appears to be a significant risk marker 

for problematic cannabis use. However, unlike the 

PTSD-alcohol literature, few studies have 

evaluated sociodemographic characteristics that 

may moderate the association between PTSD and 

cannabis use. Epidemiological work has identified 

differences in cannabis use patterns as a function 

of race/ethnicity, but the nature of the relation 

differs when trauma history and psychological 

characteristics (e.g., diagnostic status) are also 

considered. For example, in the recent National 

Surveys on Drug Use and Health, individuals 

identifying as Black or Native American reported 

the highest rates of cannabis use and related 

problems; however, in a large sample of trauma-

exposed individuals, Kevorkian and colleagues 

(2015) found the highest rates among those 

identifying as White. To date, no work has 

addressed race/ethnicity as a potential moderator 

of the association between PTSD and cannabis 

among those exposed to trauma expressing 

clinical and preclinical symptoms. The present 

study tested whether race/ethnicity moderated 

the association between PTSD symptom severity 

(PTSDSS) and past-year cannabis use. Based on 

prior studies, we hypothesized that Whites and 

Hispanics who present PTSS would have higher 

rates of cannabis use, followed by African 

Americans and Asians who would report the 

lowest rates. The final sample consists of 1,067 

college participants (M = 20. 71 years; 72.9% 

female) who completed a demographics 

questionnaire which assessed the participants’ 

race/ethnicity; a single, face valid question 

measured past-year cannabis use; and the PTSD 

Checklist which measured PTSDSS. First, 

ANOVA and multiple regression analyses were 



Conference Abstracts  84 

 
conducted to preliminarily examine the influence 

of racial/ethnic identity on past-year cannabis 

use. For primary analyses, separate moderation 

analyses were conducted using PROCESS in 

SPSS to examine specific race/ethnicity 

identification groups as moderators in the 

relationship between PTSDSS and cannabis use. 

Consistent with past literature, results of the 

ANOVA found that individuals identifying as 

White (M = 4.20) and African American (M = 4.25) 

used significantly more frequently than those 

identifying as Asian (M = 3.30) or Hispanic/Latino 

(M = 3.59); further, results of the regression 

analyses indicated that identification as Asian (β 

= -.01; p = .021) or Hispanic/Latino (β = -.12; p = 

.007) was significantly associated with less past-

year cannabis use. However, results of the 

primary analyses indicated that race/ethnic 

identity was not a significant moderator in the 

association between PTSDSS and cannabis use. 

Results indicated that race/ethnicity did not play 

an important role in the association between 

PTSS and cannabis use among trauma-exposed 

college students. Future research addressing 

race/ethnicity, as a function of diagnostic status 

(e.g., with or without PTSD, cannabis use 

disorder), and across other contextually- and 

developmentally-distinct periods, is needed. For 

example, work-addressing periods characterized 

by relevant vulnerability (e.g., puberty) could be a 

key to understanding the nature and boundaries 

of these relations. 

 

Contact: Alisa Payne 
alisapayne2014@gmail.com 

 

Does Marijuana Use Impact Academic 

Performance in the College Setting? 

 

Erica B. Peteja, Kerry D. Duck, Anastasia 

Robirds, Megan H. Smith, Alexa J. Pellegrino, 

Lillian J. Canfield, Maryia M. Schneider, 

Chelsea L. Banks, Kristina T. Phillips,  

& Michael M. Phillips 

University of Northern Colorado 

 

Background: Past research has examined the 

impact of marijuana use on college student 

academic performance. Marijuana use has been 

associated with GPA and failure to obtain a 

university degree. A recent longitudinal study 

found that as marijuana use escalated, college 

students were more likely to skip classes and 

performed more poorly academically, as measured 

by GPA. Few studies have assessed other 

academic components that might influence 

completion of one’s college degree among 

marijuana users, such as academic motivation 

and self-efficacy. The goal of the current study 

was to assess whether frequency of marijuana 

use, problem use, academic motivation, academic 

self-efficacy, and academic self-regulation impact 

university retention and performance. Methods: 

Active college-student marijuana users (n = 52) 

were recruited and completed a baseline 

assessment. Although all participants completed 

two-week ecological momentary assessment, the 

current presentation focuses on baseline data. 

Marijuana use frequency was assessed through 

days of marijuana use over the last month. 

Problem marijuana use was assessed through the 

Rutgers Marijuana Problem Index (RMPI). 

Academic self-efficacy and self-regulation were 

assessed using subscales of the Motivated 

Strategies for Learning Questionnaire (MSLQ). 

Academic motivation was assessed using a 

subscale from the Learning and Study Strategies 

Inventory (LASSI). Results: The sample was 

57.7% female, 59.6% Caucasian, and averaged 

20.04 (SD = 1.49) years of age. Mean cumulative 

GPA was 2.80 (SD = 0.65; range from 1.07-3.93). 

Participants were heavy marijuana users, 

reporting near daily use across the past 30 days 

(M = 23.56, SD = 6.85). We examined whether 

academic variables and marijuana use/problem 

use predicted current GPA via multiple 

regression. The overall model accounted for 32.9% 

of the variance. Self-efficacy (b = .05, SE= .02, p 

=.006) and academic motivation (b = .07, SE = .03, 

p = .017) contributed to current GPA.  After 

controlling for academic variables, neither the 

frequency of marijuana use, nor problem use, 

were associated with current GPA. Logistic 

regression was used to assess whether academic 

variables and marijuana use/problem use 

contribute to student retention. In the current 

sample, 86.5% (N = 45) of participants remained 

at the university one semester after their baseline 

appointment. Due to the large number of retained 

students, there were no significant predictors of 

academic retention. Conclusions: Frequency of 

marijuana use over the last 30 days did not 

contribute to GPA when controlling for academic 

factors (academic self-efficacy and motivation). 



Conference Abstracts  85 

 
There were no significant predictors for retention, 

although this was due to a lack of power to detect 

differences. Future studies with more 

participants should continue to evaluate the 

potential impact of marijuana use on college 

student success. Research reported in this 

abstract was supported by NIDA of the National 

Institutes of Health under award number 

R15DA041656 (PIs: K. Phillips, M. Phillips).  
 

Contact: Erica B. Peteja 
Erica.Peteja@colostate.edu 

 

Reduction of Benzodiazepine Use in Patients 

Prescribed Medical Cannabis 

 

Chad Purcell, Andrew Davis, Nico Moolman,  

& S. Mark Taylor 

Dalhousie University & Acadia University 

 

Background: Benzodiazepines are a class of 

medication with sedative properties, commonly 

used for anxiety and other neurological 

conditions. These medications are associated with 

several well-known adverse effects. This 

observational study aims to investigate the 

reduction of benzodiazepine use in patients using 

prescribed medical cannabis. Methods: A 

retrospective analysis was performed on a cohort 

of 146 medical cannabis patients (average age 47 

years, 61% female, 54% reporting prior use of 

cannabis) who reported benzodiazepine use at 

initiation of cannabis therapy. These data are a 

part of a database gathered by a medical cannabis 

clinic (Canabo Medical). Descriptive statistics 

were used to quantify associations between the 

proportion of benzodiazepine use with time on 

medical cannabis therapy. Results: After 

completing an average two-month prescription 

course of medical cannabis, 30.1% of patients had 

discontinued benzodiazepines. At a follow-up 

after two prescriptions, 65 total patients (44.5%) 

had discontinued benzodiazepines. At the final 

follow-up period after three medical cannabis 

prescription courses, 66 total patients (45.2%) had 

discontinued benzodiazepine use, showing a 

stable cessation rate over a six-month period. 

Conclusion: Within a cohort of 146 patients 

initiated on medical cannabis therapy, 45.2% 

patients successfully discontinued their pre-

existing benzodiazepine therapy. This observation 

merits further investigation into the risks and 

benefits of the therapeutic use of medical cannabis 

and its role relating to benzodiazepine use. 

 
Contact: Chad Purcell 

Chad.purcell@dal.ca 
Supplemental Materials Here 

 

Examining Protective Behavioral Strategies as a 

Mediator in the Relationship between Social 

Anxiety and Cannabis-Related Consequences 

 

Cody A. Raeder, Matthew R. Pearson, Adrian J. 

Bravo, & Protective Strategies Study Team 

Old Dominion University & University of New 

Mexico 

 
Individuals high in social anxiety have been 

shown to experience a greater number of alcohol- 

and cannabis-related consequences relative to 

those lower in social anxiety, despite insignificant 

differences in levels of substance use. Therefore, 

an important goal for researchers in this field is to 

determine what factors may reduce the risk of 

experiencing both alcohol- and cannabis-related 

consequences. Protective behavioral strategies 

are strategies an individual can utilize when 

using both alcohol and cannabis to reduce the 

number of negative consequences experienced as 

result of use. Previous research has shown that 

both alcohol and cannabis protective behavioral 

strategies are robustly negatively associated with 

use frequency and consequences for both 

substances. Recent work in the study of social 

anxiety and alcohol protective behavioral 

strategies has found that serious harm reduction 

strategies mediates the relationship between 

social anxiety and alcohol-related consequences. 

Given similarities in the social anxiety literature 

in regards to alcohol- and cannabis-related 

consequences, the purpose of the current research 

was to examine if cannabis protective behavioral 

strategies similarly mediates the relationship 

between social anxiety and cannabis-related 

consequences. Participants were 1022 students 

from multiple universities who completed 

measures assessing social anxiety, cannabis 

protective behavioral strategies, and cannabis-

related outcomes. Cannabis protective behavioral 

strategies were found to mediate the relationship 

between social anxiety and cannabis-related 

consequences such that those higher in social 

anxiety used fewer protective behavioral 



Conference Abstracts  86 

 
strategies, and in turn experienced more 

cannabis-related consequences when controlling 

for quantity and frequency of use. These results 

indicate that those with social anxiety who also 

use cannabis may benefit from interventions 

designed to increase the use of cannabis protective 

behavioral strategies. 

 

Contact: Cody A. Raeder 
craed001@odu.edu 

Supplemental Materials Here 

 

Differences between College Student Medical 

and Recreational Cannabis Users: More 

Similarities or More Differences? 

 

Mariah D. Sandoval, Adrian J. Bravo, Matthew 

R. Pearson, Protective Strategies Study Team, & 

Marijuana Outcomes Study Team 

University of New Mexico 

 

Currently, thirty states and the District of 

Columbia have legalized medical use of cannabis, 

with several other states allowing the use of some 

constituent of cannabis for medical purposes (i.e., 

cannabidiol). Common reasons for medical use 

include posttraumatic stress symptoms, anxiety, 

chronic pain, stress, insomnia and depression 

(Bonn-Miller et al., 2014). Few studies explicitly 

examine differences between medical and 

recreational users of cannabis (Lin et al., 2016), 

especially in the college student population. In 

some ways, one may expect some difficulties with 

separating medical from non-medical use. For 

example, many “recreational” users report using 

marijuana to cope (e.g., with negative affect), thus 

the degree to which a distinction between medical 

and non-medical use can be made is unclear. 

Using two large college student samples, the 

purpose of this study was to determine the extent 

to which college student medical users (i.e., 

individuals reporting having a medical card) 

differ systematically from recreational users. 

Using two large samples of college students 

(Study 1: 7307 college students, 4081 lifetime 

users, 2226 past month marijuana users, 46 

medical users; Study 2: 6584 college students, 

3744 lifetime users, 1969 past month marijuana 

users, 70 medical users). Although we found 

reliable differences between medical and 

recreational users in use patterns, cannabis use 

disorder symptoms, and use of protective 

behavioral strategies, we failed to find reliable 

differences on negative consequences, mental 

health symptoms, and most marijuana use 

motives. Although one may expect medical users 

to report higher coping motives or expect coping 

motives to be their strongest motive for use, they 

actually reported higher use of expansion motives 

and (like recreational users) reported 

enhancement motives as their strongest motive 

for use. Although not replicated due to not being 

assessed in both studies, the largest differences 

found were medical users reporting higher 

craving, higher marijuana identity, and lower 

perceived negative consequences of marijuana use 

Taken together, we found some striking 

similarities as well as some reliable differences 

between college student medical and recreational 

users. Future research should examine the extent 

to which these risk/protective factors operate 

similarly across medical vs. recreational users. 

