Original quantitative research – Autonomy, competence and relatedness and cannabis and alcohol use among youth in Canada: a cross-sectional analysis
This article has been peer reviewed.
Correspondence: Aganeta Enns, 136 Jean-Jacques-Lussier Pvt, Ottawa, ON K1N 6N5; Tel: 613-562-5800 ext. 4639; Email: email@example.com
Introduction: There has been increasing attention on preventing problematic youth substance use in light of concerns about rates of use and policy changes in Canada. Strengths-based approaches that emphasize protective factors, including positive mental health, are at the forefront of current prevention recommendations. However, there is a dearth of research on the association between positive mental health and substance use among youth. This study examines the associations between cannabis and alcohol use among youth and positive mental health as measured through the lens of self-determination theory.
Methods: Secondary analyses of the 2014/2015 Canadian Student Tobacco, Alcohol and Drugs Survey (CSTADS) were conducted. Participating Grade 7 to 12 students residing in Canada completed the Children’s Intrinsic Needs Satisfaction Scale (CINSS), which measures autonomy, competence and relatedness, and answered questions that measure past 30-day and more frequent cannabis use, alcohol use and binge-drinking. The associations between autonomy, competence and relatedness and substance use, stratified by sex, were examined using logistic regression.
Results: Fully adjusted models revealed that relatedness and competence were associated with lower odds of 30-day and more frequent cannabis use, alcohol use and binge-drinking. Higher autonomy was associated with higher odds of these behaviours. All associations were significant with the exception of competence and more frequent cannabis use among boys, and autonomy and more frequent alcohol use among girls.
Conclusion: The findings offer new evidence on the associations between positive mental health and substance use among youth, specifically how autonomy, competence and relatedness are associated with cannabis use, alcohol use and binge-drinking. This evidence can be used to inform health promotion and substance use prevention programs.
- In multivariate models, youth with higher levels of relatedness and competence had lower odds of 30-day and more frequent cannabis use, alcohol use and binge-drinking (except for more frequent cannabis use and relatedness among boys).
- In contrast, in fully adjusted models, youth with higher levels of autonomy had higher odds of the substance use outcomes examined (except for more frequent alcohol use and autonomy among girls).
- Higher relatedness was associated with lower odds of substance use for both girls and boys, although there was a significant interaction between a number of positive mental health concepts and substance use behaviours. This highlights the need to examine these associations separately by sex.
Keywords: adolescents, positive mental health, substance use, cannabis, alcohol
Alcohol and substance use are most commonly initiated during adolescence.Footnote 1 In 2016–2017, almost half (44%) of students in Grades 7 to 12 consumed alcohol in the past year, and one-quarter of students reported drinking five or more drinks on one occasion.Footnote 2 Just under one in five students (17%) reported cannabis use in the past year.Footnote 2 Adolescence is a critical period of cognitive and psychosocial development, and substance use may be associated with disrupted patterns of behaviours and longer-term risks such as problematic substance use later in life.Footnote 3 Alcohol and cannabis use in adolescence has been associated with increased likelihood of injury, impaired driving and negative social, psychological and legal consequences.Footnote 4Footnote 5
Changes in alcohol and drug policies in Canada, and concerns about rates of youth substance useFootnote 2 have highlighted the need to understand and prevent problematic substance use among young people. Recent changes in policies in some jurisdictions, for example, the expansion of the sale of some forms of alcohol into grocery stores and the reduction of the minimum price of beer in Ontario, have changed the availability of alcohol.Footnote 6Footnote 7 Such policy changes increase exposure to alcohol marketing in stores, which may increase the odds of adolescent alcohol consumption.Footnote 8
Preventing the sale and promotion of cannabis to youth was among the public health objectives of the recent legalization of cannabis in Canada.Footnote 9 While the Cannabis Act is the “legal framework for controlling the production, distribution, sale and possession of cannabis across Canada,” provinces and territories determine how, where and by whom cannabis can be sold. After the legalization of non-medical cannabis use in the state of Washington, the perceived harmfulness of cannabis use decreased and cannabis use increased among youth.Footnote 10 Further evidence is needed to understand youth substance use and inform prevention in the current Canadian context.
