ArticlePDF Available

Abstract and Figures

Contact with green space in the environment has been associated with mental health benefits, but the mechanism underpinning this association is not clear. This study extends an earlier exploratory study showing that more green space in deprived urban neighbourhoods in Scotland is linked to lower levels of perceived stress and improved physiological stress as measured by diurnal patterns of cortisol secretion. Salivary cortisol concentrations were measured at 3, 6 and 9 h post awakening over two consecutive weekdays, together with measures of perceived stress. Participants (n = 106) were men and women not in work aged between 35-55 years, resident in socially disadvantaged districts from the same Scottish, UK, urban context as the earlier study. Results from linear regression analyses showed a significant and negative relationship between higher green space levels and stress levels, indicating living in areas with a higher percentage of green space is associated with lower stress, confirming the earlier study findings. This study further extends the findings by showing significant gender differences in stress patterns by levels of green space, with women in lower green space areas showing higher levels of stress. A significant interaction effect between gender and percentage green space on mean cortisol concentrations showed a positive effect of higher green space in relation to cortisol measures in women, but not in men. Higher levels of neighbourhood green space were associated with healthier mean cortisol levels in women whilst also attenuating higher cortisol levels in men. We conclude that higher levels of green space in residential neighbourhoods, for this deprived urban population of middle-aged men and women not in work, are linked with lower perceived stress and a steeper (healthier) diurnal cortisol decline. However, overall patterns and levels of cortisol secretion in men and women were differentially related to neighbourhood green space and warrant further investigation.
Content may be subject to copyright.
Int. J. Environ. Res. Public Health 2013, 10, 4086-4103; doi:10.3390/ijerph10094086
International Journal of
Environmental Research and
Public Health
ISSN 1660-4601
www.mdpi.com/journal/ijerph
Article
Green Space and Stress: Evidence from Cortisol Measures in
Deprived Urban Communities
Jenny J. Roe 1,*, Catharine Ward Thompson 2,, Peter A. Aspinall 1,†, Mark J. Brewer 3,†,
Elizabeth I. Duff 3,†, David Miller 4,†, Richard Mitchell 5,† and Angela Clow 6,†
1 School of the Built Environment, Heriot-Watt University, Edinburgh EH14 4AS, UK;
E-Mail: p.a.aspinall@hw.ac.uk
2 OPENspace Research Centre, University of Edinburgh, Edinburgh EH3 9DF, UK;
E-Mail: c.ward-thompson@ed.ac.uk
3 Biomathematics and Statistics Scotland, Aberdeen AB15 8QH, UK;
E-Mails: m.brewer@bioss.ac.uk (M.J.B.); e.duff@bioss.ac.uk (E.I.D.)
4 James Hutton Institute, Aberdeen AB15 8QH, UK; E-Mail: david.miller@hutton.ac.uk
5 Centre for Research on Environment, Society and Health, Institute of Health and Wellbeing,
University of Glasgow, Glasgow G12 0XH, UK; E-Mail: richard.mitchell@glasgow.ac.uk
6 Department of Psychology, University of Westminster, London W1B 2UW, UK;
E-Mail: clowa@westminster.ac.uk
These authors contributed equally to this work.
* Author to whom correspondence should be addressed; E-Mail: jenny.roe@york.ac.uk;
Tel.: +44-(0)1904-32-3480; +44-(0)1904-32-2898.
Received: 2 August 2013; in revised form: 14 August 2013 / Accepted: 15 August 2013 /
Published: 2 September 2013
Abstract: Contact with green space in the environment has been associated with mental
health benefits, but the mechanism underpinning this association is not clear. This study
extends an earlier exploratory study showing that more green space in deprived urban
neighbourhoods in Scotland is linked to lower levels of perceived stress and improved
physiological stress as measured by diurnal patterns of cortisol secretion. Salivary cortisol
concentrations were measured at 3, 6 and 9 h post awakening over two consecutive
weekdays, together with measures of perceived stress. Participants (n = 106) were men and
women not in work aged between 3555 years, resident in socially disadvantaged districts
from the same Scottish, UK, urban context as the earlier study. Results from linear
OPEN ACCESS
Int. J. Environ. Res. Public Health 2013, 10 4087
regression analyses showed a significant and negative relationship between higher green
space levels and stress levels, indicating living in areas with a higher percentage of green
space is associated with lower stress, confirming the earlier study findings. This study
further extends the findings by showing significant gender differences in stress patterns by
levels of green space, with women in lower green space areas showing higher levels of
stress. A significant interaction effect between gender and percentage green space on mean
cortisol concentrations showed a positive effect of higher green space in relation to cortisol
measures in women, but not in men. Higher levels of neighbourhood green space were
associated with healthier mean cortisol levels in women whilst also attenuating higher
cortisol levels in men. We conclude that higher levels of green space in residential
neighbourhoods, for this deprived urban population of middle-aged men and women not in
work, are linked with lower perceived stress and a steeper (healthier) diurnal cortisol
decline. However, overall patterns and levels of cortisol secretion in men and women were
differentially related to neighbourhood green space and warrant further investigation.
