Original quantitative research – Investigating individual-level correlates of e-cigarette initiation among a large sample of Canadian high school students
HPCDP Journal Home
Published by: The Public Health Agency of Canada
Date published: October 2021
ISSN: 2368-738X
Submit a manuscript
About HPCDP
Browse
Previous | Table of Contents | Next
Gillian C. Williams, MScAuthor reference footnote 1Author reference footnote 2; Adam G. Cole, PhDAuthor reference footnote 3; Margaret de Groh, PhDAuthor reference footnote 2; Ying Jiang, MD, MScAuthor reference footnote 2; Scott T. Leatherdale, PhDAuthor reference footnote 1
https://doi.org/10.24095/hpcdp.41.10.04
This article has been peer reviewed.
Author references
Correspondence
Gillian C. Williams, School of Public Health Sciences, University of Waterloo, 200 University Ave. W., Waterloo, ON N2L 3G1; Email: gillian.williams@uwaterloo.ca
Suggested citation
Williams GC, Cole AG, de Groh M, Jiang Y, Leatherdale ST. Investigating individual-level correlates of e-cigarette initiation among a large sample of Canadian high school students. Health Promot Chronic Dis Prev Can. 2021;41(10):292-305. https://doi.org/10.24095/hpcdp.41.10.04
Abstract
Introduction: Having a better understanding of individual factors associated with e-cigarette initiation can help improve prevention efforts. Therefore, this study aimed to (1) identify baseline characteristics associated with e-cigarette initiation, and (2) determine whether changes in these baseline characteristics were associated with e-cigarette initiation.
Methods: This study used data from Year 6 (2017/18, baseline) and Year 7 (2018/19, follow-up) of the COMPASS study. The final sample included 12 315 students in Grades 9 to 11 who reported never having tried e-cigarettes at baseline. Students reported demographic information, other substance use, school behaviours, physical activity, sedentary behaviour, sleep, symptoms of anxiety and depression, and emotional regulation and flourishing scores. Hierarchical GEE models, stratified by gender, examined the association between (1) baseline characteristics and e-cigarette initiation at follow-up and (2) changes in baseline characteristics and e-cigarette initiation at follow-up.
Results: In total, 29% of students who had not yet initiated e-cigarette use reported initiating e-cigarette use at follow-up. Students in Grades 10 and 11 were less likely to initiate e-cigarette use. Other substance use, skipping school and meeting the physical activity guidelines at baseline and one-year changes to these behaviours were associated with e-cigarette initiation among both male and female students. Additionally, some differences were noted between females and males.
Conclusion: Given that other health behaviours were associated with e-cigarette initiation, prevention approaches should target multiple health-risk behaviours to help prevent youth e-cigarette use. Additionally, school-based approaches may benefit by being implemented at the beginning of high school or in junior high school.
Keywords: vaping, adolescent, alcohol drinking, cannabis smoking, cigarette smoking, mental health, exercise, sedentary behaviour
Highlights
- Twenty-nine percent of students who had not yet initiated e-cigarette use reported initiating e-cigarette use at follow-up. Other substance use (i.e. alcohol, cannabis and cigarettes) was strongly associated with e-cigarette initiation.
- Students who met the Canadian physical activity guidelines were more likely to initiate e-cigarette use, and female students who met the screen time guidelines were less likely.
- Anxiety and depression were not significantly associated with e-cigarette initiation, but there was an association with higher emotional dysregulation for females and higher flourishing for males.
- The majority of students maintained their behaviours over time; results for changes from baseline were largely consistent with findings at baseline.
Introduction
E-cigarettes are rapidly evolving devices that deliver an aerosol (or another substance), often containing nicotine, to the user in the absence of tobacco and combustion.Footnote 1 The prevalence of e-cigarette use, also known as vaping, has increased dramatically among youth in recent years.Footnote 2Footnote 3Footnote 4Footnote 5 Both Canada and the United States have seen notable increases in e-cigarette use among adolescents.Footnote 4 The US saw an increase in prevalence among high school students from 1.5% to 20.8% from 2011 to 2018, with the largest jump between 2017 and 2018 (from 12% to 21%).Footnote 5 Similarly, e-cigarette use among adolescents aged 15 to 19 in Canada doubled from 10% in 2016 to 20% in 2018.Footnote 2 Among Canadian adolescents who report e-cigarette use, 40% report daily or almost daily use and 90% report using products with nicotine.Footnote 2 E-cigarette use among youth is concerning due to the unknown effects of exposure to aerosolized chemicals and the known negative impacts of nicotine on the developing brain.Footnote 6Footnote 7
While many studies have linked e-cigarette use and cigarette initiation,Footnote 8 few have focussed on the individual factors associated with e-cigarette use initiation. Evidence suggests that tobacco use (e.g. cigars, cigarillos) and substance use,Footnote 9Footnote 10Footnote 11Footnote 12 high levels of sensation seeking,Footnote 13Footnote 14 poor mental health,Footnote 15Footnote 16 exposure to e-cigarette marketing, Footnote 9Footnote 17Footnote 18 and positive attitudes towards e-cigarettes among individuals, friends and familyFootnote 9Footnote 19Footnote 20Footnote 21 are factors associated with initiating e-cigarette use. While many studies have identified a significant association between e-cigarette initiation and gender, they have not explored further gender differences in predictive factors.Footnote 9Footnote 10Footnote 13Footnote 14Footnote 15Footnote 18Footnote 19 A better understanding of individual factors associated with e-cigarette initiation can improve prevention efforts by identifying the characteristics (both modifiable and nonmodifiable) of at-risk groups. However, there has been little investigation into the influence of other risk behaviours (e.g. truancy, poor grades), movement behavioursFootnote 22 and mental well-being on e-cigarette initiation, and how these change over time.
There is a well-established literature demonstrating that less healthy behaviour patterns among adolescents increase over time. For example, substance use and screen time tend to increase over time, while physical activity and sleep tend to decrease with age. Footnote 2Footnote 23Footnote 24Footnote 25 Although many of these changes have been well documented, there is a lack of evidence concerning how changes over time are associated with more novel experiences, including e-cigarette initiation. To date, studies examining e-cigarette initiation have examined baseline behaviours only, and it is unknown how changes in behaviour may be associated with e-cigarette initiation.
Given the novelty of e-cigarettes, there is a need to further explore the individual-level factors that are associated with e-cigarette initiation among adolescents. The objectives of this study were to identify (1) the baseline characteristics associated with e-cigarette initiation, and (2) whether changes in these baseline characteristics were associated with e-cigarette initiation among Canadian adolescents.
