Secondary Logo

Journal Logo

Psychological and Behavioural Correlates of Cannabis use among Canadian Secondary School Students

Findings from the COMPASS Study

Romano, Isabella BSc1; Williams, Gillian MSc1; Butler, Alexandra MSc1; Aleyan, Sarah MSc1; Patte, Karen A. PhD2; Leatherdale, Scott T. PhD1

doi: 10.1097/CXA.0000000000000058
ORIGINAL ARTICLES
Free
SDC

Objectives: The aim of this study was to investigate the sociodemographic, behavioural, and psychological characteristics of students who reported using cannabis in the school-year preceding cannabis legalization in Canada.

Methods: Data were derived from 36,884 students attending 122 schools that participated in year 6 (2017–2018) of the COMPASS Study, a large, prospective cohort study that administers questionnaires annually in high schools across Canada. Multilevel logistic regression models were used to estimate the odds of past-year cannabis use among students. Predictor variables used in the models included indicators of mental health and disorder, other substance use, movement, and students’ sociodemographic characteristics. We tested the moderating effects of polysubstance use with interactions between binge-drinking, cigarette use, and e-cigarette use.

Results: One in 4 students reported past-year cannabis use. Factors associated with higher odds of cannabis use included higher grade, more spending money, identifying as indigenous, use of other substances (binge-drinking, cigarettes, e-cigarettes, and opioid use), presence of depressive symptoms, and greater emotional dysregulation. Factors associated with lower odds of cannabis use included increased flourishing, recreational screen time of <2 hours per day, and adequate sleep time of 8 or more hours per night. Significant interactions between concurrent use of other substances on cannabis use were detected.

Conclusions: Cross-sectional results suggest that students reporting greater psychological wellbeing and engagement in healthy behaviours are less likely to use cannabis. Future longitudinal research should investigate these associations as protective factors, and how the identified correlates may influence changes in student cannabis use patterns postlegalization.

Objectifs: Le but de cette étude était d’examiner les caractéristiques sociodémographiques, comportementales et psychologiques des élèves ayant déclaré avoir consommé du cannabis au cours de l’année scolaire précédant la légalisation du cannabis au Canada.

Méthodes: Les données proviennent de 36 884 élèves de 122 écoles ayant participé à la 6e année (2017-2018) de l’étude COMPASS, une vaste étude de cohorte prospective qui administre des questionnaires chaque année dans les écoles secondaires du Canada. Des modèles de régression logistique à plusieurs niveaux ont été utilisés pour estimer les probabilités de consommation de cannabis au cours de la dernière année parmi les étudiants. Les variables prédictives utilisées dans les modèles comprenaient des indicateurs de la santé mentale et des troubles mentaux, de la consommation d’autres substances, des mouvements et des caractéristiques sociodémographiques des élèves. Nous avons testé les effets modérateurs de la consommation de polysomes avec les interactions entre consommation excessive d’alcool occasionnelle, usage de la cigarette et usage de la cigarette électronique.

Résultats: Un étudiant sur quatre a déclaré avoir consommé du cannabis au cours de la dernière année. Les facteurs associés à des probabilités de consommation de cannabis plus élevées incluent une classe supérieure, davantage d’argent de poche, l’identification comme étant autochtone, la consommation d’autres substances (consommation excessive d’alcool occasionnelle, cigarettes, cigarettes électroniques et consommation d’opioïdes), la présence de symptômes dépressifs et une plus grande dysfonction émotionnelle. Les facteurs associés à une probabilité plus faible de consommation de cannabis incluaient une durée accrue d’un dépistage de moins de 2 heures par jour, et de sommeil suffisant de 8 heures ou plus par nuit. Des interactions significatives entre la consommation simultanée d’autres substances et la consommation de cannabis ont été détectées.

Conclusions: Les résultats transversaux suggèrent que les étudiants qui déclarent un plus grand bien-être psychologique et un comportement plus sain sont moins susceptibles de consommer du cannabis. Les futures recherches longitudinales devraient examiner ces associations en tant que facteurs de protection et déterminer comment les corrélats identifiés peuvent influencer les changements dans les habitudes de consommation de cannabis des élèves après la légalisation.

1School of Public Health and Health Systems, University of Waterloo, Waterloo, Ontario, Canada

2Department of Health Sciences, Brock University, St. Catharines, Ontario, Canada.

Corresponding Author: Isabella Romano, BSc, School of Public Health and Health Systems, University of Waterloo, 200 University Avenue West, Lyle Hallman South, Room 1618F, Waterloo, ON, Canada N2L 3G1. Tel: +1 416 723 9561; e-mail: iromano@uwaterloo.ca

The authors report no conflicts of interest.

Supplemental digital content is available for this article. Direct URL citations appear in the printed text and are provided in the HTML and PDF versions of this article on the journal's Website (www.canadianjournalofaddiction.org).