Keywords: green space; stress; diurnal; saliva; cortisol; neighbourhood; urban;
deprivation; gender; mental health
1. Introduction
Contact with green space has been associated with benefits to mental health, particularly levels of
stress [1]. These effects are thought to operate through one or more of three mechanisms: increased
physical activity (see [2] for a systematic review) which, in turn, improves mood [3]; increased social
contact and sense of belonging within a community [4]; and psychological restoration related to
attention processes and attenuation of stress and fatigue [5,6]. Two theoretical models of psychological
restoration have been proposed. The first is related to the Kaplans’ Attention Restoration Theory [6]
which posits that natural settings, inherently rich in stimuli, invoke involuntary attention which
supports restoration from mental fatigue. The second is Ulrich’s psychoevolutionary model [7] which
posits that stress reduction arises from an immediate, affective response to the visual stimulus array of
a natural setting, impacting the brain and neuroendocrine system. However, evidence of an impact
from contact with natural environments on the neuroendocrine system is currently limited to a small
number of studies.
Contact with nature has been shown to have positive effects on blood pressure [5,7], heart rate [7],
skin conductance and muscle tension [7]. However, most studies which have used salivary cortisol as a
measure of stress in relation to contact with nature have not measured diurnal patterns, focussing
instead on measuring levels immediately before and after exposure to different environmental settings.
For example, Park and colleagues [8] have shown visiting a Japanese forest can reduce cortisol
concentrations, compared to city environments. Lee et al. [9] reported that visiting a forest had greater
benefits for stress levels than visiting an urban environment, using cortisol and pulse rates as
biomarkers of stress. Similarly, van den Berg and Custer [10] found that a period of allotment
gardening led to greater levels of stress reduction than a restful indoor task (i.e., reading),
Int. J. Environ. Res. Public Health 2013, 10 4088
using salivary cortisol as a measure. Although these studies found that contact with natural
environments reduced stress, their external validity is limited because they either had very small
numbers of participants or limited categories of participant (e.g., university students, allotment
gardeners) or used very controlled, specific settings (possibly both). Very little research has examined
the stress neuroendocrine system in relation to the long-term effects of familiar, everyday
environments, set within the context of people’s normal patterns of activity and experience. One
exception is the study by Do et al. [11] which explored the relationship between a series of
neighbourhood characteristics (deprivation, violence, disorder and social cohesion) and changes in
diurnal cortisol rhythms. In that study, higher perceived levels of neighbourhood violence were
significantly associated with lower cortisol values on awakening and with a less steep decline in
cortisol concentrations from the morning to the evening.
Some evidence points to demographic and socio-economic differences in the relationship between
green space and health. An inverse association between green space and all-cause mortality rate
appears stronger for more deprived populations than for more advantaged populations in England and
Wales [12]. Greater beneficial effects of local green space have been found in people spending more
time in their local neighbourhood (e.g., young people, older people, people not in work), perhaps due
to greater exposure to this green space [13]. Gender differences in the relationship between
environmental factors and health outcomes are beginning to emerge but remain under-explored.
For example, a population-level study in the UK has shown gender differences in the relationship of
urban green space to cause-specific mortality rates [14]. The authors suggested this difference was due
to gender differences in perceptions and usage of urban green space. Fear for safety is a common
barrier experienced by women in accessing urban green space [15] and a dimension that reduces the
likelihood of women walking in their local neighbourhoods, as compared to men [16]. A wider general
literature that has explored gender differences in the impact of general neighbourhood context on
health has generally found stronger effects on women than men [14].
Gender differences in the relationship between stress and the diurnal pattern of cortisol secretion
have also been reported. In middle-aged men, stress has been associated with higher cortisol
concentrations and a reduction in the diurnal cortisol decline (henceforth referred to as cortisol slope):
a high and flat pattern. In contrast, stressed middle-aged women presented with low cortisol
concentration and a lower and flatter cortisol slope [17]. Consistent with this finding, in older women
depression has been associated with hypocortisolemia whilst, in older men, depression was associated
with hypercortisolemia [18]. Lower cortisol levels have also been linked with negative life events,
associated with significantly lower cortisol levels in females [19]. There is increasing evidence for a
life-course influence on the cortisol cycle. For example, early life adversity has been shown to be
associated with lower cortisol profiles [20,21]. In summary, there appear to be gender differences in
the impact of neighbourhood characteristics on health as well as gender differences in the
neuroendocrine response to negative life events and stress.
This study builds on an earlier, exploratory study [22] of a socially deprived urban population that
found significant relationships between measures of neighbourhood green space, perceived stress and
the diurnal pattern of cortisol secretion. Higher levels of neighbourhood green space were linked with
lower levels of perceived stress and a steeper diurnal decline in cortisol secretion. In that study, as in
this one, neighbourhood was defined using Census Area Statistics (CAS) Ward. This is a small areal
Int. J. Environ. Res. Public Health 2013, 10 4089
unit used in the administration of the UK decennial census. It has previously been used as a definition
of neighbourhood in other research on green space and health [14,23]. Based upon this earlier study,
we generated two hypotheses for the current study:
1. Hypothesis 1: greater availability of neighbourhood green space in deprived urban communities
is linked to lower perceived stress;
2. Hypothesis 2: greater availability of neighbourhood green space in deprived urban communities
is linked to lower levels of physiological stress, as measured by the diurnal patterns of cortisol
secretion.