Methods
Host study
COMPASS is a prospective cohort study that collects data from students in Grades 9 to 12 (aged 13–18 years) in British Columbia, Alberta and Ontario, and in Secondary I–V (aged 12–17 years) in Quebec, Canada. Footnote 26 All procedures were approved by the University of Waterloo Office of Research Ethics (reference number 30118) and appropriate school board committees. A full description of the COMPASS study methods can be found in printFootnote 26 or online.
Participants
This study used data from Year 6 (2017/18, baseline) and Year 7 (2018/19, follow-up) of the COMPASS study. A total of 40 388 students in Grades 9 to 11 (and Secondary III–V in Quebec) from 111 schools participated at baseline (81.5% participation rate), and 23 168 of these (57%) were linked across both baseline and follow-up. Linked students were younger, comprised more females and had lower frequencies of substance use, including e-cigarette use, at baseline (data available upon request). Students for whom information on covariates was missing at baseline or at both baseline and follow-up (n = 5338, 23%) were removed. Students who were missing data and those who were not did not differ by frequency of e-cigarette use (data available upon request). Finally, those who had ever tried e-cigarettes at baseline were also removed from the sample (n = 5515, 31%). Thirty-three percent of students removed based on this criterion were in Grade 9 (n = 1805), 42% were in Grade 10 (n = 2304), and 25% were in Grade 11 (n = 1406). The final sample included 12 315 students in Grades 9 to 11 who reported never having tried e-cigarettes at baseline. We additionally examined a subsample of students (n = 10 727) who had complete data both at baseline and follow-up to explore whether changes in these baseline characteristics were associated with e-cigarette initiation.
Measures
Student responses were captured using the COMPASS questionnaire, which was administered during class time. Consistent with other youth health research,Footnote 27 students reported their grade, gender, ethnicity and weekly spending money.
To identify e-cigarette initiation, students were asked, “Have you ever tried an electronic cigarette, also known as an e-cigarette?” Students who indicated “yes” at baseline were removed from the sample. Students who indicated “no” at baseline and “yes” at follow-up were considered to have initiated e-cigarette use.
The questionnaire also collected information on the use of other substances, including alcohol, cannabis and cigarettes. For alcohol and cannabis use, students were categorized as “monthly” users if they indicated use once per month or more, and “infrequent” users if they indicated use less than once per month. For cigarette use and e-cigarette use, students were categorized as “ever” users or “past month” users.
The questionnaire also collected data about behaviours at school, including skipping school in the past four weeks, and English grades (French grades in Quebec).
Additionally, students were asked to report the amount of time per day spent doing moderate-to-vigorous physical activity (MVPA), engaging in sedentary screen time activities (watching or streaming TV or movies, playing video or computer games, surfing the internet and texting, messaging or emailing), and sleeping. Students were categorized as meeting or not meeting the targets for each of these movement behaviours as set by the Canadian 24-Hour Movement Guidelines for Children and Youth.Footnote 22 It is recommended that each day children and youth should accumulate at least 60 minutes of MVPA, less than 2 hours of screen time, and 8 to 10 hours of uninterrupted sleep. Footnote 22
Finally, mental health and wellbeing were assessed using the Centre for Epidemiological Studies Depression Scale (CES-D-10),Footnote 28 the Generalized Anxiety Disorder 7 (GAD-7) scale,Footnote 29 the Difficulties in Emotional Regulation Scale (DERS),Footnote 30 and the Flourishing Scale.Footnote 31 The CES-D-10 and the GAD-7 are continuous scales ranging from 0 to 30 and 0 to 21 respectively, where a score of 10 or higher is indicative of clinically relevant symptomatology; scales were dichotomized to reflect this. Footnote 28Footnote 29 The DERS is a continuous scale with a range of 6 to 30, where a higher score indicates poorer emotional regulation. The Flourishing Scale is a continuous scale with a range of 8 to 40, where a higher score indicates better flourishing. Flourishing is a state of overall wellbeing used to describe the presence of mental health; Footnote 32 the scale included level of agreement with questionnaire items such as “I lead a purposeful and meaningful life,” “I am engaged and interested in my daily activities” and “I am optimistic about my future.” We modelled a 3-unit change in both DERS and flourishing scores to capture a relevant change in score (> 1/2 a standard deviation) that was not due to chance alone. These scales have been previously validated among adolescents.Footnote 28Footnote 30Footnote 33Footnote 34Footnote 35Footnote 36Footnote 37
Analyses
The analyses were conducted in two parts. First (Part 1), we examined the association between baseline individual-level characteristics and follow-up e-cigarette initiation. Chi-squared tests compared categorical variables and t-tests compared continuous variables across e-cigarette use initiation at follow-up. Generalized estimating equations (GEE) via PROC GENMOD in SAS with an exchangeable correlation structure were used to identify baseline variables associated with e-cigarette initiation at follow-up while accounting for the nesting of students within schools. We first ran partially adjusted models examining the association between each variable and e-cigarette initiation, adjusting only for province, grade and gender. We then ran fully adjusted models adjusting for all other variables. All models were stratified by gender due to known differences in behaviour and differences identified using chi square and bivariate analyses.
Second (Part 2), we explored how changes in individual-level characteristics between baseline and follow-up were associated with e-cigarette initiation at follow-up. Students were categorized into different groups based on the change in their behaviours between baseline and follow-up. For substance use, “abstainers” did not engage in a specific behaviour at baseline or follow-up; “maintainers” continued the same level of frequency of the behaviour at baseline and follow-up; “escalators” increased the frequency of the behaviour from baseline to follow-up; and “reducers” decreased the frequency of the behaviour from baseline to follow-up. For binary variables such as skipping school, meeting movement behaviour guidelines, and depression and anxiety, students were categorized based on having “yes” responses for “both years,” “neither year,” “follow-up only” or “baseline only.” For continuous variables, including DERS and flourishing, students were categorized as “no change,” “increase” or “decrease” based on the difference between their responses at baseline and at follow-up. We used the same analytic approach as in Part 1. All analyses were performed using SAS software, version 9.4 (SAS Institute Inc., Cary, NC, USA).
Results
Descriptive characteristics
Just over half of the sample was female (56%); 41% of students were in Grade 9, 36% in Grade 10 and 23% in Grade 11; 65% were White. Between baseline (2017/18) and follow-up one year later (2018/19), 29% of students who had not yet initiated e-cigarette use reported initiating e-cigarette use (Table 1).