Received March 31, 2019

Accepted June 18, 2019

Back to Top | Article Outline

INTRODUCTION

Compared to rates of use in other countries, Canadian youth are among the world's most prevalent, frequent, and youngest cannabis users.1 Given the recent legalization of cannabis in Canada, the impact of cannabis on youth continues to be a priority area within Canada's public health landscape; however, minimal data exist to support evidence-based policy refinement and prevention strategies. Developing a more robust understanding of the factors that influence youth cannabis use is critical.

A number of demographic factors are associated with cannabis use in youth. Extensive literature highlights males as most likely to report cannabis use.2,3 Cannabis use often initiates in adolescence4 and may increase with school grade, whereby older students often use at higher frequencies.3,5 Cannabis use may also vary by ethnicity5; in Canada, First Nations, Métis, and Inuit (FNMI) youth have been identified as more likely to report using cannabis compared to other white or nonracialized youth.6 Up to 2/3 of Inuit youth report past-year cannabis use,7 compared to roughly 19% of youth in the general Canadian population.8

Youth who use cannabis typically exhibit a higher propensity to also engage in other risky health behaviours.9 Tobacco and alcohol use behaviours are strongly correlated with cannabis use.5 Similar findings have been observed in youth vaping research demonstrating a robust association between cannabis use and e-cigarette use.10 In contrast, studies examining associations between cannabis use and physical activity, sedentary behaviours, or sleep patterns among youth have yielded mixed findings, as such this area requires further investigation.5,11

Researchers have demonstrated important links between cannabis use and negative mental health experiences.12 Compared to those who report no use, youth who use cannabis more commonly report symptoms of depression and anxiety,12 and have a 3-fold greater risk of experiencing major depressive episodes.13 A recent study found that poorer levels of overall psychosocial wellbeing were most predictive of cannabis use.3 Motivation to use cannabis may be associated with impaired development of emotional regulation skills and psychosocial functioning.14 The paucity of literature investigating the associations between cannabis use and mental health, particularly among Canadian school-aged youth, warrants the need for additional research.

The aim of the present study was to investigate cannabis use among youth within the school year immediately preceding cannabis legalization in Canada (2017–2018). Our main objectives were 3-fold: (1) to estimate the prevalence of cannabis use within this sample; (2) to examine demographic, psychological, and behavioural correlates of cannabis use; and (3) to investigate the potential moderating effects contributed by concurrent use of other substances on students’ cannabis use.

Back to Top | Article Outline

METHODS

Study Design and Sample

The COMPASS Study was designed to collect data on youth health behaviours from ∼65,000 grade 9 to 12 students in Canada, allowing for the ongoing evaluation of policies, school programs, and their impact on youth health over time.15 From a convenience sample of secondary schools across Ontario, Quebec, British Columbia, and Alberta, students are recruited using active-information passive-consent procedures that have received approval from the University of Waterloo Office of Research Ethics and participating school boards. Additional details regarding the methods used within the study can be found online (www.compass.uwaterloo.ca) or in print.15

The present study used cross-sectional, student-level data from year 6 (2017–2018) of the COMPASS Study, which was the first wave to incorporate a newly developed Mental Health Model16 into the existing COMPASS Student Questionnaire (Cq). A total of 66,434 students participated in year 6 of COMPASS; we employed a complete-case analytic approach using n = 36,884 students across 122 high schools.

Missingness was primarily observed across measures of depression (23%) and anxiety (13%). Relatively few students were missing data related to their substance use (less than 2% with missing cannabis use data). Supplementary File A (Supplemental Digital Content 1, http://links.lww.com/CJA/A9) presents logistic regression results predicting the likelihood of missing cannabis use (i.e., our outcome of interest), depression, and anxiety data by students’ sociodemographic characteristics. We found some significant sociodemographic differences between students with complete versus missing data for measures of cannabis, depression, and anxiety. In general, male and non-white students had greater odds of missing data.

Back to Top | Article Outline

Measures

Cannabis Use: Dependent Variable of Interest

Students were asked, “In the last 12 months, how often did you use marijuana or cannabis?” For descriptive analyses, students were classified into 3 mutually exclusive response categories: (1) nonuse (those who reported never use or use before the past year), (2) noncurrent use (those who reported using less than once a month), and (3) current use (those who reported using once a month or more in the past month). In the multivariable analyses, past-year cannabis use was measured as a binary outcome where students were classified as either having used cannabis (yes = 1) or not (no = 0)—the latter category capturing both never use and ever use before the past year.

Back to Top | Article Outline

Sociodemographic Characteristics

Data were used from students residing in Alberta, British Columbia, Ontario, and Québec. Students were asked to report their sex (male, female), grade [9, 10, 11, 12, other (i.e., CÉGEP)], and self-identified ethnicity which was recategorized into: white, FNMI, and other or mixed. Students were also asked to report their weekly spending money ($0, $1–$20, $20–$100, over $100), which acted as a proxy measure of socioeconomic status and part-time employment.

