Health Reports
Mental health among women and girls of diverse backgrounds in Canada before and during the COVID-19 pandemic: An intersectional analysis

Abstract

Background

Mental health disparity is associated with diverse characteristics, such as gender, socioeconomic status, Indigenous identity, immigrant status, race, disability, and sexual orientation. However, intersectional studies on women’s mental health have been rare, particularly during the COVID-19 pandemic period.

Methods

Using data from two cycles of the Canadian Community Health Survey (2019 annual data and data from September to December 2020), self-reported mental health outcomes before the COVID-19 pandemic (sample size was 64,880) and during the second wave of the pandemic in the fall of 2020 (sample size of 27,246) were analyzed.

Results

After sociodemographic factors were controlled for, women and girls had higher odds of poorer self-perceived mental health and worsened mental health compared with before the COVID-19 pandemic than men and boys. Compared with 2019, the gender gap in negative self-perceived mental health increased during the pandemic. The number and type of intersections of specific socioeconomic characteristics also had an impact on mental health outcomes. During the pandemic, women and girls with the following characteristics were more likely to report low self-perceived mental health, compared with women and girls with no intersections: those with a disability (7.8 times); or who are lesbian, gay, or bisexual or have another sexual orientation than heterosexual (5.6); or who are Indigenous (3.6).

Interpretation

The intersections of gender and other sociodemographic characteristics increased the odds of negative self-perceived mental health.

Keywords

Mental health, intersectional analysis, women, COVID-19 pandemic

Authors

Jungwee Park is with the Health Analysis Division at Statistics Canada.

 

What is already known on this subject?

• Previous research has extensively documented women’s mental health and the presence of gender disparities in mental health.
• Existing knowledge highlights the varying impact of the COVID-19 pandemic on different groups across Canada.
• Recent research has examined the mental health of women and other populations during the pandemic.

What does this study add?

• This study investigates women’s and girls’ self-reported mental health before and during the COVID-19 pandemic, considering multiple characteristics such as Indigenous identity, immigrant status, racialized background, sexual orientation, disability status, and socioeconomic status.
• Taking an intersectional approach, the study seeks to understand the extent to which the intersections of various sociodemographic characteristics influenced women’s and girls’ mental health during the pandemic.
• Future intersectional analyses of mental health in the context of the pandemic could expand to encompass a wider range of mental health conditions, a more extensive array of sociodemographic characteristics, and different time periods (various waves of the pandemic and the duration of its impacts) and extend the examination to transgender and non-binary people.

Introduction

Many studies have reported increased inequities between men and women in mental health since the COVID-19 pandemic,^Note 1, ^Note 2, ^Note 3 including higher risks of post-traumatic stress symptoms^Note 4 and psychiatric disorders and loneliness.^Note 5 Moreover, extensive research has been undertaken to examine the unequal effects of the pandemic on self-perceived mental health among diverse groups such as Indigenous peoples and different gender groups.^Note 1, ^Note 3, ^Note 6, ^Note 7

As a group, women experienced unique challenges during the pandemic. At work, they were more likely to have informal and precarious jobs that faced greater layoffs or furloughs and to be frontline service providers, such as cleaning staff, cashiers, social workers, teachers, nurses, and personal support workers, bearing a greater burden of mental and physical health risks.^Note 8 At home, they carried on most of the unpaid caregiving work that increased during the pandemic, such as child care challenges, homeschooling, and other parental tasks.^Note 9, ^Note 10, ^Note 11 As a result, since the pandemic, women have reported higher levels of economic distress (e.g., job loss and income decline), household stress, social isolation, shelter-in-place rates, parental stress, substance misuse, and psychological distress.^Note 6, ^Note 12, ^Note 13 These factors also increased the risk of violence against women and girls during the pandemic.^Note 12, ^Note 14, ^Note 15

The pandemic also had a differential impact on diverse groups in Canada.^Note 1, ^Note 3, ^Note 16 A Statistics Canada study found that respondents with the following characteristics reported higher rates of experiencing discrimination during the pandemic and experienced greater consequences to their economic well-being: racialized people; immigrants; Indigenous people; or people who were lesbian, gay, bisexual, or of another sexual orientation than heterosexual (LGB+).^Note 17, ^Note 18