Variable | Total (n = 12 315) |
Female (n = 6891) | Male (n = 5424) | |||||
---|---|---|---|---|---|---|---|---|
n | % | E-cigarette initiation status | E-cigarette initiation status | |||||
No (%) (n = 4791) |
Yes (%) (n = 2100) |
p-value | No (%) (n = 3907) |
Yes (%) (n = 1517) |
p-value | |||
Grade | ||||||||
9 | 5 049 | 41.0 | 40.3 | 41.1 | 0.83 | 41.5 | 42.0 | 0.16 |
10 | 4 478 | 36.4 | 36.9 | 36.6 | 35.3 | 37.1 | ||
11 | 2 788 | 22.6 | 22.8 | 22.4 | 23.2 | 20.9 | ||
Ethnicity | ||||||||
White | 8 041 | 65.3 | 61.3 | 73.5 | < 0.01 | 63.1 | 72.4 | < 0.01 |
Non-White | 4 274 | 34.7 | 38.7 | 26.5 | 36.9 | 27.6 | ||
Weekly spending money | ||||||||
Zero | 2 627 | 21.3 | 21.6 | 13.8 | < 0.01 | 26.3 | 17.9 | < 0.01 |
$1–$20 | 3 622 | 29.4 | 29.9 | 28.1 | 29.7 | 28.9 | ||
$21–$100 | 2 612 | 21.2 | 20.3 | 26.5 | 18.8 | 23.1 | ||
$100+ | 1 361 | 11.1 | 8.9 | 15.4 | 9.8 | 15.1 | ||
Don't know/missing | 2 093 | 17.0 | 19.3 | 16.2 | 15.4 | 15.0 | ||
Alcohol use | ||||||||
None | 8 325 | 67.6 | 74.4 | 41.2 | < 0.01 | 79.9 | 51.0 | < 0.01 |
Infrequent | 2 315 | 18.8 | 17.3 | 30.9 | 11.8 | 25.1 | ||
Monthly | 1 675 | 13.6 | 8.4 | 28.0 | 8.3 | 23.9 | ||
Cannabis use | ||||||||
None | 11 597 | 94.2 | 96.6 | 85.1 | < 0.01 | 97.5 | 90.4 | < 0.01 |
Infrequent | 410 | 3.3 | 2.1 | 8.5 | 1.3 | 5.3 | ||
Monthly | 308 | 2.5 | 1.3 | 6.4 | 1.2 | 4.4 | ||
Cigarette use | ||||||||
None | 11 673 | 94.8 | 97.1 | 87.7 | < 0.01 | 97.5 | 90.2 | < 0.01 |
Ever use | 480 | 3.9 | 2.3 | 8.7 | 2.1 | 7.1 | ||
Past month use | 162 | 1.3 | 0.6 | 3.7 | 0.4 | 2.7 | ||
Skipping school | ||||||||
No | 10 182 | 82.7 | 84.6 | 71.4 | < 0.01 | 88.1 | 78.4 | < 0.01 |
Yes | 2 133 | 17.3 | 15.5 | 28.6 | 11.9 | 21.6 | ||
English gradeFootnote a | ||||||||
80%–100% | 7 489 | 60.8 | 71.8 | 63.1 | < 0.01 | 53.1 | 43.0 | < 0.01 |
70%–79% | 3 115 | 25.3 | 19.9 | 24.6 | 29.2 | 33.4 | ||
60%–69% | 1 166 | 9.5 | 6.0 | 7.9 | 11.8 | 16.6 | ||
< 60% | 545 | 4.4 | 2.4 | 4.4 | 5.9 | 7.1 | ||
Meeting PA guidelines | ||||||||
No | 7 355 | 59.7 | 66.5 | 61.6 | < 0.01 | 55.8 | 45.8 | < 0.01 |
Yes | 4 960 | 40.3 | 33.5 | 38.4 | 44.2 | 54.3 | ||
Meeting screen time guidelines | ||||||||
No | 11 250 | 91.4 | 88.3 | 93.2 | < 0.01 | 93.0 | 94.1 | 0.15 |
Yes | 1 065 | 8.7 | 11.7 | 6.8 | 7.0 | 5.9 | ||
Meeting sleep guidelines | ||||||||
No | 6 872 | 55.8 | 57.8 | 60.1 | 0.07 | 51.9 | 53.7 | 0.22 |
Yes | 5 443 | 44.2 | 42.2 | 39.9 | 48.1 | 46.3 | ||
Depression symptoms | ||||||||
No | 8 366 | 67.9 | 62.9 | 54.4 | < 0.01 | 77.6 | 77.5 | 0.93 |
Yes | 3 949 | 32.1 | 37.1 | 45.6 | 22.4 | 22.5 | ||
Anxiety symptoms | ||||||||
No | 9 484 | 77.0 | 70.6 | 65.7 | < 0.01 | 87.3 | 86.6 | 0.46 |
Yes | 2 831 | 23.0 | 29.4 | 34.3 | 12.7 | 13.5 | ||
DERS (mean, SD) |
13.8, 4.6 | 14.3, 4.8 | 15.2, 4.9 | < 0.01 | 12.8, 4.1 | 13.0, 4.1 | 0.15 | |
Flourishing Scale (mean, SD) |
32.2, 5.4 | 32.1, 5.5 | 31.8, 5.5 | 0.03 | 32.2, 5.3 | 32.9, 4.8 | < 0.01 | |
Abbreviations: DERS, Difficulties in Emotional Regulation Scale; PA, physical activity; SD, standard deviation.
|
Part 1: Association between baseline characteristics and e-cigarette initiation
After adjusting for all other factors, being in Grade 10 or 11 was associated with lower odds of e-cigarette initiation among both female and male students compared to being in Grade 9 (Table 2). More spending money was associated with increased odds of e-cigarette initiation.