 

Contact: Matthew R. Pearson 
mateo.pearson@gmail.com 

Supplemental Materials Here 

 

A Person-Centered Approach to Investigate 

Relations among Substance User Profiles and 

Impulsivity 

 

Molin Shi & Andrew K. Littlefield 

Texas Tech University 

 

Alcohol use and other substance use remain 

highly prevalent for college students in the United 

States. The extant literature has supported the 

notion that impulsivity-like facets are significant 

correlates and predictors of consumption and 

problematic usage of alcohol and other 

substances. However, there is a scarcity in the 

literature on how polysubstance use relates to 

these facets. Further, most studies in the 

literature have focused on variable-centered 

approaches to examine these relations. Although 

person-centered approaches like Latent Class 

Analysis (LCA) have been utilized to profile 

drinking patterns, and more recently, subgroups 

of marijuana users, more work is needed to 

explore the relations between profiles, impulsivity 

facets, and alcohol use outcomes. Provided these 

gaps in the literature, the present study aimed to 

use a person-centered approach to examine 

whether empirically-identified latent substance 



Conference Abstracts  87 

 
use profiles relate to impulsivity-like facets (i.e., 

positive urgency, negative urgency, lack of 

perseverance, lack of planning, and sensation 

seeking) and problem drinking differently among 

college students. A sample of 702 students from a 

large, southwestern university was used (Mage = 

19.30, SD = 1.74; 71.79% female). A latent class 

analysis of various past-month indices of alcohol 

and other substance use (e.g., marijuana use, 

binge drinking, E-cigarette use) was conducted in 

Mplus. After examining the AIC, BIC, adjusted 

BIC, Vuong-Lo-Mendell-Rubin likelihood ratio, 

and entropy model-fit statistics from one- to 

seven-class solutions, a three-class solution was 

identified as the most parsimonious and best-

fitting solution. The classes consisted of (1) a 

predominantly drinking class (n = 288), (2) a non-

users class (n = 209), and (3) a poly-substance 

users class (n = 205). A chi-square test of 

independence revealed that substance use profiles 

differed significantly on problem drinking as 

measured by AUDIT χ2 (4, N = 700) = 360.26,  

p < .001. Pairwise comparisons indicated that the 

poly-substance users reported the highest levels of 

severe and moderate problematic drinking, 

followed by predominantly drinkers, then non-

users. Conversely, poly-substance users reported 

the lowest level of minimal problematic drinking, 

followed by non-users and predominantly 

drinkers. Additionally, analysis of variance tests 

revealed significant effect of substance use 

profiles on each of the impulsivity-like facet as 

measured by the Short UPPS-P, with varied 

pairwise differences between substance use 

profiles. For example, poly-substance users 

reported greater positive urgency than either 

predominantly drinkers (p < .001) and non-users 

(p < .001). In contrast, both poly-substance users 

and predominantly drinkers endorsed higher 

sensation seeking levels than non-users (p < .001, 

p = .003, respectively), but did not significantly 

differ between one another. These findings 

indicate that different groups of substance users 

exhibit varied levels of impulsivity-like facets and 

severity of problem drinking. Future work could 

further delineate and explain differences in 

profiles, as distinct profiles could be targets of 

intervention to prevent or intervene on problem 

drinking. 

 
Contact: Molin Shi 
Molin.Shi@ttu.edu 

Supplemental Materials Here 

 

Antecedents, Contexts, and Consequences of 

Marijuana Use: A Comprehensive Review of 

Microlongitudinal Studies 

 

Sarah L. Simons, Gabriel J. Carrion-Gonzalez, & 

Matthew R. Pearson 

University of New Mexico 

 

Due to a lower level of research interest on 

marijuana over the past few decades relative to 

other drugs, there are substantial knowledge gaps 

regarding the antecedents, contexts, and sequelae 

of marijuana use. Importantly, all study designs 

are not equal with regards to closing these 

knowledge gaps. When considering cost and 

feasibility, we argue that the microlongitudinal 

design can most rapidly close these knowledge 

gaps. The microlongitudinal design includes 

repeated measurement in relatively close 

proximity to one another and is facilitated by the 

use of ecological momentary assessment (EMA; 

Shiffman, 2009) methods, also known as 

experience sampling and ambulatory assessment, 

among other things. The simplest EMA design is 

the daily diary design in which individuals 

complete a single assessment per day, but EMA 

studies can include several assessments per day 

or even continuous assessment (e.g., transdermal 

alcohol monitoring, Alessi, Barnett, & Petry, 

2017; heart rate monitoring, Wilhelm, Pfaltz, & 

Grossman, 2005). EMA studies are able to 

examine the associations among variables at both 

the between-subject and within-subject levels.  

Importantly, when examining the associations 

between a putative antecedent and outcome at the 

within-subject level, each individual is essentially 

serving as their own control. The present 

comprehensive review examines all EMA studies 

that had been published on marijuana use (or 

related outcomes) at the time of the writing of this 

review. To date, we have 47 articles published 

through April 2016, but will be presenting all 

articles published up until June 2018. The 

purpose of this review is threefold: 1) summarize 

EMA studies to better characterize the 



Conference Abstracts  88 

 
antecedents, contexts, and consequences of 

marijuana use, 2) identify methodological 

limitations of these studies that can be overcome 

in future research, and 3) identify areas of 

research that have not yet taken advantage of 

EMA methods that would help to close important 

knowledge gaps in the marijuana field. The most 

common form of EMA collection was the use of a 

PDA which emitted random signals throughout 

the day which participants would answer pre-

programmed assessments. The assessments most 

commonly measured the frequency of marijuana 

use since previous signal as well as subjective 

effects (i.e., anxiety and depression levels), 

environment, and social situation the participant 

was in. The range of signals emitted per day was 

as little as one-time (daily diaries) to as much as 

twelve times. The range of the duration of EMA 

data collection was as few as six days to as much 

as thirty days. One clear limitation across studies 

is that quantity of marijuana use is still not often 

measured in EMA studies, diminishing our ability 

to detect dose-response associations. 

 
Contact: Matthew R. Pearson 

mateo.pearson@gmail.com 
Supplemental Materials Here 

 

Marijuana Use Patterns Among College 

Students: Promise of Probing User Heterogeneity 

 

Melissa Sotelo, Matthew R. Pearson, & 

Marijuana Outcomes Study Team 

University of New Mexico 

 

The majority of marijuana research to date has 

been a comparison of marijuana users versus non-

users, assuming that the marijuana user 

population is a homogenous group. Using Latent 

Profile Analysis (LPA), previous studies have 

shown that marijuana users are a heterogeneous 

group, exhibiting several subpopulations as 

opposed to a single use pattern. Pearson et. al. 

(2017) have found four classes of college student 

marijuana users, with the largest group being 

low-quantity, infrequent users who reported the 

least number of consequences. The current study 

sought to distinguish subpopulations within 

college student marijuana users, measuring the 

quantity (in grams) of marijuana used, number of 

consequences reported, frequency of use within 

the past 30 days, and “typical frequency” of use 

during a “typical week”. A weekly grid was used 

to divide each day of the week into six 4-hour time 

blocks where the participant was asked to record 

the amount of marijuana in grams they used 

during a time block. A brief form of the Marijuana 

Consequences Questionnaire (Simons, Dvorak, 

Merrill, & Read, 2012) was used to measure 

marijuana consequences. Using LPA, we 

identified four classes of marijuana users. Class 1 

was low-quantity, infrequent users which 

contained 71.29% of users; class 2 was low-

quantity, moderate users containing 17.16% of 

users; class 3 was moderate-quantity, frequent 

users containing 9.97% of users; and class 4 was 

frequent, high quantity users containing 1.59% of 

users. Between classes, there were substantial 

differences in quantities, with class 1 reporting on 

average 2.352 grams on average per typical week, 

and class 4 reporting approximately 80 grams per 

typical week. These classes also differed on the 

reported number of consequences, with class 1 

reporting the least, and class 4 reporting the most 

consequences. Class 4 also reported the highest 

likelihood for cannabis use disorder symptoms, 

used less protective behavioral strategies, and 

identified the most strongly as a marijuana user. 

For future research, there needs to be  a more 

immediate and accurate way of reporting use and 

quantity in order to corroborate results. 

 
Contact: Matthew R. Pearson 

mateo.pearson@gmail.com 
Supplemental Materials Here 

 

Endorsing Beliefs for Using Cannabis Around 

Exercise: An Exploratory Study 

 

Alicia Stewart, Renee E. Magnan, & Benjamin O. 

Ladd 

Washington State University Vancouver & 

Washington State University 

 

There is limited knowledge concerning the 

relationship between cannabis use and exercise. 

As accessibility to cannabis increases, there is 

growing need to study how cannabis influences 

health and daily activities. Cannabis decreases 

motivation to exercise and can negatively affect 

motor control, while cannabis can also decrease 

inflammation and physical discomfort brought on 

by exercise (Gillman et al., 2015). Using a cross-

sectional survey, a community sample of at least 



Conference Abstracts  89 

 
twice weekly cannabis users (N = 88) indicated if 

they ever use cannabis one hour before engaging 

in exercise and/or within four hours after 

exercising. Participants then rated their beliefs 

about the positive influence of cannabis on 

exercise (e.g., cannabis enhances performance, 

cannabis makes it easier to recover) from 1 

(Strongly Disagree) to 7 (Strongly Agree). 

Participants also indicated the concentration of 

tetrahydrocannabinol (THC) and cannabidiol 

(CBD) in the cannabis they use most often. This 

was multiplied by average quantity (in grams) of 

use to create a dose score. Participants reported 

an average THC dose of 0.06 g (SD = 0.10) and 

average CBD dose of 0.02 g (SD = .03) per use 

episode. The majority of individuals (87.5%) used 

cannabis one hour before and/or four hours after 

exercise. Overall, participants who used cannabis 

around exercise reported higher agreement (M = 

4.89, SD = 1.21) with the positive beliefs compared 

to participants who did not report using cannabis 

around exercise (M = 3.62, SD = 1.12), t(81) = -

3.07, p < .01. Similar results were observed for 

each of the five beliefs individually. Regarding the 

specific beliefs, people were most likely to support 

the belief that cannabis makes it easier for them 

to recover from exercise (M = 5.44, SD = 1.64). 

This belief was also the only one significantly 

associated with THC dose (r = 0.27, p < .05) and 

demonstrated a trend with CBD dose (r = 0.20, p 

< .1). These results suggest that cannabis users 

who use cannabis around times of exercise may do 

so because of their positive beliefs about exercise-

related benefits of cannabis. Alternatively, the act 

of using cannabis around exercise may serve to 

enhance the positive beliefs cannabis users hold 

around exercise-related benefits of cannabis. 

Negative exercise-related beliefs of cannabis were 

not assessed in the current study, which also may 

be a contributing factor to the strength of the 

cannabis-exercise relationship. Further research 

on cannabis beliefs around exercise performance 

and recovery and THC and/or CBD consumption, 

together with biological mechanisms, could 

inform efforts to understand the risks and 

benefits of cannabis on physical activity. 

 
Contact: Benjamin O. Ladd 

benjamin.ladd@wsu.edu 

 

Cognitive and Behavioral Risk Correlates of 

Marijuana Use by Older Adults 

 

Susan A. Stoner 

University of Washington 

 

Marijuana use by older adults is on the rise. 

Published data from the National Survey of Drug 

Use and Health (NSDUH) suggested that rates of 

self-reported marijuana use increased 

significantly among adults aged 50 to 64 and 

substantially among adults aged 65 or older 

(seniors). Notably, however, of 447,196 

individuals in the combined 2006-2013 NSDUH 

dataset, seniors represented only 4.0% of the 

sample (n=17,837), and only 154 seniors (0.8%) 

reported marijuana use in the past year. Little is 

known about older marijuana users; most 

research has been conducted with young adults, in 

whom marijuana use is associated with cognitive 

impairment and risk-taking behavior. Using data 

from the Behavioral Risk Factor Surveillance 

System (BFRSS), we examined whether similar 

associations were evident in an older adult 

population. In 2016, 13 states (4 with legal non-

medical use, 3 with legal medical use, 4 with legal 

CBD use, and 2 with no legal use) optionally 

queried marijuana use in the prior 30 days as part 

of the BRFSS. Of 129,782 individuals in the 13-

state dataset, seniors comprised 36.5% of the 

sample (n=47,422) and 892 seniors (1.9%) 

reported marijuana use in the past 30 days.  

Percentages of persons aged 45-64 who self-

reported marijuana use ranged from 2.1% to 

12.8%, varying by state. Percentages of seniors 

who did so ranged from 0.3% to 6.7%.  Any 

marijuana use in the prior 30 days was negatively 

correlated with healthy days/health-related 

quality of life in both age groups and positively 

correlated with having fallen in the last 12 

months; ongoing difficulty concentrating, 

remembering, or making decisions; and cognitive 

decline over the last 12 months, particularly in the 

45-64 age group. In the small subset of senior 

marijuana users who reported cognitive decline 

and completed optional follow-up questions 

administered by 3 states (n=50), there was a 

strong association between level of marijuana use 

and the extent to which cognitive decline had led 

respondents to give up household activities or 

chores (r=.425) or interfered with work or social 

activities (r=.420). Any marijuana use in the prior 



Conference Abstracts  90 

 
30 days was positively associated with likelihood 

of alcohol use, binge or heavy drinking, driving 

under the influence of alcohol, and HIV risk-

taking in both age groups.  Among marijuana 

users, level of marijuana use was negatively 

associated with likelihood of drinking alcohol and 

HIV risk-taking.  Findings suggest that older 

adults who use alcohol and take other risks are 

more likely to use marijuana than those who do 

not, which may reflect non-medical use, and those 

who use marijuana tend to be in poorer health 

than those who do not, which may reflect medical 

use.  However, because the data are purely 

correlational and cannot demonstrate causality, 

the nature of these relationships is unclear. The 

effect of marijuana on cognition is concerning as 

older adults are vulnerable to harm from cognitive 

errors, such as errors in medication self-

administration. Cognitive and behavioral effects 

of marijuana in older adults require further 

examination with controlled studies. 

 

Contact: Susan Stoner 

sastoner@uw.edu 
Supplemental Materials Here 

 

The Role of Risk Seeking, Negative Urgency, and 

Perseverance in Marijuana Use Frequency and 

Marijuana Related Consequences 

 

Alexander J. Tyskiewicz, Jamie E. Parnes,  

& Bradley T. Conner 

Colorado State University 

 

The acceptance of marijuana as a medicine and a 

recreational drug has been increasing in the U.S. 

The number of marijuana users in the U.S. is 

increasing. The need for research that helps us 

understand variables that influence marijuana 

use, and the strategies that help people abstain 

from misuse/abuse and evade marijuana related 

consequences (MRC) is also increasing. Sensation 

seeking has been shown to predict substance use. 

Researchers have also found that impulsivity and 

emotion dysregulation have a close link to drug 

use and abuse. Research concerning the influence 

of personality traits on MRC is becoming more 

abundant. Research has supported the fact that 

impulsivity predicts MRC in frequent users. 

