Our study findings suggest that, in general, public natural space is not strongly associated with positive emotional well-being in young people. The observed associations were weak and lacked consistency. Sub-analyses with green space and blue space and the sensitivity analysis also demonstrated weak associations and inconsistent trends. However, the findings in small cities suggest a small difference in positive emotional well-being based on the context of geographic location, such that youth in small cities may benefit more from natural space. The trend detected with blue space suggests a mechanistic difference between how blue and green spaces influence emotional well-being. Nonetheless, the weak effect estimates render these observations inconclusive.
The pattern seen in the effect estimates detected in quartile 3 suggests that moderate exposure to public natural space in one’s neighborhood environment may be most beneficial. A possible explanation for a lower protective effect in areas of higher exposure (i.e. quartile 4) may be linked to the composition of natural space. Different types of natural space may have differential importance based on quality and relevance to youth. For example, quartile 4 was made up of a greater proportion of wooded areas (see Additional file 2). These types of green space may not be maintained and/or conducive to activities similar to areas such as parks. This suggests that these areas may not be as relevant for emotional well-being among youth, if a relationship exists. Alternatively, the higher effect detected in quartile 3 may suggest a more optimal environmental composition for emotional well-being. For example, these neighborhoods contained more blue space, and thus the detected relationship may be explained by exposure to water.
Several personal factors were modestly associated with emotional well-being in young people. These findings were in general consistent with existing literature. Girls may be at greater risk of poor emotional well-being [8, 31]; positive emotional well-being may decline with age through adolescence ; positive emotional well-being was associated with higher family affluence , higher levels of perceived neighborhood safety , and lower levels of perceived rundown houses in neighborhoods . There were also differences across ethnicities, with positive emotional well-being particularly low among Aboriginal students , although some sample sizes were low for non-Caucasian students. These associations indicate the importance of individual factors on emotional well-being among youth. Further, the intraclass correlation coefficient (2.85%) supports that variation in emotional well-being was more accounted for by the individual level rather than the area level.
Although previous population studies have made conclusions for an association between nature and health, many studies found weak effect estimates, similar in magnitude to those detected in our study. Research from the Netherlands has found varying degrees of associations between green space and high perceived health (β: -0.009; SE: 0.003 ; β: 0.006; SE: 0.001 ), and low prevalence of mental health illnesses (OR range: 0.95-0.98) . Similarly, a study from England showed that those from areas of higher green space reported lower rates of poor health (β: -0.021; p < 0.000; R2 = 0.8398) . A study among Japanese seniors yielded weak odds ratio of 1.13-1.17 for a relationship between green space access and longevity .
Compared to studies that have found relationships between natural space and health, there are a number of possible explanations for the null findings of our study. First, studies, such as Sugiyama et al.’s , concluding strong associations have been observed in adults, and it is possible that the effect of natural space is different in young populations. Possible reasons for differences in effect between youth and adults may lie in variations in perceptions, usage, and interactions with natural space. Youth may not perceive natural space in the same way, and thus they may use and interact with natural space differently . While adults may appreciate natural space for tranquility and reflection, youth may associate natural space more with play and socialization. Further, while adults may dictate their own decisions on where, when, and how to interact with natural space, youth may not have the same independent mobility because parents/guardians’ decisions may influence where they go and where they play [66, 67]. Alternatively, for those youth with independent mobility, natural space may serve as gathering places for delinquent and antisocial behaviors instead of retreats for health promotion .
Second, the relationship between nature and emotional well-being may be context specific. As seen through the stratified analysis, effects were more pronounced for those in small cities. This finding was similar to Maas et al.’s observation that the relationship between green space and disease was strongest in slightly urban areas . Young people may have a different relationship with public natural space based on the composition of their communities. Nature deficit disorder, a condition describing the decreasing use of natural space among youth because of increased electronic media and/or greater safety concerns surrounding young people being outside without supervision, is particularly high among urban youth [68–70]. In contrast, rural areas typically have an over-abundance of natural space, with few youth having low exposure. Therefore, the role of natural space may not be as important as other factors that affect emotional well-being. The lack of associations in rural areas in our study may be reflected by the lack of data for private and agricultural natural space, which account for a large proportion of the natural space in rural areas.
Further, our findings may be attributable to variations in geography, lifestyle, and/or culture that are specific to Canada. To our knowledge, no existing studies have assessed this relationship at a population level in Canada. Richardson et al. suggested that the high abundance and less spatial variation of green space in New Zealand, which is different compared to the Netherlands and England where much of the previous literature is based upon, accounted for the lack of association found between green space and mortality risk in their study . Similarly, there is a different spatial composition in Canada compared to these countries. Although Canada has greater total land size and natural space area, the average natural space within the 5 km buffer measured in this study was lower than those measured in the Netherlands and England [27, 28]. This may indicate that more of the natural space in Canada is located outside of the living neighborhood. The “car culture”  and tendency to drive may encourage Canadians to seek out faraway natural settings, and perhaps natural space close to home may not be as relevant. Additionally, climatic variations may play a role in how Canadians interact with the natural environment differently than those studied in other nations. For instance, patterns of usage of natural space would differ during the summer and winter months, and Canadians may have limited access to natural space due to winter conditions. This may have affected our ability to detect an effect as the HBSC was conducted during the fall, winter, and early spring.
Strengths and limitations
Strengths of this study warrant comment. In a national analysis, we integrated health and spatial data to investigate the relationship between natural space and emotional well-being. This study intentionally focused this relationship in populations of young people. It is also one of the few population studies that have employed GIS techniques to obtain objective measures of natural space. Further, the study is well powered, and the multilevel modeling allowed for examination of effects at the individual and area levels. The analysis investigated and controlled for important covariates in the relationship of interest, which also addressed a methodological gap of previous research.
This study has some shortcomings. First, the cross-sectional design does not allow for confirmation of temporality, and subsequently, causality. Second, the measure of public natural space is limited because no data were available for privately owned natural space such as yards at the home and agricultural land. Therefore, the exposure measurement may be underestimated, particularly in rural areas. As well, this study was not able to consider the quality and usage of the natural space measured, which may be a critical part in the relationship between nature and emotional well-being. Third, the use of the 5 km radius buffer around schools may lead to misclassification of the natural space measures as this was used as a proxy for home neighborhoods. However, the findings in the sensitivity analysis suggest that this was not a major concern because results for those known to live within the buffer were similar to those in the overall study population.
This type of research has the potential to inform the direction of health promotion strategies and urban planning decisions. Firstly, descriptive findings indicate that advocacy for policies and funding devoted to promotion of positive emotional well-being among youth is merited since only over half of young Canadians reported high levels. Exposure to public natural space appeared to have limited importance on positive emotional well-being of young people. There may be differences in this effect based on geographic context worthy of further consideration. Overall, focus should be placed upon more proximal variables such as individual factors and family affluence that showed stronger influences. This information is important to not only understand the health of this population, but may also help in the evaluation of current efforts that focus on emotional well-being among youth. Further, such knowledge may be useful in identifying vulnerable groups that require directed attention. Understanding factors that strongly influence emotional well-being may help to create more effective and specific public health programs and strategies.
Although findings of this study did not indicate a strong association between natural space and emotional well-being in young people, the potential health effects of natural space cannot be dismissed entirely. Natural space may, for instance, have different effects in adult populations, and it may impact other health outcomes such as physical activity that were not assessed in this study.