Current recommendations highlight strengths-based approaches to preventing problematic substance use among youth.Footnote 11Footnote 12 Strengths-based approaches include programs such as school-based prevention and comprehensive well-being initiatives. Rather than focussing on deficits, strengths-based approaches draw on protective factors, such as individual capabilities and aspects of positive development, which may include positive mental health.Footnote 11Footnote 12
The Public Health Agency of Canada defines positive mental health as “the capacity of each and all of us to feel, think, act in ways that enhance our ability to enjoy life and deal with the challenges we face. It is a positive sense of emotional and spiritual well-being that respects the importance of culture, equity, social justice, interconnections and personal dignity.”Footnote 13
The potential role of positive mental health as a protective factor for youth substance use has been inadequately studied. Furthering our knowledge of this association will help inform strengths-based prevention programs that focus on supporting positive mental health.
The present study measures positive mental health through the lens of the self-determination theory. This theory proposes that there are three basic psychological needs that contribute to overall well-being: competence, autonomy and relatedness.Footnote 14 Competence refers to a sense of self-efficacy or mastery to act in one’s environment. Autonomy refers to the perceived choice and control over the activities one completes. Relatedness involves a sense of closeness and belonging with others. Positive mental health, as conceptualized through the self-determination theory as these three needs, maps onto the psychological (i.e. autonomy and competence) and social (i.e. relatedness) well-being outcomes defined in the Positive Mental Health Surveillance Indicator Framework.Footnote 15
There is a dearth of research on how these three positive mental health needs impact behaviour, including substance use behaviours, in youth. However, some aspects of psychological well-being that overlap conceptually with the definition of positive mental health provided above, such as life satisfaction, locus of control (i.e. autonomy), resilience and positive affect, have been associated with drug and alcohol use among adolescents.Footnote 16Footnote 17Footnote 18 Two studies examined flourishing, which is a measure of overall positive mental health, among Grade 9 to 12 students in British Columbia and Ontario. These studies found that flourishing was associated with a lower likelihood of cannabis use but was not significantly associated with binge-drinking.Footnote 19Footnote 20
Overall, there has been little investigation of positive mental health and substance use, with most previous studies focussing on adults and postsecondary students.Footnote 21Footnote 22 The purpose of this study was to examine the associations between the positive mental health concepts, as conceptualized through the self-determination theory, of competence, relatedness and autonomy and substance use, including alcohol use, binge-drinking and cannabis use, among youth. Based on the authors’ knowledge, the associations between competence, relatedness and autonomy and substance use in youth across Canada have not previously been investigated; thus, this study will contribute to addressing this knowledge gap.
We analyzed data from the 2014/2015 Canadian Student Tobacco, Alcohol and Drugs Survey (CSTADS) public use microdata file.Footnote 23 CSTADS is a national, biennial survey on tobacco, alcohol and drug use among Canadian youth. This school-based survey is administered to a sample of students in Grades 6 to 12 (Grades 6 to secondary V in Quebec). In 2014/2015, a provincially generalizable sample was achieved in all provinces except New Brunswick.Footnote 23 The population consisted of students attending private or public schools. Schools in the Canadian territories, special schools (e.g. virtual schools, schools on military bases, schools on First Nations reserves or schools for students with visual or hearing impairment or special needs) and schools with fewer than 20 students enrolled in at least one eligible grade were excluded. Schools were selected using a stratified single-stage cluster design based on health-region smoking rate and type of schools (elementary or secondary). Random sampling within each stratum was used to select schools. All eligible students in the selected school were invited to participate.Footnote 23Footnote 24
A total of 336 schools and 42 094 students participated in the CSTADS, representing 47% and 66% participation rates, respectively. Data collection was completed between October 2014 and May 2015.Footnote 23Footnote 24 Participants completed a paper-and-pencil questionnaire during class-time. Students could decline participating in the survey at the time of data collection. The survey took approximately 30 minutes to complete. Teachers provided instructions on how to complete the survey, but did not circulate the room during the administration to protect confidentiality.