Based upon the accumulating evidence cited above indicating gender differences in stress responses
we expanded the present study to include a third hypothesis:
3. Hypothesis 3: patterns of cortisol secretion in this socially deprived urban population will be
differentially expressed in men and women. We did not predict what pattern these would
follow by levels of neighbourhood green space.
2. Experimental Section
2.1. Study Design
The study repeats the earlier study [22] in outcome measures, demographics, geographic location
and cross-sectional design. As before, it included a survey of participant demographics, perceived
stress, well-being and self-reported exercise levels. Repeated salivary cortisol sampling took place
over two consecutive weekdays. The study employed secondary data analysis of participants’ local
environments in order to determine percentage of green space in each participant’s neighbourhood.
The study was set in the city of Dundee, United Kingdom, selected because it contains a number of
highly deprived urban neighbourhoods with varying levels of green space. The study was carried out in
accordance with the British Psychological Society Ethical Principles for Conducting the Research
with Human Participation and with full ethical approval granted by the lead researchers’ institutional
ethics board.
2.2. Participants
To reduce age-related effects on patterns of cortisol secretion our participants were men and women
aged 3355 years of age. They were not in work for any reason (e.g., job-seeking unemployed,
on invalidity benefit, carers) and were living in socio-economically deprived areas of Dundee as
measured by the Carstairs Index of deprivation [24]. The Carstairs Index is a widely used and
well-validated indicator of area-level socio-economic deprivation based on prevalence of household
overcrowding, unemployment among men, low social class, and not having a car; the higher the
Carstairs score, the greater the deprivation. Participants were recruited from areas with a Carstairs
score of 57 (the average Carstairs score for Dundee is 3.93).
Int. J. Environ. Res. Public Health 2013, 10 4090
2.3. Recruitment
Recruitment was through a two-stage process. Firstly, specified postcode areas in the city of
Dundee were selected on the basis of Carstairs indices of 57. Within these areas a campaign of
door-to-door recruitment was undertaken over a period of five weeks, from May to June 2010.
Interested participants that fulfilled the inclusion criteria were provided with information sheets and,
following an indication that they were willing to take part, their details were recorded by the
recruitment company and followed-up by the by the research team. Potential participants were invited
to a group briefing session in the city-centre. People were excluded if they were taking oral steroids,
but more stringent exclusion criteria (e.g., use of anti-depressants, smoking) although recorded, were
waived owing to the high likelihood of finding such behaviour patterns in the target sample population.
People who had lived for less than 12 months in their neighbourhood were also excluded.
At the group briefing session, following informed consent, participants completed a short,
paper-based questionnaire and were instructed on the protocol for taking cortisol samples. Of the
158 potential participates that attended the briefing session 81 participants successfully completed the
study. As statistical analyses established no effects of season or demographic differences between this
sample and the exploratory study sample (gathered in January 2010), the databases were merged to
achieve greater statistical power with a total sample size of 106 participants, 50% male and female.
2.4. Measures
The primary outcome measures were perceived stress (the Perceived Stress Scale (PSS) [25],
the cortisol slope and average cortisol levels, derived from multiple saliva sampling across two
consecutive days. The PSS comprises 10 items measured on a 5-point response from never to very
often. The final score assesses perceived stress over the preceding month and can range from 0
(minimum level of stress) to 40 (maximum level of stress). The detailed procedure for the cortisol
sampling is described below.
Other measures included positive well-being and self-reported levels of physical activity, captured
as potential explanatory variables. Well-being was measured using a shortened version of the Warwick
and Edinburgh Mental Well-being Scale (SWEMWBS) [26]. SWEMWBS asks participants how they
had felt over the previous four weeks in relation to seven items used to measure aspects of mental
well-being (e.g., feeling relaxed, feeling useful), with responses rated on a 5-point scale from none of
the time to all of the time. Final scores can range from 7 (low well-being) to 35 (high well-being).
Physical activity was measured using one item asking for the number of days on which physical
activity (of sufficient exertion to raise breathing rate) reached or exceeded 30 min, recalled over the
past four weeks. This item is recommended by the British Heart Foundation National Centre [27].
Socio-economic deprivation was based on the Carstairs Index for population data in 2001 [24],
obtained via each participant’s postcode. Participants’ age and gender were also recorded.
Exposure to green space was measured by the percentage of the area in the Census Area Statistics
(CAS) Ward of the participant’s residence which was identified as green space. The green space
measure includes parks, woodlands, scrub and other natural environments, but not private gardens
(although participants indicated yes/no as to whether they had access to a garden). Dundee contains
Int. J. Environ. Res. Public Health 2013, 10 4091
31 CAS Wards with a mean population of 4942 at the 2001 census, and a mean proportion of green
space of 33.89%. These data were created by the Centre for Research on Environment Society and
Health (CRESH) [14, 23] and are freely available at the CRESH website [28].
