Chapter 6: The Health of Canada's Young People: a mental health focus – Neighbourhoods

Neighbourhoods

by William Pickett, Ian Janssen, and Andrei Rosu

What is a neighbourhood?

A neighbourhood is a community within a larger city, town or other geographic area. The concept of a neighbourhood however, is not just a physical or geographic construct. Neighbourhoods are also social communities, where there is opportunity for people to come together on a regular basis. They therefore consist of settings and situations where people interact socially for the mutual benefit of all that are part of that community (Bernard et al., 1997; World Health Organization, 1998).

For the purposes of this report, HBSC has defined a neighbourhood as the local setting where individual students live and go to school. We were interested in exploring the possible effects of the physical, social and economic characteristics of these neighbourhoods on various aspects of mental health in populations of young Canadians. We did this using data available at various levels, including individual student questionnaires, the HBSC administrator's questionnaire, and linked information from geographic information systems.

Why are neighbourhoods important to health?

During the recent past, there has been an increasing interest in the effects of neighbourhoods on the health of populations. Researchers and policy-makers have come to understand that the characteristics of neighbourhoods, whether measured in terms of their physical, social or economic attributes, can have important impacts on people. Neighbourhoods represent key environmental settings for youth, as policies, physical spaces and structures, and cultural, social and interpersonal interactions that occur in these environments influence their subsequent behaviour, and ultimately their health experiences (Bernard et al., 2007; Green et al., 1996; Sallis & Owen, 2002). Good examples of this include factors such as the availability and quality of affordable housing (Krieger & Higgins, 2002), the extent of poverty or socio-economic advantage (Lynch et al., 2002), safety concerns related to the presence of crime or gangs (Kee et al., 2003; Wood, 2003), or the sense of warmth and cohesion in a well organized and socially-connected neighbourhood (Gidlow et al., 2010). Each of these can contribute to states of disease or wellness.

There are many child health behaviours and outcomes that are affected by neighbourhood characteristics. Examples include dietary patterns and physical activity (Sallis & Glantz, 2009), injury (Oliver & Kohen, 2009), and violence (Bell et al., 2009), as well as more global indicators of health status including mortality (Lemstra et al., 2006). Neighbourhoods can also affect various mental health indicators at the population level, which is the focus of this report.

Physical, social and economic factors are now considered to be classic "determinants of health" at the population level. Such determinants are worthy of study as independent contributors to the health of communities, or as factors that are likely to interact to produce various states of health and disease in affected populations (Public Health Agency of Canada, 2001).

Possible effects of neighbourhoods on mental health

Strong and cohesive neighbourhoods were hypothesized to be linked with better mental health outcomes. The latter can be measured in terms of specific and global measures of mental health status, including emotional problems and emotional well‑being, and behaviour problems. A cohesive neighbourhood with strong social support may provide tangible and intangible benefits to its residents, not the least of which is an improved sense of emotional well‑being. A dangerous or highly negative neighbourhood environment can be very detrimental to health, and result in several different types of health problems, and even mortality.

Geographic information

New in 2010, school addresses for each of the participating HBSC schools were "geo-coded" and put into a geographic information system. Using this technology, measures were taken of the physical, social and economic environments in a one kilometre circular area or buffer that surrounded each HBSC school. While researchers have used buffers of varying sizes and dimensions (e.g., circular, or an area that can be travelled in a fixed time period) to characterize school neighbourhoods and their characteristics, observations made using a one to five km circular buffer have been found to be reliable for constructs such as neighbourhood socio-economic status. (Simpson et al., 2005) Neighbourhood measures taken within these buffers were either abstracted from electronic records from the 2006 Canada Census of Population (Statistics Canada, 2006), or from ArcGIS version 9.3 geographic information systems (GIS) software and the CanMap GIS database (DMTI Spatial Inc., Markham, ON). Additional neighbourhood measures were taken from the 2010 HBSC School Administrator Survey.

What are we reporting in this chapter?

In this chapter, we report on physical, social and economic factors that characterize the neighbourhoods surrounding the schools that participated in the 2010 HBSC survey. We provide summaries of these characteristics for schools that serve different grade levels (Grade 6 to 8 and lower, Grade 9 and higher, and mixed – schools that cover a larger range of grade levels). While there are examples of mixed schools in all provinces and territories, and one can find them located in both highly urban and highly rural/remote communities, they do tend to be unique. For example, 59% of the sampled schools in the three Northern Territories were mixed schools, compared with 21% of the schools sampled within the eight participating provinces. Approximately 38% of the schools sampled from rural and remote communities were mixed schools, compared with 19% of schools from suburban and 23% of schools from urban areas.

