Abstract
Background: Over the past decade, youth tobacco use has declined, and electronic cigarettes (e-cigarettes) have entered the market. The aims of this study were to describe the prevalence of e-cigarette use among youth in Canada, by province, across sociodemographic variables and smoking-related correlates; and to examine associations among e-cigarette use, sociodemographic variables and smoking-related correlates, with adjustment for other factors.
Methods: The 2014/15 Canadian Student Tobacco, Alcohol and Drugs Survey, a biennial, school-based survey, was administered to students in grades 6-12 in all Canadian provinces. Logistic regression models were fitted to estimate odds of ever and past 30-day e-cigarette use by sociodemographic variables and smoking-related correlates.
Results: A total of 336 schools from 128 school boards (47% of eligible schools approached) and 42 094 students (66% of eligible students approached) participated in the survey. In Canada, 17.7% (95% confidence interval [CI] 16.4%-18.9%) of students in grades 6-12 reported ever using e-cigarettes, and 5.7% (95% CI 5.2%-6.3%) reported past 30-day use. Substantial variation was observed across provinces. Female students had decreased odds of past 30-day use relative to male students (odds ratio [OR] 0.71, 95% CI 0.59-0.86), whereas current smokers (OR 10.0, 95% CI 6.66-15.02) and experimental smokers (OR 3.61, 95% CI 2.40-5.42) had increased odds relative to never smokers. Students who perceived that access was easy also had increased odds of using e-cigarettes relative to students who perceived that access was difficult (OR 3.86, 95% CI 2.96-5.03). Students who believed that regular use entailed slight risk (OR 0.68, 95% CI 0.52-0.88) and those who did not know risk levels (OR 0.31, 95% CI 0.21-0.46) had decreased odds compared with those perceiving no risk.
Interpretation: Our data confirm that many youth used e-cigarettes in the 30 days preceding the survey, although rates were substantially higher among current and experimental smokers than among students who had never tried smoking.
The past decade has seen a decline in tobacco smoking among youth in many countries.1-3 Over the same period, electronic cigarettes (e-cigarettes or vapour devices) were introduced into the market.4 E-cigarettes are devices in which a liquid solution, which may contain flavours or nicotine (or both), is heated to create an aerosol that is inhaled,5 mimicking the experience of conventional cigarette smoking to a greater extent than other noncombustible delivery methods, such as patches, inhalers and gum.6 Lifetime use of e-cigarettes is common among teenagers in many countries, including the United States, the United Kingdom, France and Poland; in the US, use of e-cigarettes has surpassed cigarette smoking.1,7-9 Concerns about e-cigarette use have led to fierce debate in the public health community, with advocates arguing that e-cigarettes have substantial potential for harm reduction,6,10 and others arguing that e-cigarettes could be a gateway to conventional tobacco use among youth and that they may renormalize smoking.11,12 Most studies indicate that e-cigarette use primarily occurs among conventional cigarette smokers, with few nonsmokers engaging in regular e-cigarette use.7,13-15 Some evidence suggests that a substantial number of young nonsmokers try e-cigarettes16 and that e-cigarette use among never-smoking youth could be related to initiation of use of combustible tobacco products.17,18 The evidence of a causal relationship, however, is limited.17,19
Long-term trend data on e-cigarette use are limited and not yet available in Canada. In Quebec, ever and past 30-day use of e-cigarettes among students in grades 7-11 were 34% and 6%, respectively, in 2012/13.16 Lifetime e-cigarette use was 15% among students in grades 7-12 in Ontario in 201314 and 10% among grade 9 students in the Niagara region in 2015.20 By comparison, in 2015, 16% of US students in grades 9-12 and 5% of those in grades 6-8 reported using e-cigarettes in the past 30 days, a significant increase from 1.5% and 0.6%, respectively, in 2011.1 In the UK, lifetime rates of e-cigarette use ranged from 8% to 12% in 2013/14,15,21,22 whereas more-than-monthly use ranged from 0.4% to 2%.23 Literature on e-cigarette use across Canada is lacking, because of the absence of representative data for high school and middle school students. Long-term monitoring of e-cigarette and other tobacco product use is important for surveillance and public health. To this end, Health Canada's Canadian Student Tobacco, Alcohol and Drugs Survey24 included questions on e-cigarette use beginning in the 2014/15 cycle. Our aims are to describe the prevalence of e-cigarette use among youth in Canada, by province, across sociodemographic variables and smoking-related correlates; and to examine associations among e-cigarette use, sociodemographic variables and smoking-related correlates, with adjustment for other factors.