Questions about alcohol and cannabis use were only asked of students in Grades 7 to 12,Footnote 23Footnote 24 and Grade 6 students were excluded from the analyses. The resulting sample size was 36 665.
Health Canada Research Ethics Board, the University of Waterloo and the ethics review boards of the participating school boards provided ethics approval.
Alcohol use in the past 30 days was measured with two questions in a skip pattern, beginning with the following yes/no item: “Have you ever had a drink of alcohol that was more than just a sip?” Respondents who indicated “yes” were then asked: “In the last 30 days, how often did you have a drink of alcohol that was more than just a sip?” Response options included: “I did not drink alcohol in the last 30 days (coded as no)”; “Once or twice,” “Once or twice a week,” “3–4 times a week,” “5–6 times a week,” “Every day (all coded as yes)”; and “I don’t know.” “I don’t know” responses for all substance use questions were treated as “not stated” based on guidance in the CSTADS users’ guide,Footnote 24 and excluded from analyses. A second variable representing more frequent alcohol use was created. Those who responded that they drank once or twice a week or less often were coded as less frequent use; those responding 3–4 times a week or more were coded as more frequent use. The CSTADS defined a drink of alcohol as “1 regular sized bottle, can, or draft of beer; 1 glass of wine; 1 bottle of cooler; 1 shot of liquor (rum, whisky, Baileys®, etc.); or 1 mixed drink (1 shot of liquor with pop, juice, energy drink, etc.).”
Binge-drinking in the past 30 days was measured with the following item: “In the last 30-days, how often did you have 5 or more drinks of alcohol on one occasion?” Participants chose one of the following responses: “I have never done this” or “I did not have 5 or more drinks on one occasion in the last 30 days” (coded as no); “once,” “2 times,” “3 times,” “4 times” or “5 times or more” (all coded as yes); or “I don’t know” (excluded from analysis). A second variable representing more frequent binge-drinking was also created. Those responding that they did not drink to those binge-drinking once or twice in the past 30 days were coded as less frequent; those responding 3 times or more were coded as more frequent.
Two questions measured cannabis use in past 30 days. First, respondents were asked: “Have you ever used or tried marijuana or cannabis (a joint, pot, weed, hash, or hash oil)?” Possible responses were either “yes” or “no.” Students who responded “yes” were then asked: “In the last 30 days, how often did you use marijuana or cannabis?” Response categories included: “I did not use marijuana in the last 30 days” (coded as no); “Once or twice,” “Once or twice a week,” “3–4 times a week,” “5–6 times a week” or “Every day” (all coded as yes); or “I don’t know” (excluded from analysis). A second variable representing frequent cannabis use was also created. Those responding that they did not use cannabis to those responding once or twice a week were coded as less frequent; those responding 3–4 times a week or more were coded as more frequent.
Children’s Intrinsic Needs Satisfaction Scale
The Children’s Intrinsic Needs Satisfaction Scale (CINSS)Footnote 25 is an 18-item scale that consists of three subscales (the need for autonomy, competence and relatedness), which each contain six items. The CINSS includes questions to assess the satisfaction with each need in three different contexts (with peers, at home and at school). These result in 2 items per context/concept pair. Example items include: “I feel I do things well at home” (competence/at home) and “I feel free to express myself at school” (autonomy/at school). Response categories for each statement include “really false for me,” “sort of false for me,” “sort of true for me” or “really true for me.” The CINSS has been validated in a sample of Canadian students, demonstrating good internal consistency and criterion-related and factorial validity.Footnote 26
Scores on each of the three subscales were summed and transformed to create continuous total scores for competence, relatedness and autonomy that range between 10 and 40. Higher scores indicate greater satisfaction with the measured need. Variables for high and low autonomy, competence and relatedness were created by ascribing a high value to those with a mean score of 3 or above on the six items in each scale and a low value to those with a mean score of 0 to less than 3.Footnote 27 The continuous scores are used throughout the analyses, except for the initial descriptive analyses of the prevalence of substance use behaviours by high versus low autonomy, competence and relatedness.