2.5. Cortisol Sampling Procedure
Participants were instructed to take three samples of saliva per day on two consecutive weekdays
(Monday and Tuesday) using salivette saliva sampling devices (Starstedt, Leicester, UK). Requested
sampling times were 3, 6, and 9 h after awakening. They were instructed not to smoke, eat or drink
anything but water 30 min before taking each sample and to keep a log of sample times. Wake up time
was defined as the moment a participant was first conscious of being awake. To maximize adherence,
participants were sent individualised SMS text prompts, based upon self-report predicted awakening
times, 3 times on each day to remind them to take their samples. At the end of the day, participants
were asked to freeze their samples (or place in a fridge) in a sealed bag provided. They were then
collected and shipped to the laboratory for assay analysis within five days of collection. Cortisol assays
were carried out by Salimetrics Europe using Enzyme Linked Immuno-Sorbent Assay. The lower limit
of sensitivity is 0.01638 nmol/L. The standard range in assay is 0.5138.468 nmol/L; correlation of
assay with serum: r (47) = 0.91, p < 0.0001; intra and inter-assay variations were both below 10%.
Average daily cortisol concentrations were determined by calculating the average cortisol
concentration across all sampling points on both sampling days. The cortisol slope was derived from
the change in cortisol concentration from 39 h post awakening, averaged across both sampling days.
The use of start and end data points to determine the linear slope follows methods by Thorn et al., [29]
and Power et al., [30].
2.6. Data Cleaning
When computing cortisol regression analyses, a total of 13 participants were initially removed for
data quality issues (i.e., a cortisol sample outside the required time threshold of plus or minus 45 min
from target sampling time, and/or missing data). Any single measurement greater than two standard
deviations from the mean was retained with that measure winsorised. This resulted in a sample size of
93 for mean cortisol; for cortisol slope (requiring means extant for: sample times 3, 6 and 9 h),
five further participants were removed owing to doubts about the timing of the sampling (less likely to
affect the mean cortisol), giving a sample size of 88. For the perceived stress regression, three
participants were removed for data quality issues, resulting in sample size of 103. When computing
descriptive statistics, a total of two participants were removed owing to missing data, resulting in a
sample size of 104. In summary, data were cleaned or participants removed in preparation for four
different analyses: cortisol slope, cortisol mean; perceived stress; and descriptive statistics.
2.7. Statistical Analysis
Bivariate relationships between variables were explored using Pearson correlations. Gender
differences in socio-demographic characteristics and outcome variables were assessed via
Mann-Whitney tests, as the data were not normally distributed. Multiple linear regressions examined
Int. J. Environ. Res. Public Health 2013, 10 4092
main effects and interaction effects in relationships with the primary outcome variables of perceived
stress, cortisol slope and average cortisol concentrations, and the predictors of age, gender, deprivation
level, physical activity level and the amount of neighbourhood green space; this selection of
covariates is justified by the findings of our earlier study. Since model diagnostics found the
average cortisol concentration to violate the assumption of normality, a log-transformed variable was
used to remove this. Variance inflation factors confirmed there were no collinearity issues in all final
models presented.
In initial regressions, perceived income coping was found to be a very high correlate of perceived
stress (i.e., the two share a very strong relationship with each other, in suggesting that coping with low
income levels is likely to be a main cause of stress in people living with unemployment). As a
consequence, in this particular data set, income coping can be regarded as a surrogate for stress, and
was therefore removed from the analysis. Physical activity was not significantly associated with
outcome measures in exploratory analyses, and was removed from the models for this reason. It is
possible that the number of missing values with respect to physical activity (giving a smaller resultant
sample of 88) may account for this. We found a similar pattern with the life events scale, i.e., it was
not a significant predictor of stress in exploratory regression analyses, but 25% of participants did not
answer this question. We note that removing variables from models in this manner can lead to “omitted
variable bias”, so in each case we checked that the coefficients of remaining variables were not altered
greatly as a result.
Relationships between outcome measures and green space were examined carefully; while a linear
relationship seemed appropriate for two of the measures (perceived stress and mean cortisol), there did
seem to be evidence of a step-change with respect to cortisol slope, suggesting that a binary indicator
might be more suitable than a straight line. We required an objective, scientific method for determining
where on the continuous scale to make the split into two (or more); for this purpose we used SPSS
AnswerTree to establish an optimal split of 43% green space using a CHAID (CHI squared Automatic
Interaction Detector) analysis. Note further: the algorithm used had the potential to make single or
multiple splits, but in the event, only one was found to be significant (at 43%). The term low green
space henceforth refers to areas with less than (or equal to) 43% green space; the term high green
space to areas with more than 43% green space. While associations between stress levels and access
to private or semi-private gardens are not a focus of this paper, we felt it important to ensure any
effects of a household having access to a garden were accounted for prior to assessing any effects of
access to publicly accessible green space. Reported access to a garden (yes/no) was therefore included
as a co-variate in the regression analyses.