Variable | Female | Male | ||
---|---|---|---|---|
Partially adjusted aOR (95% CI) |
Fully adjusted aOR (95% CI) |
Partially adjusted aOR (95% CI) |
Fully adjusted aOR (95% CI) |
|
Grade | ||||
9 | 1.00 | 1.00 | 1.00 | 1.00 |
10 | 0.99 (0.87–1.13) | 0.65 (0.57–0.75)Footnote * | 1.06 (0.92–1.21) | 0.81 (0.70–0.94)Footnote * |
11 | 1.10 (0.96–1.25) | 0.54 (0.46–0.63)Footnote * | 1.02 (0.87–1.21) | 0.62 (0.51–0.75)Footnote * |
Ethnicity | ||||
White | 1.00 | 1.00 | 1.00 | 1.00 |
Non-White | 0.70 (0.61–0.80)Footnote * | 0.76 (0.66–0.87)Footnote * | 0.83 (0.70–0.98)Footnote * | 0.86 (0.72–1.03) |
Weekly spending money | ||||
Zero | 1.00 | 1.00 | 1.00 | 1.00 |
$1–$20 | 1.50 (1.29–1.75)Footnote * | 1.34 (1.14–1.58)Footnote * | 1.45 (1.20–1.75)Footnote * | 1.29 (1.06–1.58)Footnote * |
$21–$100 | 2.01 (1.70–2.39)Footnote * | 1.65 (1.37–1.99)Footnote * | 1.82 (1.50–2.19)Footnote * | 1.42 (1.16–1.74)Footnote * |
$100+ | 2.63 (2.14–3.24)Footnote * | 1.89 (1.50–2.38)Footnote * | 2.27 (1.86–2.77)Footnote * | 1.65 (1.32–2.06)Footnote * |
Don't know/missing | 1.26 (1.08–1.47)Footnote * | 1.19 (1.00–1.41)Footnote * | 1.43 (1.17–1.75)Footnote * | 1.29 (1.05–1.58)Footnote * |
Alcohol use | ||||
None | 1.00 | 1.00 | 1.00 | 1.00 |
Infrequent | 3.37 (2.92–3.88)Footnote * | 2.81 (2.42–3.25)Footnote * | 3.42 (2.88–4.04)Footnote * | 2.99 (2.50–3.57)Footnote * |
Monthly | 6.48 (5.50–7.63)Footnote * | 4.12 (3.45–4.91)Footnote * | 4.51 (3.71–5.48)Footnote * | 3.16 (2.55–3.93)Footnote * |
Cannabis use | ||||
None | 1.00 | 1.00 | 1.00 | 1.00 |
Infrequent | 4.68 (3.71–5.90)Footnote * | 1.87 (1.42–2.46)Footnote * | 4.45 (3.33–5.96)Footnote * | 2.00 (1.36–2.95)Footnote * |
Monthly | 5.85 (4.29–7.97)Footnote * | 1.69 (1.17–2.45)Footnote * | 4.05 (2.56–6.42)Footnote * | 1.26 (0.75–2.14) |
Cigarette use | ||||
None | 1.00 | 1.00 | 1.00 | 1.00 |
Ever use | 4.38 (3.36–5.70)Footnote * | 2.13 (1.54–2.96)Footnote * | 3.89 (2.96–5.11)Footnote * | 2.52 (1.87–3.39)Footnote * |
Past month use | 6.38 (4.16–9.79)Footnote * | 1.72 (1.00–2.94)Footnote * | 8.35 (4.23–16.69)Footnote * | 4.28 (2.02–9.06)Footnote * |
Skipping school | ||||
No | 1.00 | 1.00 | 1.00 | 1.00 |
Yes | 2.28 (1.97–2.63)Footnote * | 1.54 (1.32–1.79)Footnote * | 2.01 (1.72–2.35)Footnote * | 1.42 (1.20–1.67)Footnote * |
English gradeFootnote a | ||||
80%–100% | 1.00 | 1.00 | 1.00 | 1.00 |
70%–79% | 1.37 (1.21–1.55)Footnote * | 1.23 (1.07–1.41)Footnote * | 1.32 (1.16–1.51)Footnote * | 1.33 (1.15–1.54)Footnote * |
60%–69% | 1.48 (1.20–1.83)Footnote * | 1.26 (0.97–1.63) | 1.57 (1.27–1.94)Footnote * | 1.55 (1.24–1.92)Footnote * |
< 60% | 2.23 (1.77–2.82)Footnote * | 1.68 (1.28–2.21)Footnote * | 1.44 (1.13–1.84)Footnote * | 1.43 (1.09–1.88)Footnote * |
Meeting PA guidelines | ||||
No | 1.00 | 1.00 | 1.00 | 1.00 |
Yes | 1.27 (1.15–1.41)Footnote * | 1.14 (1.02–1.27)Footnote * | 1.54 (1.34–1.77)Footnote * | 1.35 (1.16–1.57)Footnote * |
Meeting screen time guidelines | ||||
No | 1.00 | 1.00 | 1.00 | 1.00 |
Yes | 0.53 (0.44–0.63)Footnote * | 0.69 (0.58–0.84)Footnote * | 0.87 (0.68–1.11) | 0.97 (0.75–1.27) |
Meeting sleep guidelines | ||||
No | 1.00 | 1.00 | 1.00 | 1.00 |
Yes | 0.82 (0.73–0.92)Footnote * | 0.90 (0.80–1.02) | 0.88 (0.76–1.02) | 0.93 (0.80–1.06) |
Depression symptoms | ||||
No | 1.00 | 1.00 | 1.00 | 1.00 |
Yes | 1.51 (1.34–1.70)Footnote * | 1.13 (0.98–1.31) | 1.05 (0.88–1.25) | 1.00 (0.80–1.24) |
Anxiety symptoms | ||||
No | 1.00 | 1.00 | 1.00 | 1.00 |
Yes | 1.29 (1.15–1.44)Footnote * | 0.89 (0.75–1.05) | 1.10 (0.91–1.33) | 1.02 (0.80–1.30) |
DERS (3-unit increase) |
1.14 (1.10–1.18)Footnote * | 1.07 (1.02–1.12)Footnote * | 1.04 (0.99–1.09) | 1.05 (0.99–1.11) |
Flourishing Scale (3-unit increase) |
0.94 (0.92–0.97)Footnote * | 1.03 (1.00–1.07) | 1.06 (1.02–1.10)Footnote * | 1.09 (1.04–1.15)Footnote * |
Abbreviations: aOR, adjusted odds ratio; CI, confidence intervals; DERS, Difficulties in Emotional Regulation Scale; GEE,generalized estimating equations; PA, physical activity. Notes: Partially adjusted models controlled for grade, ethnicity, province and school-level clustering. Fully adjusted models controlled for all variables in table, province and school-level clustering. Physical activity guidelines: at least 60 minutes of moderate-to-vigorous physical activity per day. Screen time guidelines: less than 2 hours of screen time per day. Sleep guidelines: 8 to 10 hours of uninterrupted sleep. Depression and anxiety symptoms were measured using the CES-D-10 and the GAD-7, respectively. These are continuous scales where a score of 10 or higher was used to indicate clinically relevant symptomatology.
|
After adjusting for all other factors, infrequent and monthly alcohol use and ever and past month cigarette smoking were associated with e-cigarette initiation among both female and male students. Infrequent and monthly cannabis use was associated with e-cigarette initiation among female students, while only infrequent use was associated with e-cigarette initiation among male students. Skipping school and getting lower grades were associated with increased likelihood of e-cigarette initiation.
Various movement behaviours were also associated with e-cigarette initiation. After adjusting for all other factors, meeting the physical activity guidelines was associated with e-cigarette initiation for both female (adjusted odds ratio [aOR] = 1.14; 95% CI = 1.02–1.27) and male (1.35; 1.16–1.57) students. Meeting the screen time guidelines was associated with decreased odds of e-cigarette initiation only among female students (0.69; 0.58–0.84). Meeting the sleep guidelines was not significantly associated with e-cigarette initiation.