Demographic content included current grade (Grades 6–12; Quebec secondary I, II, III, IV and V were coded as Grades 7–11, respectively) and sex (male, female). Socioeconomic status was defined by the median household income for each participating school’s region. This variable was obtained using the Canadian 2011 census data on household income by the first three digits of each school’s postal code (forward sortation area). Urban/rural region was categorized based on whether a participant’s school was located in an urban or rural region.Footnote 23
We generated descriptive statistics to describe the study sample and prevalence of cannabis use, alcohol use and binge-drinking for the whole sample, by grade and by sex. The prevalence of substance use by high versus low autonomy, competence and relatedness was examined. We used chi-square tests to determine differences in substance use by sex, and the Cochran–Armitage test for trend to test whether substance use systematically increased with grade level. Bonferroni corrections were applied.
We conducted a series of logistic regressions to test the associations between autonomy, competence and relatedness and substance use variables. The first regression model included continuous CINSS subscale scores as predictor variables and past month alcohol use as the outcome variable. We then repeated the regression model for each of the remaining outcome variables: past month cannabis use; past month binge-drinking and more frequent alcohol use; cannabis use; and binge-drinking in the past month. This series of regressions were then repeated with grade, urban versus rural location and area-based household income included as control variables. We also conducted a sensitivity analysis to test for provincial status as a control; no changes in the data were found.
To identify differences between boys and girls in the associations between autonomy, competence and relatedness and substance use, interaction terms by sex were first examined in the logistic regression models. When interactions terms were significant, models were stratified by sex. Interaction terms were not interpreted; however, coefficients are presented in Table 3 to justify sex-stratified analyses.
Analyses were conducted in statistical package SAS Enterprise Guide 7.1 (SAS Institute Inc., Cary, NC, USA), using the PROC SURVEY suite of commands. Bootstrapping and survey weights were applied to account for the complex sample design of the CSTADS.Footnote 23 Complete case analysis was conducted to exclude cases with missing data on the study variables.
Prevalence of cannabis and alcohol use and binge-drinking
Based on the 2014/2015 CSTADS data, 10.8% of students in Canada between Grades 7 and 12 reported using cannabis at least once in the past 30 days (Table 1). The prevalence of alcohol use within the past month was 27.0%, and 16.0% of Grade 7–12 students reported binge-drinking in the same period. Cochran–Armitage tests for trend revealed significant associations (all p ≤ .001) between increasing substance use and grade, with the highest prevalence of cannabis use (21.3%), alcohol use in the past month (47.3%) and binge-drinking in the past month (32.6%) observed among Grade 12 students.
Table 1. Prevalence of any 30-day and more frequent cannabis and alcohol use behaviours by sex, grade and urban/rural status, CSTADS, 2014–2015 (n = 36 665)
|Description||Any 30-day use||More frequent use|
% (95% CI)
% (95% CI)
% (95% CI)
% (95% CI)
% (95% CI)
% (95% CI)
|All||10.8 (10.6–11.1)||27.0 (26.5–27.6)||16.0 (15.7–16.4)||3.7 (3.5–3.9)||1.9 (1.8–2.0)||4.9 (4.8–5.1)|
|Female||10.2 (9.8–10.5)||27.4 (26.8–27.9)||15.7 (15.3–16.1)||2.9 (2.7–3.2)||1.5 (1.4–1.6)||4.7 (4.5–4.9)|