Research also supports the notion that emotion 

dysregulation is a predictor of increased frequency 

of use. One group of researchers examined the 

influence of emotion dysregulation, impulsivity, 

and sensation seeking on marijuana use 

frequency, and MRC. Negative binomial 

regressions indicated that negative urgency, a 

facet of impulsivity, and risk seeking a facet of 

sensation seeking, significantly predicted the 

experiencing of MRC. Risk seeking was also found 

to predict past 30-day intoxication frequency and 

typical week marijuana use. Perseverance, a facet 

of impulsivity, showed to be a protective factor 

that decreases MRC, past 30-day intoxication 

frequency, and typical week marijuana use 

frequency. Further research concerning variables 

that either cause or interfere with these relations 

will help us understand how to predict marijuana 

outcomes. Past research has supported the notion 

that protective behavioral strategies (PBS) 

mediates the relation between sensation seeking 

and MRC. Therefore, PBS may also mediate the 

relation between other personality traits and 

MRC. Since PBS serve to reduce MRC, continued 

research on this factor is warranted. In this study, 

we will test if there are significant effects of risk 

seeking, perseverance, and negative urgency on 

three marijuana outcomes; MRC, past 30-day 

intoxication frequency, and typical week 

marijuana use frequency. We will also test to see 

if the use of PBS mediates those interactions. 

Based on previous literature, we hypothesize that 

the use of PBS will mediate the effect risk seeking 

has on MRC, past 30-day intoxication frequency, 

and typical week marijuana use frequency. 

Further, we hypothesize that PBS use will 

mediate the interaction between perseverance 

and all three outcomes being analyzed. Lastly, we 

hypothesize that PBS use will mediate the 

interaction between negative urgency and MRC. 

We expect results to uphold our hypothesis. The 

data that shows the relations between these traits 

and outcomes will increase our ability to predict 

marijuana outcomes. If the mediation of PBS is 

shown to be significant in these relations, useful 

clinical implications could be made. This study  

could support an emphasis on PBS in clinical and 

counseling settings. 

 

Contact: Alexander Tyskiewicz 
Alexander.Tyskiewicz@colostate.edu 

Supplemental Materials Here 

 



Conference Abstracts  91 

 
Examining Predictors of Initial Marijuana 

Engagement Using Recursive Partitioning 

 

Gemma T. Wallace, Bradley T. Conner,  

& Audrey M. Shillington  

Colorado State University 

 

While several constructs have been associated 

with initiating marijuana use, most studies have 

examined only a few risk factors at a time, often 

due to statistical constraints of a priori analyses. 

Exploratory analyses that use machine learning 

allow for the joint analysis of hundreds of 

predictor variables without concern for statistical 

power. Thus, exploratory analyses may identify 

risk factors for marijuana use that have not been 

examined in previous literature. We used 

recursive partitioning to identify sets of variables 

that are associated with lifetime presence of 

marijuana use. Data are repeated cross-sectional 

survey responses to the American College Health 

Association’s National College Health Assessment 

(ACHA NCHA) from 2011, 2013, and 2015. 

Participants were students at a Colorado 

university (N = 4052, 77% White, 61% female, 

mean age = 22.77). The ACHA NCHA measures a 

wide range health-related constructs, including 

alcohol and substance use behaviors and norms, 

sexual behaviors, physical and mental health, 

negative and positive affect, academic difficulties, 

and stressful and/or traumatic experiences. Thus, 

the ACHA NCHA provides an opportunity for a 

broad assessment of factors that may be 

associated with marijuana use. Given 

multicollinearity between marijuana and other 

substance use variables, non-alcohol substance 

use measures were not included in these analyses. 

We tested two recursive partitioning models to 

examine correlates of ever having used 

marijuana. Our first model evaluated all 

constructs within the ACHA NCHA dataset that 

had an endorsement frequency of ≥ 50 (193 

variables). Alcohol consumption amount and 

frequency were the strongest correlates, and this 

first tree explained 39% of the variability in user 

status (relative error = 0.61, cross-validation error 

= 0.62, SE = 0.02). In our second model, we 

removed alcohol variables to identify constructs 

that were most correlated to user status in the 

absence of substance and alcohol measures (160 

variables). In this second tree, sexual behavior 

variables were the strongest correlates, and this 

model accounted for 25% of the variability in this 

sample (relative error = 0.75, cross-validation 

error = 0.75, SE = 0.02). Both recursive 

partitioning models produced parsimonious 

decision trees that accounted for a substantial 

amount of variance in the lifetime presence of 

marijuana use. Among a wide-ranging set of 

predictor variables, higher engagement in alcohol 

use and sexual behaviors seem to be the most 

salient correlates of ever having used marijuana. 

These exploratory analyses corroborate previous 

literature suggesting that individuals who engage 

in multiple types of health-risk behavior are more 

likely to initiate marijuana use. 

 

Contact: Gemma T. Wallace 

Gemma.wallace@colostate.edu 
 

Perceived Risk, Perceived Addictive Potential, 

and Protective Behavioral Strategies’ Impact on 

Marijuana Outcomes 

 

Andrew P. Weinstein, Matthew R. Pearson, & 

Marijuana Outcomes Study Team 

University of New Mexico 

 

Perceptions about the risk associated with 

marijuana use behaviors have been changing over 

time. A recent study examining the perception of 

how risky marijuana use behaviors are found that 

the perceived risk of marijuana use behaviors (e.g. 

regularly using marijuana) has changed between 

1991 and 2016 while a protective effect remained 

stable among most populations of US 12th grade 

students (Terry-McElrath et al., 2017). Large 

national samples, like the National Survey of 

Drug Use and Health, typically use these 

questions to assess the perceived risk of 

marijuana use. Another protective factor for 

marijuana outcomes are protective behavioral 

strategies (PBS), which are cognitive behavioral 

strategies used proximal to substance use, which 

limit consumption or mitigate negative 

consequences (Pedersen et al., 2016). In the 

present study we sought to 1) examine Perceived 

risk an alternative measure of perceived risk 

among college students and 2) assess the 

relationship between marijuana protective 

behavioral strategies and perceptions of risk with 

marijuana-related outcomes. We used the data 

from the Marijuana Outcomes Study Team, which 

is a large multisite study examining a myriad of 



Conference Abstracts  92 

 
marijuana related topics at 8 US universities in 8 

states.  We found that a traditional measure of 

perceived risk and perceptions of addictive 

potential were constitutionally different factors 

and interacted with PBS to influence marijuana 

outcomes. 

 
Contact: Matthew R. Pearson 

mateo.pearson@gmail.com 
Supplemental Materials Here 

 

Sleep Difficulties, Marijuana Use and Suicidal 

Behaviors in Adolescents 

 

Maria M. Wong, Ph.D. 

Idaho State University 

 

This study examined the relationship among sleep 

difficulties, marijuana use and suicidal behaviors 

(ideation and attempts) in adolescents. 

Additionally, we also tested whether marijuana 

use mediated the relationship between sleep 

difficulties and suicidal behaviors. Previous 

research reported that sleep difficulties and 

substance use predicted suicidal thoughts and 

attempts (Roane & Taylor, 2008; Wong & Brower, 

2012). Here we extended the literature by 

specifically testing whether marijuana use had 

any relationship with suicidal behavior, after 

controlling for other prominent risk factors of 

suicide such as depressive symptoms and alcohol 

related problems. Study participants were 6504 

adolescents from the National Longitudinal Study 

of Adolescent Health (ADD HEALTH) (Harris et 

al., 2009). Data were collected from interviews 

and questionnaires. The current study analyzed 

data from both Times 1 and 2 (T1 and T2), which 

were one year apart.  Sleep difficulties were 

measured by the question, “Please tell me how 

often you have had each of the following 

conditions in the past 12 months - Trouble falling 

asleep or staying asleep?” (0=never, 1=a few 

times, 2=once a week, 3=almost every day, 

4=every day). Lifetime marijuana use was 

examined by the question, “During your life, how 

many times have you used marijuana?” Suicidal 

thoughts were measured by the question, “During 

the past 12 months, did you ever seriously think 

about committing suicide?” Suicidal attempts 

were measured by one question, “During the past 

12 months, how many times did you actually 

attempt suicide?” Marijuana use and suicide 

variables were recoded as dichotomous variables 

(0=no; 1=yes). Data were analyzed by logistic 

regression. All analyses controlled for 

demographic variables that are associated with 

suicidal behaviors (i.e., sex, age, school grade, 

poverty, and chronic health problems) as well as 

depressive symptoms and alcohol-related 

problems (e.g., driving under the influence of 

alcohol, getting into a fight with family due to 

drinking). In T1, controlling for demographic 

variables, depressive symptoms and alcohol-

related problems, both sleep difficulties (OR = 

1.26, p<.001) and marijuana use (OR = 2.45, 

p<.001) were associated with suicide ideation. 

Controlling for T1 suicide ideation and all other 

covariates, both sleep difficulties (OR = 1.13, 

p<.05) and marijuana use (OR = 1.33, p<.05) were 

significant predictors of T2 suicide ideation. 

Marijuana use at T1 significantly mediated the 

relationship between T1 sleep difficulties and T2 

suicide ideation (95% asymmetric confidence 

interval = .004 to .08, p<.05). Both sleep 

difficulties (OR = 1.35, p<.001) and marijuana use 

(OR = 3.05, p<.001) were associated with suicide 

attempts at T1. Controlling for T1 suicide attempt 

and all other covariates, sleep difficulties were not 

a significant predictor (OR = 1.14, p=.09) while T1 

marijuana use was a significant predictor (OR = 

1.81, p<.01) of T2 suicide attempt. Both sleep 

difficulties and marijuana use were concurrently 

and longitudinally associated with suicide 

ideation. Suicide prevention programs for 

adolescents could include information on 

management of insomnia symptoms, as well as 

the health effects of both acute and chronic 

marijuana use. 

 
Contact: Maria M. Wong 

wongmari@isu.edu 
 

Insomnia Symptoms, Protective Behavioral 

Strategies for Marijuana and Hazardous 

Marijuana Use among College Students 

 

Maria M. Wong, Bianca Montoya, Elizabeth 

Craun, & Protective Strategies Study Team 

Idaho State University 

 

Previous research has reported that sleep 

problems longitudinally predicted both onset of 

substance use and substance-related problems 

(Wong, Brower, & Zucker, 2009; Wong, Robertson, 



Conference Abstracts  93 

 
& Dyson, 2015). However, the mediators of this 

relationship remain unclear. The main goals of 

this study were to examine (i) the relationship 

between insomnia symptoms and cannabis use 

problems and (ii) whether protective behavioral 

strategies for marijuana mediated this 

relationship. Study participants were 984 college 

students from ten different universities in the 

U.S. These students participated in an 

anonymous survey on substance use and related 

risk and protective factors. Insomnia symptoms 

refer to problems falling asleep, problems staying 

asleep, and early morning awakening. These 

symptoms were measured by the Insomnia 

Severity Index (ISI) (Morin, Belleville, Bélanger, 

& Ivers, 2011). Protective behavioral strategies 

for marijuana were measured by the Protective 

Behavioral Strategies for Marijuana Scale 

(PBSM) (Pedersen, Huang, Dvorak, Prince, & 

Hummer, 2017). Cannabis use problems were 

measured by the Cannabis Use Disorders 

Identification Test Revised (CUDIT-R) (Adamson 

et al., 2010). Cut-off scores of 8 and 12 were used 

to indicate hazardous cannabis use and possible 

cannabis use disorder respectively. All analyses 

controlled for three demographic variables that 

have been shown to associate with substance use, 

i.e., gender, age and race. Insomnia symptoms 

were associated with an increase in the odds of 

hazardous cannabis use (OR=1.02, p<.05) and 

cannabis use disorder (OR=1.04, p<.01). Insomnia 

symptoms also had a significant relationship with 

PBSM (b=-.02(.01), =-.08, t=-2.51, p=.01). 

Severity of insomnia symptoms negatively 

predicted the use of protective behavioral 

strategies for marijuana. Controlling for insomnia 

symptoms and demographics variables, PBSM 

significantly predicted lower odds of hazardous 

cannabis use (OR=.50, p<.001) and cannabis use 

disorder (OR=.49, p<.001). PBSM significantly 

mediated the relationship between insomnia 

symptoms and hazardous cannabis use (95% 

ACI=.002 to .018, p<.05), as well as the 

relationship between insomnia symptoms and 

possible cannabis use disorder (95% ACI=.002 to 

.020, p<.05). We discussed the implications of 

these findings on the prevention of cannabis use 

problems among college students in the U.S. Past 

research showed that insomnia symptoms were 

associated with alcohol and drug-related 

problems. However, in those studies, the use of 

cannabis was not examined separately from other 

drugs. This study extended past research by 

showing that insomnia symptoms were associated 

with both hazardous cannabis use and possible 

cannabis use disorder. Additionally, protective 

behavior strategies significantly mediated the 

relationship between insomnia symptoms and 

cannabis use. Higher insomnia symptoms were 

associated with lower scores of protective 

strategies, which in turn was associated with 

higher cannabis consumption. The use of 

protective strategies (e.g., limit use to weekends, 

avoid using marijuana to cope with emotions such 

as sadness or depression) requires planning and 

thinking in advance before marijuana use. 

Problems falling or staying asleep may make it 

difficult for individuals to have the mental energy 

and self-control to engage in these strategies. 

Prevention and intervention programs of 

marijuana use could consider the potential impact 

of insomnia symptoms on protective behavioral 

strategies.  

 

Contact: Maria M. Wong 
wongmari@isu.edu 

 



Conference Abstracts  94 

 

SYMPOSIA 
 

Symposium: Co-Use of Marijuana with Alcohol 

and Tobacco among Young Adults: Current 

Research and Future Directions 

 

Chair: Adrian J. Bravo 

University of New Mexico 

 

Concurrent polysubstance use can be defined as 

the use of two or more substances within a given 

time period (e.g., over the past month or year) and 

can lead to dramatic health consequences due to 

the additive or interactive effects of combining 

substances. Within the present symposium, we 

feature five researchers who have examined 

diverse research questions on the concurrent use 

of marijuana with either alcohol or tobacco among 

young adults. First, Dr. Pedersen will present on 

findings examining the effects of differing co-use 

patterns (e.g., simultaneous use vs using both, but 

on separate occasions) of marijuana and tobacco 

and their effects on negative consequences. 