After adjusting for all other factors, reporting clinically relevant symptoms of anxiety or depression was not associated with e-cigarette initiation among either female or male students. However, among females, each 3-point increase in the DERS, representing poorer emotional regulation, was associated with slightly increased odds of e-cigarette initiation (1.07; 1.02–1.12), whereas among males, each 3-point increase in the Flourishing Scale, representing stronger flourishing, was associated with a slightly increased odds of e-cigarette initiation (1.09; 1.04–1.15).
Part 2: Association between changes in covariates and e-cigarette initiation
Across both genders, most students maintained their behaviours over time between baseline and follow-up, although, notably, 29% of students reported increasing their alcohol use (Table 3).
Variable | Total (n = 10 727) |
Female (n = 6032) | Male (n = 4695) | |||||
---|---|---|---|---|---|---|---|---|
n | % | E-cigarette initiation status | E-cigarette initiation status | |||||
No (%) (n = 4191) |
Yes (%) (n = 1841) |
p-value | No (%) (n = 3399) |
Yes (%) (n = 1296) |
p-value | |||
Alcohol use | ||||||||
Maintainers | 2045 | 19.1 | 14.9 | 35.7 | < 0.01 | 11.7 | 28.1 | < 0.01 |
Abstainers | 4977 | 46.4 | 55.6 | 13.1 | 62.9 | 20.5 | ||
Escalators | 3053 | 28.5 | 24.1 | 43.4 | 20.0 | 43.4 | ||
Reducers | 652 | 6.1 | 5.4 | 7.8 | 5.3 | 7.9 | ||
Cannabis use | ||||||||
Maintainers | 340 | 3.2 | 1.5 | 9.1 | < 0.01 | 1.0 | 6.0 | < 0.01 |
Abstainers | 8950 | 83.4 | 91.8 | 60.3 | 93.2 | 63.6 | ||
Escalators | 1259 | 11.7 | 5.3 | 27.3 | 4.9 | 28.3 | ||
Reducers | 178 | 1.7 | 1.4 | 3.3 | 1.0 | 2.1 | ||
Cigarette use | ||||||||
Maintainers | 338 | 3.2 | 1.6 | 8.1 | < 0.01 | 1.1 | 6.5 | < 0.01 |
Abstainers | 9511 | 88.7 | 95.1 | 71.4 | 95.7 | 74.0 | ||
Escalators | 721 | 6.7 | 2.4 | 17.8 | 2.1 | 17.3 | ||
Reducers | 157 | 1.5 | 1.0 | 2.7 | 1.1 | 2.2 | ||
Skipping school | ||||||||
Skipped both years | 1169 | 10.9 | 9.1 | 20.9 | < 0.01 | 6.7 | 13.4 | < 0.01 |
Skipped neither year | 7009 | 65.3 | 68.4 | 46.2 | 75.2 | 56.7 | ||
Skipped follow-up only | 1882 | 17.5 | 16.3 | 25.4 | 13.2 | 21.8 | ||
Skipped baseline only | 667 | 6.2 | 6.2 | 7.6 | 4.9 | 8.0 | ||
English gradeFootnote a | ||||||||
No change | 7050 | 65.7 | 73.3 | 67.1 | < 0.01 | 60.8 | 52.2 | < 0.01 |
Increase in grade | 1779 | 16.6 | 13.0 | 16.1 | 19.2 | 22.0 | ||
Decrease in grade | 1898 | 17.7 | 13.8 | 16.8 | 19.9 | 25.8 | ||
Meeting PA guidelines | ||||||||
Met neither year | 4884 | 45.5 | 52.7 | 46.8 | < 0.01 | 42.0 | 29.9 | < 0.01 |
Met both years | 2551 | 23.8 | 17.9 | 21.1 | 27.8 | 36.0 | ||
Met follow-up only | 1519 | 14.2 | 13.6 | 14.8 | 14.0 | 15.6 | ||
Met baseline only | 1773 | 16.5 | 15.8 | 17.3 | 16.2 | 18.5 | ||
Meeting screen time guidelines | ||||||||
Met neither year | 9408 | 87.7 | 83.3 | 90.3 | < 0.01 | 90.2 | 91.7 | 0.11 |
Met both years | 333 | 3.1 | 4.8 | 1.8 | 2.3 | 1.5 | ||
Met follow-up only | 400 | 3.7 | 5.1 | 3.0 | 3.0 | 2.2 | ||
Met baseline only | 586 | 5.5 | 6.8 | 4.9 | 4.4 | 4.7 | ||
Meeting sleep guidelines | ||||||||
Met neither year | 4887 | 45.6 | 48.7 | 50.1 | 0.09 | 40.9 | 41.1 | 0.22 |
Met both years | 2784 | 26.0 | 25.1 | 22.1 | 28.7 | 27.0 | ||
Met follow-up only | 1096 | 10.2 | 9.0 | 9.9 | 10.8 | 12.8 | ||
Met baseline only | 1960 | 18.3 | 17.2 | 17.9 | 19.6 | 19.1 | ||
Depression symptoms | ||||||||
Neither year | 5617 | 52.4 | 46.6 | 37.3 | < 0.01 | 64.3 | 61.1 | 0.04 |
Both years | 2536 | 23.6 | 29.1 | 34.7 | 14.4 | 14.6 | ||
Follow-up only | 1647 | 15.4 | 16.0 | 17.8 | 12.9 | 16.1 | ||
Baseline only | 927 | 8.6 | 8.3 | 10.3 | 8.3 | 8.2 | ||
Anxiety symptoms | ||||||||
Neither year | 6948 | 64.8 | 57.3 | 48.9 | < 0.01 | 78.6 | 75.2 | 0.04 |
Both years | 1604 | 15.0 | 20.1 | 23.2 | 7.0 | 7.4 | ||
Follow-up only | 1286 | 12.0 | 13.2 | 16.7 | 8.4 | 10.7 | ||
Baseline only | 889 | 8.3 | 9.4 | 11.1 | 5.9 | 6.7 | ||
DERS | ||||||||
No change | 5399 | 50.3 | 48.9 | 45.4 | 0.05 | 54.1 | 52.1 | 0.44 |
Increase | 2884 | 26.9 | 28.6 | 30.3 | 23.7 | 25.0 | ||
Decrease | 2444 | 22.8 | 22.5 | 24.3 | 22.2 | 22.9 | ||
Flourishing Scale | ||||||||
No change | 5567 | 51.9 | 52.3 | 50.3 | 0.26 | 52.5 | 51.3 | 0.77 |
Increase | 2276 | 21.2 | 19.9 | 21.6 | 22.1 | 22.6 | ||
Decrease | 2884 | 26.9 | 27.7 | 28.1 | 25.4 | 26.1 | ||
Abbreviations: DERS, Difficulties in Emotional Regulation Scale; PA, physical activity. Notes: Changes in DERS and Flourishing Scale reflect a 3-unit change.
|
Similar to the results from Part 1, female and male students who abstained from other substance use, specifically alcohol use, cannabis use and cigarette smoking, at both baseline and follow-up had lower odds of initiating e-cigarette use compared to those who maintained their frequency of substance use in the fully adjusted models (Table 4).