Methods
Source of data
We obtained data from a cross-sectional, biennial paper-and-pencil school-based survey administered to elementary and secondary school students across Canada between October 2014 and May 2015. The target population was students in grades 6-12 (grade 6 to secondary 5 in Quebec) attending private, public and Catholic schools in all 10 Canadian provinces, excluding schools in the 3 territories and special schools (e.g., schools for students with visual impairment and schools on First Nations reserves). A generalizable sample was achieved at the national and provincial levels, with the exception of New Brunswick (because of low response rate). However, national estimates do include data from participating New Brunswick schools.
Study design
The survey employed a stratified, single-stage cluster design, with strata based on health-region smoking rate and type of school. In each province, 2 or 3 health-region smoking rate strata and 2 school-level strata were defined. Random selection of schools within each stratum ensured a generalizable sample within each province. Classroom teachers administered the survey. Questionnaires were available in English and French. The survey's website offers details (www.cstads.ca).
Measures
The primary outcome measures of interest were self-reported ever and past 30-day e-cigarette use (yes/no). Smoking-related correlates included ever use of other tobacco products (little cigars or cigarillos, cigars, roll-your-own cigarettes, bidis, smokeless tobacco, water pipe, blunt wraps), smoking status (current smoker, former smoker, experimental smoker or never tried smoking), perceived risk of harm from using e-cigarettes on a regular basis (no risk, slight, moderate, great risk, don't know) and perceived ease of access of e-cigarettes (very difficult, fairly difficult, fairly easy, very easy, don't know). We also analyzed sociodemographic variables of sex (male, female), grade (6-12; grade 6 to secondary 5 in Quebec), province of residence, ethnicity (white, black, West Asian/Arab, South Asian, East/Southeast Asian, Latin American/Hispanic, Aboriginal, other), socioeconomic level of school neighbourhood (obtained from school postal code) and school location (urban, rural). All measures are defined in Appendix 1 (available at www.cmajopen.ca/content/5/2/E460/suppl/DC1).
Statistical analysis
We used descriptive statistics to estimate the weighted prevalence of ever and past 30-day e-cigarette use across sociodemographic variables, use of other tobacco products, smoking status, perceived ease of access to e-cigarettes and perceived risk of harm from regularly using e-cigarettes. We fitted unadjusted and adjusted logistic regression models to estimate odds of ever and past 30-day e-cigarette use by the listed sociodemographic and smoking-related variables. We selected these covariables because other studies have shown that they are significant predictors of e-cigarette use and smoking onset in adolescents.25,26 The calculation of tolerances and variance inflation factors indicated no problems of multicollinearity among explanatory variables. We report our results as unadjusted and adjusted odds ratios (ORs) with 95% confidence intervals (CIs). Participants with missing data pertaining to the variables of interest were excluded from all analyses.
We used survey weights to adjust for sample selection (school and class levels), nonresponse (school, class and student levels) and poststratification of the sample population relative to grade and sex distribution in the total population. Bootstrap weights, which we used to calculate CIs of the prevalence estimates and regression analyses, account for the effects of survey design on variance estimates. We performed all analyses using SAS version 9.4 (SAS Institute Inc.).
All references to "students" refer specifically to students in grades 6-12, unless specified otherwise.
Ethics approval
Multiple bodies provided ethics review and approval: Health Canada Research Ethics Board, University of Waterloo Office of Research Ethics and ethics review boards at participating institutions and school boards. Depending on school board requirements, parents or guardians provided active or passive permission (with active information) for their child to participate. Individual student consent to participate was also obtained.
Results
A total of 336 schools from 128 school boards (47% of eligible schools approached) and 42 094 students (66% of eligible students approached) were recruited for the survey.
In Canada, 17.7% (95% CI 16.4%-18.9%) of students in grades 6-12 reported ever using e-cigarettes, which represents about 441 900 students, and 5.7% (95% CI 5.2%-6.3%) reported past 30-day use (representing about 142 900) (Table 1). Past 30-day use was reported by 46.6% (95% CI 39.3%-53.9%) of current smokers, 20.4% (95% CI 11.8%-28.9%) of former smokers, 19.4% (95% CI 17.7%-21.2%) of experimental smokers and 1.8% (95% CI 1.5%-2.0%) of students who had never tried smoking. The prevalence of ever and past 30-day use was higher among male students (compared with female students) and students in grades 10-12 (compared with those in grades 6-9). Rates were highest among students identifying as Aboriginal and lowest among students identifying as Asian. E-cigarette use varied across provinces, with the lowest rates found in Ontario and Alberta and the highest in Newfoundland and Labrador.
Ever and past 30-day e-cigarette use was more prevalent among students who had tried other tobacco products, those who perceived getting e-cigarettes was easy and those who thought regular use entailed no risk. Prevalence of e-cigarette use did not vary by socioeconomic level of school neighbourhood, but rate of ever use was higher among students attending a school located in a rural area. Missing data for self-reported ever and past 30-day e-cigarette use was 1.8% and 2.1%, respectively, and ranged from 0% for smoking status, sex, grade, location, socioeconomic level of school neighbourhood and province to 3.9% for perceived risk of harm.
Results presented in Table 2 describe factors associated with ever and past 30-day use of e-cigarettes (with only past 30-day use being discussed below). In terms of sociodemographic indicators, adjusted logistic regression models showed that female students had decreased odds of using e-cigarettes compared with male students (OR 0.71, 95% CI 0.59-0.86), while students in grades 10-12 had decreased odds of e-cigarette use compared with students in grades 6-9 (OR 0.78, 95% CI 0.63-0.98). In terms of provincial variations, students in British Columbia, Manitoba, Newfoundland and Labrador, Nova Scotia, Prince Edward Island and Saskatchewan all had higher odds of e-cigarette use compared with Ontario. Students attending a school located in a high socioeconomic level neighbourhood had increased odds of e-cigarette use (OR 1.55, 95% CI 1.26-1.92) compared with students whose school was located in a middle-income neighbourhood. No difference in e-cigarette use was observed across ethnic groups or by location of school.
In terms of cigarette and other tobacco use, current smokers (OR 10.0, 95% CI 6.66-15.02), experimental smokers (OR 3.61, 95% CI 2.40-5.42) and former smokers (OR 2.75, 95% CI 1.31-5.80) all had elevated odds of e-cigarette use compared with students who had never tried smoking, as did students who had tried any other tobacco product (OR 4.33, 95% CI 3.20-5.87). Perceptions and access also matter; students who perceived that access to e-cigarettes was easy (OR 3.86, 95% CI 2.96-5.03) had increased odds of use compared with those who perceived that access was difficult. Meanwhile, students who perceived that regular use of e-cigarettes was at all risky had lower odds of using e-cigarettes.
Interpretation
This study provides national estimates of the prevalence of e-cigarette use among students in grades 6-12 in Canada. Almost 18% of students reported ever having tried an e-cigarette, and almost 6% indicated use in the past 30 days; these rates correspond to about 441 900 and 142 900 student users, respectively. Patterns in e-cigarette use observed across sociodemographic indicators suggest that male students, those attending schools located in high socioeconomic neighbourhoods and those in younger grades had higher odds of use. Reported rates of e-cigarette use among current, experimental and former cigarette smokers were all higher than rates among students who never tried smoking, as were rates among those who used other tobacco products. Perceived ease of access to e-cigarettes and perceived harm also affected use, with higher odds of use among those students who perceived that e-cigarettes were fairly easy or very easy to access, and lower consumption among those who perceived that e-cigarettes pose a health harm.
Overall, lifetime and past 30-day rates of e-cigarette use among students were similar to those observed for smoking cigarettes (17.6% [447 100] and 6.2% [158 900], respectively).24 The Canadian Tobacco, Alcohol and Drugs Survey, based on telephone interviews of participants 15 years and older, is the only other national-level data against which to compare these results.27 According to that survey, e-cigarette prevalence among participants 15-19 years of age in 2015 was 26% for ever use and 6% for past 30-day use, an increase from 2013 (20% and 3%, respectively).27 The rate observed in our study among students of the same age (those in grades 10-12) were similar for ever use (27.1%) but slightly higher for past 30-day use (8.9%). Rates of past 30-day use are higher in the US (16% of high school students and 5% of middle school students in 2015)1 and lower in the UK (2% of those 11-18 years old in Great Britain in 201415 and 3% of those 15 years old in Scotland21). International comparisons should be interpreted with caution because of differences in the timing of surveys, age ranges and designs.
In the current study, we observed substantial variations across provinces, with the lowest rate found in Ontario and the highest rate in Newfoundland and Labrador. Reasons for these provincial differences are unknown and should be investigated. Consistent with the literature, use of e-cigarettes was much higher among current and former smokers than among never smokers.28 A recent study in Ontario's Niagara region found that only 6% of adolescents reported using e-cigarettes for reduction or cessation of smoking,20 with several studies finding that young smokers are less likely than older smokers to report using e-cigarettes to quit smoking.13,29,30 Consistent with other studies, use is more prevalent among males;1,13,14,20 users of other tobacco products such as little cigars, water pipes and smokeless tobacco;16,31 users of alcohol and marijuana;10 and students who perceive that access is easy.13 Use is less prevalent among students who perceive they risk harming themselves by using e-cigarettes.13
Limitations
This study had several limitations. First, the results do not indicate the proportions of students who used e-cigarettes with and without nicotine. This information was not assessed in the questionnaire because, at the time of data collection, e-cigarettes with nicotine, although widely available, had not been authorized to be legally sold on the Canadian market, and because the labelling of nicotine content on e-cigarettes has been shown to be inaccurate.32,33 Second, the response rate of eligible schools approached was just under 50%. Schools have competing priorities, including other provincial, regional and local surveys. In 2014/15, recruitment in New Brunswick was limited by concurrent administration of its census Student Wellness Survey. Third, our results are not representative of young people who do not attend school, who live in institutions, who attend school on First Nations reserves, or who attend special schools or schools located on military bases. Finally, cross-sectional results do not allow the inference of causal relationships between e-cigarette use and factors associated with such use.
Conclusion
Our results suggest that a substantial number of Canadian students use e-cigarettes. There is considerable debate and uncertainty surrounding the role of e-cigarettes as a gateway to smoking; our cross-sectional study cannot be used to inform this debate. Although e-cigarette use is substantially less likely among never smokers and experimental smokers than among current smokers, in absolute numbers never smokers and experimental smokers make up the majority of grade 6-12 students in Canada who reported using e-cigarettes in the past 30 days. More research is needed to document the frequency of and reasons for use of e-cigarettes among youth, which are likely to differ among current, experimental and never smokers.
Since these data were collected in 2014/15, many provinces have banned sales of e-cigarettes to minors, promotion and advertising of e-cigarettes, and use of these devices where smoking is prohibited, all of which are likely to rufther decrease perceptions that e-cigarettes are easily accessible to minors. Studies will be needed to monitor use of, access to and perceptions of e-cigarettes in the coming years, given that Canada has introduced major changes in the regulation of e-cigarettes, plain packaging for tobacco products and legalization of marijuana.
Supplemental information
For reviewer comments and the original submission of this manuscript, please see www.cmajopen.ca/content/5/2/E460/suppl/DC1
Acknowledgements
The following Canadian Student Tobacco, Alcohol and Drugs Survey investigators contributed to survey design and data collection: Antony Card, Jo-Ann MacDonald, Donna Murnaghan, Richard Bourhis, Steve Brown, Tara Elton Marshall, Nazeem Muhajarine and Donna Turner. The authors thank Aaron Bonham, Tammy Cumming, Vicki Rynard and Noel Gruber for their contributions to survey implementation and data analysis and for their help in preparing the manuscript.
Footnotes
Competing interests: None declared.
Contributors: Annie Montreuil and Marjorie MacDonald reviewed the literature. Annie Montreuil, Marjorie MacDonald, Mark Asbridge, Cameron Wild, Steve Manske and Erin Rutherford contributed to data collection. Annie Montreuil, Marjorie MacDonald, Mark Asbridge, Cameron Wild and David Hammond contributed to the data analysis, and all of the authors contributed to the interpretation of results. Annie Montreuil and Marjorie MacDonald cowrote the article, with contributions from Mark Asbridge and Cameron Wild. All of the authors reviewed the article critically for important intellectual content, approved the final version and agreed to act as guarantors of the work.
Funding: This study was supported by the Canadian Cancer Society (grant 2011-701019), through the Propel Centre for Population Health Impact (Propel Centre) at the University of Waterloo. Data used for this research were obtained from Health Canada's Canadian Student Tobacco, Alcohol and Drugs Survey (formerly the Youth Smoking Survey), which was conducted by the Propel Centre.
References
- Copyright 2017, Joule Inc. or its licensors