Second, Dr. Prince will present on findings 

exploring the relations among a variety of alcohol 

and cannabis co-use patterns (i.e., use of alcohol 

and cannabis on the same day; alcohol first; 

cannabis first; alcohol last; cannabis last; 

simultaneous use) and alcohol and cannabis 

related consequences (both separately and 

combined). Third, Dr. Jackson will present on the 

predictive utility of simultaneous 

alcohol/marijuana (SAM) motives as compared to 

motives for drinking and marijuana use on 

substance use. Fourth, Dr. Villarosa-Hurlocker 

will present on findings testing motivational 

models of alcohol and marijuana use problems 

among college students who endorsed both past-

month marijuana and alcohol use, and endorsed 

varying levels of social anxiety. Fifth, Dr. Bravo 

will present on findings comparing etiological 

models of protective behavioral strategies for both 

alcohol and marijuana among college student co-

users. Finally, Dr. White will discuss the clinical 

and research implications from these studies. 

 
Contact: Adrian J. Bravo 

ajbravo@unm.edu 
 

Co-Use of Marijuana with Tobacco and Nicotine 
Products among Young Adults 

 
Eric R. Pedersen, Joan S. Tucker, Rachana 

Seelam, & Elizabeth J. D’Amico  

RAND Corporation 

 

As an increasing number of states within the 

United States legalize marijuana for recreational 

sale and possession, it is important to understand 

if certain methods of use are associated with 

increased risk for heavy use and consequences 

among young adults. One manner by which 

marijuana is used is through co-use with tobacco 

and/or nicotine products (e.g., using marijuana 

and combustible cigarettes right after one 

another, vaping marijuana together with nicotine-

containing e-liquid in electronic cigarettes). There 

is a significant public health concern that changes 

in recreational marijuana policy may have the 

unintended consequence of increasing use of 

tobacco/nicotine products and impeding smoking 

cessation efforts among young people who are co-

using the products. However, to date, there is 

little information on methods of tobacco/nicotine 

and marijuana co-use, as well as how co-use may 

associate with negative outcomes among the 

young adult population. Existing research on the 

correlates and consequences of tobacco and 

marijuana co-use has been limited in that most 

studies in this area (1) were conducted at a time 

when these substances were almost exclusively 

smoked and do not reflect the recent proliferation 

in product types (e.g., electronic cigarettes, 

marijuana edibles) and methods of use (e.g., 

ingesting, vaporizing, dabbing) and (2) did not 

have the ability to differentiate between tobacco 

and marijuana concurrent use, simultaneous use, 

and co-administration. In this study, we examined 

the marijuana and tobacco/nicotine co-use 

behavior of a sample of over 1,200 young adult 

(mean age of 21 years old) marijuana users living 

in California. Participants completed an online 

survey about their use of marijuana and 

tobacco/nicotine products in the past year. The 

most prevalent co-use behaviors were using an e-

cigarette to vape marijuana, smoking a cigarette 

and then smoking a joint on the same occasion, 

and smoking a joint that contained both tobacco 

and marijuana. We compared marijuana users 

who reported no co-use of tobacco/nicotine 

products to marijuana users who reported co-use 



Conference Abstracts  95 

 
of marijuana and tobacco/nicotine as (a) 

concurrent use (i.e., using both, but on separate 

occasions), (b) simultaneous use (i.e., using one 

right after the other, such as smoking a marijuana 

joint and then smoking a combustible cigarette), 

and (c) co-administration use (i.e., using both 

through the same delivery system, such as 

smoking tobacco and marijuana together in a 

blunt). Compared to marijuana users who 

reported no co-use, marijuana users who reported 

concurrent use, simultaneous use, or co-

administration use reported more frequent 

marijuana use, had higher rates of screening for 

cannabis use disorder, and reported greater 

frequency of marijuana consequences. Co-

administration use emerged as the riskiest of the 

three co-use behaviors. Those who reported 

engaging in all three types of co-use were the most 

at-risk for frequent use and consequences. 

However, those who reported marijuana use 

without any co-use reported a higher quantity of 

marijuana use than those in the co-use groups. 

These findings help to illuminate both the 

prevalence and risks associated with co-use of 

marijuana and tobacco/nicotine products and can 

be used to inform policies for states considering 

regulation of marijuana and tobacco/nicotine 

products. 

 

Alcohol and Cannabis Co-Use and Consequences: 
The Role of Timing 

 

Mark A. Prince, Kristina T. Phillips, Michael M. 

Phillips, & Trent L. Lalonde  

Colorado State University & University of 

Norther Colorado 
 

Approximately one-fourth of U.S. college students 

report use of cannabis in the last month, one-third 

report heavy drinking, and up to 90% report co-use 

of both substances. Recent data shows that 

cannabis use among college students in U.S. states 

with legal recreational use is higher than in states 

without legal use. College students are an 

important sub-group to assess, as cannabis and 

alcohol use peak in young adulthood and can still 

have an impact on the developing brain. Both 

substances are associated with lower academic 

achievement, cognitive impairment, addiction, 

driving impairment and injury, and mental health 

effects. Past research has shown that simultaneous 

or concurrent use of cannabis and alcohol can lead 

to more negative consequences than use of either 

substance alone. Method: The present study 

examined college students (N = 632) who reported 

both alcohol and cannabis co-use across 2-

semesters of cross-sectional data collection. We 

explored the relations among a variety of alcohol 

and cannabis co-use patterns (i.e., use of alcohol 

and cannabis on the same day; alcohol first; 

cannabis first; alcohol last; cannabis last; 

simultaneous use), and alcohol and cannabis 

related consequences both separately and 

combined (i.e., a sum of the two scales) using Path 

Analysis. Results: Alcohol and cannabis 

consequences were negative binomial distributed 

when examined separately, and the combined 

consequences outcome was Poisson distributed. 

Patterns of significance and the direction of the 

effects were the same when alcohol and cannabis 

consequences were modeled separately or together 

(both using a composite score or separate variables 

in the same model). Using cannabis last was a 

robust positive predictor of alcohol, cannabis, and 

combined consequences. Using alcohol and 

cannabis on the same day, but not necessarily 

simultaneously, was a significant and negative 

predictor of alcohol, cannabis, and combined 

consequences. Surprisingly, simultaneous use was 

not associated with any of the three consequences 

outcomes. Using cannabis first negatively 

predicted alcohol consequences and was not 

associated with either cannabis or combined 

consequences. Similarly, using alcohol first was 

negatively associated with cannabis consequences 

but not associated with either alcohol or combined 

consequences. Finally, using alcohol last had a 

trend towards a positive effect on alcohol 

consequences and was not related to either 

cannabis or combined consequences. Discussion: 

The most salient clinical implication of the current 

study is that using cannabis after alcohol portends 

risk for both alcohol and cannabis related 

problems. While it is possible that there may be a 

manner of using both alcohol and cannabis on the 

same day that results in fewer consequences, 

future studies using person-centered and episode-

level approaches are needed to identify risky and 

safer patterns of co-use. Understanding how the 

timing of alcohol and cannabis co-use relates to 

alcohol and cannabis consequences provides 

insights that can be used in prevention and 

intervention efforts aimed at college students. 



Conference Abstracts  96 

 
Using Marijuana and Alcohol Together: 

Specificity in Substance Use Motives among 
College Students 

 
Kristina M. Jackson, Alexander W. Sokolovsky, 

Kerri L. Hayes, & Helene R. White 

Brown University & Rutgers University 

 

The majority of college students who use both 

alcohol and marijuana have used them 

simultaneously on at least one occasion, but the 

motivations underlying simultaneous alcohol-

marijuana (SAM) use have been understudied. 

Although SAM motives may overlap conceptually 

with alcohol and marijuana motives, the degree to 

which simultaneous use is uniquely driven by 

SAM motives is unclear. The purpose of this study 

is to examine the predictive utility of SAM 

motives as compared to motives for drinking and 

marijuana use. Additionally, we explore the 

moderating role of negative affect and peer 

descriptive norms on the relationships between 

SAM use and conceptually related coping motives 

and social motives, respectively. Past-year alcohol 

and marijuana users age 18-24 (N=1,390; 62% 

female; 69% White; 12% Hispanic) recruited from 

three college campuses in the U.S. completed a 30-

45-minute web survey assessing alcohol and 

marijuana use, motivations for use, affect, and 

peer descriptive norms at two time points 

separated by three months. Past 30-day alcohol 

use was reported by 93% of the sample and 75% 

reported past 30-day marijuana use, with at least 

one occasion of SAM use in the past 30 days 

reported by 71% of users of both alcohol and 

marijuana. The most frequently endorsed SAM 

motives were coping motives, social motives, 

desire to attain greater positive effects with 

combined use, and boredom. In a series of 

regression models we predicted past 30-day use of 

each type of substance from coping and social 

motives pertaining to each substance or SAM. 

Coping and social motives were significant 

predictors of use with strongest associations 

observed in models where there was 

correspondence between type of motive and type 

of substance use behavior. Further, accounting for 

SAM-specific coping motives, using marijuana to 

cope was still predictive of SAM, but using alcohol 

to cope was not; alcohol and marijuana social 

motives were not significant predictors of SAM 

when SAM-specific social motive were included 

the model. Finally, in a set of models testing 

interactions with affect and norms, the 

interaction between coping motives and negative 

affect was significant for alcohol, marijuana, and 

SAM use. Somewhat unexpectedly, coping 

motives were greater for those low on negative 

affect. There was also a significant interaction 

between social motives and friend norms such 

that the effect for social motives was most 

pronounced for those who reported SAM-using 

friends; the same effect (although less 

pronounced) was observed for marijuana use but 

was not significant for alcohol use. For the most 

part, use of alcohol and marijuana together 

appears to be driven by motives specific to 

simultaneous use. Additionally, the strength of 

these motivational influences depends on salient 

individual and contextual characteristics. Future 

work should examine the extent to which motives 

may indicate particular vulnerability for some 

young adults to established contextual 

antecedents of substance use. 

 

The Relationship between Social Anxiety and 
Alcohol and Marijuana Use Outcomes: A 

Motivational Model of Substance Use 
 

Margo Villarosa-Hurlocker, Adrian J. Bravo, 

Matthew R. Pearson, & Protective Strategies 

Study Team 

University of New Mexico 

 
Background: College students with social anxiety 

are particularly vulnerable to problematic alcohol 

and marijuana use given their susceptibility for 

elevated anxiety symptoms in social settings 

combined with the normative nature of substance 

use. Existing research has supported coping-

motivated substance use for these students when 

examining alcohol and marijuana use problems 

separately. The next step is to determine whether 

students with social anxiety who use both 

substances, do so for similar or different reasons. 

Objectives: The current study tested motivational 

models of alcohol and marijuana use in a sample 

of college students from 10 universities across the 

U.S. who endorsed both past-month alcohol and 

marijuana use. Method: Among college student 

dual alcohol/marijuana users (n = 2,034), the 

majority of participants identified as being either 

White, non-Hispanic (67.95%) or of 

Hispanic/Latino ethnicity (15.88%), female 



Conference Abstracts  97 

 
(69.08%), and reported a mean age of 20.24 (SD = 

3.16) years. Results: Path analyses revealed that 

negatively reinforcing drinking motives (coping 

with anxiety and conformity) significantly 

mediated the positive relationship between social 

anxiety symptoms and alcohol-related problems. 

Further, coping and expansion marijuana use 

motives significantly mediated the positive 

association between social anxiety and 

marijuana-related problems. Conclusions: Taken 

together, students with social anxiety who are 

focused on anxiety management may use either 

alcohol or marijuana; however, these students 

may be more inclined to drink to fit in with peers 

and use marijuana to expand their perceptual 

experiences. 

 

The Relationship between Distal Antecedents 
and Alcohol and Marijuana Use Outcomes among 

Dual Users: A Comprehensive Examination of 
Protective Behavioral Strategies 

 
Adrian J. Bravo, Andrew Weinstein, Matthew R. 

Pearson, & Protective Strategies Study Team 

University of New Mexico 

 

Background: Protective behavioral strategies 

(PBS) are behaviors that are used immediately 

prior to (e.g., setting a limit on consumption), 

during (e.g., using only in a safe context), and/or 

after substance use (e.g., using a designated 

driver) that reduce consumption, intoxication, 

and/or substance-related harm. Among college 

samples, increasing evidence suggests that both 

alcohol and marijuana PBS use are robust 

protective factors and have been shown to mediate 

the effects of known risk factors (i.e., gender, age 

of drinking onset, college substance use beliefs, 

drinking motives, and impulsivity-like traits) on 

alcohol (Bravo et al., 2015, 2016, 2017a) and 

marijuana outcomes (Bravo et al., 2017b). 

However, it is unknown whether PBS use would 

operate similarly for both substances among dual 

users. Objectives: In an extension of previous 

research, the present study examined which distal 

antecedents uniquely (i.e., controlling for other 

factors) relate to alcohol/marijuana outcomes via 

alcohol/marijuana PBS use and among a large 

group of college students who use both alcohol and 

marijuana. Method: Participants were college 

students that consumed both alcohol and 

marijuana at least one day in the previous month 

(n=2,034, 69.09% female). Across two models (one 

for each substance), proposed distal antecedents 

(i.e., gender, age of first use, college substance use 

beliefs, impulsivity-like traits, and motives) were 

modeled as predictors of negative consequences 

via PBS use and substance use consumption. 