Physical factors that were measured included characteristics of housing, traffic, and land use, and accessibility to parks and recreational facilities in the 1 kilometre neighbourhood surrounding each participating school. Some data on physical factors were collected from school administrators in the 2010 HBSC survey. Other data were obtained using the geographic information systems.

Social factors included measures of the social climates of school neighbourhoods. These were reported by school administrators, who summarized their perceptions of crime, racial and religious tensions, the presence of gangs, the aesthetics of the buildings, streets, roads and land, and the presence of drugs and alcohol in the neighbourhoods that surrounded their schools.

Economic factors that were measured in school neighbourhoods included standard indicators of formal education, total household income levels, and housing ownership. These were measured using 2006 data from the Canada Census of Population (Statistics Canada, 2006).

We also examined neighbourhood characteristics in relation to the four indicators of mental health: (1) emotional well‑being; (2) prosocial behaviours; (3) emotional problems; and (4) behavioural problems.

Physical characteristics of school neighbourhoods

Housing and land use

6.1 Leading types of housing in the 1 km buffer surrounding Canadian schools, by school type (%)

[Click to enlarge Figure 6.1]

Figure 6.1 shows the types of housing that characterize the neighbourhoods surrounding the 436 Canadian schools that participated in the 2010 HBSC survey. These are divided into the types of schools that serve Grade 6 to 8 (or lower), Grade 9 to 10 (and higher), and "mixed grade" student populations. Single detached houses and various types of apartment buildings were the leading types of housing for each of the three categories of schools. Mixed schools had a different mix of housing types. These were consistent with housing types in rural and Northern centres in the three Territories, as well as communities in parts of some provinces (e.g., Quebec, Alberta) that employ mixed grade school systems.

6.2 Leading types of land use mix in the 1 km buffer surrounding Canadian schools, by school type (%)

[Click to enlarge Figure 6.2]

The majority of participating schools were situated in neighbourhoods with a mixture of different types of land use. Fewer schools were situated in neighbourhoods that were exclusively residential (Figure 6.2).

Large proportions of young people are exposed to non-residential neighbourhood environments when travelling to or from school, or during recess and other school breaks. This can have both positive and negative effects on health. Non-residential environments may expose students to: physical hazards that endanger their safety, access to businesses such as fast food outlets that may affect poor nutritional choices, and vehicular traffic that limits the opportunity for active transportation to school (Sallis & Glantz, 2009). On the other hand, residential-commercial land use mixes can facilitate participation in physical activity, in that young people can walk to local stores and food outlets. Hence, land use can have several different influences on both the health and safety of young people, and is an important neighbourhood characteristic.

6.3 Vacant or shabby houses and buildings are a problem in the neighbourhood where the school is located, by school type (%)

[Click to enlarge Figure 6.3]

Vacant or shabby housing in the school neighbourhood was perceived by school administrators to be at least a minor problem in 29% of the elementary schools, 22% of high schools, and 40% of the schools serving mixed grades (Figure 6.3). The peak surrounding mixed schools is important, as it reflects lower socio-economic conditions surrounding those school communities that are required to combine school grades. This is consistent with the more challenging economic conditions in many Northern and remote communities.

Traffic

6.4 Heavy traffic is a problem in the neighbourhood where the school is located, by school type (%)

[Click to enlarge Figure 6.4]

Approximately one-half of the school administrators that served Grade 6 to 8 students reported heavy traffic as a problem, with 14% reporting heavy traffic to be a moderate or major problem (Figure 6.4). This percentage rose to 24% in high schools (Grade 9-10), and was 16% in schools that served mixed grade levels. The presence of heavy traffic may be a factor that influences the health of young people. First, heavy traffic is a cause of child pedestrian injury (Oliver & Kohen, 2009). Second, due to such threats to safety, heavy traffic may influence parental and student decisions to walk or bicycle to school, or to engage in physical activity in school neighbourhoods.

Parks and recreational facilities

6.5 Parks located in the 1 km buffer surrounding Canadian schools, by school type (%)

[Click to enlarge Figure 6.5]

A strong majority of the schools serving each of the different grade levels were located within 1 kilometre of at least one public park (Figure 6.5). During favorable weather conditions, this distance is considered to be accessible by most young people in ten minutes or less. Access to public parks offers one potential opportunity for young people to engage in physical activity, although safety concerns may limit their use in some Canadian settings (Nichol et al., 2010).