Int. J. Environ. Res. Public Health 2013, 10 4093
3. Results and Discussion
3.1. Results
A repeated measures ANOVA factoring day (sampling days 1 and 2), and sample (cortisol
concentrations at 3, 6, 9 h post-awakening) revealed no significant main effect for day. As expected,
there was a highly significant main effect of sampling time (F = 26.62, df = (1,54), p < 0.001),
indicating that cortisol means varied across the day. Both results suggested participant adherence to the
required sampling protocol and legitimised the strategy of averaging both cortisol measures (average
concentration and cortisol slope) across the two sampling days to give the most reliable measures.
The demographic profile of participants is provided in Table 1. Women reported higher perceived
stress, lower well-being and lower levels of physical activity and their average cortisol concentrations
were significantly lower than men’s.
Table 1. Characteristics of the study population (n = 104).
Total
Mean (SD)
Male
Mean (SD)
Female
Mean (SD)
Percentage sample
50%
50%
Mean age (years)
44.75 (6.91)
44.21 (6.65)
45.28 (7.20)
Level of deprivation (mean Carstairs score)
6.64 (2.21)
6.28 (2.36)
7.0 (2.02)
Work status %
Working part-time (less than 30 h/week)
3%
2%
4%
Job-seeking
67%
79%
57%
In education
4%
4%
4%
Not in work owing to sickness, disability
11%
11%
11%
Not in work owing to caring for family
13%
4%
22%
Unknown
1%
0%
2%
Percentage of green space in residential environment
Low green space < 43%
73%
77%
70%
High green space 43%
27%
23%
30%
Perceived Stress (PSS)
14.28 (5.8)
18.66 (5.9) *
21.87 (5.40) *
Perceived well-being (SWEMWBS)
22.35 (4.94)
23.55 (4.92) *
21.10 (4.67) *
Reported physical activity (days/month)
8.20 (8.71)
9.70 (8.9) *
6.67 (8.29) *
Cortisol mean concentration (over 2 days)
6.00 (2.85)
6.54 (2.54)
5.58 (3.08)
Cortisol slope (mean slope over 2 days)
4.00 (5.00)
4.09 (4.74) *
3.91 (5.27) *
* statistically significant gender difference between male and females p < 0.05 (Mann-Whitney U Test).
Bivariate correlations revealed that, within the total population, higher mean cortisol concentrations
were positively associated with a higher cortisol slope (Table 2). This atypical relationship indicates
that, in this socially disadvantaged sample, low cortisol levels were associated with flat diurnal cortisol
profiles, i.e., low cortisol levels were associated with less healthy cortisol profiles. The cortisol slope
was negatively associated with perceived stress and positively associated with physical activity and
more green space (measured on the binary variable). Higher percentage neighbourhood green space
was associated with lower perceived stress and higher mean cortisol secretion.
Int. J. Environ. Res. Public Health 2013, 10 4094
Gender differences in stress levels were found when taking account different levels of green space
in the residential environment (Table 3). In both men and women, perceived stress was higher in low
green space areas, but women’s perceived stress was significantly higher in low green space areas than
men’s (p < 0.05). Also, women had significantly lower mean cortisol concentrations in lower green
space areas compared to men (p < 0.05), indicative of greater hypocortisolemia (low flat cycle). It is
notable that the differences in stress levels between men and women were only significant in low green
space areas, with evidence of greater comparability in stress levels in high green space areas.
Table 2. Relationships between cortisol patterns, health measures and percentage green space.
1
Cortisol
mean
2
Cortisol
Slope
3
Stress
(PSS)
4
Well-
being
5
PA
6
% GS
7
Bin
GS
1. Cortisol Mean
1
2. Cortisol Slope
0.458 ***
1
3. Stress (PSS)
0.112
0.173 *
1
4. Well-being (SWEMWBS)
0.099
0.109
0.616 ***
1
5. PA (Physical Activity)
0.113
0.207 *
0.069
0.003
1
6. Percentage Green Space (% GS)
0.311 ***
0.121
0.286 **
0.080
0.014
1
7. Binary green space (BinGS)
0.136
0.174 *
0.144
0.107
0.033
0.812 ***
1
* p < 0.10; ** p < 0.05; *** p < 0.01. The reported correlation coefficients are Pearson’s (relatively similar
Spearman rank correlations were found). Log transformed data was used for cortisol mean analysis.
Table 3. Gender differences in stress levels by level of green space.
Variable
Male mean (SD)
Female mean (SD)
Low Green
Space
High Green
Space
Low Green
Space
High Green
Space
Perceived Stress (PSS)
13.24 * (5.80)
10.92 (5.83)
16.49 * (5.94)
14.38 (3.38)
Cortisol mean concentration
(nmol/L) (over 2 days)
7.62 * (10.22)
4.61 (1.14)
4.24 * (2.03)
6.43 (3.50)
Cortisol slope (nmol/L)
(mean slope over 2 days)
3.87 (4.89)
4.98 (4.16)
3.25 (4.49)
5.67 (6.85)
* statistically significant difference between males and females p < 0.05 (Mann-Whitney U Test).