Results: Across both models and controlling for 

the effects of all other predictors, PBS use was 

significantly negatively associated with all 

substance use outcomes. Across both substances, 

PBS use significantly mediated the associations 

between gender (women reported higher PBS 

use), age of first use (having an older age of first 

use was associated with more PBS use), and 

college substance use beliefs (higher beliefs was 

associated with lower PBS use) on all substance 

use outcomes. In the alcohol model only, alcohol 

PBS use significantly mediated the associations of 

both negative urgency and enhancement motives 

(both associated with lower PBS use) on each 

alcohol outcome. Within the marijuana model 

only, marijuana PBS use significantly mediated 

the associations between coping (associated with 

lower PBS use) and conformity motives 

(associated with more PBS use) on all marijuana 

outcomes including the double-mediated paths. 

Discussion: Our results suggest that PBS use is a 

good candidate to be considered as a mechanism 

by which dual alcohol/marijuana users moderate 

their substance use and attenuate their risk of 

experiencing related consequences. Based on the 

current sample, we can identify certain 

characteristics that might be particularly good 

targets for PBS based intervention efforts. For 

example, our results indicate that male college 

students and those with higher college substance 

use beliefs may benefit from increasing their PBS 

use for both substances. Moreover, several 

differences were found across substances and 

targeting specific characteristics for a particular 

substance may improve the efficacy of PBS 

interventions. 

 

Discussant: Helene R. White 
Rutgers University 

 



Conference Abstracts  98 

 
Symposium: Dealing with the Frequently 

Comorbid Condition of Depression and Cannabis 

Use Disorder in Adolescents 

 

Chair: Albert J. Arias 

Yale University 

 

This symposium will deal with the frequently 

comorbid conditions of depression and cannabis 

use disorder (CUD). Dr. Chris Hammond will 

present on the literature linking depression and 

CUD, including epidemiology, and including the 

possible pathophysiological basis for comorbidity. 

CUD in adolescents is associated with clinically 

significant depression. Reliable data estimating 

the prevalence of significant depressive symptoms 

and major depressive disorder (MDD) in this 

population are scarce, although one estimate 

based on literature review is about 20 to 30% of 

adolescents with CUD are depressed. Depression 

and co-occurring drug use disorder in adolescents 

are associated with higher rates of alcohol use and 

alcohol use disorders, as well as greater risk of 

suicidal behavior. Multiple pathways to 

developing depression and CUD are possible, and 

there is data suggesting that some adolescents 

develop CUD after becoming depressed, and some 

develop depression after CUD. Another important 

clinical question is whether or not having 

depressive symptoms foils the treatment of 

substance use disorder. The second talk by Dr. 

Albert Arias will review a secondary analysis of 

the Cannabis Youth Treatment study of N=600 

adolescents in treatment for CUD which asks that 

question and examines the relationship between 

depression and cannabis use over time in that 

sample. Lastly, Dr. Yifrah Kaminer will discuss 

research on treatment of CUD in adolescents, and 

will discuss how to deal with depression in this 

population, and with commentary on directions 

for further research. 

 

Contact: Yifrah Kaminer 
kaminer@uchc.edu 

 

The Link between Affective Disorders and 
Cannabis Use in Adolescents 

 

Christopher J. Hammond 

Johns Hopkins University 

 

This talk will review the scientific evidence 

linking cannabis use and mood disorders in 

adolescents. The talk will include epidemiology, 

as well as a review of longitudinal studies 

examining the risk of developing mood disorders 

with cannabis use. Pathways to developing dual 

diagnosis will be discussed. A clinical case 

presentation will be included. 

 

 

The Impact of Depressive Symptoms on 
Cannabis Use Treatment Outcomes in 

Adolescents 
 

Albert J. Arias 

Yale University 

 

Objectives: Depressive symptoms are common in 

adolescents with cannabis use disorder (CUD), 

and can complicate assessment and treatment. A 

question remains about whether depressive 

symptoms interfere with cannabis use treatment. 

This is a report of a retrospective analysis of a 

data set where we examine the response of 

depressive symptoms to treatment of only the 

CUD. Method: N= 600 adolescents (age 12-18) 

with CUD received three months of active 

treatment for substance use with one of five types 

of psychotherapeutic psychosocial treatment. 

None of these treatments was aimed specifically 

at treating depressive symptoms. Self-reported 

number of days having used cannabis in the past 

90, as well as self-reported level of depression was 

assessed at Baseline (BL) and again at 3-, 6-, 9-, 

and 12-months. A dual change Latent Change 

Score (LCS) model was used to examine the 

trajectory of frequency of cannabis use and 

depressive symptoms. Results: Rates of major 

depressive disorder (MDD) and any depressive 

symptoms at BL were 18% and 70.2% 

respectively. Depressive symptoms improved over 

the course of treatment, as did cannabis use 

(reduction). Greater baseline depression was a 

significant associated with attenuated 

improvement in subsequent cannabis use (less 

improvement for those with greater depressive 



Conference Abstracts  99 

 
symptoms). Change in cannabis was not 

associated with subsequent change in depression, 

suggesting that the improvement in depressive 

symptoms was not specifically due to the 

improvement in cannabis use. Conclusion: In this 

cohort of adolescents with CUD, depressive 

symptom severity improved over time, however 

there was a measurable impact of depression on 

the treatment effect for cannabis use. 

 

Research on the Treatment of Youth with 
Cannabis Use Disorder and Comorbid Conditions 

 
Yifrah Kaminer 

University of Connecticut Healthcare Center 

 

Dr. Kaminer will describe his work in developing 

treatments for adolescents with CUD, as well as 

its common dual diagnoses of depression and 

anxiety disorders. Also reviewed will be the 

common strategies for clinical treatment in this 

population. Additionally, directions for future 

research will be reviewed. 

 

Discussant: Yifrah Kaminer 
University of Connecticut Healthcare Center 

 

Symposium: Rehashing Approaches to 

Marijuana-Related Consequences: A Budding 

Area of Research 

 

Chair: Jamie E. Parnes 

Colorado State University 

 

Understanding physical and mental health risks 

of marijuana use is critical to harm reduction 

interventions. As such, research has extensively 

focused on understanding factors associated with 

negative consequences of use. While many 

antecedents of harm have been identified, there 

remain unexamined factors that may account for 

significant variance in negative consequences. 

Moreover, most literature to-date has measured 

consequences as a count of the number of negative 

consequences experienced. However, other 

consequence-related factors, such as the severity 

of reported consequences, or positive 

consequences, have yet to be examined. Other 

unexamined influences on negative consequences, 

such as price of marijuana, may help inform 

interventions. Four symposium talks will address 

unexamined factors and alternative ways of 

examining consequences of marijuana use. The 

first talk will focus on incorporating the 

marijuana purchasing task into understanding 

the relation between marijuana use motives 

(MUMs) and number of negative marijuana-

related consequences experienced. This study 

examined the influence of marijuana price 

breakpoint (i.e., price at which one no longer 

purchases marijuana) as a potential mediating 

variable. The second talk examined how use 

relates to negative consequence count, 

consequence severity, and individual differences. 

The third talk examined if classes based on MUMs 

were associated with marijuana-related negative 

consequences and severity. The final talk will 

discuss the relation between MUMs and positive 

consequences of use. The four talks will provide 

greater understanding of marijuana consequences 

from various perspectives. Results may inform 

harm reduction interventions seeking to reduce 

negative consequence number and severity, as 

well as increase positive outcomes. 

 

Contact: Jamie Parnes 

Jamie.parnes@colostate.edu 
 

A Behavioral Economic Approach to Analyzing 
the Relations between Marijuana Use Motives 
and Marijuana Related Consequences in Young 

Adults 
 

Kirstyn N. Smith-LeCavalier, Samuel R. Davis, 

& Mark A. Prince 

Colorado State University 

 

As the landscape of marijuana changes with 

greater social acceptance, prevalence of use, and 

more permissive legislation, the demand for a 

deeper understanding of the factors that relate to 

marijuana misuse and related negative 

consequences is increasing. Previous studies have 

described the direct positive association between 

marijuana use motives (MUMs) (i.e., social, 

coping, enhancement, conformity, expansion), 

marijuana use, and marijuana-related 

consequences (MRC). Moreover, MUMs account 

for unique variance in marijuana related 

consequences that is not fully explained by 

marijuana use alone. Recently, purchase tasks 

have been used to establish economic demand for 

marijuana use. The marijuana purchase task 

(MPT) assesses estimated consumption of 



Conference Abstracts  100 

 
marijuana at specific prices per “hit” of 

marijuana. From a behavioral economics view, 

MPT has been used to demonstrate a negative 

association between demand for marijuana and 

price. Additionally, greater demand for marijuana 

and lower sensitivity to price fluctuations have 

been observed in persons who display marijuana 

dependence symptoms. However, little research 

has examined the MPT in relation to psychological 

constructs such as motives and negative 

consequences. This study focused on whether 

breakpoint (i.e., cost where individuals are no 

longer willing to purchase marijuana) mediated 

the relation between MUMs and MRCs. The 

present study is a secondary data analysis of the 

initial Marijuana Outcome Study Team (MOST) 

data collection. The analytic sample consisted of 

the young adults who reported using marijuana in 

the past 30 days (n = 2044). All study hypotheses 

were tested using path analysis. Number of 

consequences was specified as a negative 

binomially distributed count variable. Monte 

Carlo Confidence Intervals will be used to assess 

the significance of the indirect effects in each 

model. Results indicated that when controlling for 

other motives, enhancement, coping, and 

expansion motives positively predicted MPT 

breakpoint. Furthermore, controlling for MUMs 

and breakpoint, all MUMs were associated with 

MRC. With regard to the direct effects of the 

individual models, social motives are negatively 

related to MRC. In addition, coping, 

enhancement, conformity, and expansion motives 

are positively associated with experiencing 

negative consequences. Findings for the indirect 

effects suggest that breakpoint may partially 

mediate the relation between social, 

enhancement, expansion, and coping motives. 

Results from the current study can contribute to 

the understanding of factors that are associated 

with marijuana demand characteristics and 

purchasing behaviors among college student 

marijuana users. 

 

Negative Consequence Severity: What’s the 
Worst That Can Happen? 

 
Theodore J. Fetterling, Jamie E. Parnes, Samuel 

R. Davis, Mark A. Prince, & Bradley T. Conner 

Colorado State University 

 

Marijuana use rates are highest among college-

aged individuals for both annual (41%) and past 

30-day (25%) use compared to all other age 

groups. Additionally, marijuana use has been 

linked to a variety of negative consequences (e.g., 

feeling sluggish, dependence). While the 

prevalence of negative consequences related to 

marijuana use has been widely studied, research 

has yet to examine the perceived severity of 

marijuana-related consequences. Although 

heavier users report experiencing significantly 

more marijuana-related consequences, they are 

not more likely to anticipate future consequences 

than individuals with low use. Thus, 

consideration of consequence severity is 

warranted. Studies have shown perceived 

consequence severity can decrease likelihood of 

engagement in various health risk behaviors. 

Further, sex differences have been noted for 

engagement in risky activities. The present study 

has two aims. The first is to test the relations 

between recent marijuana use, endorsement of 

negative consequences, and reported negative 

consequence severity. The second aim is to explore 

potential descriptive differences (e.g., sex, age) on 

consequence severity ratings. We hypothesized 

that past 30-day marijuana use would predict 

greater average consequence severity and highest 

severity consequence, and that number of 

consequences endorsed would mediate this 

relation. We also explored descriptive differences 

among severity ratings of consequences. College 

student marijuana users (n = 695) completed a 

survey containing items from the Marijuana 

Consequences Questionnaire (MACQ) with an 

associated 5-point Likert scale to rate 

consequence severity (“minimally negative” to 

“extremely negative”). Hypotheses were tested 

using path analysis. Number of consequences 

endorsed was a negative binomially distributed 

count variable and Monte Carlo Confidence 

Intervals were used to test the indirect effects. 

Findings indicated that the direct effect of past 30-

day marijuana use significantly predicted 

increased average and highest consequence 



Conference Abstracts  101 

 
severity ratings. When including number of 

endorsed consequences as a mediator, significant 

indirect effects were detected for both average, 

and highest consequence severity. Exploratory 

analyses were used to test for differences in 

perceived consequence severity among descriptive 

variables. Results suggest the effects of past 30-

day marijuana use on perceived consequence 

severity are transmitted through number of 

endorsed consequences. We expected the 

consequences rated as more negative to vary by 

sex as well as other demographic variables. The 

current study provides preliminary evidence that 

marijuana users who endorse greater negative 

consequences are likely to experience higher 

severity of negative consequences. The combined 

findings may inform interventions aimed at 

reducing specific marijuana-related 

consequences, informed by demographic 

differences, to attenuate the frequency and 

severity of experienced consequences. 

 

Marijuana Use Motives Relation to Frequency 
and Severity of Negative Consequences 

 
Samuel R. Davis, Mark A. Prince, & Bradley T. 

Conner 

Colorado State University 

 

Marijuana use motives (MUMs) are positively 

associated with marijuana use and marijuana-

related negative consequences. In particular, 

motives to enhance positive affect are associated 

with greater use and fewer consequences, while 

motives to cope are associated with greater 

frequency of negative consequences. However, no 

research has examined the relation between 

MUMs and severity of these negative 

consequences. This study examines MUMs and 

their relation to frequency and severity of 

marijuana-related negative consequences. This 

study is a secondary data analysis of college-aged 

marijuana users at a large public University (N = 

668, Mage = 20.40). A series of Latent Class 

analyses (LCAs) were conducted to discern latent 

classes of MUMs. Once latent classes were 

identified, differences among latent classes of 

frequency of negative consequences experienced 

in the last 30 days, frequency of consequences in 

a typical week of use, frequency of consequences 

in a heavy week of use, and severity of negative 

consequences experienced in the last 30 days were 

tested using the BCH method in MPlus version 8. 

