Canadian Postsecondary Education Alcohol and Drug Use Survey, 2021-2022: Technical notes:

Data notes

The number of schools recruited for CPADS 2021-22 represents 23% of all colleges and universities in Canada (total=196), all of which were contacted for recruitment if they met eligibility criteria. The overall survey sample includes representation from all Canadian regions except for Nunavut, Northwest Territories, and Yukon (who chose not to participate).

• In the data tables, results are reported as a percentage and are accompanied by their associated 95% confidence interval.
• Results with moderate sampling variability are indicated with the symbol ‘*’ and should be interpreted with caution.
• Some data are not reportable due to low numbers. Data are only reportable if the number of observations in a cell’s numerator was >=10, the number of observations in the cell’s denominator was >= 50, and the coefficient of variation for the result was < 33.3%. Unreportable numbers are suppressed to ensure that participants cannot be identified or to avoid reporting unstable results. These are indicated with the ‘#’ symbol in the data tables.
• Due to rounding, percentages may not add to 100.
• The pooled two-proportion z-test was used to analyze differences between cycles, genders, and self-reported mental health groups. All reported differences in the web summary text are statistically significant at the 95% confidence level.
• While statistical significance examines whether the findings are likely to be due to chance, effect size examines the magnitude of any differences found. Cohen's h was used to measure effect size where h values less than 0.2 are considered to have a trivial effect size with no meaningful difference between the proportions being compared. The h value criteria for small, medium, and large effect sizes are 0.2, 0.5, and 0.8, respectively. Examining effect size in addition to statistical significance helps with understanding and interpreting findings.
• Weights were applied to ensure that the data are representative of the student population at each school by sex and age. Results from each school were then further weighted to match the overall student population of the four regions captured in the CPADS 2021-22 cycle.
• Most responses were analysed by gender. However, responses related to heavy drinking were analysed by sex (the respondents’ biological sex assigned at birth), as the definition for heavy drinking is based on sex.
• Questions about cannabis did not differentiate between medical and non-medical use, anticipating that the proportion of young adults who use cannabis exclusively for medical purposes would be small and would not greatly impact resultsⁱⁱ.

Key definitions

• Transgender/non-binary: includes transgender males, transgender females, non-binary gender, gender-fluid, and other write-in responses.
• Heavy drinking: defined as having 4 or more drinks for females and 5 or more drinks for males on one occasion in the past 30 days. ‘On one occasion’ means consuming drinks consecutively or within a couple of hours of each other.
• Canada’s Low-Risk Alcohol Drinking Guidelines 2011: The CPADS calculates the proportion of respondents who exceed the Guidelines based on alcohol consumption in the 7 days prior to the survey. Those who follow this guidance must consume "no more than 10 drinks a week for women, with no more than 2 drinks a day most days and 15 drinks a week for men, with no more than 3 drinks a day most days. It is recommended to plan non-drinking days every week to avoid developing a habitⁱ.” Note: The 2011 drinking guidelines were those in effect during the 2021-22 school year.
• ASSIST: An international group of researchers specialised in higher-risk substance use developed the Alcohol, Smoking and Substance Involvement Screening Test (ASSIST) for the World Health Organization (WHO) to help detect and manage substance use and related problems in primary and general medical care settings. The CPADS cannabis module adopts questions from ASSIST to screen respondents for higher-risk cannabis use. Respondents are categorised based on their consumption as follows:
  1. Low risk of developing health and other problems (score of 0-3)
  2. Moderate risk of developing health and other problems (score of 4-26)
  3. High risk of developing health and other problems and indicative of a substance dependence (score of 27+)
• Higher-risk psychoactive pharmaceutical use is defined as using more (quantity), using more often (frequency) or using substances in a way other than prescribed (for example, to get high).

Survey limitations

The following survey design limitations may have influenced the CPADS results:

• Results are based on self-reported data, which are subject to recall bias, data entry errors, and prone to under or over reporting. Respondents may forget actual use with increasing consumption and those who drink more heavily have been found to have higher rates of non-participation in surveysⁱⁱⁱ. In addition, desirability bias (where respondents answer in a way that they think is socially acceptable) and lack of knowledge of standard serving sizes may contribute to reporting errors^iv.
• The survey topic is shown in the title of this project, which could lead to a skewed sample of respondents that are by nature more interested and knowledgeable on the topic of substance use. This could lead to an over or under estimation of prevalence.
• No data were available from Nunavut, Northwest Territories, or Yukon, who chose not to participate in the survey.