Regression analyses were performed to predict perceived stress, the cortisol slope and average
cortisol secretion over the day (see Table 4).
Int. J. Environ. Res. Public Health 2013, 10 4095
Table 4. Linear regression model analyses predicting perceived stress, cortisol slope and
cortisol mean over 2 days.
Main Effects
Model 1
Perceived stress (n = 103)
Model 2
Cortisol Slope (n = 88)
Model 3
Cortisol mean (n = 93)
Variable
β
t
95% CI
β
t
95% CI
β
t
95% CI
Gender (female)
5.47 ***
3.11
1.98, 8.95
0.04
0.04
1.79, 1.72
0.62 ***
3.06
1.02,
0.22
Age
0.14 *
1.70
0.29, 0.02
0.13 **
2.05
0.00, 0.26
0.01 **
2.29
0.00, 0.03
Deprivation level
0.30
1.20
0.19, 0.78
0.00
0.00
0.39, 0.38
0.01
0.29
0.03, 0.04
% green space
(continuous)
0.08 **
2.28
0.14,
0.01
-
-
-
0.001
0.21
0.01, 0.01
binary green space
(≥43%)
-
-
-
3.00 ***
2.79
0.86, 5.14
-
-
-
No Garden
3.48 **
2.06
0.13, 6.82
-
-
-
-
-
-
Interaction Effects
No Garden
with gender (female)
4.31 *
1.93
8.75, 0.13
-
-
-
-
-
-
% green space
(continuous) with
gender (female)
-
-
-
-
-
-
0.01 **
2.05
0.00, 0.02
Note cells with “- denotes terms removed from the model for that particular response variable. * p < 0.10;
** p < 0.05; *** p < 0.01.
Model 1
Perceived Stress: being female and living with lower neighbourhood green space was associated
with higher perceived stress. Not having a garden was associated with higher stress levels but,
as the interaction term shows, only for males (to clarify: the main effect of “no garden” for males
corresponds to the main effect for “no garden”, and total effect for females is 3.48−4.31 = 0.83,
t = 0.71, p = 0.683, hence not significant). The overall model was significant at p = 0.003 with
adjusted r2 value of 13%.
Model 2
Cortisol Slope: Increasing age and living with high rather than low levels of green space (binary
variable) was associated with a greater diurnal cortisol decline. This pattern is illustrated in Figure 1
which shows that participants living in areas of higher green space had a steeper cortisol slope profile
whilst those participants living with lower neighbourhood green space have a flatter slope profile.
There was no evidence of gender or interaction effects of gender with green space in the models,
suggesting the positive association between green space and cortisol slope was consistent in both men
and women. The overall model was marginally significant at p = 0.052 with adjusted r2 of 6.3%.
Int. J. Environ. Res. Public Health 2013, 10 4096
Figure 1. Difference in cortisol slope between participants living in high versus low green
space areas.
Model 3
Cortisol Mean Concentration: Being female and of younger age was associated with lower average
cortisol concentrations (Table 4). There was no significant overall association between neighbourhood
green space and average cortisol concentrations. There was a significant interaction between
percentage green space and gender such that higher neighbourhood green space was associated with
higher average cortisol levels among females, with no effect for males. The overall model was
significant at p = 0.004 with adjusted r2 value of 13%. The interaction effect between gender and green
space (F = 7.03, df = (1,94), p = 0.009) was confirmed in a repeated measures (multilevel) analysis
(using the original separate cortisol readings within individuals), although we prefer to present the
separate, simpler analyses here, as there are fewer problems caused by missing values. In low green
space, women showed a low and flat slope, illustrated in Figure 2(a). For men, lower green space
was associated with a high flat cortisol slope associated as illustrated in Figure 2(b).
In all three models, age was a significant predictor of stress: with increasing age perceived stress
decreased and mean cortisol concentrations were higher with a healthierand steeperdiurnal slope.
We found no evidence of omitted variable bias in any of our three models.
3.2. Discussion
As hypothesised, the study found that higher levels of neighbourhood green space were associated
both with lower levels of perceived stress (Hypothesis 1) and a steeper diurnal decline in cortisol
secretion, for men and women (Hypothesis 2). Green space was found to be an independent predictor
of the circadian cortisol cycle. This confirms the earlier study [22] and establishes further evidence of
an association between levels of neighbourhood green space and neuroendocrine function, which may
point to better understanding of any link between the observed beneficial effects of green space
and health.
369
3
6
9
12 High green space
Low green space
Hours post awakening
Cortisol (nmol/l)
Int. J. Environ. Res. Public Health 2013, 10 4097
Figure 2. (a) Differences in mean cortisol slope in women living in high versus low green
space areas. (b) Differences in mean cortisol slope in men living in high versus low green
space areas.
(a)
(b)
A new findingand atypical of a more socially advantaged middle-aged populationwas that
lower cortisol levels were associated with flat diurnal cortisol profiles, suggestive of a prevalence of
hypocortisolemia and chronic stress. Dysregulation of the diurnal pattern of cortisol secretion is
associated with an array of negative health outcomes [31].