A four-class solution provided the best overall 

model fit to the data. Class 1, labeled as the 

enhancement class (ENH), was characterized by 

higher enhancement motives and low conformity, 

coping, social, and expansion motives. Class 2 was 

characterized by low MUMs and was labeled the 

low motives class (LM). Class 3, labeled as the low 

conformity class (LCON), was characterized by 

high enhancement, coping, social, and expansion 

motives and low conformity motives. Class 4 was 

characterized by moderate marijuana use motives 

and was labeled the moderate motives class (MM). 

The BCH test was significant for past 30-day 

consequences, typical week consequences, heavy 

week consequences, and severity of consequences. 

Comparisons revealed that the LCON and MM 

classes reported greater severity of negative 

consequences, and greater frequency of negative 

consequences compared to the ENH and LM 

classes. The ENH class reported significantly 

lower severity of negative consequences than the 

LM, LCON, and MM classes. Results suggest that 

among college marijuana users, profiles for 

moderate-to-strong overall motivation to use 

marijuana are associated with greater negative 

consequences and severity of those consequences 

compared to low overall motive users. However, a 

profile of just high enhancement users is 

associated with less severity of the negative 

consequences compared to low, moderate, and 

high overall motive users. Future interventions 

should target persons with greater coping, social, 

and expansion motives compared to persons with 

greater enhancement motives when working to 

attenuate severity and frequency of marijuana-

related negative consequences. 

 

Motivated to Feel Good: Marijuana Motives 
Predict Positive Consequences of Use 

 
Jamie E. Parnes  

Colorado State University 

 

Past research has extensively examined negative 

consequences of marijuana use. However, despite 

experiencing negative consequences, many 

individuals continue using. Therefore, it is 

important to understand the factors that 

maintain use. One of these factors is positive 

consequences of use, such as activity 

enhancement, euphoria, and stress reduction. 



Conference Abstracts  102 

 
Another influencing factor is marijuana use 

motives. Motives for use, including social, coping, 

enhancement, conformity, and expansion motives, 

all uniquely relate to negative consequences of 

use. However, it is not known how these motives 

relate to positive consequences. Each motive may 

uniquely relate to positive outcomes. Motives 

associated with greater positive outcomes likely 

become more reinforced over time than other 

motives. The present study seeks to examine how 

motives predict positive consequences of 

marijuana use. Given the lack of previous 

research, exploratory analysis was conducted to 

examine these relations. Undergraduate 

marijuana-using students (N = 676) completed a 

survey measuring past 30-day marijuana use, 

marijuana use motives, and positive consequences 

of use. Multiple linear regression was conducted 

with all five marijuana use motives predicting 

positive outcomes of use, controlling for past 30-

day use. Given the multiple comparisons made in 

this study, alpha was set to .01 to reduce 

likelihood of type 1 errors. Results indicate that 

coping motives, enhancement motives, expansion 

motives, and past-30 day use positively predict 

positive consequences of use. Additionally, 

conformity motives negatively predict positive 

consequences. Social motives were unrelated to 

positive consequences. Overall, results support 

that positive outcomes experienced by users vary 

by use motive. Those using to conform with peers 

experience fewer positive consequences, possibly 

as these users may feel extrinsic pressure to use 

rather than an intrinsic desire. Alternatively, 

those using for their own desires, such as coping, 

enhancement, or expansion, experience greater 

positive outcomes as they may be using for 

internally motived reasons. Moreover, positive 

experiences related to specific motives may 

reinforce continued use for certain motives rather 

than others. For example, positive consequences 

from use to cope with negative affect may 

reinforce using to cope. However, using to cope 

also predicts greater negative consequences of use 

that pose various health risks. Harm reduction 

interventions seeking to maximize the benefits of 

use may target marijuana use motives. 

Intervening on those using for conformity motives 

may help reduce use when it is likely to lead to 

fewer positive consequences. Future research 

should continue to examine positive consequences 

as both an outcome and predictor of continued use. 

Doing so may help increase understanding of how 

to help users maximize the benefits of use and 

reduce occasions associated with fewer positive 

consequences. 

 

Discussant: Jamie E. Parnes  
Colorado State University 

 

Symposium: Brain Imaging in Relation to both 

Acute and Chronic Cannabis Use 

 

Chair: Godfrey Pearlson 

Yale University 

 

Consequent on widespread medical marijuana 

legalization and recreational marijuana 

decriminalization/legalization, there is a growing 

need to understand cannabis’ acute and longer-

term effects on brain function and behavior, many 

facets of which are under-documented. This 

symposium will examine brain and behavioral 

consequences of acute cannabis use relevant to a 

variety of different contexts by describing 

emerging findings from our studies that employ 

task-based fMRI, resting state fMRI, and proton 

spectroscopy. To examine cognitive and brain 

consequences following acute use, we were 

interested in how single doses of vaporized 

cannabis impact behavioral performance on time-

estimation and three separate measures of 

simulated driving behavior that represented 

different crucial aspects of motor vehicle 

operation. In particular, we wanted to document 

the time courses of these behavioral and brain 

changes, their relationship to administered dose 

and blood levels of THC and its metabolites and 

any differences attributable to baseline marijuana 

use frequency. Because of concerns about the 

drug’s effect in younger individuals, particularly 

on memory function, we were especially 

interested in measuring brain metabolic effects 

using proton spectroscopy in relationship to 

robust functional MRI assessment of memory 

encoding. Overall, this symposium will attempt to 

showcase novel uses of MRI-based neuroimaging 

in the context of human marijuana neuroscience. 

 
Contact: Godfrey Pearlson 
godfrey.pearlson@yale.edu 

 



Conference Abstracts  103 

 
Alterations in Resting State Functional MRI 

Connectivity Following Acute Doses of Cannabis 
. 

Shashwath Meda  

Olin Research Center, Institute of Living 

 

Resting state functional MRI (rs-fMRI) examines 

how activation in numerous distal brain regions 

engage in synchrony over time in the absence of 

formal experimentally-driven performance 

demands.  As such, rs-fMRI characterizes 

frequently engaged-neural systems and can 

quantify many different types of network 

connectivity abnormalities in those circuits.  rs-

fMRI have not only been used to characterize 

physiological perturbations such as drowsiness 

and a large number of different neuropsychiatric 

illnesses such as schizophrenia and Alzheimer’s 

disease, but fs-fMRI has demonstrated its 

sensitivity to acute intoxication with a variety of 

medications.  Numerous studies suggest the 

sensitivity to THC, but the evidence comes 

indirectly from studies that examine samples with 

different types of regular THC use, dependence, or 

abstinence.  There is an absence of cannabis 

challenge studies that directly characterize the 

acute effects of THC intoxication or its persistence 

over time. As part of a NIDA-funded study, we 

examined the differences in rs-fMRI between 

placebo marijuana (MJ) and two doses of active 

drug (0.5 G of ~6% and ~13% THC -containing 

floral MJ) administered using a Volcano vaporizer 

in N = 14 subjects. Data were acquired from 6m 

43s of resting fMRI signal acquired on a 3-T MRI 

scanner. Data were preprocessed using the 

minimal preprocessing guidelines prescribed by 

the Human Connectome Project (HCP) that 

include distortion correction, surface generation, 

cross-modal registration and spatial alignment. 

Low frequency drift was then removed from the 

preprocessed data and analyzed using the Group 

ICA toolbox to derive temporally coherent 

intrinsic resting networks. Within-network 

spatial differences in resting connectivity across 

different doses and over time within sessions were 

assessed using a repeated measure ANCOVA 

design (adjusted for age, sex and motion) 

implemented in the PALM package. We will 

present preliminary results from analyses that 

quantify functional connectivity within the nodes 

of brain systems that are highly reproducible 

across people and different task contexts (e.g., the 

frontoparietal executive network, cingulo-

opercular salience network, etc.), as well as ICA-

identified neural systems that integrate those 

brain regions that have a high density of CB 

receptors (e.g., striatal, cerebellum).  Results are 

pending and will be discussed at the time of the 

symposium. 

 

 

Effects of Chronic Cannabis Use on Brain Proton 
Spectroscopy Measures, and Working Memory 

fMRI  
 

Alecia Dager 

Yale University 

 

Emerging adults show the highest rates of heavy 

marijuana use. Marijuana use is associated with 

memory impairments and altered functional 

magnetic resonance imaging (fMRI) response, but 

the neurochemical underpinnings remain 

unclear. To better understand the relationship 

between marijuana use and memory dysfunction, 

we collected information on hippocampal and 

frontal lobe neurochemistry, brain activation 

levels during an active memory encoding and 

recognition fMRI task, and performance on 

neurocognitive measures of verbal list learning in 

current heavy marijuana users and controls, ages 

18-22.  The California Verbal Learning Task 

(CVLT) assessed verbal learning. During fMRI, 

participants performed an object-pair encoding 

and recognition task (Ragland et al., 2012). 

Magnetic resonance spectroscopy yielded N-

acetylaspartate (scaled to total creatine, 

NAA/tCr), a marker of neuronal integrity, in left 

hippocampus and left prefrontal cortex.  

Marijuana users performed worse than controls 

during CVLT immediate recall (t=2.28, p=0.049), 

and trended toward poorer CVLT delayed recall 

(p=0.051). Groups did not differ on frontal or 

hippocampal fMRI response during encoding or 

recognition. However, left hippocampal response 

during encoding correlated with better CVLT 

delayed recall (rho=0.991, p=0.001). Although 

there were no group differences in NAA/tCr, 

groups differed in their relationship between 

NAA/Cr and fMRI response in hippocampus 

(t=2.5, p=.042). Here, controls demonstrated a 

trend for a positive relationship (t=2.5, p=.069), 

whereas marijuana users demonstrated no 

relationship.  This within-subject comparison of 



Conference Abstracts  104 

 
fMRI and magnetic resonance spectroscopy 

provides preliminary evidence of altered 

relationships between hippocampal 

neurochemistry, memory-related fMRI response, 

and verbal learning. Ongoing data collection will 

further characterize these relationships, 

providing critical insight into the neural impact of 

marijuana use in this vulnerable population. This 

work has important implications as marijuana 

use increases with legalization and 

medicalization. 

 

Acute Cannabis Exposure Alters Behavior and 
Brain Activation on an fMRI Time Estimation 

Task 
 

Michael Stevens 

 Olin Research Center, Institute of Living 

 

Time estimation is impaired by acute cannabis 

use.  As part of an ongoing NIDA-funded study 

(R01DA038807), fMRI data were collected from 

n=10 regular MJ users. On three separate days, 

participants used a vaporizer and paced 

inhalation method to smoke marijuana, randomly 

receiving 0.5gm of either NIDA herbal 13.4% 

THC, 5.9% THC, or placebo cannabis.  On each 

visit, participants were administered drug using 

paced inhalation by 9:00 a.m., then underwent 

fMRI 3 separate times post-dose (1½ h, 3½ h and 

5½ h), in a randomized, double-blind, counter-

balanced design across visits.  After dosing, 

participants underwent fMRI while performing a 

rapid, event-related time estimation task, where 

they were presented with a sequence of boxes and 

were instructed to decide which box was on the 

screen longest.  The task presented short (500-900 

msec) or long (1500-1900 msec) stimuli durations 

within either of these ranges to assess drug 

impact on both neural systems.  Both Short and 

Long intervals for all 9 fMRI runs for each 

participant were included in analyses to 

determine the extent of regional activation 

regions during the task. As expected from prior 

studies, time estimation task performance 

engaged bilateral cerebellum and putamen, plus 

an extended set of bilateral cortical regions 

comprising anterior insula, inferior frontal gyri, 

middle frontal gyri (BA 9/46), and dorsal anterior 

cingulate and medial frontal gyrus.  Short interval 

time estimation elicited greater cerebellum 

activity, whereas lateral and medial prefrontal 

regions were more engaged during Long interval 

estimation.  An analysis of reaction time (RT) and 

accuracy found significant changes over time 

since THC use only for Short interval RT for the 

moderate MJ dose condition.  Here, RT decreased 

with continued recovery from drug use (T1 vs. T3 

p = .029).  As behavioral effects were greater for 

Short interval trials, we focused fMRI analyses on 

this condition.  The predominant effect of recent 

THC use was to reduce brain activation.  Common 

to both drug doses, less activity was elicited in 

basal ganglia and frontoparietal network regions. 

Cerebellum was relatively unaffected in by 

moderate, but was hypo-functional following low 

dose.  There were different dose effects in SII 

somatosensory processing regions.  A large effect 

size change was observed in dorsal anterior 

cingulate (dACC), where over time the amount of 

activation decreased for low dose days (xyz = 0, 32, 

30, Cohen’s d = 1.57).  In addition, throughout low 

dose days, more BOLD signal change was noted in 

left putamen and left amygdala.  By 5½ h post-

drug, right putamen activation had not yet 

returned to normal in either the low or moderate 

dose conditions.  Interestingly, after 5 ½ h post-

MJ use, activation in frontoparietal regions 

consistent with the dorsal attention network 

showed greater activation relative to the 

comparable placebo fMRI for both drug doses.  

Overall, MJ use resulted in relative 

hypoactivation compared to placebo for striatal 

brain regions known to play an important role in 

mental timekeeping cognitive processes, as well 

as for several distributed neural systems engaged 

for higher-order cognition. 