6.6 Recreational facilities located in the 1 km buffer surrounding Canadian schools, by school type (%)

[Click to enlarge Figure 6.6]

There was more limited access to public recreational facilities, with 38% (Grade 9-10) to 51% (Grade 6-8) of schools reporting no such facilities within the 1 kilometre circular buffer (Figure 6.6).

Social characteristics of school neighbourhoods

6.7 Tensions based on racial, ethnic, or religious differences are a problem in the neighbourhood where school is located, by school type (%)

[Click to enlarge Figure 6.7]

6.8 Tensions based on racial, ethnic, or religious differences are a problem in the neighbourhood where school is located, by community type (%)

[Click to enlarge Figure 6.8]

The HBSC School Administrator survey contained a series of items used to describe the social climate of neighbourhoods that surround participating Canadian schools. Perceived tensions surrounding racial, ethnic, or religious differences were identified as at least a "minor problem" in about 30% of the schools (Figure 6.7). The proportions of school administrators who viewed these tensions as at least "moderate" problems did not vary strikingly by school type. Schools in rural centres and large urban centres experienced the most problems of this type (Figure 6.8).

6.9 Garbage, litter, or broken glass in the street or road, on the sidewalks, or in yards are a problem in the neighbourhood where school is located, by school type (%)

[Click to enlarge Figure 6.9]

6.10 Garbage, litter, or broken glass in the street or road, on the sidewalks, or in yards are a problem in the neighbourhood where school is located, by community type (%)

[Click to enlarge Figure 6.10]

The physical appearance of streets, roads and sidewalks is a common social indicator used to describe the quality of a school neighbourhood. The majority of school administrators reported such aesthetics to be at least a minor problem (Figure 6.9). However, such factors were considered to be more of a problem in high schools and schools serving mixed grades, as well as in schools located in rural communities (Figure 6.10). This may be attributed to socio-economic challenges in some of these more remote settings.

6.11 Selling or using drugs or excessive drinking are problems in the neighbourhood where school is located, by school type (%)

[Click to enlarge Figure 6.11]

6.12 Selling or using drugs or excessive drinking are problems in the neighbourhood where school is located, by community type (%)

[Click to enlarge Figure 6.12]

Illicit drug use and excessive drinking were identified as major problems for the neighbourhoods surrounding 6% of the high schools, and 8% of the schools that served mixed grade levels (Figure 6.11). Administrators in about 20% of the schools serving Grade 6 to 8 reported that selling or using drugs or excessive drinking were problems in their immediate neighbourhoods. These were only identified as major problems in the neighbourhoods of schools in rural and large urban centres (Figure 6.12).

6.13 Gangs are a problem in the neighbourhood where school is located, by school type (%)

[Click to enlarge Figure 6.13]

6.14 Gangs are a problem in the neighbourhood where school is located, by community type (%)

[Click to enlarge Figure 6.14]

Gangs were not perceived to be a problem in the neighbourhoods surrounding most participating schools, with less than 10% reporting gangs as a moderate or major problem (Figure 6.13). Gangs were reported to be a major problem in 5% of the participating high schools. These problems were most notable for school neighbourhoods in large urban centres (Figure 6.14).

6.15 Crime is a problem in the neighbourhood where school is located, by school type (%)

[Click to enlarge Figure 6.15]

Larger percentages of school administrators reported neighbourhood crime as a problem, with the highest prevalence levels indicated for high schools and schools serving mixed grade levels (Figure 6.15). The fact that the percentages of schools where crime was reported as a problem exceeded reports of gangs, suggests that much of the perceived crime is not related to gang activity.

Economic characteristics of school neighbourhoods

6.16 Education of population aged 15 and over in the 1 km buffer surrounding Canadian schools, by school type (%)

[Click to enlarge Figure 6.16]

Levels of formal education provide one standard indicator used to describe socioeconomic status and how it varies between populations. Distributions of observed education levels were roughly equivalent in the neighbourhoods of schools defined by school types (Figure 6.16). Roughly one-quarter of the population in school neighbourhoods had achieved each of four categories of education: university, college or an apprenticeship or trade, high school, or less than high school education.

Similarly, no substantial differences in the levels of total household income were observed in the three different school types (Table 6.1). The most common level of household income was less than $40,000.

6.17 Ownership arrangements for housing in the 1 km buffer surrounding Canadian schools, by school type (%)

[Click to enlarge Figure 6.17]

The housing observed in the neighbourhoods of participating schools was primarily owned or rented, with lower observed percentages of ownership surrounding the schools serving mixed grade levels (Figure 6.17).

6.18 Family structure of population in the 1 km buffer surrounding Canadian schools, by school type (%)

[Click to enlarge Figure 6.18]

Finally, the percentage of single parent families is often used as a standard indicator of lower socio-economic status for neighbourhoods and communities. Approximately one in five families in the school neighbourhoods reported such family structures (Figure 6.18).

Relationships between neighbourhood characteristics and mental health

It is plausible that young people from advantaged and disadvantaged school neighbourhoods may also report different types of health outcomes, both positive and negative. A neighbourhood that is well planned, socially cohesive, and economically affluent would be expected to contribute to better health. This argument might extend to indicators of mental health in populations of young people.

While some evidence of possible neighbourhood effects on emotional health were found, in general the analysis of these associations did not turn up consistent or clear patterns.

For example, Figure 6.19 relates reports of high levels of emotional problems to the housing conditions that surround school neighbourhoods. There is little evidence of a strong association between this measure of neighbourhood disadvantage and the mental health of young people, although a modest association was observed among girls, who generally report higher levels of emotional problems. Behavioural problems did not appear to increase with increasing levels of perceived housing problems among either gender (Figure 6.20).

6.19 Students reporting high levels of emotional problems, by the presence of vacant and shabby housing in their neighbourhood, by gender (%) ¹

[Click to enlarge Figure 6.19]

6.20 Students reporting high levels of behavioural problems, by the presence of vacant and shabby housing in their neighbourhood, by gender (%)

[Click to enlarge Figure 6.20]

6.21 Students reporting high levels of emotional well‑being, by the presence of recreational facilities in their neighbourhood, by gender (%)

[Click to enlarge Figure 6.21]

6.22 Students reporting high levels of emotional well‑being, by the presence of parks in their neighbourhood, by gender (%)

[Click to enlarge Figure 6.22]

Similar regression analyses were used to examine positive outcomes such as emotional well‑being. There was a similar lack of consistency in these associations. For example, there did not appear to be an association between access to recreational facilities and emotional well‑being (Figure 6.21). Increased access to parks was inconsistently related to higher levels of emotional well‑being (Figure 6.22).

6.23 Students reporting high levels of behavioural problems, by the presence of racial, ethnic or religious tensions in their neighbourhood, by gender (%)

[Click to enlarge Figure 6.23]

There was some evidence that relationships between neighbourhood factors and mental health can occur in the expected direction of more neighbourhood problems coinciding with more mental health problems. Figure 6.23 shows that neighbourhoods with high reported levels of social tension also contained larger numbers of students with behavioural problems. While these relationships are not necessarily causal, it is clear that suboptimal social environments and problem behaviours do coexist, and this is very pronounced in some Canadian school communities.

What young people thought about these findings

Based upon the youth engagement workshop, young people clearly recognize the importance of the environments that surround them as possible determinants of health. They understand that many positive and negative health behaviours are developed within the context of local neighbourhood settings. They understand that safe and socially cohesive environments may contribute to healthy choices and behaviours, while neighbourhoods that endanger their safety can have the opposite effect, and put the health of young people at risk.

During the workshop, the young people and adults in attendance (mostly researchers and government representatives) were given four choices of environments to select from as being "most influential to their ongoing health". These four choices were home, school, peers and neighbourhood and would provide priorities for policy development at the government level. The students were asked to stand by a pillar that represented their decision.

At the start of the exercise, everyone in the room seemed to acknowledge that all settings were important for government action. As things continued, everyone was urged to make a single choice. A few students and adults clustered around the "school environment" and "peer" pillars. Quite a number of adults but only one student selected "neighbourhoods" as their initial choice. A few students remained standing in the middle of the room, and stated that all of these environments were of importance to their health, including mental health outcomes, and they did not wish to make a single choice. But by far the majority of students stood together at the "home" pillar part of the room. The lesson to be taken from this exercise was that, while students recognize the importance of all contextual environments as determinants of health, the home remained, by far, the most influential in their opinion. Neighbourhoods, while an obvious focus of government policy and recent research efforts, remained a lower priority for the students in attendance.

Summary and implications

Commentary

Neighbourhoods can and do have an important influence on the health of populations, including young people. For this reason, it was felt that an exploration of the physical, social and economic environments surrounding Canadian schools would be of value. This chapter therefore profiles the selected characteristics of school neighbourhoods, and subsequently relates specific characteristics to indicators of mental health reported by students. Two major highlights emerged from this exploratory analysis.

First, it is clear that location matters. Schools in highly urban areas have different neighbourhood influences than schools in rural and remote regions of Canada. This is particularly true for social indicators of neighbourhood climate, such as tensions related to racial, ethnic or religious differences, or the presence of gangs or problems with crime.