In addition, we found different levels of perceived stress and different cortisol secretion patterns
in men and women (Hypothesis 3) that also varied by amount of green space in the local area.
Firstly, women in low green space areas had significantly higher levels of perceived stress. Secondly,
the diurnal cortisol pattern between men and women was different in areas of low green space but not
in high green space areas. Women showed evidence of greater hypocortisolemia (i.e., a low flat
slope cycle) and men a high flat slope cycle in low green space areas. A low flat cycle is associated
with exhaustion and chronic stress [19,3234], whereas a high flat cyclewhilst also indicative of
severe stresssuggests better overall stress regulation [35,36]. This finding is consistent with reports
of hypocortisolemia in middle-aged females exposed to stress, depression and negative life events [1719].
3 6 9
3
6
9
12
Women low green space
Women high green space
Hours post awakening
Cortisol (nmol/l)
3 6 9
3
6
9
12
Men low green space
Men high green space
Hours post awakening
Cortisol (nmol/l)
Int. J. Environ. Res. Public Health 2013, 10 4098
It suggests that high green space may contribute to lessening the differential effect of stress on women
compared with men. Whilst associations between stress levels and access to gardens were not the focus
of this study, it is interesting to note that perceived stress was higher for those with no garden, but that
this effect was significantly weaker for females. This suggests further research should explore types of
green space, including access to gardens, and gender effects on stress levels.
We found a positive relationship between higher green space levels and mean cortisol levels in
women, but not in men. It seems that higher neighbourhood green space may contribute in some
way to higher but healthier cortisol levels in this female population and, conversely, to moderating
unhealthily high cortisol levels in men (i.e., cortisol concentrations in men come down in higher
green space areas to a healthier level). This finding provides further evidence of gender differences
in neuroendocrine associations with psychosocial factors and points to gender differences in the
neuroendocrine response to the environment. However it is interesting that in males and females,
irrespective of overall cortisol levels, a greater diurnal decline in cortisol was associated with more
neighbourhood green space. High green space levels therefore appear to offer some degree of stress
buffering in this socially disadvantaged population, reducing or moderating the differences in stress
levels perceived in men and women compared with those living with low green space.
The participants in this study were mostly middle-aged and long-term residents in urban, deprived
neighbourhoods. It is plausible that the impact of unemployment and long-term socio-economic
adversity in the participants resulted in long-term HPA axis dysregulation and the cortisol profiles
presented here. Recently it has been shown that adult only city living is associated with greater
stress-related activity in the amygdala, which activates the HPA axis [37], Unfortunately, the impact of
lifetime exposure to city dwelling was not directly explored in this study although, in the
neighbourhoods surveyed, it is typical for residents to have been life-long inhabitants of the same city.
For the purpose of this current study, what is of most interest is that those residing in city areas with
higher levels of green space report less perceived stress and appear to have been more resilient to the
negative effects of urban deprivation as evidenced by a steeper cortisol slope profile.
These results are potentially important in understanding how neighbourhood green space might
contribute to public health improvement. Stress is known to impact on cardiovascular disease via
interaction with a variety of risk factors that include genetics, early experience, age, race, diet and
physical activity [38], with women at greater risk [39,40] but little is known about the environmental
factors that might also contribute. At a population level, higher levels of green space are associated
with reduced cardio-vascular mortality in England and Wales [12]. Our data indicate that
neighbourhood green space is associated with perceptions of stress as well as the stress neuroendocrine
system and this may be a pathway by which the environment can impact health. Whilst more research
is needed to understand these mechanisms, including a better understanding of how use of nearby
green space may affect the pattern of stress associated with higher neighbourhood green space,
our study represents a valuable step in establishing a biological pathway linking green space with
stress levels in deprived urban environments.
Int. J. Environ. Res. Public Health 2013, 10 4099
3.3. Limitations
Our study was cross-sectional in design and therefore cannot demonstrate causality.
Whilst demographic and socio-economic factors were accounted for, it is possible that other individual
and environmental factors, for example smoking status, or social contact, may have influenced
outcomes. The percentage of smokers in our sample wasbased on pilot screeningknown to be
high, with smoking levels in deprived areas of Dundee over 50% [41]. It was not realistic, therefore, in
this population, to exclude smokers from our study. Whilst smoking increases levels of salivary
cortisol it has a short-term effect on the neuroendocrine system. We controlled for this with text
promptsissued to participants at timed intervals during the day advising them to refrain from
smoking (and other stimulants) half an hour prior to sampling procedures. We have no reason to
believe participants did not comply with this instruction: the linear decline data suggests they did not
smokeas requestedotherwise we would have seen a spike in cortisol. All spikes detected were
managed in the data set as described in Section 2.6.
We were unable to confirm the significant relationship found in the earlier study between cortisol
slope and physical activity, although we found a positive correlation between these two variables,
showing a positive effect of increased physical activity on cortisol slope patterns. We suspect this is
owing to a high proportion of missing data in answering the physical activity question (12% of our
sample did not provide data for this variable). In future, more objective measures might usefully be
used to establish daily physical activity levels in participants e.g., accelerometers. We would also
recommend the inclusion of a social connectedness scale to further explore relationships between
stress and levels of green space butin the current studythis was outwith our agenda.