Women in these groups tended to be particularly affected by the pandemic. For instance, from March 2020 to August 2021, Indigenous women’s employment recovery was slower than that of Indigenous men and non-Indigenous men and women.^Note 19, ^Note 20 The unemployment rate among Indigenous women remained above its pre-pandemic level in the first year following the beginning of the pandemic.20 Also, recent immigrant women displayed the largest gap in employment recovery with their Canadian-born counterparts. Their recovery was lower by 5 percentage points in both May and June 2020, and 2 percentage points in July 2020.^Note 18

It has been also reported that the pandemic might have exacerbated the vulnerabilities of LGB+ people in terms of job loss, financial insecurity, and homelessness.^Note 21 Stonewall^Note 22 reported that lower educational attainment, poverty, housing, and food insecurity among lesbian and bisexual women and transgender people was related to stigma and discrimination worldwide—across 24 countries.

The pandemic had a significant impact on the livelihood of individuals with disabilities as well. According to Trudell and Whitmore,^Note 23 the economic and housing precarity already experienced by women with disabilities was exacerbated during the pandemic. Despite the financial support introduced by the Government of Canada, the economic needs of women with disabilities tended to remain unmet.^Note 24 In addition, many formal services and supports, including home and personal care support, were limited during the COVID-19 pandemic. A result of these socioeconomic impacts and limits on services was an increased vulnerability of women with disabilities as they relied on informal support networks potentially including abusers or gender-based violence.^Note 24

Mental health disparities are not only observed between men and women but also associated with certain characteristics, including socioeconomic status, race, disability, and sexual orientation.^Note 6, ^Note 16, ^Note 25, ^Note 26, ^Note 27 As discussed, the pandemic has brought about health, economic, and various social repercussions, which, in turn, are impacting the mental health of the population.^Note 28 Existing mental health inequalities experienced by diverse groups might have been intensified during the pandemic. It has been recognized that gender inequity in mental health outcomes has widened since the pandemic.^Note 1, ^Note 2, ^Note 3 During the pandemic period, certain groups face a higher risk of experiencing more severe challenges. If only one characteristic is studied, its impact may be overgeneralized, and the complex and interacting nature of the inequity cannot be fully understood. It is important to take an intersectional approach, examining multiple sources of inequalities together.^Note 29, ^Note 30, ^Note 31 Rather than studying the mental health of women as one general group, it is crucial to explore the mental health of women at the intersections of certain sociodemographic characteristics that contribute to structural inequalities (e.g., women with severe disabilities, Indigenous women, and LGB+ women).

Intersectional studies on women’s mental health have been rare, particularly during the pandemic period.^Note 25, ^Note 32 Moreover, among the previous intersectional studies on mental health, only a few studies involved multiple intersectional characteristics.^Note 33, ^Note 34 Most intersectional studies tend to focus on a couple of characteristics.^Note 26, ^Note 27, ^Note 35, ^Note 36, ^Note 37, ^Note 38, ^Note 39, ^Note 40, ^Note 41 For example, many studies focused on the intersection between gender and ethnic identity, but rarely investigated the intersections between gender, ethnicity, sexual orientation, and disability. To fill this research gap, this study examines women’s and girls’ self-reported mental health before and during the COVID-19 pandemic using seven characteristics, including Indigenous identity, immigrant status, racialized background, LGB+ sexual orientation, disability, and socioeconomic status (low income and unemployment). Because of the small sample size, this study only provided estimates for LGB+ women, not specifically for lesbian women, bisexual women, or women with another sexual orientation that is not lesbian, bisexual, or heterosexual. It attempts to answer the following research questions:

• To what extent do the intersections of various characteristics affect women’s and girls’ self-perceived mental health during the COVID-19 pandemic?
• How does this effect of the intersections on women’s and girls’ self-perceived mental health during the COVID-19 pandemic compare with that before the pandemic?
• To what extent have the intersections of various characteristics affected the change in women’s and girls’ self-perceived mental health (worsened or not) since the COVID-19 pandemic compared with before the pandemic?

Methods

Data sources

This study used the annual cycles of the Canadian Community Health Survey (CCHS). The CCHS is a cross-sectional survey that collects information related to health status, health care utilization, and health determinants for the Canadian population. This analysis focused on the data on self-reported mental health outcomes before the COVID-19 pandemic (2019 CCHS annual data) and during the second wave of the pandemic in the fall of 2020 (2020 CCHS—September to December 2020) for people living in the 10 provinces. After a brief pause in collection in mid-March, near the end of the first collection period, because of lockdowns and public health guidelines, the 2020 CCHS resumed collection in September 2020.^Note 3 The second, third, and fourth quarterly samples were collected during very short collection periods of about five weeks each from September to December. The impossibility of conducting in-person interviews, the shorter collection periods, and collection capacity issues resulted in a significant decrease in response rates.^Note 42 The collection from September to December 2020 provided information reflecting respondents’ experience with the COVID-19 pandemic. A detailed description3 of the data validation and methodology change in the 2020 CCHS is available elsewhere.

Sample

The sample size of the 2020 CCHS (September to December) was 27,246 (12,078 men and 15,168 women), representing 32,342,696 people aged 12 years or older living in the 10 provinces in Canada. The sample size of the 2019 CCHS was 64,880, representing 31,837,719 people aged 12 years or older living in the 10 provinces in Canada. Table 1 shows the distribution of selected characteristics of Canadians aged 12 years or older.



Measures

Gender

In this analysis, gender is based on a question asking, “What is your gender?” Responses were reported under three categories: man, woman, and non-binary. Non-binary refers to people whose reported gender is not exclusively man or woman. Because of the small sample size, individuals who were categorized as non-binary were excluded from the analysis. 

Place of residence

All population centres, including small (1,000 to 29,999 people), medium (30,000 to 99,999), and large (100,000 or greater), were classified as urban regions. Rural areas inside and outside a census metropolitan area or census agglomeration were classified as rural regions.

Low income

In this analysis, individuals whose total household income was within the lowest quintile were categorized as individuals with low income.

Employment status

Employment status was measured based on the working status in the previous week. Those respondents of working age (aged 15 to 74 years) who reported that they did not have a job last week were classified as unemployed. Those aged younger than 15 and older than 74 were not considered as unemployed.

Immigrant status

All non-Canadian-born individuals, including landed immigrants and non-permanent residents, were classified as immigrants.

Racialized group

The concept of “racialized group” is measured with the “visible minority group” variable in the CCHS as defined by the Employment Equity Act. The Employment Equity Act defines visible minorities as “persons, other than Aboriginal peoples, who are non-Caucasian in race or non-white in colour.” The visible minority population consists mainly of those who self-reported belonging to the following groups: South Asian, Chinese, Black, Filipino, Arab, Latin American, Southeast Asian, West Asian, Korean, and Japanese.

Indigenous identity

In this article, the term “Indigenous people” refers to those who self-identified as First Nations people, Métis, or Inuit. The CCHS does not collect data on reserves and other Indigenous settlements in the provinces. Consequently, the results discussed for First Nations people exclude those living on reserves; results also exclude Indigenous peoples in the territories or remote northern regions of the provinces, which include Inuit Nunangat.

Sexual orientation

Sexual orientation of respondents was determined based on a question asking about self-reported sexual orientation. Responses are reported using the following categories: heterosexual; lesbian or gay; bisexual; and people whose reported sexual orientation is not heterosexual, lesbian, gay, or bisexual. “LGB+ sexual orientation” includes people who are lesbian, gay, bisexual, or of another sexual orientation that is not heterosexual. 

Disability

In this article, disability was measured based on Health Utilities Index (HUI) global scores, which provide a summary index of health-related quality of life on a 0.0 to 1.0 scale. HUI was based on utility scores of eight attributes: vision, hearing, speech, mobility (ability to get around), dexterity (use of hands and fingers), emotion (feelings), cognition (memory and thinking), and pain. The level of disability was determined by the score’s cut-off points: none (1.00), mild (0.89 to 0.99), moderate (0.70 to 0.88), and severe (less than 0.70).^Note 43, ^Note 44 In this analysis, the mental health outcomes of individuals with severe disabilities were examined in comparison with those of other individuals.