Back to Top | Article Outline

Psychological Characteristics: Mental Disorder and Wellbeing

The Generalized Anxiety Disorder 7-item Scale (GAD-7)17 was included in the Cq to capture students’ self-reported symptoms of anxiety. Using a 4-point Likert scale (0 = not at all, 3 = nearly every day), students reported how often they had been bothered by problems (e.g., trouble relaxing, uncontrollable worrying) in the past 2 weeks. Sum scores (ranging from 0 to 21) were dichotomized and indicated the presence of clinically relevant anxiety symptoms (GAD-7 score ≥10), as previously validated for nonclinical youth samples.18 Internal consistency of the GAD-7 was high (α = 0.97).

To assess self-reported depressive symptoms, students responded to questions from the Centre for Epidemiological Studies Depression Scale (Revised)-10 (CESD-R-10).19 The CESD-R-10 asked students how often, in the past 7 days, they felt symptoms such as loneliness, sadness, and trouble concentrating. Students were asked to respond on a 4-point Likert scale (0 = none or less than 1 day, 3 = 5–7 days) with scores ranging between 0 and 30; CESD-R-10 scores ≥10 indicated clinically relevant depressive symptoms.20 The CESD-R-10 was found to have high internal consistency (α = 0.97).

Students’ overall psychosocial wellbeing (or, flourishing) was assessed using a modified 8-item Flourishing Scale (FS).21 Students were asked to rate their level of agreement along a 5-point Likert scale (0 = strongly disagree, 4 = strongly agree) with statements, such as “I am optimistic about my future” and “people respect me.” Total possible scores ranged on a continuum from 8 to 40, with higher scores indicating greater flourishing and lower scores indicating poor psychosocial wellbeing (or, languishing). The FS had high internal consistency (α = 0.97).

Socioemotional skills were assessed using 6 items from the Difficulties in Emotion Regulation Scale (DERS).22 The highest loading item in each subscale was used based on previous factor analytic studies among adolescent populations.16 The internal consistency across items used from the DERS was high (α = 0.97). Students were asked to endorse the frequency at which statements, such as “I pay attention to how I feel” and “when I am upset, I lose control over my behaviour” using a 5-point Likert scale (0 = almost never, 4 = almost always), with total scores ranging from 6 to 30. Higher scores indicated greater impairment (i.e., emotional dysregulation).

Back to Top | Article Outline

Behavioural Characteristics: Substance Use and Movement

Students were asked to report the frequency at which they participated in binge drinking (operationalized as 5 drinks of alcohol or more on 1 occasion) in the past 12 months. Consistent with national surveillance measures,23 a current (past 30 days) binge-drinking variable was created in which students fell into a binary yes or no response category. The Cq also asked individuals, “on how many of the last 30 days did you smoke 1 or more cigarettes?” Participants who reported cigarette use at least once within the last 30 days were classified as yes to current (past 30 days) cigarette use.24 The same question was repeated, asking individuals whether they had used electronic cigarettes (e-cigarettes) within the last 30 days. Consistent with previous research,25 student responses to current (past 30 days) e-cigarette use were dichotomized as yes or no. Students also responded to the following question: “have you used or tried any of the following medications to get high?” and selected applicable items across a list of opioids that included past-year use of oxycodone (e.g., “oxies” and “percs”), fentanyl, and other prescription pain relievers (i.e., codeine, morphine). Students who reported any opioid use in the last 12 months were labelled yes to current (past 12 months) opioid use.

Respondents were also asked questions in relation to physical activity, recreational screen time, and sleep. In accordance with the Canadian 24-Hour Movement Guidelines,26 students were considered to have met the recommendations if, on average, they reported: (1) engaging in moderate-to-vigorous physical activity at least 60 minutes per day (“meets physical activity guidelines” = yes); (2) engaging in no more than 2 hours of total recreational screen time per day (“meets screen time guidelines” = yes); and (3) obtaining at least 8 hours of sleep each night (“meets sleep guidelines” = yes).

Back to Top | Article Outline

Analyses

Descriptive statistics were used to examine the sociodemographic, psychological, and behavioural characteristics of the student sample, by cannabis use status (nonuse, noncurrent use, current use), using chi-square and 1-way analysis of variance (ANOVA) tests at 99% significance (Table 1). Five multivariable logistic regression models were used; the first model (Table 2, Model I) examined associations between sociodemographic characteristics and the outcome of interest: past-year cannabis use (yes/no). The next 3 models (Table 3, Models II–IV) involved the stepwise addition of grouped behavioural and psychological characteristics in the following order: substance use, mental disorder and wellbeing, and movement, while holding sociodemographic variables constant. The last model (Table 3, Model V) included 2- and 3-way interaction terms testing the moderating effects of several combinations of polysubstance use on cannabis use alone.

Table 1

Table 1

TABLE 2

TABLE 2

TABLE 3

TABLE 3

We calculated the intraclass correlation coefficient estimating the amount of within-school variation in past-year cannabis use. A significant amount of school-level variability was detected, accounting for roughly 7% of the total variation in students’ cannabis use (intraclass correlation coefficient = 0.072). To account for the nested structure of the data (i.e., students clustered within schools), we fit generalized estimating equation models with the GENMOD procedure using SAS version 9.4 software.27 Beta estimates were exponentiated in order to obtain adjusted odds ratios, which we reported alongside 99% confidence intervals.

Back to Top | Article Outline

RESULTS

Characteristics of the Student Sample

Our sample included a total of 36,884 COMPASS students, 50% of which identified as female. The majority of students (70%) identified as white; roughly 3% self-identified as FNMI. Fewer than 20% of students reported having no weekly spending money. Roughly half of students (49%) within the study sample resided in Ontario.

Over 1/3 (36%) of students reported past 30-day binge drinking. Within the study sample, 10% reported past 30-day cigarette use while 23% reported past 30-day e-cigarette use. A small proportion of students (2%) reported past-year use of opioids for the purpose of getting high. A large proportion of students in our sample did not meet the recommended guidelines for physical activity (61%), sleep (59%), or screen time (94%). In addition, many students reported clinically relevant symptoms of anxiety (22%) and depression (32%). Among all students, there was a mean FS score of 32.0 [standard error (SE) = 5.8] and mean DERS score of 14.1 (SE = 4.8).

Back to Top | Article Outline

Prevalence of Cannabis Use

Twenty-four percent of students in our sample reported noncurrent cannabis use, while 14% reported current (past 30 days) cannabis use. Table 1 presents student-level descriptive characteristics by cannabis use status; significant differences in cannabis use were observed across all student characteristics. More males engaged in current cannabis use compared to females. A greater proportion of students who reported current cannabis use also reported past 30-day binge drinking, cigarette use, e-cigarette use, and 12-month opioid use. The proportion of students who scored ≥10 on each of the GAD-7 and CESD-R-10 was highest among those who indicated current (past 30 days) cannabis use. Students who reported nonuse of cannabis had the highest average FS score and lowest average DERS score, and were more apt to meet the recommended guidelines for screen and sleep time.

Back to Top | Article Outline

Factors Associated with Past-year Cannabis Use among Students

Logistic regression results predicting the likelihood of students’ past-year cannabis use are presented in Tables 1 and 2. Male students were more likely to report past-year cannabis use compared to females and, generally, the likelihood of cannabis use increased with increasing school grade and weekly spending money (Table 2, Model I). Compared to students who identified as white, FNMI students were more likely to report cannabis use (Table 2, Model I). In Table 3, Models II through IV present results of a step-wise addition of student characteristics while controlling for sex, grade, ethnicity, weekly spending money, and province. While male students were more likely to report past-year cannabis use in Table 1, Model I, lower likelihood of male cannabis use [odds ratio = 0.87, 99% confidence interval (0.82, 0.93)] was detected after controlling for concurrent substance use in Table 3, Model II.

In Table 3, Model III, students who reported clinically relevant symptoms of depression (CESD-R-10 score ≥10) were approximately 20% more likely to report past-year cannabis use. For every 1-point increase along the FS scale, students were on average 3% less likely to report past-year cannabis use. Similarly, students were 2% more likely to report past-year cannabis use with every 1-point increase in DERS Items score. These estimates remained consistent while adjusting for students’ physical activity, screen time, and sleep (Table 3, Model IV).

Back to Top | Article Outline

Moderating Effects of Polysubstance Use

Model V (Table 3) displays results produced by the addition of 2- and 3-way interaction terms. Two-way interactions between substances used were shown to each be significantly associated with cannabis use, indicating that, for example, the relationship between concurrent reports of binge-drinking and cannabis use varied according to whether or not students also reported cigarette use. Model IV suggests students were more likely to use cannabis if they also reported concurrent: binge drinking and cigarette use (depicted in Fig. 1), binge drinking and e-cigarette use (Fig. 2), and cigarette use and e-cigarette use (Fig. 3). The 3-way interaction (addressing concurrent use of binge drinking, cigarettes, and e-cigarettes) yielded a null association.

FIGURE 1

FIGURE 1

FIGURE 2

FIGURE 2

FIGURE 3

FIGURE 3

Back to Top | Article Outline

DISCUSSION

Among a large sample of students across 122 high schools in Canada, a substantial proportion reported cannabis use during the school year preceding its legalization (2017–2018). Of the nearly 1 in 4 students who used cannabis in the past year, over half reported using it within the past month. Our findings are notably higher than other nationally representative estimates demonstrating that the prevalence of past-year cannabis use among Canadian adolescents in 2017 was 19%.8 Traditional active consent procedures have been shown to substantially underestimate youth substance use, thereby introducing bias; the COMPASS Study's use of passive consent may have minimized this risk.28 We also detected several significant associations between student-level characteristics (sociodemographic, behavioural, and psychological) and odds of cannabis use. Seeing as evidence following cannabis liberalization (i.e., legalization of medicinal cannabis, decriminalization) in some US states suggests a moderate but significant increase in cannabis use is to be expected among youth in Canada,29 ongoing cannabis intervention and prevention efforts should consider the psychological and behavioural correlates identified here.

The results of our study are supported by existing evidence linking mental health and cannabis use. Youth who use cannabis during adolescence have been previously shown to exhibit symptoms of affective and mood disorders,12 suggesting cannabis use poses adverse effects on mental health. In particular, more frequent use may be associated with heightened risk for depressive episodes later on in early adulthood.12 Our results indicate that within a nonclinical sample of youth, those who report symptoms of depression may have an increased propensity to use cannabis during high school years. The role of cannabis as a coping tool for youth who experience psychological and emotional distress has been noted by other researchers, often citing the perception that self-medication alleviates anxious and depressive symptoms.30 Similarly, high-risk cannabis use has been hypothesized to damper one's ability to process emotions healthily.31 Our findings suggest that an increased likelihood to use cannabis may be associated with emotional dysregulation, further supporting theories that some individuals with poor regulation skills may use cannabis to manage their negative emotions, affect, and as a method of coping.3 The ability to identify and cope with negative emotions may also be protective against the development of mood and substance use disorders, whereas the presence of mental disorder oftentimes hinders emotion regulation processes.32 We did not detect a significant association between past-year cannabis use and clinically relevant symptoms of anxiety among youth. This null finding is inconsistent with the existing body of literature, perhaps because the present study does not differentiate between potential subtypes of cannabis use behaviours or high-risk subgroups of youth. Further analyses are required to better understand the specific relationship, if any, between cannabis and anxiety among youth. The ways in which these constructs are interrelated with and affected by cannabis use warrant further temporal investigation.

Along a flourishing-languishing continuum, higher overall psychosocial wellbeing may be protective against engagement in risky health behaviours. Recent evidence from the COMPASS Study has demonstrated a robust and sustained protective effect of flourishing on cannabis use regardless of mental disorder symptoms being present.3 The impact of positive psychology on other modifiable behaviours (i.e., physical activity, sleep, etc.) has been investigated previously,33 while the area of flourishing and cannabis use remains underresearched. Interestingly, sedentary behaviours such as screen time may also be associated with cannabis use.34 Our own findings demonstrate that students who reported using no more than 2 hours of recreational screen time per day were less likely to report past-year cannabis use. We found a similar, yet modest, association among students who met the recommended sleep time of at least 8 hours per night. Our findings suggest that engaging in healthy behaviours may be protective against cannabis use during adolescence. Within the context of psychosocial wellbeing, further research is required to investigate whether these factors contribute to enhanced resilience; possessing positive psychological resources such as resilience is thought to mitigate risks of psychopathology.35 More evidence concerning the protective aspects of behavioural and psychological health on cannabis use is needed.

Substantial research has illustrated the phenomenon of clustered substance use behaviours in youth.36 Before testing for interaction effects, students within our sample were in general between 4 and 6 times more likely to report using cannabis in the past year if they had also reported use of other substances such as alcohol (5 drinks or more on 1 occasion), cigarettes, e-cigarettes, and opioids. Compared to students who reported nonuse across substances, analysis with these interactions revealed upwards of 50 times greater relative odds of cannabis use if students indicated yes to binge drinking and cigarette use (Fig. 1), binge drinking and e-cigarette use (Fig. 2), or cigarette use and e-cigarette use (Fig. 3). Due to successful tobacco cessation and prevention efforts, Canada has seen significant reductions in tobacco use across subpopulations, including among adolescents.37 Despite this, recent evidence suggests that recent increases in e-cigarette use may pose risk for future smoking susceptibility among nonsmoking youth.25 Examining changes in youth substance use patterns may contribute important knowledge to our understanding of cannabis legalization impacts in Canada.

This study contributed important knowledge about the sociodemographic correlates of Canadian youth who reported cannabis use in a survey administered during the school year preceding legalization. The likelihood of using cannabis was higher among those in higher grades and greater weekly spending money. These associations have been noted previously3 and are likely a function of normalization of cannabis use and greater accessibility of cannabis to older students, as well as greater means through which to purchase it. Consistent with the results of this study, male sex is known to be a predictor of cannabis use5; however, our findings uniquely demonstrated that when controlling for use of other substances, female students were more likely to report using cannabis in the past year. Our findings also indicate higher odds of cannabis use among students who self-identified as being of FNMI ethnicity. In Canada, where there is a colonial history of indigenous marginalization and poverty, FNMI youth face many inequitable health outcomes related to substance use, mental health, and suicidality.38 Future cannabis research and prevention efforts would be remiss not to prioritize the needs and perspectives of indigenous communities. Finally, it should also be noted that we detected significant within-school variability in student cannabis use, indicating students within a given school may share common socioenvironmental exposures that influence their behaviours. A growing body of research points to the potential effectiveness of school-based resilience interventions on youth substance use,39 but additional work is needed to evaluate existing programs in practice. Robust evaluations of school-based substance use prevention programs can be made possible through use of quasiexperimental designs and hierarchical COMPASS data.

Back to Top | Article Outline

Strengths and Limitations

Notable strengths of this study include its large sample size (n = 36,884) and 2-level design (i.e., students clustered within schools). There are also some limitations. First, there is possibility of confounding introduced by social desirability bias. It is possible that students underreported their substance use due to perceived consequences; at the time these data were collected, cannabis was an illicit substance in Canada. This limitation was partially mitigated by the COMPASS Study's data collection protocols, which guarantee student anonymity and do not require active parental consent.15 An additional limitation pertains to our use of nondiagnostic measures within a nonclinical sample. The GAD-7 and CESD-R-10 tools were operationalized using clinically relevant cutoff scores; however, our findings are not meant to be generalized toward any clinical population, such as individuals with psychiatric or substance-related diagnoses. Potential weakness of our physical activity, screen time, and sleep variables should be acknowledged. The movement behaviour measures were self-reported, and thus prone to social desirability and recall bias, and were operationalized to align with the Canadian 24-Hour Movement Guidelines. Researchers seeking to further investigate the association between movement and cannabis use may consider additional measures (e.g., sleep quality, types of screen time, etc.). Future research should also incorporate multidimensional measures of cannabis use beyond a past-year binary response (e.g., monthly or daily use, amount used, mode of use, potency, etc.). The cross-sectional design of this study prevents inference about the directionality or temporality of associations underlying the psychological, behavioural, and sociodemographic correlates we assessed in relation to cannabis use. Given the plausible bidirectional nature between cannabis use and these correlates, additional research is necessary to further understand these effects. Plans for future longitudinal analyses using pre- and postlegalization COMPASS Mental Health Model data are underway. To this point, we were also limited by our use of cannabis data collected a single year before it being legal; previous researchers have noted increasing trends initiating some years earlier, likely as a function of normalized discourse leading up to cannabis legalization.40 Nonetheless, these data represent an important baseline timepoint from which robust evaluations can be conducted once postlegalization data become available. Lastly, limitations with respect to missing data should be noted as well. The decision to conduct a complete-case analysis resulted in a nearly 50% decrease in our analytic sample size and increased risk of bias. We retained sufficient power given that our sample remained large (n = 36,884) despite the missingness, but did detect that students with certain sociodemographic characteristics (e.g., male sex, non-white ethnicity) had a higher likelihood of missing cannabis use, anxiety, and depression data (Supplementary File A, Supplemental Digital Content 1, http://links.lww.com/CJA/A9).

Back to Top | Article Outline

Recommendations and Future Directions

Cannabis use among Canadian high school students is associated with other substance use behaviours, depressive symptoms, poor psychosocial and emotional functioning, higher screen time, and less sleep. Cannabis use prevention strategies may benefit from different targeted approaches toward male and female students, older students, and students from Indigenous communities. However, broader universal school-based substance use interventions may be most effective; namely, those aimed at fostering resilience and teaching emotional regulation skills. Interventions rooted in positive psychology are becoming recognized as more sustainable for preventing substance use at the population level,39 especially compared to ineffective antidrug campaigns of the past (e.g., “say no to drugs” messaging). At an individual level, integrated efforts between schools and clinicians may benefit from identifying students with problematic substance use and screening for symptoms of mental disorder. Future research should seek to investigate the role that school environments play in influencing students’ cannabis use behaviours.

The findings of this study provide additional knowledge regarding factors associated with youth cannabis use, before it became legal for adult consumption in Canada. In addition to identifying subsets of students who may benefit from targeted cannabis intervention and prevention efforts, this research provides the foundation for pre–post analyses of the impact cannabis legalization may have on youth health in Canada in the coming years. The COMPASS Study offers a unique research platform to also evaluate the influence of differing cannabis policies across provinces, and the effectiveness of school-based cannabis prevention programs for youth.

Back to Top | Article Outline

CONCLUSIONS

Using data collected the year before cannabis legalization in Canada, we found that cannabis use was highly prevalent among a large sample of high school youth. We observed significant associations between past-year cannabis use and several behavioural, psychological, and sociodemographic characteristics of students. The question of whether cannabis legalization in Canada will impact future cannabis use behaviours, and for whom, remains unknown for now. In the meantime, our findings suggest that psychosocial prosperity, emotional resilience, and engagement in healthy behaviours may be protective factors against students’ cannabis use. As such, school-based interventions focused on addressing emotional resilience and mental health should be considered within future programming efforts. As we progress with the legalization of cannabis, these findings may primarily be used to inform future research, as well as assist in detecting factors associated with youth cannabis use within a Canadian context.

Back to Top | Article Outline

ACKNOWLEDGEMENTS

The COMPASS Study was supported by a bridge grant from the Canadian Institutes of Health Research (CIHR) Institute of Nutrition, Metabolism and Diabetes (INMD) through the “Obesity—Interventions to Prevent or Treat” priority funding awards (OOP-110788; grant awarded to STL), an operating grant from the Canadian Institutes of Health Research (CIHR) Institute of Population and Public Health (IPPH) (MOP-114875; grant awarded to STL), a CIHR Project Grant (PJT-148562; grant awarded to STL), and by a research funding arrangement with Health Canada (#1617-HQ-000012; contract awarded to STL). The COMPASS Mental Health Module was supported by a CIHR Bridge Grant (PJT-149092; grant awarded to KAP/STL) and a CIHR Project Grant (PJT-159693; awarded to KAP).

Back to Top | Article Outline

REFERENCES

1. United Nations Office on Drugs and CrimeWorld Drug Report 2016. 2016;United Nations Office on Drugs and Crime, New York: Available from: https://www.unodc.org/doc/wdr2016/WORLD_DRUG_REPORT_2016_web.pdf. Accessed November 6, 2018
2. Hango DW, LaRochelle-Côté S. Association between the Frequency of Cannabis Use and Selected Social Indicators. Insights on Canadian Society. Statistics Canada Catalogue no. 75-006-X.
3. Butler A, Patte KA, Ferro MA, et al Interrelationships among depression, anxiety, flourishing, and cannabis use in youth. Addict Behav 2019;89:206–215.
4. Griffin KW, Botvin GJ. Evidence-based interventions for preventing substance use disorders in adolescents. Child Adolesc Psychiatr Clin N Am 2010;19:505–526.
5. Sampasa-Kanyinga H, Hamilton H, LeBlanc A, et al Cannabis use among middle and high school students in Ontario: a school-based cross-sectional study. CMAJ Open 2018;6:E50–E56.
6. Sikorski C, Leatherdale ST, Cooke M. Tobacco, alcohol and marijuana use among Indigenous youth attending off-reserve schools in Canada: cross-sectional results from the Canadian Student Tobacco, Alcohol and Drugs Survey. Health Promot Chronic Dis Prev Can 2019;39:207–215.
7. Brunelle N, Plourde C, Landry M, et al Patterns of psychoactive substance use among youths in Nunavik. Indittera 2010;2:1–12.
8. Health CanadaCanadian Tobacco, Alcohol and Drugs Survey (CTADS): Summary of Results for 2017. 2019;Government of Canada, Available from: https://www.canada.ca/en/health-canada/services/canadian-tobacco-alcohol-drugs-survey/2017-summary.html#n5. Accessed June 18, 2019
9. Zuckerman M. Grinevald C, Wright J. Sensation seeking: behavioural expressions and biosocial bases. International Encyclopedia of the Social & Behavioural Sciences Second EditionOxford: Elsevier; 2015;607–614.
10. Tavolacci M-P, Vasiliu A, Romo L, et al Patterns of electronic cigarette use in current and ever users among college students in France: a cross–sectional study. BMJ Open 2016;6:e011344.
11. Terry-Mcelrath YM, O’Malley PM, Johnston LD. Exercise and substance use among American youth, 1991–2009. Am J Prev Med 2011;40:530–540.
12. Patton GC, Coffey C, Carlin JB, et al Cannabis use and mental health in young people: cohort study. BMJ 2002;325:1195–1198.
13. Rey JM, Sawyer MG, Raphael B, et al Mental health of teenagers who use cannabis. Results of an Australian survey. Br J Psychiatry 2002;180:216–221.
14. Zimmermann K, Walz C, Derckx RT, et al Emotion regulation deficits in regular marijuana users. Hum Brain Mapp 2017;38:4270–4279.
15. Leatherdale ST, Brown KS, Carson V, et al The COMPASS Study: a longitudinal hierarchical research platform for evaluating natural experiments related to changes in school-level programs, policies and built environment resources. BMC Public Health 2014;14:1–7.
16. Patte KA, Bredin C, Henderson J, et al Development of a Mental Health Module for the COMPASS System: Improving Youth Mental Health Trajectories Part 1: Tool Development and Design. COMPASS Technical Report Series. 2017;University of Waterloo, Waterloo, Ontario: 4; Available from: www.compass.uwaterloo.ca. Accessed November 6, 2018
17. Spitzer RL, Kroenke K, Williams JB, et al A brief measure for assessing generalized anxiety disorder: the GAD-7. Arch Intern Med 2006;166:1092–1097.
18. Mossman SA, Luft MJ, Schroeder HK, et al The Generalized Anxiety Disorder 7-item Scale in adolescents with generalized anxiety disorder: signal detection and validation. Ann Clin Psychiatry 2017;29:227–234A.
19. Van Dam NT, Earleywine M. Validation of the Center for Epidemiologic Studies Depression Scale-Revised (CESD-R): pragmatic depression assessment in the general population. Psychiatry Res 2011;186:128–132.
20. Haroz E, Ybarra ML, Eaton WW. Psychometric evaluation of a self-report scale to measure adolescent depression: the CESDR-10 in two national adolescent samples in the United States. J Affect Disord 2014;158:154–160.
21. Diener E, Wirtz D, Tov W, et al New well-being measures: short scales to assess flourishing and positive and negative feelings. Soc Indic Res 2010;97:143–156.
22. Neumann A, van Lier PA, Gratz KL, et al Multidimensional assessment of emotion regulation difficulties in adolescents using the Difficulties in Emotion Regulation Scale. Assessment 2010;17:138–149.
23. Elton-Marshall T, Leatherdale ST, Manske SR, et al Research methods of the Youth Smoking Survey (YSS). Chron Dis Inj Can 2011;32:47–54.
24. Wong SL, Shields M, Leatherdale ST, et al Assessment of validity of self-reported smoking status. Health Rep 2012;23:47–53.
25. Aleyan S, Cole A, Qian W, et al Risky business: a longitudinal study examining cigarette smoking initiation among susceptible and non-susceptible e-cigarette users in Canada. BMJ Open 2018;8:e021080.
26. Tremblay MS, Carson V, Chaput J-P. Introduction to the Canadian 24-Hour Movement Guidelines for children and youth: an integration of physical activity, sedentary behaviour, and sleep. Appl Physiol Nutr Metab 2016;41:311–327.
27. SAS Institute Inc.SAS® 9.4 Statements: Reference. Cary, NC: SAS Institute Inc; 2013.
28. Thompson-Haile A, Bredin C, Leatherdale ST. Rationale for Using an Active-Information Passive-Consent Permission Protocol in COMPASS. COMPASS Technical Report Series. 2013;University of Waterloo, Waterloo, Ontario: 1; Available from: www.compass.uwaterloo.ca. Accessed November 6, 2018
29. Miech RA, Johnston L, O’Malley PM, et al Trends in use of marijuana and attitudes toward marijuana among youth before and after decriminalization: the case of California 2007–2013. Int J Drug Policy 2015;26:336–344.
30. Bonn-Miller MO, Vujanovic AA, Zvolensky MJ. Emotional dysregulation: association with coping-oriented marijuana use motives among current marijuana users. Subst Use Misuse 2008;43:1653–1665.
31. Bonn-Miller MO, Zvolensky MJ, Bernstein A. Marijuana use motives: concurrent relations to frequency of past 30-day use and anxiety sensitivity among young adult marijuana smokers. Addict Behav 2007;32:49–62.
32. Picó-Pérez M, Radua J, Steward T, et al Emotion regulation in mood and anxiety disorders: a meta-analysis of fMRI cognitive reappraisal studies. Prog Neuropsychopharmacol Biol Psychiatry 2017;79:96–104.
33. Hinkley T, Verbestel V, Ahrens W, et al Early childhood electronic media use as a predictor of poorer well-being: a prospective cohort study. JAMA Pediatr 2014;168:485.
34. Vesna L, Olivera SJ. Physical activity, sedentary behavior and substance use among adolescents in slovenian urban area. Zdr Varst 2015;54:168–174.
35. Zimmer-Gembeck MJ, Skinner EA. The development of coping: implications for psychopathology and resilience. Developmental Psychopathology: Risk, Resilience, and Intervention 2016;John Wiley & Sons Inc, Hoboken, NJ: 1–61.
36. Agrawal A, Budney AJ, Lynskey MT. The co-occurring use and misuse of cannabis and tobacco: a review. Addiction 2012;107:1221–1233.
37. Reid JL, Hammond D, Rynard VL, et al Tobacco use in Canada: Patterns and Trends. 2017;Propel Centre for Population Health Impact, University of Waterloo, Waterloo, Ontario: 1–112.
38. Lehti V, Niemelä S, Hoven C, et al Mental health, substance use and suicidal behaviour among young indigenous people in the Arctic: a systematic review. Soc Sci Med 2009;69:1194–1203.
39. Hodder RK, Freund M, Wolfenden L, et al Systematic review of universal school-based ‘resilience’ interventions targeting adolescent tobacco, alcohol or illicit substance use: a meta-analysis. Prev Med 2017;100:248–268.
40. Zuckermann AME, Battista K, de Groh M, et al Prelegalisation patterns and trends of cannabis use among Canadian youth: results from the COMPASS prospective cohort study. BMJ Open 2019;9:e026515.
Keywords:

cannabis; depression; emotional regulation; flourishing; substance use; cannabis; consommation de substances; épanouissement; dépression; gestion des émotions

Supplemental Digital Content

Back to Top | Article Outline
© 2019 by Lippincott Williams & Wilkins, Inc.