|Male||11.5 (11.1–11.9)Footnote * of Table 1||26.7 (26.1–27.3)Footnote * of Table 1||16.4 (15.9–16.8)Footnote * of Table 1||4.5 (4.3–4.7)Footnote * of Table 1||2.4 (2.2–2.5)Footnote * of Table 1||5.2 (4.9–5.4)Footnote * of Table 1|
|7||0.9 (0.8–1.1)||5.8 (5.3–6.3)||1.4 (1.2–1.6)||0.2 (0.1–0.2)||0.2 (0.1–0.2)||0.2 (0.2–0.2)|
|8||3.4 (3.0–3.7)||11.3 (10.6–12.0)||4.4 (4.0–4.8)||1.1 (0.9–1.2)||0.8 (0.1–0.9)||1.1 (0.9–1.3)|
|9||7.4 (6.8–8.0)||21.3 (20.1–22.4)||10.3 (9.7–10.9)||2.4 (2.0–2.8)||1.2 (1.1–1.3)||2.7 (2.5–2.9)|
|10||12.4 (11.8–13.0)||33.1 (32.4–33.9)||19.1 (18.4–19.9)||4.5 (4.2–4.8)||2.0 (1.8–2.2)||4.4 (4.1–4.7)|
|11||18.9 (18.4–19.4)||41.5 (40.4–42.7)||28.0 (27.1–28.9)||5.8 (5.5–6.1)||3.3 (3.0–3.6)||8.8 (8.2–9.4)|
|12||21.3 (20.3–22.4)Footnote ** of Table 1||47.3 (46.0–48.7)Footnote ** of Table 1||32.6 (31.5–33.6)Footnote ** of Table 1||8.1 (7.6–8.6)Footnote ** of Table 1||3.9 (3.3–4.5)Footnote ** of Table 1||12.4 (11.7–13.1)Footnote ** of Table 1|
|Urban/rural statusFootnote a of Table 1|
|Urban||10.4 (10.1–10.8)||24.1 (23.2–24.9)||14.1 (13.5–14.6)||3.6 (3.4–3.8)||1.7 (1.6–1.9)||4.2 (4.0–4.4)|
|Rural||12.4 (11.7–13.1)Footnote * of Table 1||38.4 (37.0–40.0)Footnote * of Table 1||23.8 (22.7–25.0)Footnote * of Table 1||4.1 (3.6–4.6)Footnote * of Table 1||2.6 (2.4–2.8)Footnote * of Table 1||8.0 (7.5–8.5)Footnote * of Table 1|
The prevalence of more frequent use was lower than that of any 30-day use, at 3.7%, 1.9% and 4.9% for more frequent cannabis use, alcohol use and binge-drinking, respectively. Consistent with past 30-day use, the highest prevalence of more frequent use was among Grade 12 students, at 8.1%, 3.9% and 12.4% for cannabis use, alcohol use and binge-drinking, respectively.
In all cases, the prevalences of past 30-day and more frequent cannabis use, alcohol use and binge-drinking were higher among those with low levels of autonomy (p ≤ .001), relatedness (p ≤ .001) and competence (p ≤ .001) than among those with high levels (Table 2). However, the magnitude of these differences varied by substance and by measured need. For example, the prevalence of more frequent binge-drinking was almost twice as high among those with low competence (7.9%) than those with high competence (4.1%), while the difference was smaller between the prevalence of more frequent cannabis use among those with low versus high autonomy (3.4% and 3.0%, respectively).
Table 2. PrevalenceFootnote a of Table 2 of any 30-day and more frequent cannabis use, alcohol use and binge-drinking by low versus high autonomy, competence and relatedness, CSTADS, 2014–2015 (n = 36 665)
|Characteristic||Any 30-day use||More frequent use|
% (95% CI)
% (95% CI)
% (95% CI)
% (95% CI)
% (95% CI)
% (95% CI)
|LowFootnote b of Table 2||14.1 (13.7–14.5)||29.9 (29.3–30.5)||18.0 (17.5–18.6)||3.4 (3.1–3.7)||5.1 (4.9–5.4)||6.2 (5.8–6.6)|
|HighFootnote b of Table 2||9.3 (9.0–9.6)Footnote * of Table 2||25.9 (25.2–26.5)Footnote * of Table 2||15.2 (14.8–15.6)Footnote * of Table 2||3.0 (2.8–3.2)Footnote * of Table 2||1.4 (1.3–1.5)Footnote * of Table 2||4.4 (4.3–4.6)Footnote * of Table 2|
|LowFootnote b of Table 2||18.5 (17.8–19.2)||35.3 (34.7–36.0)||22.0 (21.7–22.7)||6.6 (6.2–7.0)||3.9 (3.7–4.2)||7.9 (7.5–8.3)|
|HighFootnote b of Table 2||8.7 (8.5–9.0)Footnote * of Table 2||24.8 (24.2–25.5)Footnote * of Table 2||14.5 (14.0–14.9)Footnote * of Table 2||2.7 (2.6–2.9)Footnote * of Table 2||1.4 (1.3–1.5)Footnote * of Table 2||4.1 (4.0–4.3)Footnote * of Table 2|
|LowFootnote b of Table 2||19.0 (18.4–19.7)||36.2 (35.5–36.8)||23.3 (22.8–23.8)||7.6 (7.2–8.0)||4.2 (3.9–4.5)||7.9 (7.5–8.4)|
|HighFootnote b of Table 2||8.7 (8.5–9.0)Footnote * of Table 2||25.0 (24.4–25.6)Footnote * of Table 2||14.4 (14.0–14.8)Footnote * of Table 2||2.6 (2.5–2.8)Footnote * of Table 2||1.4 (1.3–1.5)Footnote * of Table 2||4.3 (4.1–4.4)Footnote * of Table 2|
