Discussion

The results here suggest that nature experiences in urban green spaces may be having a considerable impact on population health, and that these benefits could be higher if more people were engaged in nature experiences. Specifically, our results suggest that up to a further 7% of depression cases and 9% of high blood pressure cases could be prevented if all city residents were to visit green spaces at least once a week for an average duration of 30 minutes or more. The societal costs of depression are estimated at AUD$12.6 billion per annum for employed Australians alone34, and the direct costs of hypertension in the United States have been estimated at US$48 billion35. Given that our results show nature experiences, if causal in nature, could simultaneously lead to a suite of health benefits for mental health (depression), physical health (high blood pressure), social health (social cohesion), and a positive health behaviour (physical activity), the cumulative cost savings across all health outcomes could be immense if this behavioural change was targeted.

Our finding that the duration, and frequency of nature interactions are varyingly associated with the four health outcomes has potentially important implications for the design of health interventions, and also reveals new hypotheses that warrant further attention. For example, while provision and quality of green spaces is undoubtedly important, health programs aiming to reduce the prevalence of depression or high blood pressure might also focus on behavioural interventions, for example, promoting longer duration green space visits. In contrast, improved social cohesion in communities is a well-known benefit of public green spaces36,37, and interventions that aim to enhance social cohesion might fruitfully focus on increasing residents’ frequency of visits38. Social cohesion is itself important for public health, as it is positively associated with physical and mental wellbeing39. These flow-on benefits are likely to add considerably to the economic and social value of urban green space.

Here physical activity was associated with both higher duration and frequency of green space visits, which is important given it can reduce the risk of a wide range of non-communicable diseases such as diabetes, cardiovascular disease and obesity40. Green spaces are often considered settings that directly facilitate exercise41, and visiting green spaces can incidentally entail walking, running or cycling. Vegetated areas also offer shade and improved temperature regulation42, providing a pleasant location for physical activity. This is particularly relevant in cities such as Brisbane, a sub-tropical location with hot summers and a mean of 113 cloudless days per year43. However, while many studies have found that more people undertake physical activity (e.g. cycling and walking) in greener neighbourhoods17, the results are sometimes mixed; for example, these patterns could be due to other activities such as gardening44, or because active people self-select into greener neighbourhoods45. While our results add to the body of knowledge on this subject, these varying explanations require further attention.

Our measure of nature intensity (vegetation complexity) showed no association with any of the health outcomes measured. Other studies have found that higher levels of plant, butterfly and bird species richness (or perceived species richness) can enhance a person’s feelings of restoration13,14, and future work might fruitfully explore the effect of such measures within the nature dose framework. There are also other hypotheses describing relationships between health and vegetation complexity; for example, studies have found that more people tend to visit public green spaces with moderate levels of vegetation cover (rather than high or low)46, and vegetation is also likely to influence the perception of safety of an area25. Systematic consideration of nature dose-response relationships will therefore be critical to understanding how to enhance health outcomes from exposure to nature.

We observed significantly fewer cases of depression and high blood pressure in people who spent an average of 30 minutes or more visiting green space in the survey week, and there was some indication that longer duration visits may be associated with an even lower prevalence of depression. However, here we traded-off accuracy in detecting differences across the incremental increases in dose for achieving a high level of representation across the population (i.e. sampling did not target respondents with varying durations of nature exposure). Given that this type of dose-response relationship could contribute further evidence for causality according to Hill’s criterion47, future studies would benefit from achieving relatively even sampling representation across the relevant nature dose levels. An added consideration when interpreting the results outlined here is that the effects of depression itself can influence a person’s activity levels48, and so could reduce the likelihood that a person visits green-space. The same effect could also occur for high blood pressure, where people who have other risk factors such as obesity might also be less likely to visit green spaces (note, BMI and physical activity were considered as covariates here, so these effects are somewhat accounted for). Thus, studies that explore changes over time within individuals and across populations could be a particularly powerful way to further elucidate dose-response relationships between nature and health.

This study used a self-report online survey, an approach which brings a number of benefits (such as the large sample size and a high level of stratification across the population), as well as limitations. For example, recalling events can pose challenges, question order can affect responses, and many other factors can affect how well a person responds to questions49. While we used measures to minimize these limitations, other methods such as longitudinal studies using tracking technologies might provide complementary understanding of nature-dose relationships. Future research exploring the role of a broader range of socio-demographic and community factors related to health outcomes, but which also have the potential to influence interaction with nature (e.g. marital status and crime) will also shed light on the mechanistic pathways linking nature exposure to health.

Nature relatedness, or the differences in the way people view their connection with the natural world, could both drive interactions with nature and enhance wellbeing in its own right50. We found that higher levels of nature relatedness predicted greater feelings of social cohesion and higher levels of physical activity. This supports other research which has found that people with higher nature relatedness scores also often report better wellbeing, happiness and life satisfaction33,51, and lower levels of anxiety52. A limitation of studies so far within this area is that they are often single time-point studies, and research is needed to whether actively altering this trait might influence health and wellbeing.

Interactions with nature simultaneously deliver mental, physical and social health outcomes for a population through multiple pathways22. By harnessing the synergistic potential of these pathways, contact with nature has the potential to lower not just the prevalence of single chronic conditions, but also multiple chronic or acute medical conditions that co-occur within one person. However, here we have also shown that the different components of experiences of nature (the frequency, duration or intensity) variously influence the health outcomes. This has important implications for the design of health interventions targeting improvements in the four health domains examined here. Ongoing efforts to unpack the nature-health relationship will be vital to combat the emerging public health challenges associated with urbanization, and to ensure that investment in green space provides value for money21,22,23.

How to cite this article: Shanahan, D. F. et al. Health Benefits from Nature Experiences Depend on Dose. Sci. Rep. 6, 28551; doi: 10.1038/srep28551 (2016).

Results

The first stage of our analysis was to examine the relationship between individual-level experiences of nature and four health outcomes in a population sample of 1538 residents of Brisbane City, Australia. These health outcomes included whether the respondent scored as having mild or worse depression determined from an established 7 item questionnaire28, whether the respondent reported being under treatment for high blood pressure, perceptions of social cohesion derived from three survey questions29,30,31, and the self-reported number of days on which physical exercise occurred for more than 30 minutes during the survey week.

We measured experiences of nature across three components, including the usual frequency of outdoor green space visits across a year, the average duration of visits to green space across a week, and the intensity of nature (measured as the highest level of vegetation complexity within any of the green spaces that a respondent visited, following a hypothesis that higher levels of vegetation lead to greater health outcomes; Table 1, Fig. 2). Multivariate analyses revealed that a longer duration of individual nature experiences was significantly linked to a lower prevalence of depression and of high blood pressure, and increased physical activity. A higher frequency of green space visitation was an important predictor for increased social cohesion, and both duration and frequency showed a significant positive relationship with higher levels of physical activity (Table 1). These multivariate analyses accounted for key covariates including age, gender, Body Mass Index (BMI; weight in kilograms/square of height in meters), and socio-economic indicators including the income, education, and neighborhood socio-economic disadvantage (Index of Socio-economic Disadvantage, IRSD; Table 1)32. We also found that people with a stronger self-reported connection to nature (measured using the Nature Relatedness scale33) had greater levels of social cohesion and physical activity, but did not show a reduced prevalence of depression or high blood pressure (Table 1).

Table 1 The relationship between four health outcomes (the response variables), socio-demographic covariates and nature experience predictor variables.

The bivariate relationships between health responses (A–D) and nature experiences, comprising (i) the average duration of visits to green space; (ii) the normal reported frequency of visits to green space; and (iii) the nature intensity, measured as vegetation complexity within the best visited public green space

We examined the dose-response relationship between the odds of a respondent being recorded as having high blood pressure or depression and incremental increases in the duration of nature experiences, while accounting for covariates (Fig. 3, Table 2). We found that the odds were significantly lower than the null model for depression when reported green space visits were an average of 30 minutes or more (i.e. the confidence interval did not overlap with an odds ratio of one; Fig. 3A), with a slight increase in mean gains until a duration of 1 hour 15 minutes. For high blood pressure, there was also a significant health improvement after 30 minutes of exposure, though the dose-response curve showed high variability at higher exposure levels (Fig. 3B). The power of the test for high blood pressure and depression was reduced at higher durations (indicated by wider 95% confidence intervals).

Figure 3: Dose-response graphs showing the adjusted odds ratio from logistic regression for incrementally increasing average duration of green space visits.

95% confidence intervals are shown. An odds ratio above one indicates an individual is more likely to have the disease where the threshold of green space visitation is not met.

Table 2 The odds ratios for a person having depression or high blood pressure where specific risk factors are present (the result for each variable was calculated while accounting for all their other risk factors; i.e. multivariate analyses), and the proportion of disease cases in the study population attributable to various risk factors (average population attributable fraction).

We found that the proportion of cases of depression and high blood pressure in the population that can be attributed to city residents failing to spend an average of 30 minutes or more during a green space visit across the course of their week (the ‘population attributable fraction’) was 0.07 for depression, and 0.09 for high blood pressure (Table 2); that is, there could be up to 7% fewer cases of depression and 9% fewer cases of high blood pressure if the entire sampled population met the minimum duration criteria of 30 minutes or more.

How to cite this article: Shanahan, D. F. et al. Health Benefits from Nature Experiences Depend on Dose. Sci. Rep. 6, 28551; doi: 10.1038/srep28551 (2016).