Second, whether assumed causal or not, it was striking that the physical and economic factors measured around schools appeared to have little influence on student perceptions of their mental health. Emotional problems and behavioural problems did not vary significantly with the perceived quality of local housing. Proximity of schools to recreational facilities was not associated with emotional well‑being, and emotional health actually appeared to decline in neighbourhoods with greater access to park facilities. Such findings were supported by student perceptions reported at the national workshop, where other factors such as the home environment were perceived as being more important determinants of mental health. These determinants are complex, and go far beyond simplistic explanations that attempt to tie neighbourhood factors to specific health outcomes in simple causal models. The linkage of HBSC survey data with population health data from other sources provides tremendous opportunity to further explore these complex models.

References

• Bell, N., Schuurman, N. & Hameed, S.M. (2009). A multilevel analysis of the socio-spatial pattern of assault injuries in greater Vancouver, British Columbia. Canadian Journal of Public Health, 100:73-77.
• Bernard, P., Charafeddine, R., Frohlich, K.L., Daniel, M., Kestens, Y. & Potvin, L. (2007). Health inequalities and place: a theoretical conception of neighbourhood. Social Science & Medicine, 65:1839-1852.
• Gidlow, C., Cochrane, T., Davey, R.C., Smith, G. & Fairburn, J. (2010). Relative importance of physical and social aspects of perceived neighbourhood environment for self-reported health. Preventative Medicine, 51:157-163.
• Green, L.W., Richard, L. & Potvin, L. (1996). Ecological foundations of health promotion. America Journal of Health Promotion, 10:270-281.
• Kee, C., Sim, K., Teoh, J., Tian, C.S. & Ng, K.H. (2003). Individual and familial characteristics of youths involved in street corner gangs in Singapore. Journal of Adolescence, 26:401-412.
• Krieger, J. & Higgins, D.L. (2002). Housing and health: time again for public health action. American Journal of Public Health, 92:758-768.
• Lemstra, M., Neudorf, C. & Opondo, J. (2006). Health disparity by neighbourhood income. Canadian Journal of Public Health, 97:435-439.
• Lynch, J.W., Smith, G.D., Kaplan, G.A. & House, J.S. (2000). Income inequality and mortality: importance to health of individual income, psychosocial environment, or material conditions. British Medical Journal, 320:1200-1204.
• Nichol, M., Janssen, I. & Pickett, W. (2010). Associations between neighborhood safety, availability of recreational facilities, and adolescent physical activity among Canadian youth. Physical Activity & Health, 7:442-450.
• Oliver, L. & Kohen, D. (2009). Neighbourhood income gradients in hospitalisations due to motor vehicle traffic incidents among Canadian children. Injury Prevention, 15:163-169.
• Public Health Agency of Canada. What is the population health approach? Available at: http://www.phac-aspc.gc.ca/ph-sp/approach-approche/appr-fra.php. Accessed April 2011.
• Sallis, J.F. & Glantz, K. (2009). Physical activity and food environments: solutions to the obesity epidemic. Milbank Quarterly, 87:123-154.
• Sallis, J. F. & Owen, N. (2002) Ecological models of health behavior. In K. Glanz, F.M. Lewis, &, B.K. Rimer (Eds.), Health Behaviour and Health Education: Theory, Research, and Practice, 3rd Edition (pp.462-484). San Francisco: Jossey-Bass.
• Simpson, K., Janssen, I., Craig, W.M. & Pickett, W. (2005). Multilevel analysis of associations between socioeconomic status and injury among Canadian adolescents. Epidemiology & Community Health, 59:1072-1077.
• Statistics Canada. (2006). Census of Canada: Profile Data. Ottawa, Canada. Statistics Canada (producer); Tetrad Computer Applications, Vancouver, BC (distributor). Available: Queen's University Social Science Data Centre.
• Statistics Canada. (2006). 2006 Census, Analysis series. Ottawa.
• Wood D. (2003). Effect of child and family poverty on child health in the United States. Pediatrics, 112:707-711.
• World Health Organization. (1998). Social capital. In D. Nutbeam (Ed.), Health Promotion Glossary (p.19). Geneva: WHO, Division of Health Promotion, Education and Communication.

1. 28% of boys who feel that vacant or shabby housing in their neighbourhood are not a problem report relatively high levels of emotional health problems, compared with 55% of girls who feel that vacant or shabby housing in their neighbourhood are not a problem. A full explanation of how to interpret the figures that relate to mental health is provided in Chapter 1.