Intersectionality of socioeconomic characteristics

In this analysis, the following socioeconomic characteristics of individuals were studied: low income, unemployment, immigrant status, Indigenous identity, racialized group, LGB+ sexual orientation, and disability. When individuals present two or more of these characteristics, these characteristics intersect and may be related to different experiences and outcomes.

Number of intersections of socioeconomic characteristics

To look at the intersectionality of characteristics, individuals were classified into four groups, reflecting how many of the seven socioeconomic characteristics mentioned above they presented: zero, one, two, and three or more. 

Self-perceived mental health

Self-perceived mental health was measured by asking respondents, “In general, would you say your mental health is excellent? Very good? Good? Fair? Poor?” If the responses were fair or poor, the respondent was considered to have low self-perceived mental health. If the responses were excellent, very good, or good, the respondent was considered to have high self-perceived mental health.

Worsened mental health

Worsened mental health was based on a retrospectively reported measure of self-perceived mental health during the COVID-19 pandemic compared with before the pandemic started. Respondents were asked, “Compared to before the pandemic started, how would you say your mental health is now? Would you say much better now? Somewhat better now? About the same? Somewhat worse now? Much worse now?” Those who answered somewhat worse now or much worse now were classified as having worsened mental health. Those who answered much better now, somewhat better now, or about the same were classified as not having worsened mental health.

Analysis

Descriptive statistical analyses were conducted to provide prevalence rates of negative mental health outcomes—low self-perceived mental health reported in 2019 and 2020 and worsened mental health since the COVID-19 pandemic began reported retrospectively in 2020. Estimates for women and girls were compared with those for men and boys. In addition, an intersectional approach was taken to compare the mental health outcomes of diverse groups of women and girls.

Multivariate logistic regression analyses were conducted to examine the relationships of various types of sociodemographic characteristics with mental health, while controlling for several confounding factors. Adjusted odds ratios of low self perceived mental health and worsened mental health since the COVID-19 pandemic began for men and boys and women and girls were presented separately. Age, marital status, family arrangement, and place of residence were confounding factors that were controlled.

In this analysis, missing cases in main variables were very low, ranging from 0% (in income, age, and place of residence) to 5% (in sexual orientation). Descriptive statistics were based on available case analysis, and multivariate regression was based on complete case analysis following listwise deletion of the missing cases.

The symbol E next to an estimate indicates that the coefficient of variation (CV) for this estimate was from 15.1% to 35.0% and the data quality was marginal. Users should interpret these results with caution. Estimates were suppressed and indicated by the symbol F when the CV was greater than 35.0%. Statistical significance was indicated based on the tests with a p-value of less than 0.05. For estimates produced from the CCHS data to be representative of the Canadian population, individual sampling weights were used. Bootstrap weights were used for variance estimation and significance tests. The number of bootstrap replicates was 1,000. All analyses used SAS version 9.4 (SAS Institute, North Carolina, United States).

Results

Differences in mental health outcomes between women and girls and men and boys before and during the COVID-19 pandemic

Both before and during the COVID-19 pandemic, women and girls were more likely than men and boys to report low self-perceived mental health. During the pandemic from September to December 2020, 11.5% of women and girls reported low self-perceived mental health compared with 9.0% of men and boys (Appendix Table 1). Similarly, women’s and girls’ overall prevalence rates of worsened mental health during the pandemic were higher than those of their male counterparts (Appendix Table 2). From September to December 2020, 37.1% of women and girls reported that their current mental health was somewhat or much worse compared with before the start of the pandemic. Men’s and boys’ rate of worsened mental health (28.5%) was about 9 percentage points lower than women’s and girls’. Women’s and girls’ rate of low self-perceived mental health was higher in 2020 than in 2019, before the pandemic (11.5% in 2020 vs. 8.7% in 2019). Similarly, men’s and boys’ rates went up from 7.3% in 2019 to 9.0% in 2020.

Chart 1 presents the odds ratios of a series of logistic regressions showing women’s and girls’ negative mental health outcomes compared with men’s and boys’ during the pandemic. Model 1 shows the effect of gender on mental health after controlling for age; model 2 controls for age, marital status, family arrangement, and place of residence (rural vs. urban); model 3 controls for the number of intersections of socioeconomic characteristics (low income, unemployment, immigrant status, Indigenous identity, racialized group, LGB+ sexual orientation, and disability); and model 4 controls for all aforementioned factors together. Women and girls consistently showed higher odds of negative mental health outcomes compared with men and boys, including lower self-reported mental health and worsened mental health since the COVID-19 pandemic, in all models. A similar pattern was found for both mental health outcomes. For example, compared with men and boys, women and girls had 1.5 times higher odds of reporting worsened mental health during the pandemic after controlling for sociodemographic characteristics.

Description of Chart 1



Data table for Chart 1
Table summary
This table displays the results of Data table for Chart 1 Adjusted odds ratio (appearing as column headers).

	Adjusted odds ratio
Low self-perceived mental health
Model 1 - adjusted for age	1.34
Model 2 - adjusted for age and sociodemographic factors	1.32
Model 3 - adjusted for age and number of intersections	1.29
Model 4 - adjusted for age, sociodemographic factors and number of intersections	1.26
Self-reported worsened mental health since the pandemic began
Model 1 - adjusted for age	1.507
Model 2 - adjusted for age and sociodemographic factors	1.489
Model 3 - adjusted for age and number of intersections	1.501
Model 4 - adjusted for age, sociodemographic factors and number of intersections	1.484
Source: Canadian Community Health Survey annual cycle 2020 (September to December).

Number of intersections of socioeconomic characteristics

The number of intersections has been of interest to previous research on intersectionality.45,46 In this analysis, this was measured by looking at how many of the following selected sociodemographic characteristics individuals reported: low income, unemployment, immigrant status, Indigenous identity, racialized group, LGB+ orientation, and disability. In 2020 (since the start of the COVID-19 pandemic), about 15.6% of women and girls, and 12.5% of men and boys, reported three or more of the selected characteristics. For example, LGB+ racialized adult women with no job or low-income immigrant women with a disability belong to this “three or more” category. In 2020, about 30.3% of women and girls presented none of these characteristics, and 29.5% presented one of them (data not shown). Those percentages were lower than in 2019 (32.4% presenting none and 30.5% presenting one) primarily because of an increase in the number of women who were unemployed in 2020.

Not surprisingly, women and girls characterized by a higher number of intersections tended to show a higher prevalence rate of poor self-reported mental health in 2020. Compared with women and girls who did not report being part of any of the selected sociodemographic groups (i.e., Canadian-born, White, heterosexual women with no severe disability who had a job and whose household income belonged to the second or higher quintiles), women and girls who reported one of the seven sociodemographic characteristics were 2.0 times more likely to report low self-perceived mental health; women and girls with two intersections were 2.5 times more likely, and women and girls with three or more intersections were almost 3 times more likely to report low self-perceived mental health (Chart 2, Appendix Table 1). The number of intersections did not show a significant effect on worsened mental health since the pandemic began.

Examining specific intersections

Description of Chart 2



Data table for Chart 2
Table summary
This table displays the results of Data table for Chart 2 Low self-perceived mental health and Worsened mental health since pandemic began, calculated using adjusted odds ratio units of measure (appearing as column headers).

	Low self-perceived mental health	Worsened mental health since pandemic began
	adjusted odds ratio
One	1.96Data table for Chart 2 Note †	0.92
Two	2.46Data table for Chart 2 Note †	0.83
Three or more	2.86Data table for Chart 2 Note †	0.95
Note † significantly different from reference category (women and girls with no intersection) Return to note † referrer Notes: Adjusted for age, marital status, family arrangement, and place of residence. Out of seven intersections: low income, unemployment, Indigenous identity, immigrant status, racialized group, non-heterosexual orientation, and severe disability. Source: Canadian Community Health Survey annual cycle 2020 (September to December).