Variable | Female | Male | ||
---|---|---|---|---|
Partially adjusted aOR (95% CI) |
Fully adjusted aOR (95% CI) |
Partially adjusted aOR (95% CI) |
Fully adjusted aOR (95% CI) |
|
Alcohol use | ||||
Maintainers | 1.00 | 1.00 | 1.00 | 1.00 |
Abstainers | 0.09 (0.07–0.11)Footnote * | 0.19 (0.15–0.23)Footnote * | 0.13 (0.11–0.16)Footnote * | 0.26 (0.21–0.33)Footnote * |
Escalators | 0.69 (0.58–0.82)Footnote * | 0.87 (0.72–1.05) | 0.87 (0.71–1.06) | 1.13 (0.91–1.40) |
Reducers | 0.60 (0.47–0.77)Footnote * | 0.70 (0.54–0.90)Footnote * | 0.63 (0.48–0.83)Footnote * | 0.80 (0.58–1.09) |
Cannabis use | ||||
Maintainers | 1.00 | 1.00 | 1.00 | 1.00 |
Abstainers | 0.10 (0.07–0.14)Footnote * | 0.31 (0.21–0.45)Footnote * | 0.10 (0.07–0.15)Footnote * | 0.30 (0.19–0.49)Footnote * |
Escalators | 0.81 (0.57–1.16) | 1.01 (0.68–1.49) | 0.87 (0.56–1.34) | 1.12 (0.68–1.83) |
Reducers | 0.40 (0.24–0.66)Footnote * | 0.60 (0.33–1.07) | 0.35 (0.19–0.63)Footnote * | 0.52 (0.25–1.08) |
Cigarette use | ||||
Maintainers | 1.00 | 1.00 | 1.00 | 1.00 |
Abstainers | 0.14 (0.11–0.20)Footnote * | 0.37 (0.25–0.53)Footnote * | 0.12 (0.08–0.19)Footnote * | 0.27 (0.17–0.43)Footnote * |
Escalators | 1.45 (1.01–2.08)Footnote * | 1.55 (0.98–2.45) | 1.35 (0.84–2.17) | 1.26 (0.73–2.17) |
Reducers | 0.56 (0.33–0.97)Footnote * | 0.58 (0.31–1.07) | 0.38 (0.20–0.73)Footnote * | 0.39 (0.19–0.80)Footnote * |
Skipping school | ||||
Skipped both years | 1.00 | 1.00 | 1.00 | 1.00 |
Skipped neither year | 0.26 (0.22–0.31)Footnote * | 0.58 (0.49–0.69)Footnote * | 0.35 (0.28–0.43)Footnote * | 0.72 (0.58–0.90)Footnote * |
Skipped follow-up only | 0.64 (0.53–0.78)Footnote * | 0.84 (0.68–1.05) | 0.82 (0.66–1.01) | 1.05 (0.82–1.35) |
Skipped baseline only | 0.51 (0.39–0.66)Footnote * | 0.65 (0.50–0.85)Footnote * | 0.80 (0.60–1.06) | 1.14 (0.82–1.59) |
English gradeFootnote a | ||||
No change | 1.00 | 1.00 | 1.00 | 1.00 |
Increase | 1.37 (1.18–1.59)Footnote * | 0.98 (0.81–1.19) | 1.31 (1.11–1.54)Footnote * | 1.17 (0.96–1.42) |
Decrease | 1.39 (1.19–1.62)Footnote * | 1.06 (0.87–1.29) | 1.46 (1.22–1.74)Footnote * | 1.21 (0.99–1.49) |
Meeting PA guidelines | ||||
Met neither year | 1.00 | 1.00 | 1.00 | 1.00 |
Met both years | 1.39 (1.20–1.61)Footnote * | 1.15 (0.97–1.36) | 1.91 (1.58–2.30)Footnote * | 1.48 (1.18–1.84)Footnote * |
Met follow-up only | 1.30 (1.10–1.54)Footnote * | 1.09 (0.92–1.30) | 1.64 (1.34–2.01)Footnote * | 1.28 (0.98–1.68) |
Met baseline only | 1.30 (1.12–1.50)Footnote * | 1.13 (0.95–1.34) | 1.69 (1.39–2.06)Footnote * | 1.36 (1.08–1.71)Footnote * |
Meeting screen time guidelines | ||||
Met neither year | 1.00 | 1.00 | 1.00 | 1.00 |
Met both years | 0.32 (0.22–0.47)Footnote * | 0.58 (0.39–0.84)Footnote * | 0.60 (0.38–0.96)Footnote * | 0.94 (0.57–1.54) |
Met follow-up only | 0.54 (0.39–0.74)Footnote * | 0.82 (0.59–1.14) | 0.69 (0.43–1.11) | 0.79 (0.48–1.32) |
Met baseline only | 0.64 (0.49–0.83)Footnote * | 0.88 (0.66–1.16) | 1.13 (0.82–1.55) | 1.39 (0.92–2.09) |
Meeting sleep guidelines | ||||
Met neither year | 1.00 | 1.00 | 1.00 | 1.00 |
Met both years | 0.74 (0.63–0.86)Footnote * | 0.84 (0.71–0.98)Footnote * | 0.87 (0.72–1.04) | 1.01 (0.83–1.23) |
Met follow-up only | 0.98 (0.78–1.23) | 0.96 (0.75–1.22) | 1.14 (0.90–1.43) | 1.26 (0.99–1.62) |
Met baseline only | 0.96 (0.82–1.12) | 1.02 (0.87–1.20) | 0.92 (0.75–1.14) | 1.02 (0.81–2.19) |
Depression symptoms | ||||
Both years | 1.00 | 1.00 | 1.00 | 1.00 |
Neither year | 0.62 (0.54–0.72)Footnote * | 1.04 (0.85–1.27) | 0.88 (0.71–1.09) | 1.13 (0.82–1.57) |
Follow-up only | 0.91 (0.75–1.10) | 1.19 (0.94–1.51) | 1.21 (0.94–1.55) | 1.31 (0.90–1.89) |
Baseline only | 1.02 (0.83–1.26) | 1.11 (0.86–1.45) | 0.94 (0.72–1.24) | 1.02 (0.70–1.49) |
Anxiety symptoms | ||||
Both years | 1.00 | 1.00 | 1.00 | 1.00 |
Neither year | 0.72 (0.62–0.84)Footnote * | 0.91 (0.72–1.14) | 0.88 (0.69–1.12) | 0.99 (0.69–1.44) |
Follow-up only | 1.09 (0.89–1.33) | 1.13 (0.88–1.46) | 1.20 (0.89–1.63) | 1.26 (0.82–1.93) |
Baseline only | 1.03 (0.81–1.30) | 0.96 (0.72–1.28) | 1.09 (0.82–1.45) | 1.11 (0.77–1.61) |
DERS (3-unit change) | ||||
No change | 1.00 | 1.00 | 1.00 | 1.00 |
Increase | 1.16 (1.03–1.30)Footnote * | 1.05 (0.89–1.23) | 1.09 (0.93–1.26) | 0.98 (0.81–1.18) |
Decrease | 1.19 (1.03–1.38)Footnote * | 1.12 (0.94–1.33) | 1.08 (0.92–1.27) | 1.03 (0.86–1.24) |
Flourishing Scale (3-unit change) | ||||
No change | 1.00 | 1.00 | 1.00 | 1.00 |
Increase | 1.13 (1.00–1.28)Footnote * | 1.04 (0.88–1.23) | 1.07 (0.90–1.27) | 1.00 (0.82–1.23) |
Decrease | 1.06 (0.95–1.18) | 0.92 (0.80–1.33) | 1.04 (0.90–1.20) | 0.97 (0.81–1.16) |
Abbreviations: aOR, adjusted odds ratio; CI, confidence interval; DERS, Difficulties in Emotional Regulation Scale; GEE, generalized estimating equations; PA, physical activity. Notes: Partially adjusted models controlled for grade, ethnicity, province and school-level clustering. Fully adjusted models controlled for grade, ethnicity, spending money, all variables in table, province and school-level clustering. For substance use, “abstainers” did not engage in a specific behaviour at baseline or follow-up; “maintainers” continued the same level of frequency of the behaviour at baseline and follow-up; “escalators” increased the frequency of the behaviour from baseline to follow-up; and “reducers” decreased the frequency of the behaviour from baseline to follow-up.
|
Significant results were found for students who did not skip classes either year and for female students who skipped class in the baseline year only.