Results of logistic regression models
In models of autonomy, competence and relatedness and substance use that included sex as an interaction term, many associations demonstrated a significant interaction with sex. Interactions between sex and competence, as well as between sex and relatedness, for all three 30-day substance use behaviours were significant. (See Table 3 for unexponentiated estimates and their standard errors.) The interaction between sex and autonomy was also significant for 30-day binge-drinking. Interactions between sex and competence were significant for all more frequent substance use behaviours. In addition, the interaction between sex and autonomy was significant for more frequent cannabis use. No interactions between sex and relatedness were significant for more frequent substance use variables. All subsequent models are stratified by sex.
Table 3. Unexponentiated estimates, standard errors and p values for logistic regression interaction terms between sex and autonomy, competence and relatedness for 30-day and more frequent cannabis use, alcohol use and binge-drinking, CSTADS, 2014–2015 (n = 36 665)
|Behaviour and characteristic||Cannabis||Alcohol||Binge-drinking|
|Estimate||SE||p value||Estimate||SE||p value||Estimate||SE||p value|
|Any 30-day use|
|Competence||−0.030||0.002||< .0001||−0.014||0.002||< .0001||−0.021||0.003||< .0001|
|More frequent use|
|Competence||−0.042||0.006||< .0001||0.008||0.009||< .0001||−0.010||0.003||.01|
In unadjusted models that examined autonomy, competence and relatedness separately with 30-day cannabis use, alcohol use and binge-drinking, all odds ratios were significant for both girls and boys; higher scores were associated with lower odds of each of these behaviours (Table 4). The same pattern was observed for more frequent cannabis use, alcohol use and binge-drinking (Table 5).
Table 4. Unadjusted and fully adjusted odds ratios of any 30-day cannabis use, alcohol use and binge-drinking by autonomy, competence and relatedness, stratified by sex, CSTADS, 2014–2015 (n = 36 665)
OR (95% CI)Footnote * of Table 4
OR (95% CI)Footnote * of Table 4
OR (95% CI)Footnote * of Table 4
|Unadjusted odds ratios|
|Fully adjusted odds ratios|
Table 5. Unadjusted and fully adjusted odds ratios of more frequent cannabis, alcohol use and binge-drinking by autonomy, competence and relatedness, stratified by sex, CSTADS, 2014–2015 (n = 36 665)
|Characteristics||More frequent cannabis use
OR (95% CI)
|More frequent alcohol use
OR (95% CI)
|More frequent binge-drinking
OR (95% CI)
|Unadjusted odds ratios|
|Fully adjusted odds ratios|
|Girls||Autonomy||1.07||(1.05–1.08)||1.03||(1.00–1.07)Footnote ** of Table 5||1.04||(1.03–1.06)|
|Competence||1.01||(0.99–1.03)Footnote ** of Table 5||0.98||(0.96–1.00)||0.95||(0.94–0.96)|
After adjusting for grade, income and urban versus rural school location and including continuous scores of autonomy, competence and relatedness in the model simultaneously, higher competence and relatedness continued to be associated with a significantly lower odds of 30-day substance use, while autonomy was associated with a significantly higher odds, for both boys and girls (Table 4). For each unit increase in relatedness, the odds of cannabis use, alcohol use and binge-drinking decreased by 8%, 5% and 4% for girls and 2%, 3% and 2% for boys, respectively. For each unit increase in competence, the odds of cannabis use, alcohol use and binge-drinking decreased by 7%, 4% and 4% for girls and 9%, 5% and 7% for boys, respectively. In contrast, for each unit increase in autonomy, the odds of cannabis use, alcohol use and binge-drinking increased by 6%, 4% and 4% for girls and 5%, 4% and 6% for boys, respectively (Table 4).
Results of models of more frequent cannabis use, alcohol use and binge-drinking were similar to those for 30-day substance use (Table 5). Again, higher levels of competence and relatedness were associated with significantly lower odds of more frequent cannabis use, more frequent alcohol use and more frequent binge-drinking, for both girls and boys. The only exception was for the association between competence and more frequent cannabis use, which was not significant for boys (Table 5). Higher levels of autonomy were associated with significantly higher odds of more frequent cannabis use, more frequent alcohol use and more frequent binge-drinking for boys and girls. The exception was more frequent alcohol use among girls, where the association was not significant.
This study provides national estimates of past 30-day and more frequent cannabis use, alcohol use and binge-drinking use in relation to levels of autonomy, competence and relatedness among youth. Higher levels of competence, a sense of self-efficacy or mastery to act in one’s environment, was associated with a lower odds of each of three 30-day substance use behaviours and a lower odds of all but one more frequent substance use behaviours.
These results align with those of studies that have suggested an association between greater self-efficacy and lower levels of substance use among adolescents.Footnote 28Footnote 29 Our finding that relatedness, a greater sense of closeness and belonging with others, was associated with lower odds of all 30-day and more frequent substance use behaviours is consistent with previous research that has noted the importance of social relationships with peers and family in preventing youth substance use.Footnote 30Footnote 31 The findings for relatedness and competence are consistent with studies demonstrating a link between higher overall scores of positive mental health, such as flourishing, and lower levels of substance use.Footnote 19Footnote 21Footnote 22
Given the large number of significant interactions between positive mental health concepts and sex for substance use behaviours, examining these differences is an important area of future research, as the small body of past research on this subject have not noted any sex/gender differences.Footnote 19Footnote 21Footnote 22
Without accounting for grade, urban/rural location, income, competence and relatedness, higher autonomy was associated with a lower prevalence of substance use. However, after controlling for covariates, higher autonomy was significantly related with higher odds of all substance use behaviours except for more frequent alcohol use among girls.
The change in the association between higher autonomy and lower odds of substance use in the unadjusted findings to higher autonomy and higher odds of substance use in the adjusted model was unexpected. Conclusions are mixed on whether youth autonomy and substance use is significantly associated and whether higher autonomy is associated with higher or lower substance use.Footnote 32Footnote 33Footnote 34 Having a sense of choice and control is an aspect of positive mental health and fosters a healthy transition through adolescence to adulthood. The timing of adolescent autonomy may be pertinent, as one study suggested that early development is linked with increased substance use among youth.Footnote 34 However, autonomy may also be a factor that could be leveraged in substance use prevention.
Some prevention efforts have focussed on supporting youth autonomy in decision-making about substance use. A systematic review of mass media campaigns that aim to prevent substance use revealed four interventions that had some evidence of beneficial effects.Footnote 35 Two of these four interventions emphasized non-use of drugs as a means to support autonomy. The two interventions included the “Be Under Your Own Influence” and “Above the Influence” media campaigns in the United States, which have been associated with lower use of cannabis.Footnote 36 However, another media campaign that emphasized self-efficacy (“My Anti Drug”), which was included in the same systematic review by Allara et al., was among the interventions associated with harmful effects.Footnote 35 Overall, the evidence supporting the effectiveness of mass media campaigns to prevent drug use is weak.Footnote 35 The association between autonomy and youth substance use appears to be a nuanced one that may involve developmental and contextual factors and warrants further research.
