Mental health of youth and young adults: Technical notes

---

About the data source

Data for this analysis come from the Canadian Community Health Survey (CCHS)^{Footnote 3}. Statistics Canada collected data from:

• January to December for the 2015 to 2019 cycles
• January to March, and then again from September to December for the 2020 cycle
• January 2021 to February 2022 (apart from May 2021 due to census data collection) for the 2021 cycle
• February to December for the 2022 cycle

The target population for the survey was those aged 12 years or older, living in the 10 provinces and 3 territories in Canada. Excluded from the survey were:

• those living on reserve or other Aboriginal settlements
• full-time members of the Canadian Armed Forces
• youth aged 12 to17 in foster homes
• those living in institutions
• residents in certain remote regions of Quebec

---

Survey questions

Descriptions and notes for questions in the CCHS.

Positive mental health

CCHS cycles 2015 to 2022 measured self-rated mental health by asking, “In general, would you say your mental health is…?” and lists response options:

• “Excellent”
• “Very good”
• “Good”
• “Fair”
• “Poor”

People with “Excellent” or “Very good” mental health were categorized as having high self-rated mental health^{Footnote 1Footnote 4}.

CCHS cycles 2015 to 2022 measured life satisfaction by asking, “Using a scale of 0 to 10, where 0 means “Very dissatisfied” and 10 means “Very satisfied”, how do you feel about your life as a whole right now?”. Life satisfaction was kept as a numerical value to determine mean life satisfaction rating^{Footnote 1Footnote 4}.

CCHS cycles 2015, 2019 and 2020 measured happiness by asking, “Would you describe yourself as being usually…?” and lists response options:

• “Happy and interested in life”
• “Somewhat happy”
• “Somewhat unhappy”
• "Unhappy with little interest in life”
• “So unhappy, that life is not worthwhile”

People who responded they are usually “Happy and interested in life” were categorized as having high happiness^{Footnote 1Footnote 4}.

Mental illness

CCHS cycles 2015 to 2021 measured "long-term conditions”. Conditions included mood disorders and anxiety disorders. The survey told respondents that they would be asked about “long-term conditions which are expected to last or have already lasted 6 months or more and that have been diagnosed by a health professional”. The survey then asked about each type of condition.

The presence of a mood disorder was measured by asking, “Do you have a mood disorder such as depression, bipolar, mania or dysthymia (Yes or No)?”. The presence of an anxiety disorder was measured by asking, “Do you have an anxiety disorder such as a phobia, obsessive-compulsive disorder or a panic disorder (Yes or No)?”. Respondents who answered “Yes” to either question were categorized as having a mood and/or anxiety disorder.

In CCHS cycle 2022, the mood disorder question was modified to include manic depression. The anxiety disorder question was modified to include generalized anxiety disorder and exclude obsessive-compulsive disorder. These content changes could bias the prevalence estimate. Therefore, this study did not consider CCHS cycle 2022 in the analysis for this outcome.

---

Analysis

For our study, the population was limited to those 12 to 25 years old living in the 10 provinces. The sample size for each year of the survey was:

• 2015 n=7,821
• 2016 n=8,225
• 2017 n=8,210
• 2018 n=7,439
• 2019 n=7,145
• 2020 n=3,946
• 2021 n=5,361
• 2022 n=5,660

Prevalence or mean of each mental health outcome was calculated by year for the overall population and by sex (males and females). All estimates were weighted to be nationally representative, and variance was calculated using bootstrap weights. Non-overlapping 95% confidence intervals (CIs) were used to examine statistically significant differences across years and between sociodemographic groups.

---

Limitations

Since all data are self-reported and cross-sectional, findings are likely subject to social desirability bias. Therefore, causation cannot be inferred based on the findings. Trends are at the population-level and not at the individual level.

The COVID-19 pandemic had major impacts on the data collection operations for the CCHS in 2020. Data collection was stopped mid-March and did not resume until September. The impossibility of conducting in-person interviews, the shorter collection periods and collection capacity issues resulted in a significant decrease in the response rates. As a result, data from 2020 CCHS should be interpreted with caution, especially when examining small sub-populations or comparing to other CCHS years.

The 2022 redesign of the CCHS changed the collection mode from computer-assisted telephone interview (CATI)/computer-assisted personal interview (CAPI) to an electronic questionnaire (EQ) format with CATI and CAPI follow-up for non-response. Based on an internal study conducted by Statistics Canada, the changes seen in 2022 for self-rated mental health and life satisfaction could be due to mode effects. As a result, caution should be used when comparing estimates from CCHS 2022 to estimates from CCHS cycles 2015-2021. Taking this into consideration, comparisons on the overview tab are only made from CCHS cycles 2015-2021, but 2022 estimates are still provided in the data tool to show the most recent data.

The mood and/or anxiety disorder outcome only captures those who self-reported being diagnosed by a medical professional. It does not include individuals with symptoms of a mood and/or anxiety disorder without a formal diagnosis. It also does not include those who have a formal diagnosis but are unwilling to disclose it. The results likely underrepresent the true number of individuals dealing with anxiety and mood disorders.

These estimates do not consider other contextual factors. For example, when looking at Immigrant status, this study did not consider length of residency in Canada^{Footnote 2}. Therefore, these estimates should be interpreted with caution.

The mental health outcomes analyzed in this study were developed for use in a Western context. They may not resonate with all groups identified in our cultural/racial data breakdowns, such as First Nations off reserve, Inuit or Métis.

We used non-overlapping 95% CIs to determine statistical significance. This is a conservative approach and could have overlooked some statistically significant differences.