Male students who met the physical activity guidelines both years, who started meeting the guidelines and who stopped meeting the guidelines were at increased risk of e-cigarette initiation compared to those who did not meet the guidelines either year. Female students who met the screen time and sleep guidelines both years were less likely to start using e-cigarettes than those who met the guidelines neither year.
Changes in mental health and well-being indicators were not significantly associated with e-cigarette initiation.
Discussion
Over the course of one year, almost one-third (29%) of the Canadian secondary school students who had not yet initiated e-cigarette use reported initiation. This is consistent with research showing a rapid increase in the popularity of e-cigarette use among students, Footnote 3 and highlights the importance of investigating e-cigarette initiation. We identified multiple demographic and behavioural factors that were associated with e-cigarette initiation. Furthermore, we explored how changes to baseline behavioural factors among both female and male students were associated with e-cigarette initiation. The stratified findings illustrate some differences in factors associated with e-cigarette initiation between the genders that could inform tailored e-cigarette use prevention programs.
This study adds to the current literature by exploring an expanded range of factors associated with e-cigarette initiation. Younger students were more likely to initiate e-cigarette use, possibly because they are less resistant to peer influence.Footnote 38 This suggests that e-cigarette prevention efforts are needed prior to starting Grade 9 and may need to be reinforced in secondary school. Consistent with previous evidence for cigarette smoking Footnote 39 and e-cigarette initiation,Footnote 16 female and male students with more spending money were more likely to initiate e-cigarette use. The cost of devices can be a deterrent for price-sensitive youth; therefore, taxation policies that increase the cost of e-cigarette devices and accessories (e.g. e-liquid, pods) may help to reduce e-cigarette initiation among youth.
As expected, participation in other substance use (i.e. alcohol, cannabis and cigarettes) was strongly associated with e-cigarette initiation. At baseline, monthly alcohol use posed the greatest risk for females and past month cigarette use posed the greatest risk for males, followed by monthly alcohol use. Due to the relatively high number of students who reported alcohol use (33% for alcohol use vs. 6% for cannabis use and 5% for cigarette use) and the high odds of initiation, prevention efforts in this domain may also help prevent e-cigarette use, although additional evaluation evidence is required.
Results for changes in substance-use behaviours over time were similar. Many earlier studies have noted the clustering of health-risk and substance-use behaviours among adolescents, and it is likely that impulsivity and high levels of sensation seeking are underlying risk factors for these behaviours. Footnote 9Footnote 10Footnote 12Footnote 13Footnote 14Footnote 15Footnote 21 Prevention programs should therefore address multiple substances and the underlying reasons that students use these substances, although additional evaluation of such programs on multiple health-risk behaviours is necessary.
Other health related behaviours were also associated with e-cigarette initiation, though results were sometimes complex and some differences between male and female students were observed. For example, students who met the Canadian physical activity guidelines were more likely to initiate e-cigarette use. An earlier Canadian study also identified a link between physical activity and e-cigarette use;Footnote 40 however, other US-based studies have identified no link.Footnote 41Footnote 42Footnote 43 There is also evidence that youth view e-cigarettes as a less harmful alternative to cigarettes.Footnote 44 While students are increasingly aware of the harms of regular nicotine vaping, fewer students perceive harms with non-nicotine vaping or occasional nicotine vaping. Footnote 2Footnote 45 This may explain their appeal to youth participating in sport who have been found to avoid other inhaled substances such as cannabis and cigarettes.Footnote 46
Meeting screen time guidelines, however, was negatively associated with e-cigarette initiation among female students who met screen time guidelines at baseline and follow up (i.e. maintainers). Previous research has identified a link between exposure to e-cigarette advertising and e-cigarette initiation, Footnote 9Footnote 17Footnote 18 and students who meet Canadian screen time recommendations could have lower levels of exposure to advertising, particularly online. E-cigarette promotion is prevalent online, and youth who report exposure are more likely to initiate use. Footnote 47Footnote 48 This result may have only been found among female students due to gender differences in how students engage in screen time: male students are more likely to spend time playing video games while female students are more likely to spend time on their mobile phone.Footnote 49 However, these results should be interpreted with caution, given that the vast majority of both female and male students did not meet screen time guidelines.
Finally, meeting the sleep guidelines both years among females was negatively associated with e-cigarette initiation. These results are intuitive and add to the literature. While cannabis use and initiating binge drinking have been associated with not meeting the sleep guidelines,Footnote 50Footnote 51 other research has found no association.Footnote 52 This finding may be due to a shift in lifestyle less conducive to sleep, although more research is warranted.
In contrast to previous studies,Footnote 15Footnote 16Footnote 53 mental health indicators were not significantly associated with e-cigarette initiation. Two previous studies identified internalizing problems (e.g. anxiety, depression) were associated with initiation of e-cigarettes only but not dual use of e-cigarettes and conventional cigarettes,Footnote 15Footnote 16 while another identified depressive symptoms as associated with initiation of e-cigarette use, cigarette use and dual use.Footnote 53 However, these studies used different measures, accounted for dual use of cigarettes and e-cigarettes, and two did not control for cannabis use,Footnote 16Footnote 53 which has also been associated with depression Footnote 54 and could confound results. Additional studies are necessary to further explore the association between mental health indicators and e-cigarette initiation.
Results for the Difficulties in Emotional Regulation Scale (DERS) were significant for female students. This is consistent with previous research that has found emotional dysregulation to be associated with cigarette initiation.Footnote 55 Higher DERS indicates lower levels of emotional regulation, suggesting that female students are potentially using e-cigarettes as a coping strategy. Therefore, teaching alternative positive coping strategies could be an important component of e-cigarette prevention programs for female students.
In contrast, higher flourishing was associated with increased odds of e-cigarette initiation among males. This is opposite to results that have linked higher flourishing with less substance use or no effect,Footnote 56Footnote 57Footnote 58 suggesting that e-cigarettes are not being used as a coping mechanism among males. Curiosity about a novel product is a leading reason adolescents try e-cigarettes and could be motivating this group, along with marketing that broadly appeals to youth.Footnote 59Footnote 60 Previous research has been cross-sectional,Footnote 56Footnote 57Footnote 58 and it is possible that the direction of association changed after students transitioned to more regular use. Based on these results and the positive associations seen between physical activity and e-cigarette use, e-cigarette use may be more of a social activity, but more research is needed to explore this hypothesis.
Strengths and limitations
The main strength of this study is the use of a large, school-based longitudinal dataset to examine factors associated with e-cigarette use. In particular, the use of passive consent procedures maximizes the student participation rate and limits selection bias that is common in youth substance-use studies that use active consent procedures.Footnote 26Footnote 61Footnote 62Footnote 63Footnote 64 This study is the first of its kind to examine e-cigarette initiation in the Canadian context and how changes in behaviour may impact e-cigarette initiation among Canadian students. The COMPASS study includes questions that assess a range of health behaviours, allowing a comprehensive examination of the influence of demographic characteristics, behavioural factors and mental health indicators on e-cigarette initiation.
Although the COMPASS study has a large sample size, it was designed to evaluate changes in school programs and policies using natural experiment methodology. Therefore, it is not representative of all Canadian secondary school students. Also, because the questionnaire neither defines “e-cigarette” nor lists brands, and because of the changing language used by youth to refer to e-cigarette devices (e.g. “vaping,” “Juuling”), this study may underreport e-cigarette use. It is also likely that the relationship between risk factors and e-cigarette initiation is influenced by other factors not measured in the COMPASS survey, such as exposure to e-cigarette marketingFootnote 9Footnote 17Footnote 18 or e-cigarette susceptibility. Footnote 12
Additionally, as is the case with most self-report surveys, there could be reporting bias with respect to substance use; however, students are assured of the anonymity of their responses. Furthermore, participant drop-out and limitations with linking students across both waves may have resulted in an underestimation of e-cigarette initiation rates and their associations with demographic and behavioural variables, because students who are linked over time are more likely to be younger, female and less likely to use substances.Footnote 65Footnote 66 There were also some differences between the complete case sample and those removed, as well as differences between the samples in Part 1 and Part 2 that may have resulted in bias. Finally, the use of two time points prevented us from assessing the temporal order between changes in covariates at follow-up and e-cigarette initiation (i.e. e-cigarette initiation could have occurred before or after covariate status at follow-up). Our analysis of changes in characteristics on e-cigarette use should therefore be considered exploratory.
Conclusion
This prospective study examining factors associated with e-cigarette initiation provides novel evidence to support the need for stronger e-cigarette prevention efforts aimed at youth populations. Over the span of just one year, almost one-third of the sample of previous nonusers initiated e-cigarette use. Prevention approaches should target multiple health-risk behaviours to help prevent youth e-cigarette initiation. Additionally, given that Grade 9 students were at higher risk of initiation, school-based approaches may benefit from being implemented before high school.
Acknowledgements
The COMPASS study has been supported by a bridge grant from the Canadian Institute of Health Research (CIHR) Institute of Nutrition, Metabolism and Diabetes (INMD) through the “Obesity – Interventions to Prevent or Treat” priority funding awards (OOP-110788, awarded to SL); an operating grant from the CIHR Institute of Population and Public Health (IPPH) (MOP-114875, awarded to SL); a CIHR project grant (PJT-148562, awarded to SL); a CIHR bridge grant (PJT-149092, awarded to KP/SL); a CIHR project grant (PJT-159693, awarded to KP); and a research funding arrangement with Health Canada (#1617-HQ-000012, contract awarded to SL). The COMPASS-Quebec project additionally benefits from funding from the Ministère de la Santé et des Services sociaux of the province of Quebec, and the Direction régionale de santé publique du CIUSSS de la Capitale-Nationale. This work was supported by an operating grant from CIHR (#170256, awarded to AC). GCW is supported by the Ontario Graduate Scholarship (OGS) and by the Public Health Agency of Canada through the Federal Student Work Experience Program.
Conflicts of interest
Scott Leatherdale is an Associate Scientific Editor with the HPCDP Journal but has recused himself from the review process for this paper. The authors have no other conflicts of interest to declare.
Authors’ contributions and statement
GCW collaborated on the study methodology, conducted statistical analysis, interpreted the results and drafted the original manuscript. AGC conceived of the study research questions, collaborated on the study methodology, interpreted study results, contributed to the original manuscript draft and reviewed the manuscript for important intellectual content. MdG, YJ and STL collaborated on the study methodology, interpreted study results and revised the manuscript for important intellectual content. STL is the principal investigator of the COMPASS study, wrote the funding proposal, developed the tools and led study implementation and coordination. All authors read and approved the final manuscript.
The content and views expressed in this article are those of the authors and do not necessarily reflect those of the Government of Canada.
Page details
- Date modified: