Skip to main content
NCBI home page
Canadian Journal of Public Health = Revue Canadienne de Santé Publique logoLink to Canadian Journal of Public Health = Revue Canadienne de Santé Publique
. 2021 Apr 20;112(4):629–637. doi: 10.17269/s41997-021-00498-4

Trends in mental health inequalities in urban Canada

Sharalynn Missiuna 1,✉, Charles Plante 2, Punam Pahwa 1, Nazeem Muhajarine 1, Cordell Neudorf 1
PMCID: PMC8225713  PMID: 33877585

Abstract

Objectives

Canadians do not all enjoy equal levels of health. The presence of income-related health inequalities has been well established in Canada, but there is a lack of consistent reporting of mental health inequalities in Canada’s largest cities. This study reports the prevalence and inequalities in mental health outcomes at the city, provincial, and national levels over time.

Methods

Self-reported poor mental health, life stress, and physician-diagnosed self-reported mood and anxiety disorder from the Canadian Community Health Survey were pooled over five-year intervals and combined with neighbourhood income information from the Canadian Census. First, prevalence rates were calculated for each interval at the neighbourhood level for urban communities. Second, the distributions of these neighbourhood rates were summarized at the city level and for Canada as a whole using overall prevalence rates and concentration indices of inequality. Finally, trends in these city- and country-level outcomes were also explored.

Results

At the national level, starting from 2001 to 2005, the prevalence of poor mental health (27.9%), mood disorder (7.3%), and anxiety disorder (6.8%) had significantly increased by 2011–2015. Inequalities were present in 2001–2005 and worsened over time. The prevalence rate at the national level of life stress was 66.6% in 2001–2005 and decreased over time.

Conclusion

The large and increasing values of inequalities and the difference in prevalence rates and inequalities in cities highlight the necessity for mental disorder-specific data and for city-level analysis of inequalities. The next steps in reducing inequalities involve deconstructing the health inequalities, and continued monitoring.

Keywords: Mental health, Inequalities, Social determinants of health, Cities, Canada

Introduction

Canada is among the healthiest countries in the world, but not all Canadians enjoy the same high levels of health (Miller and Lu 2019). In recent years, we have seen mounting research highlighting disparities in health, and we have seen much less attention paid to mental health inequalities. This paper explores income-related mental health inequalities in Canada’s largest cities, that is, cities that are classified as census metropolitan areas (CMAs) (Statistics Canada 2016). This paper aims to provide key descriptive findings and provide a foundation for more in-depth city-level research on mental health inequalities and their determinants.

Health inequalities describe how some populations, due to their biological or socio-demographic characteristics, experience better health than others (World Health Organization 2013). When health inequalities are shown to be unjust, avoidable, and unfair, it is also an important indicator of health inequity (Asada 2006). Socio-economic status (SES), often measured using income, is an important determinant of health inequalities that exists in an intersectional relationship with other determinants, such as gender, ethnicity, and race (López and Gadsden 2016; Marmot 2002). It is an important place to start when advancing health inequalities research. Research has shown that health inequalities due to income are widespread in Canada and that poorer populations tend to have worse health outcomes than richer populations (Public Health Agency of Canada 2018).

Health inequalities are also shaped in important ways by where people live. The growing majority of Canadians live in cities, and public health in Canada is organized and administered at the local level. Studies have found that the mental health of populations who live in urban and rural areas is not the same (Kulig and Williams 2011; Wiens et al. 2017). Due to a variety of factors, including differences in age, working opportunities, access to healthcare, and demographics (Kulig and Williams 2011), the lived experience of rural and urban populations differs to an extent that could obfuscate comparisons of geographies that include both populations. These findings point to the need to explore health inequalities at the city level independently.

Particularly in urban settings, the social and material aspects that make up an individual’s neighbourhood impact an individual’s mental health (Gariepy et al. 2014; Generaal et al. 2019). That is, the health of people who share a neighbourhood is affected by similar social features. For this reason, we operationalize income using an area-based measure of income that uses neighbourhoods, rather than individuals, as units of analysis when comparing health outcomes (Denny and Davidson 2012). We reserve the study of individual-level income on mental health to future analysis.

To date, there have been few studies that examine SES-related mental health inequalities in urban Canada (Dunn and Hayes 2000; Lemstra et al. 2006; Steele et al. 2006; Wilson et al. 2004). These studies have used a variety of methods to quantify inequalities that have made a comparison between cities and/or analysis over time difficult. For example, SES can be quantified as government-defined low-income cut off within neighbourhood (Lemstra et al. 2006), an average education level (Steele et al. 2006), and by a combination of neighbourhood income and diversity characteristics (Wilson et al. 2004). This study used neighbourhood income as a proxy for SES.

In this article, we report on mental health inequalities measured by self-reported poor mental health, life stress, mood disorders, and anxiety disorders in cities across Canada between 2001 and 2015. In doing so, it provides answers to the following two questions: “How has the prevalence of self-reported mental health outcomes changed in Canada’s largest cities from 2001 to 2015?” and “How have income-related self-reported mental health inequalities changed from 2001 to 2015?”

Methods

Data

We utilized data from two sources maintained by Statistics Canada and accessed via the Research Data Centre at the University of Saskatchewan. Individual-level data on self-reported health outcomes were collected from the Canadian Community Health Survey (CCHS). The CCHS is a national cross-sectional survey that has been administered since 2000 to a representative sample of Canadians over the age of 12. The CCHS was administered in two-year cycles for 2001–2002, 2003–2004, and 2005–2006, and in single-year cycles from 2007 onward. Briefly, the CCHS covers the population aged 12 and over living in one of the 10 Canadian provinces, excluding those living on reserves and other Aboriginal settlements, full-time members of the Canadian Forces, and residents of institutions. Together, these exclusions make up about 3% of the Canadian population. We also used neighbourhood income data from the 2001, 2006, and 2011 Canadian Censuses of Population (Census). Detailed sampling methods are provided by Statistics Canada (Statistics Canada 2011a).

Data management and processing occurred in a sequential fashion. First, CCHS surveys were combined into three large datasets in five-year increments (2001–2005, 2006–2010, 2011–2015) in order to align with the 2001, 2006, and 2011 Census. Data were then recoded so each of the outcome variables was in dichotomous form. Each respondent was assigned to a dissemination area (DA) using the postal code conversion file (PCCF+) provided by Statistics Canada (Statistics Canada 2017). The PCCF+ corresponding to the 2001, 2006, and 2011 Census were used to reflect the 15-year time period. The CCHS data were then merged with neighbourhood-level income information acquired from the Census.

Population

The main unit of analysis was the DA, which is defined as an area containing approximately 400–700 people (Statistics Canada 2016) and often thought of as a neighbourhood. A DA is defined as urban if it has a population density of above 400 persons per square kilometre and forms part of a population centre that has a total population of more than 1000 people (Statistics Canada 2016). Not all DAs in a CMA are considered urban. If a DA is located within the geographical boundaries of a CMA but does not meet the population density requirements, it can be considered non-urban (Statistics Canada 2016). These DAs were excluded from the analysis. There are many DAs contained in each of the 33 CMAs included in this study. CMAs are defined by Statistics Canada as consisting of one or more municipalities situated around a core that has a population of at least 100,000 people (Statistics Canada 2016). We restricted our analysis to urban residents in CMAs. This left us with approximately 130,000 individuals for each biannually reported CCHS (2001, 2003, 2005), and 65,000 for each annually reported survey (2007–2015). Over the 15-year study period, a total of approximately 962,000 responses were collected; of those, approximately 405,000 lived in urban DAs and were included in this study.

We analyzed DAs to get estimates at the CMA and national levels. We carried out our analysis separately for DAs in eight CMAs that have populations over 750,000 people: Vancouver, Calgary, Edmonton, Winnipeg, Toronto, Ottawa-Gatineau, Montreal, and Quebec. We also combined the urban DAs from each of the 33 urban CMAs in Canada to make up estimates at the national level. That is, the urban DAs in Toronto were analyzed to make the Toronto CMA estimate and were also combined with the DAs from the other 32 CMAs to make up the national estimate.

Mental health variables

Four dichotomous mental health outcome variables were constructed using data in the CCHS: poor mental health, life stress, mood disorder, and anxiety disorder. Respondents were classified as having poor mental health if they, on a five-category scale from excellent to very good, good, fair, and poor, did not report having excellent or very good mental health. Respondents were classified as having life stress if they reported that they, on a five-category scale, found most days quite a bit or extremely stressful. Respondents were classified as having a mood disorder if they reported that they had been diagnosed by a physician as having any mood disorder, including depression, bipolar disorder, mania, or dysthymia, that was expected to last or had already lasted 6 months or more (yes vs no). Respondents were classified as having an anxiety disorder if they reported that they had been diagnosed by a physician as having an anxiety disorder, such as a phobia, obsessive-compulsive disorder, or a panic disorder, that was expected to last or had already lasted 6 months or more (yes vs no).

Neighbourhood income

Health inequalities were examined by stratifying health outcomes by the adjusted average neighbourhood income. The adjusted average neighbourhood income was calculated using the after-tax income variable from the Census. Total household income was adjusted by dividing it by the square root of the number of people living in the household (Statistics Canada 2011b). Neighbourhood income was then calculated as the average adjusted household income in each DA.

Analysis

Prevalence rates and concentration indices were calculated for each CMA and at the national level for each five-year time period. Prevalence rates were calculated by taking the ratio of the number of individuals with a mental health disorder over the total number of people within each DA to approximate the distribution of mental health in neighbourhoods in cities across Canada. Concentration indices (CI) were then calculated to summarize inequalities within these distributions. Both the prevalence rates and concentration indices in cities and at the national level were compared across time periods. Bootstrapping techniques were used to develop robust standard errors. Finally, comparisons using 2-tailed t tests were made between prevalence rates and CI.

The concentration index is a bivariate measure, calculated using the SES and the cumulative health of the population. The measure of SES and health varies based on the research interest; in this paper, variables describing income and mental health were used. The CI is calculated by ordering the population (in this study, DA) by SES, then determining the relative proportion of health attributed to each level of SES. If the cumulative SES of a population is plotted on the x-axis, and the cumulative percentage of health is plotted on the y-axis, in a completely equal situation, a 45° line would be formed, known as the line of perfect equality. The concentration curve is plotted using the actual data. The CI is then calculated to be twice the area between the concentration curve and the line of perfect equality (Wagstaff et al. 1991).

To interpret the CI, we consider both its magnitude and direction. The magnitude of the CI ranges from −1 to 1, with 0 indicating perfect equality and 1 indicating perfect inequality, wherein 100% of the health belongs to the richest or poorest of the population. If the CI is greater than zero, then a greater proportion of health belongs to those with the highest SES; if it is less than zero, the greatest proportion of health belongs to those of lesser SES (O’Donnell et al. 2007). The CI can also be interpreted as the percentage of the health of the population that needs to be transferred from the wealthiest groups in our sample to the poorest in order to reduce the CI to zero (O’Donnell et al. 2007).

The CI is a leading measure of health inequalities widely used by researchers (Wagstaff et al. 1991). Although it is technically more complex than other measures of health inequalities like the disparity rate ratio or disparity rate difference, the CI uses the entire population to determine inequality estimates, rather than fractions of the available data as used by these other measures (Wagstaff et al. 1991). It also has desirable statistical properties that make it more conducive to future more elaborate statistical analysis like multivariate decomposition (Speybroeck et al. 2010)

All data analysis was conducted using Stata 14 software within the Saskatchewan Research Data Centre (SKY-RDC).

Results

Table 1 reports the prevalence rate for each indicator at the national and city levels from 2011 to 2015. At the national level, 6.8% of urban Canadians reported living with anxiety disorder, 7.3% reported living with a mood disorder, 27.9% reported poor mental health, and 65.5% reported living with high levels of life stress.

Table 1.

Change in prevalence rates (%) of mental health indicators over time

Anxiety disorder Mood disorder
2001–2005 Δ 2006–2010 Δ 2011–2015 Δ 2001–2005 Δ 2006–2010 Δ 2011–2015 Δ
  Canada 4.1 ↗ 5.2 ↗ 6.8 ↑ 5.3 ↗ 6.4 ↗ 7.3 ↑
  Vancouver 3.8 ↗ 4.7 6.1 ↑ 5.5 ↗ 6.5 ↗ 7.9 ↑
  Calgary 3.6 4.4 ↗ 5.6 ↑ 5.9 6.7 6.8
  Edmonton 4.0 4.7 7.1 ↑ 6.8 6.7 8.1
  Winnipeg 4.1 ↗ 6.0 ↗ 7.2 ↑ 5.3 ↗ 7.5 8.1 ↑
  Toronto 3.4 ↗ 4.2 5.6 ↑ 4.7 ↗ 5.5 ↗ 6.2 ↑
  Ottawa-Gatineau 5.3 ↗ 7.1 ↗ 8.3 ↑ 6.5 ↗ 8.1 8.9 ↑
  Montreal 4.1 ↗ 5.0 ↗ 5.6 ↑ 4.1 ↗ 4.9 5.0 ↑
  Quebec 3.7 ↗ 5.3 ↗ 6.3 ↑ 3.5 3.9 4.2
Poor mental health Life stress
2001–2005 Δ 2006–2010 Δ 2011–2015 Δ 2001–2005 Δ 2006–2010 Δ 2011–2015 Δ
  Canada 25.8 25.3 ↗ 27.9 ↑ 66.6 ↘ 65.5 65.5 ↓
  Vancouver 29.6 28.4 ↗ 32.2 ↑ 64.6 ↘ 62.7 63.8
  Calgary 23.1 23.6 25.1 68.4 ↘ 65.9 65.4 ↓
  Edmonton 26.6 26.2 27.3 67.5 66.1 66.3
  Winnipeg 25.8 ↗ 28.7 30.6 ↑ 66.2 65.1 63.8
  Toronto 27.1 ↘ 25.2 ↗ 27.8 68.2 67.0 66.4 ↓
  Ottawa-Gatineau 25.3 24.6 ↗ 28.6 ↑ 67.5 66.9 65.0 ↓
  Montreal 23.6 23.9 ↗ 26.0 ↑ 66.2 66.6 68.0 ↑
  Quebec 21.0 19.7 ↗ 22.7 65.3 64.8 62.5 ↓
Open in a new tab

Statistical test compares each city to itself over time; p < 0.05 (two-tailed tests); “↗” indicates a statistically significant increase in prevalence rate between subsequent time periods, “↘” indicates a statistically significant decrease in prevalence rate between subsequent time periods, “↑” indicates a statistically significant increase in prevalence rate between 2001–2005 and 2011–2015, and “↓” indicates a statistically significant decrease in prevalence rate between 2001–2005 and 2011–2015. Rates are percentages. “Δ” indicates change. Data source: Canadian Community Health Survey, 2001 to 2005 and 2011 to 2015, Statistics Canada

Table 1 also shows that the prevalence rates of anxiety disorder, mood disorder, and poor mental health at the national level and in many of the cities have increased by 2011–2015. Though the magnitude and statistical significance of the change varied by city, all of the cities showed an increase in the prevalence rate of anxiety disorder, mood disorder, and poor mental health over the three time periods studied. Interestingly, there was a difference in the magnitude of the change by indicator between time periods. For instance, more cities saw a statistically significant increase in mood disorder from the 2001–2005 to the 2006–2010 time period than from the 2006–2010 to the 2011–2015 time period, though the opposite was true for poor mental health.

In contrast with the prevalence rates of poor mental health, mood disorder, and anxiety disorder, the prevalence rate of life stress did not increase. However, the prevalence rate of high self-reported stress was 65.5%, which was over twice as high as the prevalence rate for poor mental health and nearly 10 times as high as the prevalence rates for mood disorder and anxiety disorder. These differences were similar at the city level. At the national level, and for four of the eight cities, life stress also showed a decrease in prevalence rate from the first time period in 2001–2005 to 2011–2015.

Figure 1 illustrates the graded increase of anxiety disorder, mood disorder, and poor mental health from the richest quintile to the poorest quintile at the national level. For explanatory purposes, it is helpful to present health inequality as a gradient among income quintiles. Quintiles represent the population divided into equal groups after they are ordered from highest to lowest (or lowest to highest) on a certain measure. There are significantly higher levels of anxiety disorder, mood disorder, and poor mental health in DAs categorized as quintile one, the poorest quintile, as compared with DAs of the higher income quintiles. However, differences in health between quintiles were not only between the extremes, Q1 and Q5, but also apparent in the more moderate quintiles. The graded effect suggests that there is a meaningful relationship between all five of the income quintiles and the mental health indicator. That is, the effect of income on mental health is not only apparent between individuals who live in neighbourhoods with extreme wealth or poverty. Notably, the effect of income quintiles on mental health is not seen when examining life stress.

Fig. 1.

Fig. 1

Open in a new tab

Prevalence in mental health measures by income quintiles in urban Canada, 2011 to 2015. Data source: Canadian Community Health Survey, 2011 to 2015, Statistics Canada

The CI measures that are seen in Table 2 are consistent with the results of Fig. 1. Table 2 reports the CI at the national and city levels for each mental health indicator in 2001–2005, 2006–2010, and 2011–2015. In the most recent time period, the concentration index was −0.353 for anxiety disorder, −0.340 for mood disorder, −0.219 for poor mental health, and −0.063 for life stress. In other words, in order for the CI describing anxiety disorders to be zero, not indicating income-based inequalities, 35.3% of the poor health related to anxiety disorder, 34.0% of the poor health related to mood disorders, 21.9% of the poor health related to poor mental health, and 6.3% of the poor health related to life stress must be shifted from the poor half to the rich half of the population. The magnitude of the CI can be observed to be greatest in anxiety disorder and mood disorder, which had the lowest prevalence rates (Table 1) and the lowest in life stress.

Table 2.

Change in concentration index of mental health indicators over time

Anxiety disorder Mood disorder
2001–2005 Δ 2006–2010 Δ 2011–2015 Δ 2001–2005 Δ 2006–2010 Δ 2011–2015 Δ
  Canada −0.24 ↗ −0.26 ↗ −0.35 ↑ −0.23 ↗ −0.26 ↗ −0.34 ↑
  Vancouver −0.26 −0.25 ↗ −0.31 ↑ −0.24 −0.22 ↗ −0.31 ↑
  Calgary −0.25 ↗ −0.30 −0.35 ↑ −0.26 ↗ −0.32 −0.33 ↑
  Edmonton −0.26 ↗ −0.34 ↗ −0.43 ↑ −0.23 ↗ −0.31 ↗ −0.46 ↑
  Winnipeg −0.18 −0.16 ↗ −0.19 −0.18 ↘ −0.08 ↘ 0.00 ↓
  Toronto −0.33 −0.34 ↗ −0.42 ↑ −0.33 ↗ −0.36 ↗ −0.39 ↑
  Ottawa-Gatineau −0.24 ↗ −0.35 −0.31 ↑ −0.24 ↗ −0.29 ↗ −0.34 ↑
  Montreal −0.30 ↗ −0.34 ↗ −0.44 ↑ −0.29 ↗ −0.34 ↗ −0.41 ↑
  Quebec −0.24 ↗ −0.35 −0.38 ↑ −0.30 ↗ −0.38 −0.36 ↑
Poor mental health Life stress
2001–2005 Δ 2006–2010 Δ 2011–2015 Δ 2001–2005 Δ 2006–2010 Δ 2011–2015 Δ
  Canada −0.12 ↗ −0.14 ↗ −0.22 ↑ −0.02 ↗ −0.03 ↗ −0.06 ↑
  Vancouver −0.10 ↗ −0.16 ↗ −0.18 ↑ −0.02 ↗ −0.03 ↗ −0.08 ↑
  Calgary −0.15 −0.16 −0.23 −0.03 −0.03 −0.08
  Edmonton −0.12 ↗ −0.16 ↗ −0.27 ↑ 0.00 ↗ −0.02 ↗ −0.09 ↑
  Winnipeg −0.09 ↘ −0.05 ↗ −0.16 ↑ −0.01 ↘ 0.02 ↗ −0.06 ↑
  Toronto −0.17 −0.18 −0.23 −0.03 ↗ −0.04 ↗ −0.08 ↑
  Ottawa-Gatineau −0.12 ↗ −0.17 ↗ −0.18 ↑ −0.02 ↗ −0.03 ↗ −0.05 ↑
  Montreal −0.14 −0.19 ↗ −0.25 ↑ −0.02 ↗ −0.04 ↗ −0.04 ↑
  Quebec −0.12 ↗ −0.17 ↗ −0.29 ↑ −0.01 ↗ −0.03 ↗ −0.07 ↑
Open in a new tab

Statistical test compares each city to itself over time; p < 0.05 (two-tailed tests); “↗” indicates a statistically significant increase in concentration index between subsequent time periods, “↘” indicates a statistically significant decrease in concentration index between subsequent time periods, “↑” indicates a statistically significant increase in concentration index between 2001–2005 and 2011–2015, and “↓” indicates a statistically significant decrease in concentration index between 2001–2005 and 2011–2015. “Δ” indicates change. Data source: Canadian Community Health Survey, 2001 to 2005 and 2011 to 2015, Statistics Canada

The results of the inequality measures found in Table 2 show that inequalities are found at the national and city levels for every indicator. Employing a simplified method for interpreting the CI, the negative value for nearly every value demonstrates that there is a surplus of poor health in the poorer half of the neighbourhoods studied as compared with the richer half of the neighbourhoods.

Table 2 also provides evidence of the increasing inequality found in almost every city. The increasing arrows in Table 2 highlights that the CI is growing further from zero, that is that inequalities are widening over time in all mental health indicators, including life stress. In terms of the CI, this indicates that a lesser portion of the poor mental health of the population is owned by the richer half of the population in 2011–2015 as compared with in 2001–2006.

Figure 2 illustrates the change in prevalence of each mental health indicator by income quintile over time. The positive slope of the lines in the anxiety disorder, mood disorder, and poor mental health graphs indicates that the prevalence of each of the disorders is increasing over time. Looking from left to right on each of the four graphs, the increased separation in the lines representing the quintiles shows increased inequality.

Fig. 2.

Fig. 2

Open in a new tab

Mental health prevalence by income quintile over time. Data source: Canadian Community Health Survey, 2001 to 2005 and 2011 to 2015, Statistics Canada

Discussion

These results confirm and expand on what is already known about health inequalities in Canada, namely that mental health outcomes are significantly worse for urban Canadians living with less income, and that these inequalities are increasing over time. This study also highlights how widespread mental health inequalities are in Canada; however, these are not the same in every city. More research is needed to understand these city-level differences. When considering anxiety disorder, the CI index of the most equal cities is interpreted to show that 19.2% of the poor health related to anxiety disorders transferred from the poor half to the rich half of the population to reach a neutral CI, while in the most unequal cities, this number reaches over 43%.

In contrast to previous literature based on hospitalization data (Public Health Agency of Canada 2016), the prevalence rates in mood disorder, anxiety disorder, and poor mental health increased over time. This study found a statistically significant increase in poor mental health, mood disorder, and anxiety disorder with national-level prevalence rates of mood disorder increasing from 4.1% to 6.3% and anxiety disorder increasing from 5.3% to 7.3% (as seen in Table 2). However, reports from the Public Health Agency of Canada found no change with prevalence rates of mood and anxiety disorders combined ranging from 9.4% to 10.5% between 1996 and 2010 (Public Health Agency of Canada 2016). The explanations for this discrepancy include this study’s use of self-reported data and exclusion of rural populations. The CCHS phrases questions concerning mood and anxiety disorders to include only physician-diagnosed disorders. The use of these measures introduces the possibility that the increased prevalence in these mental health disorders could be due to shifting diagnosis criteria (Surís et al. 2016). Combined with reduced stigmas surrounding mental health resulting in increased help-seeking behaviours and mental health diagnoses (Chiu et al. 2020), it is possible that the increased prevalence found in this paper reflects an increase in self-reporting of mental health disorders that does not correspond to a clinical increase in mental health disorders. Additionally, previous research found the prevalence of poor mental health to be lower in rural areas than in urban areas (Wiens et al. 2017). As PHAC’s report used national data that contained both rural and urban populations, it is possible that this study’s exclusion of rural results that typically contain lower levels of poor mental health elucidated an urban-specific trend in which the incidence of poor mental health was concentrated, producing higher prevalence rates.

Absences of inequality in life stress

The measure of life stress did not follow the same patterns as those of poor mental health, anxiety disorder, or mood disorder. While large inequalities were seen in poor mental health, anxiety disorder, and mood disorder, the level of inequality in life stress was noticeably lower. There are two theories as to why this variable did not follow the expected pattern. First, the shifting social stigma surrounding mental health does not encompass the societal notions of life stress (Welte and Russell 1993), in part due to socially desirable notions of stressful lives in higher SES as it is associated with business, productivity, and therefore wealth (Bellezza et al. 2017). Second, stress can be subdivided into categories of eustress (good) and distress (bad) depending on how an individual perceives the situation, that is, not all types of stress are equated with poor mental health (Branson et al. 2019). Individuals with higher SES may experience and report higher levels of stress with less impact on either their social status or their mental well-being.

Limitations

This study used cross-sectional self-reported data, specifically single-question variables, to determine the presence of mental health disorders. This is an efficient method, but lacks the validity of other, multi-question measures (Wittchen 1994). Of note, the variables selected as indicators of mental health in this study were the most detailed variables that were available consistently from Statistics Canada from 2001 to 2015. More in-depth indicators were available in the CCHS mental health special topic surveys in 2002 and 2012; however, these were not used in this analysis due to a lack of similar data available consistently for all study years. Social desirability is a well-known limitation when using self-reported data with sensitive subjects such as mental health disorders (Latkin et al. 2017). The use of a blunt measure such as the single-question self-reported data could have caused an underestimation of the prevalence of the mental health indicators. Additionally, the presence of mood disorder and anxiety disorder was determined based on reported physician diagnosis. It is possible that the change in prevalence rates of mood disorder and anxiety disorder is due to changes in diagnostic criteria and shifting social stigma that leads to increased help-seeking behaviour and physician diagnosis. A change in prevalence of mental health indicators due to any of these factors that are concentrated in poorer or richer populations would drive or minimize inequalities. However, this information was not available with the current data.

This study also used an area-based measure for income and health outcomes. The benefits and drawbacks of using individual vs area-based methods have sparked much discussion (Denny and Davidson 2012). Area-based measures are able to better capture the environmental contributions or collective experiences that can be beneficial or harmful to mental health. However, using area-based measures also introduces the possibility of committing an ecological fallacy—it would be inadvisable to apply the results of this study at an individual level.

Finally, this is a broad overview of mental health that does not take into account potential confounding factors such as age and sex. These should be addressed in future work discussing the determinants of inequalities.

Implications

This paper compared the prevalence rates and mental health inequalities in Canada’s cities over three time periods in order to highlight the severity of this issue and to act as a foundation for future work in deconstructing and then reducing inequalities. Growing mental health inequalities will have large emotional and economic implications for cities if not addressed (Mental Health Commission of Canada 2017). As it is unethical to decrease the health of one population to achieve equality, mental health-specific interventions targeting areas of low socio-economic status in order to improve those populations’ mental health need to be implemented in cities across Canada. There are many systemic factors that need to be addressed that could be underlying growing mental health inequalities. These factors include consequences from shifting political climates, changes in economic environments, and transitioning mental health stigma. Further research into the underlying causes of the inequalities needs to be conducted at the city level in order to determine where inequities exist, and form actionable processes to reduce and eliminate their presence in Canada.

Conclusion

This study serves as evidence of mental health inequalities that emphasize the need for continued city-level surveillance of these indicators and for further work in deconstructing the determinants of mental health inequality in urban Canada.

Acknowledgements

This research was conducted at The Saskatchewan Research Data Centre which is part of the Canadian Research Data Centre Network (CRDCN). This service is provided through the support of the University of Saskatchewan, the Province of Saskatchewan, the Canadian Foundation for Innovation, the Canadian Institutes of Health Research, the Social Sciences and Humanities Research Council, and Statistics Canada. The authors would also like to specially thank our RDC analyst Ruben Mercado for his help and support and Dr. Anne Leis for her help in completing the French translation of the abstract. We also thank our anonymous reviewers for their helpful comments and suggestions.

Author contributions

SM was involved in all aspects of the study. CN contributed study conception and design. CP advised on study methods, analysis, and interpretation of data. PP, NM, and CN were involved in planning and supervised the work. All authors provided critical feedback and helped shape the research, analysis and manuscript. All authors approved the final version for publication.

Funding

This research was funded in part by the Urban Public Health Network. No other financial support was received for the research, authorship, and/or publication of this article.

Declarations

Ethics approval

Ethics approval was exempted by the University of Saskatchewan Behavioural Research Ethics Board.

Conflict of interest

The authors declare no competing interests.

Disclaimer

All views expressed in this work are our own.