 
Disruption of Brain Circuits Involved in a 

Simulated Driving fMRI Task Following Acute 
Cannabis Exposure 

 
Godfrey Pearlson 

Yale University 

 

Although numerous studies provide evidence that 

recent marijuana use can impair performance on 

tests of cognitive abilities thought to be important 

for optimal motor vehicle operation, there is little 

understanding of exactly how the drug affects the 

brain to cause such impairments.  fMRI data were 

collected from n=12 regular marijuana users. On 

three separate days in a randomized, double-

blind, counter-balanced design across visits, 



Conference Abstracts  105 

 
participants used a vaporizer and paced 

inhalation method to smoke marijuana, randomly 

receiving 0.5gm of either NIDA herbal 13.4% 

THC, 5.9% THC, or placebo cannabis. Although 

the overall protocol collected fMRI data 3 separate 

times following marijuana use at 9:00 a.m., here 

we focused on fMRI data collected 1½ hours post-

dose to observe dose effects related to virtual 

driving performance not only after peak THC 

intoxication.  Using Real Time, Inc. realistic 

virtual driving simulation software, multiple 

instances of different driving demands were 

naturalistically embedded into three ~10-minute 

fMRI driving tasks administered during each in-

scanner drive.  Event onsets were extracted, then 

modeled to create activation maps for each 

condition.  We contrasted study doses in repeated-

measures statistical comparisons to identify 

which brain regions had greater or lesser BOLD 

signal response for the events in that task relative 

to the implicit baseline formed by the remainder 

of the timeseries for each paradigm:  Gap 

Acceptance is a strategic planning task where 

participants decide when to accelerate from a stop 

to overtake a parked car by merging into a lane of 

oncoming traffic. The task engaged diverse 

prefrontal cortex regions within both 

frontoparietal executive and ventral attention 

networks, disengaged motor planning regions, 

lateral orbitofrontal cortex, and regions within 

dorsal attention and default mode networks.  Car-

Following measures tactical decisions when 

participants respond to the 

acceleration/deceleration of a lead car that 

pseudo-randomly alters speed.  The task elicited 

greater activation in motor planning/execution 

brain regions, motor anterior cingulate, posterior 

dorsal attention network and right putamen.  The 

caudate disengaged to these events, as did 

bilateral SII somatosensory cortex.  Lane Keeping 

assesses operational vehicle control where 

participants maintain car position despite 

unpredictable wind gusts that elicit Vehicle 

correction. The task elicited greater left 

hemisphere lateral prefrontal cortex activation in 

superior frontal sulcus, mid-dlPFC, anterior 

vlPFC and bilateral cerebellum and relative 

decreases in activity in right caudate, dorsal 

cingulate and bilateral precentral gyri.  

Marijuana use altered brain function in every 

driving context.  Across all driving tasks, bilateral 

putamen was less engaged by active study drug.  

All other effects were diverse and differed by THC 

dose. During Gap Acceptance, both doses showed 

extensive right hemisphere frontoparietal deficits 

and lower anterior cingulate cortex activation.  

For Lane Keeping, supplemental motor area and 

secondary visual cortex activity were reduced 

after both low and moderate MJ doses, as were 

greater right dorsolateral, ventrolateral, and 

ventromedial prefrontal cortex activity.  For Car 

Following, putamen deficits were dose specific, 

with higher doses linked to lower activation.  

Other dose-specific effects included bilateral 

precentral gyri and left frontoparietal cortex 

deficits.  After both doses, there was greater 

activity in visual association, motor, premotor, 

and supplementary motor cortices. 

 
Discussant: Godfrey Pearlson 

Yale University 
 

Symposium: Measuring Marijuana: Assessment 

Challenges and Our Partial Solutions 

 

Chair: Matthew R. Pearson 

University of New Mexico 

 

Every known drug of abuse has dose-dependent 

effects. In the context of increased medicalization 

and legalization of marijuana (or cannabis) in the 

United States and throughout the world, there is 

a rising pressure to quantify the risks and benefits 

of marijuana use. However, the crude assessment 

of marijuana use is at best obscuring and at worst 

preventing our ability to detect the associations 

between marijuana use and important outcomes. 

A review of the marijuana literature reveals that 

there are two primary ways that marijuana (or 

cannabis) use is assessed: 1) user status (a 

comparison of marijuana users and non-users 

with various time windows: past month, past 

year, lifetime, etc.) and 2) marijuana use 

frequency (i.e., number of days used in the past 30 

days). Unfortunately, marijuana use quantity is 

not typically measured and there is no clear gold 

standard for assessing marijuana use quantity. 

The lack of a gold standard measure of marijuana 

is not befuddling. There are several challenges 

with assessing marijuana use. Marijuana is 

available in many different preparations 

including flower, concentrate, and edible 

preparations. These products can have different 

routes of administration (e.g., smoking, vaping, 



Conference Abstracts  106 

 
eating) and there is not a standard apparatus for 

most of these (e.g., bowl/bong sizes vary). Even 

within each preparation, potency of marijuana 

varies significantly. In the present symposium, 

each presentation will measure marijuana use in 

a unique way to overcome some of the limitations 

mentioned above, and we will discuss how the 

field can improve upon the measurement of 

marijuana use. 

 
Contact: Matthew R. Pearson 

mateo.pearson@gmail.com 
 

Marijuana Use Grid: A Brief, Comprehensive 
Measure of Marijuana Use 

 

Matthew R. Pearson  

University of New Mexico 

 

The present study introduces a brief, yet 

comprehensive retrospective self-report measure 

of frequency and quantity of marijuana use: the 

Marijuana Use Grid (MUG). A large sample of 

college student marijuana users were recruited 

from 8 universities throughout the United States 

(total sample n = 6,584, past month marijuana 

users n = 1,969), including states that reflect each 

of the categories of legal marijuana: 1) no legal 

marijuana, 2) legal medical marijuana, and 3) 

legal medical and recreational marijuana. Using a 

weekly grid, the MUG asks participants to report 

the amount of marijuana that they use (in grams) 

during each of 6 4-hours blocks of time each day 

(12a-4a, 4a-8a, 8a-12p, 12p-4p, 4p-8p, 8p-12a) 

during a typical week of marijuana use (Monday 

– Sunday). From this measure, we can calculate a 

large number of variables including frequency 

(i.e., counting all non-zero values) and quantity 

(i.e., summing number of grams) of marijuana use 

for each day of the week, for each time of day, and 

for the entire week. We used Spearman rank 

correlations to examine the associations between 

these marijuana use measures and marijuana 

consequences and symptoms of cannabis use 

disorder (CUD). We found moderate associations 

between overall marijuana frequency and 

consequences/CUD symptoms (strongest r = .546 

for CUDIT-R, weakest r = .326 for CAST). We also 

found moderate associations between overall 

marijuana quantity and consequences/CUD 

symptoms (strongest r = .466 for CUDIT-R, 

weakest r = .274 for CAST). To explore whether 

frequency and/or quantity of use on particular 

days or during particular time periods were most 

predictive of consequences, we conducted four 

stepwise regressions: 1) day of week frequency, 2) 

time of day frequency, 3) day of week quantity, 

and 4) time of day quantity. Week-end use (i.e., 

Friday/Saturday) tended to be most predictive of 

consequences, but time of day analyses were 

inconsistent across frequency and quantity 

estimates. Lastly, to test for incremental validity, 

we examined with MUG frequency or quantity 

measures predicted consequences above and 

beyond the effect of a simple past month frequency 

measure. Although marijuana use frequency 

assessed by the MUG predicted consequences 

above and beyond the effect of a simple past 

month frequency measure, marijuana use 

quantity did not. This finding suggests either that 

quantity is less important than frequency in the 

determination of marijuana consequences, or that 

significant error in the assessment of quantity 

diminishes our ability to detect such 

relationships. Although improvements on 

marijuana use measures need to continue, the 

MUG has shown utility in its association with 

important outcomes and given its brief nature, the 

MUG can easily be integrated in future marijuana 

studies. 

 
Quantifying Cannabis: A Field Study of 

Marijuana Quantity Estimation 
 

Mark A. Prince, Bradley T. Conner,  

& Matthew R. Pearson 

Colorado State University & University of New 

Mexico 

 

The assessment of marijuana use quantity poses 

unique challenges. These challenges have limited 

research efforts on quantity assessments. The 

goal of the present study was to examine regular 

and heavy marijuana users’ ability to estimate 

quantity (e.g., weight in grams) of both flower and 

concentrated marijuana to narrow the gap 

between tightly controlled lab studies and survey-

based observational studies. This study had four 

primary aims: 1) determine the typical dose (i.e., 

grams used) for two common types of marijuana 

products (i.e., flower and concentrates), 2) 

determine the relative accuracy of marijuana 

users’ estimations of quantities of marijuana, 3) 

identify salient predictors of quantity of 



Conference Abstracts  107 

 
marijuana prepared for a single dose, and 4) 

identify salient predictors of estimation 

(in)accuracy. We recruited a sample of 128 

regular-to-heavy marijuana users (48% female) 

for a field study wherein they prepared and 

estimated quantities of marijuana flower in a joint 

or a bowl as well as marijuana concentrate using 

a dab tool. For Aim 1, we found that participants 

typically packed .25 grams of marijuana into a 

bowl, rolled about .58 grams into a joint, and 

loaded about .08 grams of concentrate using a dab 

tool. For Aim 2, we found that the vast majority of 

participants overestimated the quantity of 

marijuana that they used in their preparations 

(71% - 81% across preparations). Effect sizes 

indicate these overestimations were medium-to-

large in size (.557 < ds <. 823). In terms of 

proportional difference, the estimates of 

concentrated marijuana were the most inaccurate 

(estimated quantity was 113.2% higher than 

actual quantity) followed by flower in bowls 

(estimated quantity was 68.3% higher than actual 

quantity) and flower in joints (estimated quantity 

was 37.2% higher than actual quantity). For Aim 

3, we used Spearman’s rank correlations and 

found that males used significantly more 

marijuana flower in their preparations than 

females (ρ= .259), an effect driven by their 

tendency to prepare larger joints (ρ = .463). 

Individuals who reported being a recreational 

user only (compared to individuals who endorsed 

being a medical user, both a recreational and 

marijuana user, or a CBD only user) used 

significantly more marijuana flower generally, 

significantly more flower in a joint, marginally 

more flower in a bowl, and significantly more 

concentrate (.348 < ρs < .587). Compared to 

individuals not employed in the cannabis 

industry, individuals employed in the cannabis 

industry also used significantly more flower in a 

joint, in a bowl, and more concentrate (.267 < ρs < 

.412). For Aim 4, we used Spearman’s rank 

correlations and failed to find robust predictors of 

estimation accuracy. Self-reported quantity 

estimates are inaccurate, which has implications 

for studying the link between quantity and 

marijuana use outcomes. 

 

Introducing the Daily Sessions, Frequency, Age 
of Onset, and Quantity of Cannabis Use 

Inventory (DFAQ-CU) 
 

Carrie Cuttler, Alexander Spradlin, & Protective 

Strategies Study Team 

Washington State University 

 

We created the Daily Sessions, Frequency, Age of 

Onset, and Quantity of Cannabis Use Inventory 

(DFAQ-CU) because the current lack of 

psychometrically sound inventories for measuring 

these dimensions of cannabis use has impeded 

research on the effects of cannabis in humans. In 

Study 1, we recruited a sample of 2,062 cannabis 

users to assess the DFAQ-CU’s factor structure 

and reliability. To assess validity, a subsample of 

645 participants completed additional measures 

of cannabis dependence and problems. A six-factor 

structure was revealed, with factors measuring: 

daily sessions, frequency, age of onset, marijuana 

quantity, cannabis concentrate quantity, and 

edibles quantity. The factors were reliable, with 

Cronbach’s alpha coefficients ranging from .69 

(daily sessions) to .95 (frequency). Results further 

provided evidence for the factors’ convergent 

(MSHQ, TLFB), predictive (CAST, CUDIT-R, 

CUPIT), and discriminant validity (AUDIT). In 

Study 2, we recruited 1,976 cannabis users, of 

which 1,285 completed the DFAQ-CU. To assess 

validity, we included other measures of marijuana 

use (Marijuana Use Grid [MUG]), cannabis-

related problems, symptoms of cannabis use 

disorder, marijuana use motives, marijuana 

refusal self-efficacy, and marijuana protective 

behavioral strategies. Supporting its convergent 

validity, DFAQ-CU’s frequency measure was 

strongly correlated with MUG frequency (r = .74) 

and past month frequency (r = .86); DFAQ-CU’s 

age of onset measure was strongly correlated with 

an alternative measure of age of first use (r = .82); 

DFAQ-CU’s marijuana quantity measure was 

most strongly associated with MUG quantity (r = 

.47). Among the facets of the DFAQ-CU, its 

frequency measure was most strongly positively 

associated with cannabis-related problems, 

symptoms of cannabis use disorder, and 

marijuana use motives, and most strongly 

negatively associated with refusal self-efficacy 

and use of protective behavioral strategies. 

Supporting its discriminant validity, none of the 

DFAQ-CU subscales were significantly correlated 



Conference Abstracts  108 

 
with eating disorder behaviors. Overall, the 

DFAQ-CU is a psychometrically sound inventory 

that measures frequency, age of onset, and 

quantity of cannabis use. It contains pictures of 

marijuana to facilitate the measurement of 

quantity of marijuana used, as well as questions 

to assess the use of different forms of cannabis 

(e.g., concentrates, edibles), methods of 

administering cannabis (e.g., joints, hand pipes, 

vaporizers), and typical THC levels. As such, the 

DFAQ-CU should help facilitate research on 

frequency, quantity, and age of onset of cannabis 

use. 