Chart 3 shows variations in mental health outcomes during the pandemic between women and girls with specific intersecting sociodemographic characteristics in comparison with women and girls who did not report any of the selected characteristics. It shows adjusted odds ratios of two negative mental health outcomes. After the analysis controlled for age, marital status, family arrangement, and place of residence (rural or urban), all seven selected characteristics were associated with low self-perceived mental health during the pandemic. Overall, 11.5% of women and girls reported low self-perceived mental health during the pandemic. Further disaggregation revealed that 30.2% of women and girls with severe disabilities, 35.9% of LGB+ women and girls, and 24.4% of Indigenous women and girls reported low self-perceived mental health during the pandemic. On the other hand, immigrant women and girls (6.7%) and racialized women and girls (8.5%) were less likely to report low self-perceived mental health (Table 2, Appendix Table 1). Notably, women and girls with severe disabilities were 6.3 times more likely to report poor mental health than other women and girls with no disabilities or less severe (mild or moderate) disabilities.

Description of Chart 3



Data table for Chart 3
Table summary
This table displays the results of Data table for Chart 3 Self-perceived mental health and Worsened mental health since the pandemic began, calculated using adjusted odds ratio units of measure (appearing as column headers).

	Self-perceived mental health	Worsened mental health since the pandemic began
	adjusted odds ratio
Low income	2.44	0.81Data table for Chart 3 Note †
Unemployment	2.55	0.92Data table for Chart 3 Note †
Indigenous identity	3.64	1.11Data table for Chart 3 Note †
Immigrant status	1.02	0.79Data table for Chart 3 Note †
Racialized group	1.09	0.69Data table for Chart 3 Note †
LGB+ sexual orientation	5.60Data table for Chart 3 Note †	1.00
Severe disability	7.77Data table for Chart 3 Note †	1.48Data table for Chart 3 Note †
Note † significantly different from reference category (women and girls with no intersection) Return to note † referrer Notes: Adjusted for age, marital status, family arrangement, and place of residence. "LGB+ sexual orientation” includes people who are lesbian, gay, bisexual, or of another sexual orientation that is not heterosexual. Source: Canadian Community Health Survey annual cycle 2020 (September to December).



Compared with other women and girls, racialized women and girls and immigrant women and girls were less likely to report that their mental health became worse during the pandemic. However, women and girls with severe disabilities were 1.5 times more likely than other women and girls to report worsened mental health (Chart 3). Compared with 37.1% of women and girls overall, 47.1% of women and girls with a disability reported worsened mental health during the pandemic (Table 2, Appendix Table 2).

Women and girls with multiple intersecting characteristics

Table 2 presents mental health outcomes for women and girls with multiple intersecting characteristics. For example, this table provides statistics for LGB+ women and girls overall, as well as low-income LGB+ women and girls, unemployed LGB+ women and girls, and LGB+ women and girls with severe disabilities, while controlling for intersections with all other sociodemographic variables in the model. The reference group for these analyses was women and girls who had none of the seven selected sociodemographic characteristics (i.e., low income, unemployment, immigrant status, Indigenous identity, racialized group, LGB+ orientation, or disability).

Compared with the reference group, the odds of low self-perceived mental health were 5.6 for LGB+ women and girls overall, 5.4 for low-income LGB+ women and girls, 6.2 for unemployed LGB+ women and girls, and 18.7 for LGB+ women and girls with severe disabilities.

The odds of low self-reported mental health for immigrant women and girls and racialized women and girls were statistically lower than those of the reference group. For immigrant women and girls and racialized women and girls with disabilities, however, the odds of low mental health became much greater than those of the reference group (5.5 for immigrant women and girls with disabilities; 7.2 for racialized women and girls with disabilities).

When LGB+ women and girls or women and girls with severe disabilities had one of the other sociodemographic characteristics studied, their self-perceived mental health was impacted. For example, women and girls with severe disabilities who were unemployed were 9.2 times more likely than the reference group to report low mental health. The odds of reporting low self-perceived mental health for LGB+ women and girls with disabilities were 18.7 times greater than the reference group with none of seven intersecting characteristics (Table 2).

Almost 40% of women and girls reported that their current mental health was somewhat or much worse compared with before the pandemic (Table 2). In particular, unemployed Indigenous women and girls and unemployed women and girls with disabilities showed higher odds than the reference group of reporting worsened mental health. Overall, immigrant and racialized women and girls were less likely to report worsened mental health since the pandemic began. However, there was no statistically significant difference in reporting worsened mental health between the reference group and immigrant and racialized women and girls with another selected sociodemographic characteristic, such as low income, unemployment, LGB+ sexual orientation, or severe disability. 

Discussion

This study highlighted certain mental health outcomes of diverse groups of women and girls before (2019) and during (September to December 2020) the COVID-19 pandemic. As expected, gender disparities in mental health outcomes between men and boys and women and girls were persistent. After controlling for selected sociodemographic factors, women’s and girls’ odds of certain mental health outcomes in this analysis (low self-perceived mental health and worsened mental health compared with before the pandemic) were higher than men’s and boys’. Compared with 2019, the gap in negative self-perceived mental health between men and boys and women and girls increased during the pandemic. This finding was consistent with previous research suggesting the exacerbation of mental health challenges experienced by certain populations, specifically women.^Note 1, ^Note 2 It is widely recognized that the COVID-19 pandemic has exacerbated various pre-existing socioeconomic and health challenges faced by women, who have long endured systemic oppression, discrimination, and inequality.^Note 6, ^Note 7, ^Note 15, ^Note 17, ^Note 18, ^Note 21, ^Note 22, ^Note 23, ^Note 24, ^Note 47, ^Note 48

The findings of this study were consistent with previous research that has found lower self-perceived mental health among Indigenous people^Note 3, ^Note 15 and other vulnerable populations, such as LGB+ people and individuals with long-term conditions or disabilities.^Note 3 More importantly, however, this analysis illustrated that the exacerbation of these challenges became more pronounced when multiple vulnerable characteristics intersected. In particular, when two of these three characteristics—Indigenous identity, LGB+ sexual orientation, and severe disabilities or conditions—interacted with each other, the likelihood of experiencing low self-perceived mental health significantly increased.

As well, the intersection of those characteristics with certain economic conditions, such as low income or unemployment, was associated with high odds of poor self-perceived mental health. Indigenous women and girls who were unemployed were most likely to say that their mental health worsened since the pandemic started. About 55% of them reported this, compared with 37% of women and girls overall and 29% of men and boys.

Disability status consistently increased the odds of poor self-perceived mental health when it intersected with all other characteristics included in this analysis. Also, women and girls with severe disabilities showed higher odds of reporting worsened mental health since the pandemic began. However, it is important to note that the eight components of the HUI measure of disability used in this analysis included emotion as well as cognition (see “Measures” section). This may be related to the high effect of disability on mental health outcomes.

Since this analysis did not further disaggregate data by specific subgroups, the results reflect the averages for only the overall populations of immigrant and racialized women and girls. Generally, however, racialized and immigrant status showed some “protective” effects on mental health outcomes. For example, the odds of reporting worsened mental health were lower than the reference group with none of the seven selected sociodemographic characteristics included in this study. For racialized women, this may be related in part to the Black–White mental health paradox, pointing out racialized people’s subjective mental health advantage over White people.^Note 49, ^Note 50 Positive mental health outcomes found among immigrant women and girls suggest that the healthy immigrant effect extends to mental health.^Note 51 Their relatively positive mental health outcomes may also be related to immigrants’ and racialized people’s tendency to underreport mental health issues and underuse mental health care.^Note 52

Interestingly, the low odds of low self-perceived mental health among immigrant and racialized women and girls compared with the reference group were reversed for women and girls with disabilities. In other words, when racialized and immigrant women and girls had a severe disability or condition, the protective effects on self-reported mental health disappeared. Compared with the reference group, immigrant women and girls with severe disabilities were five times more likely to report poor mental health, and racialized women and girls with disabilities were seven times more likely. These results may indicate that the effects of the pandemic on the well-being of racialized or immigrant women and girls with disabilities^Note 23, ^Note 24 were so significant that any protective factors associated with their racialized or immigrant status against poor mental health were nullified.

While women and girls reported poorer mental health during the pandemic than in the year preceding it, the patterns found during the two periods studied were similar. That is, the characteristics associated with women’s and girls’ mental health outcomes during the pandemic, and the effect of their intersections, were also observed before the pandemic. There were more significant findings in the 2019 cycle since its sample size was more than twice as large as that of the 2020 cycle.

Limitations

The data pertaining to the pandemic period were collected as part of the 2020 CCHS annual cycle. As a result, the portion of the cycle that covers the pandemic period was limited in its sample size. Because of the small sample size in 2020, it was not possible to further examine subgroups of the population groups included in this analysis. For example, because of the limited sample size, it was not possible to provide a distinctions-based analysis of the mental health outcomes of First Nations people living off reserve, Métis, and Inuit. Similarly, certain characteristics of immigrant women and girls, such as time since immigration (recent vs. long-term immigrants), age at immigration, and country of birth, which have impacts on mental health,^Note 51 could not be examined. Again, because of small sample sizes, the analysis could not examine different groups of racialized women and girls. Each specific racialized group may possess unique cultural and ethnic characteristics and warrant further exploration on how characteristics impact associations with mental health outcomes. Sample size issues also limited the opportunities to examine the relationship between mental health outcomes and the intersectionality of selected sociodemographic characteristics included in this analysis. No more than three intersecting characteristics were analyzed.

The data collection for the period during the pandemic was conducted from September to December 2020. Thus, this analysis could not examine the entire impact of the pandemic. Additionally, in the 2020 cycle of the CCHS, some bias was discovered, which is likely attributable to the limitations in survey collection during the pandemic, including the decreased response rate and the use of telephone interviews only.^Note 3 This makes it difficult to compare the findings of this study with those from other cycles.

Since the CCHS is a cross-sectional survey, no causal relationships can be inferred based on the associations found in this analysis. Some of the mental health outcomes might precede or lead to given socioeconomic conditions. For example, economic disadvantages included in this analysis, such as low income or unemployment, may as well be a consequence of poor mental health. Also, the two mental health outcomes in this analysis were self-reported, which are different from clinical diagnoses and should not be interpreted as such.

Despite these limitations, this study, which uses the initial CCHS data encompassing both pre-pandemic and pandemic periods, provides a unique analysis of how different sociodemographic characteristics of women and girls in Canada can intersect and be associated with different mental health outcomes. This contributes to a more comprehensive understanding of the experiences of women and girls of diverse backgrounds.

Conclusion and future studies

The intersectional analysis in this current study highlights the compounding impact of existing social inequalities on the mental health of women and girls with diverse backgrounds before and during the COVID-19 pandemic. It is crucial to consider the intersections of certain characteristics of women and girls for a more comprehensive understanding of women’s and girls’ mental health outcomes, particularly during the COVID-19 pandemic. This approach is important not only for identifying groups of women at higher risk of experiencing poor mental health but, more importantly, for comprehending the specific interrelationships among diverse characteristics that are pertinent to mental health outcomes. The findings of this study may help clinicians in recognizing risk factors and provide decision makers with insights into the implications for policies aimed at directing services and resources to individuals who are more vulnerable to poor mental health outcomes.

Future intersectional analyses of mental health in the context of the pandemic could benefit from an examination of the impact across different time periods, including various waves of the pandemic, and the duration of its effects. Furthermore, data collection should be designed to make possible a more comprehensive examination of a wide range of sociodemographic characteristics and their intersections with sufficient sample sizes, allowing for disaggregation into more specific subgroups. These intersectional analyses can also be extended to include more mental health outcomes, such as stress and diagnosed mental disorders, as well as their long-term association with physical health.^Note 28, ^Note 53 Additionally, future intersectional analysis may delve into more detailed aspects of men’s mental health. Combining data from several cycles of the CCHS can enhance sample sizes, enabling these analyses to be carried out effectively.

Acknowledgement

This paper was funded by Women and Gender Equality Canada.