Footnotes

Publisher’s note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

References

  1. Asada Y. Is health inequality across individuals of moral concern? Health Care Analysis: Journal of Health Philosophy and Policy. 2006;14(1):25–36. doi: 10.1007/s10728-006-0008-6. [DOI] [PubMed] [Google Scholar]
  2. Bellezza S, Paharia N, Keinan A. Conspicuous consumption of time: when busyness and lack of leisure time become a status symbol. The Journal of Consumer Research. 2017;44(1):118–138. [Google Scholar]
  3. Branson V, Palmer E, Dry MJ, Turnbull D. A holistic understanding of the effect of stress on adolescent well-being: a conditional process analysis. Stress and Health: Journal of the International Society for the Investigation of Stress. 2019;35(5):626–641. doi: 10.1002/smi.2896. [DOI] [PubMed] [Google Scholar]
  4. Chiu M, Amartey A, Wang X, Vigod S, Kurdyak P. Trends in objectively measured and perceived mental health and use of mental health services: a population-based study in Ontario, 2002-2014. Canadian Medical Association Journal. 2020;192(13):E329–E337. doi: 10.1503/cmaj.190603. [DOI] [PMC free article] [PubMed] [Google Scholar]
  5. Denny K, Davidson MJ. Area-based socio-economic measures as tools for health disparities research, policy and planning. Canadian Journal of Public Health. 2012;103(8):4–6. doi: 10.1007/BF03403822. [DOI] [PMC free article] [PubMed] [Google Scholar]
  6. Dunn JR, Hayes MV. Social inequality, population health, and housing: a study of two Vancouver neighborhoods. Social Science & Medicine. 2000;51(4):563–587. doi: 10.1016/S0277-9536(99)00496-7. [DOI] [PubMed] [Google Scholar]
  7. Gariepy G, Blair A, Kestens Y, Schmitz N. Neighbourhood characteristics and 10-year risk of depression in Canadian adults with and without a chronic illness. Health & Place. 2014;30:279–286. doi: 10.1016/j.healthplace.2014.10.009. [DOI] [PubMed] [Google Scholar]
  8. Generaal E, Timmermans EJ, Dekkers JEC, Smit JH, Penninx BWJH. Not urbanization level but socioeconomic, physical and social neighbourhood characteristics are associated with presence and severity of depressive and anxiety disorders. Psychological Medicine. 2019;49(1):149–161. doi: 10.1017/S0033291718000612. [DOI] [PMC free article] [PubMed] [Google Scholar]
  9. Kulig JC, Williams AM. Health in Rural Canada. Vancouver: UBC Press; 2011. [Google Scholar]
  10. Latkin CA, Edwards C, Davey-Rothwell MA, Tobin KE. The relationship between social desirability bias and self-reports of health, substance use, and social network factors among urban substance users in Baltimore, Maryland. Addictive Behaviors. 2017;73:133–136. doi: 10.1016/j.addbeh.2017.05.005. [DOI] [PMC free article] [PubMed] [Google Scholar]
  11. Lemstra M, Neudorf C, Opondo J. Health disparity by neighbourhood income. Canadian Journal of Public Health. 2006;97(6):435–439. doi: 10.1007/BF03405223. [DOI] [PMC free article] [PubMed] [Google Scholar]
  12. López, N., & V.L. Gadsden. (2016). Health Inequities, Social Determinants, and Intersectionality. Discussion Paper, National Academy of Medicine, Washington, DC.
  13. Marmot M. The influence of income on health: views of an epidemiologist. Health Affairs. 2002;21(2):31–46. doi: 10.1377/hlthaff.21.2.31. [DOI] [PubMed] [Google Scholar]
  14. Mental Health Commission of Canada. (2017). Strengthening the case for investing in Canada’s mental health system: economic considerations. Ottawa: Mental Health Commission of Canada.
  15. Miller, L.J., & Lu, W. (2019). These are the world’s healthiest nations. Bloomberg News.
  16. O’Donnell O, van Doorslaer E, Wagstaff A, Lindelow M. Analyzing health equity using household survey data. Washington: The World Bank; 2007. [Google Scholar]
  17. Public Health Agency of Canada. (2016). Report from the Canadian Chronic Disease Surveillance System: mood and anxiety disorders in Canada, 2016. Ottawa: Public Health Agency of Canada.
  18. Public Health Agency of Canada . Key health inequalities in Canada. Edmonton: Public Health Agency of Canada; 2018. [Google Scholar]
  19. Speybroeck N, Konings P, Lynch J, Harper S, Berkvens D, Lorant V, Geckova A, Hosseinpoor AR. Decomposing socioeconomic health inequalities. International Journal of Public Health. 2010;55(Issue 4):347–351. doi: 10.1007/s00038-009-0105-z. [DOI] [PubMed] [Google Scholar]
  20. Statistics Canada. (2011a). Canadian Community Health Survey - Mental Health (CCHS).  Ottawa: Statistics Canada.
  21. Statistics Canada . Low income measures. Ottawa: Statistics Canada; 2011. [Google Scholar]
  22. Statistics Canada. (2016). Census Dictionary. Ottawa: Statistics Canada. 
  23. Statistics Canada. (2017). Postal code OM conversion file. Ottawa: Statistics Canada.
  24. Steele LS, Glazier RH, Lin E. Inequity in mental health care under Canadian universal health coverage. Psychiatric Services. 2006;57(3):317–324. doi: 10.1176/appi.ps.57.3.317. [DOI] [PubMed] [Google Scholar]
  25. Surís A, Holliday R, North CS. The evolution of the classification of psychiatric disorders. Behavioral Science. 2016;6(1):5. doi: 10.3390/bs6010005. [DOI] [PMC free article] [PubMed] [Google Scholar]
  26. Wagstaff A, Paci P, van Doorslaer E. On the measurement of inequalities in health. Social Science & Medicine. 1991;33(5):545–557. doi: 10.1016/0277-9536(91)90212-U. [DOI] [PubMed] [Google Scholar]
  27. Welte JW, Russell M. Influence of socially desirable responding in a study of stress and substance abuse. Alcoholism, Clinical and Experimental Research. 1993;17(4):758–761. doi: 10.1111/j.1530-0277.1993.tb00836.x. [DOI] [PubMed] [Google Scholar]
  28. Wiens K, Williams JVA, Lavorato DH, Bulloch AGM, Patten SB. the prevalence of major depressive episodes is higher in urban regions of Canada. Canadian Journal of Psychiatry. 2017;62(1):57–61. doi: 10.1177/0706743716659246. [DOI] [PMC free article] [PubMed] [Google Scholar]
  29. Wilson K, Elliott S, Law M, Eyles J, Jerrett M, Keller-Olaman S. Linking perceptions of neighbourhood to health in Hamilton, Canada. Journal of Epidemiology and Community Health. 2004;58(3):192–198. doi: 10.1136/jech.2003.014308. [DOI] [PMC free article] [PubMed] [Google Scholar]
  30. Wittchen HU. Reliability and validity studies of the WHO--Composite International Diagnostic Interview (CIDI): a critical review. Journal of Psychiatric Research. 1994;28(1):57–84. doi: 10.1016/0022-3956(94)90036-1. [DOI] [PubMed] [Google Scholar]
  31. World Health Organization. (2013). Key concepts. In Social Determinants of Health. Geneva: World Health Organization.

Articles from Canadian Journal of Public Health = Revue Canadienne de Santé Publique are provided here courtesy of Springer

RESOURCES