The findings we present in this article provide new insight into the associations between autonomy, competence and relatedness and recent substance use among youth, based on analysis of a large, representative survey. This information is particularly relevant in the current context of changing alcohol and drug policies and trends in youth substance use.
The 2018 Chief Public Health Officer report highlighted the role of prevention, including the importance of promoting resilience and protective factors, in addressing problematic substance use among youth.Footnote 37 Strengths-based approaches to prevention have come to the forefront of many recommendations, emphasizing the need to better understand protective factors in relation to problematic substance use among youth in Canada.Footnote 12Footnote 38 The findings from the present study further support the potential roles of competence and relatedness as protective factors for substance use. They are consistent with, for example, the Ontario Ministry of Health and Long-Term Care guideline on substance use prevention and harm reduction, which gives examples of protective factors that may decrease the likelihood of substance use and related harms.Footnote 38 Examples include competence and factors that align with relatedness (i.e. positive parent relationships and positive teacher and social connectedness at school).Footnote 38
A systematic literature review of strengths-based youth development programs that promote the positive development of “skills, attitudes, relationships and identities” and include factors that align with relatedness and competence was recently conducted. This review identified several mechanisms through which these programs may reduce substance use. These include enhancing protective factors in general to buffer against risk factors for substance use, enhancing a specific protective factor to reduce a specific risk behaviour and “piling up” multiple protective factors.Footnote 39 Whether relatedness and competence act through one or several of these mechanism warrants future research.
Strengths and limitations
There are limitations that should be considered when interpreting the results of this study.
The findings presented are observational, and further investigation is needed to draw conclusions on the causal role of autonomy, competence and relatedness in problematic substance use prevention. Each of the three subscales combined scores across three contexts (home, school and with peers) and the present study was limited to include scores combined across contexts. Further investigation into the associations between substance use and autonomy between contexts may be warranted.
The present study was limited to cannabis and alcohol use, including binge-drinking; other forms of substance use were not examined.
Complete case analysis, which confines analyses to cases with complete data, was used to handle missing data. This may have resulted in bias due to the loss of information.
The analyses were conducted with the 2014/2015 cycle of the CSTADS, rather than the 2016/2017 cycle, due to data availability at the time the project was initiated. However, these analyses examined associations, not prevalence; we would not expect notable changes in the associations between variables between data collection cycles. Future research may investigate if associations are moderated by time.
Items on substance use and positive mental health relied on self-report and may have been subject to social desirability bias. The active consent process may have introduced bias into the sample, or may have affected participants’ responses.Footnote 40
Less than half of schools participated, which may have also biased the sample. As this analysis was conducted with a cross-sectional survey, no inferences about causality can be made. However, to our knowledge, this is the first research using a large, representative sample of Canadian youth to examine the association between positive mental health and substance use, using a measure that has been validated in this age group.Footnote 26
These findings provide evidence to further elucidate the associations between aspects of positive mental health and substance use among youth in Canada. The results of this study provide contextual evidence that can be used to further our understanding of risk and protective factors associated with substance use, inform health promotion and substance use prevention programs and prevent or reduce problematic substance use among youth.
Data used for this research were taken from Health Canada’s Canadian Student Tobacco, Alcohol and Drugs Survey (CSTADS), which was conducted for Health Canada by the Propel Centre for Population Health Impact at the University of Waterloo. Health Canada has not reviewed, approved nor endorsed this research. Any views expressed or conclusions drawn herein do not necessarily represent those of Health Canada or the Public Health Agency of Canada.
We appreciate the helpful comments provided by Dr. Katelyn Godin on an earlier draft of this article and the assistance of Adena Pinto and Matthew Hunt with tables and data verification.
Conflict of interest
The authors have no conflicts of interest to disclose.
Authors’ contributions and statement
AE and HO conceived the project.
AE conducted the analysis.
AE and HO interpreted the results.
AE drafted the article and HO critically revised the article.
The content and views expressed in this article are those of the authors and do not necessarily reflect those of the Government of Canada.
Report a problem or mistake on this page
- Date modified: