Tobacco Use in Canada
Patterns and Trends - 2015 Edition

end material: APPENDIX C

Appendix C: Data Analysis

Data analysis was completed by Vicki Rynard, MSc, and Robin Burkhalter, MMath, of the Propel Centre for Population Health Impact, using datasets made available by Statistics Canada and Health Canada. We are grateful to Rashid Ahmed for statistical contributions to previous editions. Statistical guidance for previous editions was provided by K. Stephen Brown, PhD, of the Propel Centre for Population Health Impact and the Department of Statistics & Actuarial Science, University of Waterloo.

This report and the views expressed herein do not necessarily reflect the views or opinions of Statistics Canada or Health Canada.

Estimates
The data presented in this report are weighted estimates, unless otherwise noted. The CTUMS/CTADS survey weights assigned by Statistics Canada in the annual datasets were used for CTUMS/CTADS analyses, and YSS survey weights were used for YSS analyses; CTUMS/CTADS and YSS were not analysed together and there was no overlap of the survey weights between the two surveys. Estimates for categorical measures were generated using the SURVEYFREQ procedure in SAS statistical software (Version 9.4), while estimates for continuous variables (e.g., cigarettes per day) were generated using the SURVEYMEANS procedure in SAS. Estimates and confidence intervals were generated using the statistical software STATA 10.1 using the bootstrap weights where they were available (CTADS 2013, CTUMS 2001 to 2012, and YSS 2004-05 to 2012-13).

Reporting
Confidence intervals are available as supplementary material on the website, but are not included in this report; caution should be used when making comparisons without first checking the confidence intervals. Estimates are not reported where specific categories included less than 30 individuals (unweighted), except where noted as not meeting Statistics Canada’s quality standards. In addition to this rule, Health Canada also recommends calculating the coefficient of variation to determine the quality level of the estimate (for further information, please refer to the documentation for specific surveys and waves/years). As a result, some estimates included in this report may be reported “with caution” or not reported by Health Canada in their releases.

Rounding
Estimates in figures and the associated data tables have been rounded to one decimal place. Provincial estimates for numbers of smokers reported in sections 2.1-2.10 have been rounded to the nearest thousand.

Significance Testing
Statistical comparisons between groups/years were tested using regression analysis, with p<0.05 as the cut-off for significance. Bootstrap weights were used to perform significance testing between groups or between the latest two years, where they were available (CTADS 2013, CTUMS 2001 to 2012, and YSS 2004-05 to 2012-13). The statistical comparisons were performed using the SURVEYLOGISTIC procedure in SAS statistical software (Version 9.4) for all binary response variables. For the continuous variable of cigarettes per day, comparisons were performed using the SURVEYREG procedure in SAS. Comparisons of prevalence rate and cigarettes per day (CPD) over the time period 1999-2013 (#4, #5, #13 and #14) were tested using a dataset of the prevalence rates and CPD by year. The GLM procedure in SAS was used with the yearly prevalence rate regressed on year for #4, the log of yearly prevalence rate regressed on year for #5, yearly CPD regressed on year for #13 and log of yearly CPD regressed on year for #14.

Where statistical testing has been performed, comparisons are marked with a superscript number, which refers to a p-value that can be found in the "Index of Statistical Tests" (page 2). Throughout the report, the term “significant” has been reserved for instances where statistical testing has been performed at the 5% of level of significance (i.e., p<0.05).

Data for Section III (Youth)
Both YSS and CTUMS/CTADS data were used for the youth analysis: YSS data were used for youth in grades 6-9, who were approximately aged 10-14, and CTUMS/CTADS was used for youth aged 15-19. Earlier waves of the YSS included students in grade 5, who are not included in this report for purposes of comparability between survey waves. The more recent waves of the YSS (from 2006-07 onward) also included students in grades 10-12, but these students were not included in the analysis due to their overlap in age with the CTUMS/CTADS sample; CTUMS/CTADS was selected as the data source for older youth since the sampling frame includes youth both in and out of school, whereas the YSS only samples youth who are attending school.

YSS and CTUMS/CTADS data have been integrated where possible. However, differences in the questions asked on each survey and the timing of the surveys does not allow for parallel reporting of all measures. The most recent wave was 2013 for CTADS, and 2012-13 for the YSS. The YSS runs on school years (data collection between September and June), while CTUMS/CTADS runs on calendar years (data collection from February to December). Data collected via the YSS (grades 6-9) are presented by grade rather than age, as the survey was school-based and sampling was done by grade. CTUMS/CTADS is not school-based, so data are presented by age.

The YSS and CTUMS/CTADS base their definition of a current smoker on different items: the YSS defines a current smoker as having smoked at least 100 cigarettes in his/her lifetime and smoked in the 30 days preceding the survey; CTUMS/CTADS defines a current smoker using their response to the question "At the present time do you smoke cigarettes every day, occasionally, or not at all?"