 

Supplemental Materials Here 

 

Examining the Role of Potency on Marijuana Use 
and Problems Utilizing Retrospective Self-Report 

and Ecological Momentary Assessment 
 

Benjamin O. Ladd & Renee E. Magnan 

Washington State University, Vancouver 

 

Tetrahydrocannabinol (THC) potency has 

increased by as much as tenfold in the past 40 

years and with the greater availability of 

concentrated marijuana products this trend is 

likely to continue. One perspective is that with 

increased potency, users may titrate their 

consumption; on the other hand, increased 

potency may lead to greater adverse 

consequences. Due to a variety of ethical, legal, 

and practical reasons, assessment of marijuana 

use has largely overlooked variability in 

marijuana potency. The purpose of the current 

study was to examine relationships among self-

reported marijuana use and preferred 

cannabinoid potency at the aggregate and event 

levels. Recreational marijuana users (N = 94) aged 

21 and older were recruited from the community 

in two bordering states with legal recreational 

markets. At baseline, participants completed a 

past-month quantity-frequency measure that 

included items on typical THC and cannabidiol 

(CBD) potency. Participants then completed a 14-

day ecological momentary assessment (EMA) 

protocol using their personal smartphone, during 

which they self-initiated a brief survey after each 

use episode. Results from the baseline assessment 

suggest that higher THC was associated with 

greater frequency, r = .32, p < .01, but not quantity 

of use. CBD levels were not associated with either 

frequency or quantity. Additionally, THC and 

CBD levels were not significantly correlated. A 

similar pattern was observed in the EMA data. 

Participants reported similar THC levels across 

the baseline and EMA assessments, t(91) = -1.35, 

p = .18, M = 21.3%, SD = 5.1% and M = 20.6%, SD 

= 5.2%, respectively. Significantly higher baseline 

CBD levels were reported compared to EMA 

reports, t(91) = -4.41, p < .001, M = 9.1%, SD = 

6.5% and M = 6.4%, SD = 3.7%, respectively. 

Baseline THC was not associated with marijuana 

problem severity. Baseline CBD was negatively 

associated with problem severity, B = -7.57, SE = 

3.77, p < .05, however this effect became a trend 

(p = .06) when quantity and frequency of use were 

also entered into the regression model. When 

considering the EMA data, an opposite effect 

emerged such that THC was positively predictive 

of problem severity, B = 9.42, SE = 4.36, p < .05, 

while CBD was not a significant predictor. 

However, again the effect of THC was reduced (p 

= .07) when quantity and frequency were 

included. Results suggest that contrary to the 

titration hypothesis, users selecting higher THC 

marijuana also report more frequent use. 

Additionally, relationships among consumption 

variables and marijuana problem severity differed 

based on the measurement modality. These 

findings further support the need for greater 

consideration and refinement of marijuana use 

measurement. As with other substances, 

researchers should carefully consider the goals 

and demands of a given research project when 

selecting a measurement approach. 

 

Discussant: Matthew R. Pearson 
University of New Mexico 

 

Symposium: Examining Both Sides of the Coin: 

Cannabis as Treatment Target and Potential 

Therapeutic Agent 

 

Chair: Eric R. Pedersen 

RAND Corporation 

 

Cannabis occupies a peculiar space among drugs. 

By some metrics, it has a lower addictive 

potential/abuse liability than most other drugs of 

abuse; however, given higher prevalence rates, 

cannabis use disorder is more prevalent than any 

of other illicit drug use disorder. In the United 

States, use/sale/possession of cannabis is federally 



Conference Abstracts  109 

 
illegal and it is listed as a Schedule I drug, which 

indicates that it is without medical value; 

however, 30 of 50 states permit the use of 

cannabis for medical purposes. In this 

symposium, researchers explore the different 

sides of cannabis in the context of treatment. Dr. 

Ozechowski reports the effects of enhanced 

supervision of Functioning Family Therapy, an 

evidence-based family-centered intervention, on 

marijuana-related outcomes among adolescents. 

In a sample of adolescents with alcohol use 

disorder (and majority with a cannabis use 

disorder), Dr. Kaminer examines how 

commitment to a harm reduction vs. abstinence 

goal affects cannabis outcomes. Dr. Houck 

observes incidental effects of treatment for alcohol 

use disorder on cannabis outcomes in patients 

from Project MATCH and COMBINE. Dr. Sisley 

reviews the barriers and discusses the 

implications of such barriers to conducting 

research examining the therapeutic potential of 

cannabis. 

 

Preventing Adolescent Marijuana Use with a 
Family-Centered Behavioral Health 

Intervention: A 12-month Follow-up Study 
 

Holly B. Waldron, Michael S. Robbins, Timothy 

Ozechowski, Hyman Hops, Charles W. Turner, & 

Janet Brody 

Oregon Research Institute 

 

Background: With an increasing number of US 

states legalizing adult recreational and/or medical 

marijuana use, with a concomitant increase in 

teen use, the need for effective prevention efforts 

is clear and imperative. Most efforts to date have 

been school-based with mixed results. Another 

potential avenue for marijuana use prevention is 

via community-based treatment for adolescent 

mental health problems. The primary aim of the 

current study was to examine implementation 

processes associated with Functional Family 

Therapy (FFT), an evidence-based family-

centered intervention for youth disruptive 

behavior, and widely disseminated in community 

practice settings. A critical feature of efficacious 

interventions as shown in a number of studies is 

maintaining fidelity to the program. In research 

based interventions, fidelity is maintained via 

observation-based methods, whereas in most 

community based interventions, supervision is 

based on the therapists’ self-report, the latter a 

much less expensive approach. Yet, evidence for 

the clear superiority of the different methods has 

not been established. We compared observation-

based supervision (BOOST Building Outcomes 

with Observation-Based Supervision: An FFT 

Effectiveness Trial) using session audio 

recordings uploaded to the supervisor, with the 

FFT dissemination program’s supervision as 

usual (SAU; using self-reports) for youth referred 

for behavioral health treatment in community 

practice settings. None of the adolescents was 

referred specifically for drug-related problems, 

thus, the rates of drug use were relatively low 

with a number of non-users (approximately 65%). 

However, supervisors were knowledgeable in 

techniques specific to drug treatment. Hence, in 

the context of this study, we examined the extent 

to which FFT, implemented using BOOST versus 

SAU, would have preventive effects on adolescent 

substance use over a 1-year period. Sample: 

BOOST (26 therapists, 105 families) and SAU (21 

therapists, 59 families) supervision was 

implemented in eight community mental health 

settings with FFT experienced therapists (77% 

female). Method: Participating youth (41% 

female) had been referred for treatment for 

diverse problem behaviors (e.g., delinquent and/or 

disruptive behaviors, depression, noncompliance). 

Families included those with Hispanic (62%), 

African American (19%), Non-Hispanic White 

(12%) or other (7%) ethnic/racial origins. Therapy 

sessions were audio-recorded. Clinical process 

variables included treatment fidelity, 

engagement, and retention. Clinical outcomes, 

including adolescent behavior, family functioning, 

and substance use, were measured at baseline, 5 

months, and 12 months after treatment initiation. 

Results: The BOOST therapists did not differ 

significantly from SAU therapists in engagement, 

retention rates, or rated adherence. However, 

BOOST adolescents with a prior history of use had 

significantly lower rates of marijuana use at the 

5th and 12th month assessment. Moreover, the 

rates for adolescents in the SAU condition 

continued to increase through the 12-month 

assessment. In contrast, teens in the BOOST 

condition decreased from baseline to the 5-month 

assessment and maintained these levels through 

to 12 months. Conclusion: The results suggest 

that enhanced and more expensive therapy 

supervision of family-based therapy can prevent 



Conference Abstracts  110 

 
the normal escalation of marijuana use in high-

risk adolescents. Moreover, the results also 

suggest that family-based therapy needs to focus 

on drug-specific components in order to achieve 

success. 

 

Adolescents in Treatment for Cannabis Use: Goal 
Commitment Predicts Outcome 

 

Yifrah Kaminer, Christine M. Ohannessian, & 

Rebecca H. Burke  

University of Connecticut School of Medicine 

 

Objective: Commitment to change is an innovative 

potential mediator and mechanism of behavior 

change (MOBC) that has not been examined in 

adolescents with cannabis use disorders (CUD). 

The Adolescent Substance Abuse Goal 

Commitment (ASAGC) questionnaire is a reliable 

and valid 2-scale measure developed to assess the 

adolescent’s commitment to either abstinence 

(i.e., recovery model) or harm reduction (HR) 

model, that in addition to decrease in negative 

consequences, includes consumption reduction as 

a stated treatment goal. The objective of this 

paper is to examine the ASAGC’s ability to predict 

treatment outcome of youth with CUD. Method: 

During sessions three and nine of a 10-week 

treatment program, therapists completed the 

ASAGC for 170 adolescents 13-18 years of age 

with alcohol use disorder (AUD), the majority of 

whom (82%) were diagnosed with co-occurring 

CUD. Cannabis use during treatment and 

aftercare was confirmed by drug urinalysis and 

self-reports until 12-month from study onset. 

Results: Results from logistic regression analyses 

assessing goal commitment at Session 3 indicated 

that both HR and recovery predicted marijuana 

use at Session 3. However, only recovery assessed 

predicted later marijuana use (assessed at 

Session 9 and during aftercare). When goal 

commitment at Session 9 was examined, only 

recovery predicted marijuana use, concurrently 

and longitudinally. These results indicated that 

adolescents who had higher scores for recovery 

were less likely to be positive for marijuana use. 

HR was not a significant predictor for outcome. 

Conclusions: Study findings demonstrate that 

goal commitment consistently predicts CUD 

treatment outcome. HR may not be an attainable 

goal for youth due to delayed neurodevelopmental 

processes of inhibitive behaviors. Commitment to 

recovery specifically is a salient predictor for 

concurrent and future marijuana use.   

 

Incidental Effects of Treatment for Alcohol Use 
Disorder on Cannabis Use: Results from Project 

MATCH and the COMBINE Study 
 

Jon M. Houck 

University of New Mexico 

 

Reciprocal functional interactions between the 

endogenous cannabinoid and opioid systems are 

well known. THC can reduce pain perception 

(Vivian et al., 1998), and rhesus monkeys 

administered THC reduced their self-

administration of heroin (Li et al., 2015). 

Consistent with this, recent work in humans 

found that joint administration of cannabis and 

low doses of oxycodone produce analgesia similar 

to that of larger doses of oxycodone alone (Cooper 

et al., 2018). Naltrexone is an opioid antagonist 

that also acts on cannabinoid receptors. In 

nonhuman primates, acute naltrexone reduces 

THC self-administration (Justinova et al., 2004). 

In humans, the effects of naltrexone on cannabis 

use appear to vary with the experience of the 

individual. In non-cannabis-smokers, acute 

naltrexone potentiates the effects of low cannabis 

doses (Haney, 2007), while in daily cannabis 

smokers, maintenance doses of naltrexone are 

associated with decreased cannabis self-

administration (Haney et al., 2015). The 

COMBINE Study was a large multi-site 

medication (naltrexone, acamprosate) clinical 

trial for alcohol use disorder that did not exclude 

participants on the basis of cannabis dependence 

(Anton et al., 2006). COMBINE participants who 

used cannabis generally had higher alcohol use at 

the end of treatment did than those who did not 

use cannabis (Subbaraman et al., 2016). However, 

cannabis users randomized to receive naltrexone 

and the Combined Behavioral Intervention (CBI) 

reported end-of-treatment drinking equivalent to 

those randomized to receive naltrexone who did 

not use cannabis. The goal of the present study 

was to text the incidental effects of naltrexone 

maintenance for alcohol use disorder on self-

reported cannabis use in the COMBINE Study. 

Data on cannabis use days were obtained from the 

Structured Clinical Interview for DSM-IV (SCID-

I). Naltrexone dose estimates in COMBINE were 

generated by inspection of used blister packs. 206 



Conference Abstracts  111 

 
participants (19.9% female) reported cannabis use 

during the 16-week treatment period. Cannabis 

use was significantly lower in the naltrexone-

taking groups than in other groups (X2(1)=3.865, 

p=.049). In a follow-up analysis comparing 85 

participants assigned to receive either naltrexone 

or placebo, negative binomial regression 

controlling for baseline use indicated that 

participants randomized to receive naltrexone 

had significantly fewer cannabis use days during 

treatment than did those assigned to receive 

placebo (b=-1.969, p=.043). There was not a 

significant effect of CBI (b=-.563, p=.078). At the 

12-month follow-up, there were no significant 

effects on cannabis use days. Results of the 

present analysis suggest that in the COMBINE 

study, naltrexone had incidental effects on 

participant cannabis use. That is, although 

COMBINE was not designed to influence 

cannabis use, participants who used cannabis and 

took naltrexone had fewer cannabis use days 

during the treatment period than did those who 

did not take naltrexone, and participants who 

took the prescribed dose of naltrexone had fewer 

cannabis use days than those who did not. After 

participants stopped taking naltrexone, cannabis 

use did not differ between groups. Although 

limited data on cannabis use reduce confidence in 

the study findings, these results suggest that 

naltrexone should be examined further as a 

treatment for cannabis use disorder, particularly 

in populations that also use alcohol. 

 

Slaying Dragons for Science--Navigating the 
Inexplicable Barriers to US Marijuana Efficacy 

Research 
 

Suzanne Sisley 

Scottsdale Research Institute 

 

This lecture, led by Sue Sisley, MD, will discuss 

the barriers to Cannabis efficacy research and 

myriad ways US government has systematically 

impeded this work. We will use our recent battle 

to implement the veterans research as prime 

example of this research blockade trying to use 

medical marijuana to treat patients with 

treatment-resistant Post-Traumatic Stress 

Disorder (PTSD). 

 

 

 

 

 

 

Copyright:  © 2019 Authors et al. These are open 

access abstracts distributed under the terms of 

the Creative Commons Attribution License, which 

permits unrestricted use, distribution, and 

reproduction, provided the original author and 

source are credited, the original sources is not 

modified, and the source is not used for 

commercial purposes. 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

https://creativecommons.org/licenses/by/4.0/