We chose to examine the diurnal decline in cortisol secretion (3, 6 and 9 h post awakening) as
opposed to the cortisol awakening response (CAR). It was not practical in this deprived population to
impose the intensive post awakening saliva sampling regime required for determination of the CAR.
In addition we did not have objective means of verifying adherence with the required regime which is
essential for accurate determination of this measure (recently it has been shown that post awakening
non-adherence of just 515 min can lead to misleading CAR estimates) [42]. Diurnal cortisol patterns
have been established using as few as two post-awakening samples (at 3 and 6 h); establishing the
slope pattern over the day between 39 h post awakening was therefore felt to be an appropriately
rigorous measurement in this context.
Owing to limitations in UK land use classifications systems our objective measure of percentage
green space was unable to pick up the finer grain detail of front gardens and street trees. Improving the
measurement of different types of green space is therefore one important area to develop in future
research to test hypotheses in relation to gender differences in relation to associations between access
to private/semi-private gardens, other types of green space and health.
4. Conclusions
Higher levels of neighbourhood green space in deprived urban communities are linked with lower
perceived stress and a steeper diurnal decline in cortisol secretion, confirming the earlier study [22].
This new study extends the evidence for a biological pathway which may explain the link between
Int. J. Environ. Res. Public Health 2013, 10 4100
neighbourhood green space and health outcomes, for which stress has an aetiological role. In addition,
we found evidence of hypocortisolemia (i.e., lower daytime levels of cortisol) in this socially deprived
urban population, an atypical pattern and indicative of chronic stress, with greater hypocortisolemia
evident in women living in low green space. Furthermore, we have shown that higher levels of green
space are associated with a lower level of this pattern in women, with a positive association of
increased green space with mean cortisol concentrations, warranting further investigation. Through
successful replication of earlier methodology, we have demonstrated that this experimental approach is
an ecologically valid and reliable method for furthering evidence of salutogenic environment-body
interactions within a real-world context.
Acknowledgments
We thank Sue Morris, Keith Marshall, Jane Morrice and Paula Horne from the James Hutton
Institute, for assisting with recruitment protocol; Lisa Thorn, University of Westminster, for
exploratory statistical analysis. This study is part of the Green Health project funded by the
Scottish Government project No. MLU/ECA/UGW/847/08, with the James Hutton Institute in
collaboration with the Universities of Edinburgh, Glasgow and Heriot-Watt, and with Biomathematics
and Statistics Scotland.
Conflicts of Interest
The authors declare no conflict of interest.
References
1. Van den Berg, A.E.; Maas, J.; Verheij, R.A.; Groenewegen, P.P. Green space as a buffer between
stressful life events and health. Soc. Sci. Med. 2010, 70, 12031210.
2. Lachowycz, K.; And Jones, A.P. Greenspace and obesity: A systematic review of the evidence.
Obes. Rev. 2011, 12, e183e189.
3. Barton, J.; Pretty, J. What is the best dose of nature and green exercise for improving mental
health? A multi-study analysis. Environ. Sci. Technol. 2010, 44, 39473955.
4. Maas, J.; van Dillen, S.M.E.; Verheij, R.A.; Groenewegen, P.P. Social contacts as a possible
mechanism behind the relation between green space and health. Health Place 2009, 15, 586595.
5. Hartig, T.; Evans, G.W.; Jamner, L.D.; Davies, D.S.; rling, T. Tracking restoration in natural
and urban field settings. J. Environ. Psychol. 2003, 23, 109123.
6. Kaplan, R.; Kaplan, S. The Experience of Nature: A Psychological Perspective; Cambridge
University Press: New York, NY, USA, 1989.
7. Ulrich, R.S.; Simons, R.F.; Losito, B.D.; Fiorito, E.; Miles, M.A.; Zelson, M. Stress recovery
during exposure to natural and urban environments. J. Environ. Psychol. 1991, 11, 201230.
8. Park, B.J.; Tsunetsugu, Y.; Kasetani, T.; Hirano, H.; Kagawa, T.; Sato, M.; Miyazaki, Y.
Physiological effects of Shinrin-yoku (taking in the atmosphere of the forest)Using salivary
cortisol and cerebral activity as indicators. J. Physiol. Anthropol. 2007, 26, 123128.
Int. J. Environ. Res. Public Health 2013, 10 4101
9. Lee, J.; Park, B.J.; Tsunetsugu, Y.; Ohira, T.; Kagawa, T.; Miyazaki, Y. Effect of forest bathing
on physiological and psychological responses in young Japanese male subjects. Public Health
2011, 125, 93100.
10. Van den Berg, A.E.; Custer, M.H.G. Gardening promotes neuroendocrine and affective
restoration from stress. J. Health Psychol. 2010, 16, 311.
11. Do, D.P.; Diez Roux, A.V.; Hajat, A.; Auchincloss, A.; Merkin, S.; Ranjit, N.; Shea, S.;
Seeman, T. Circadian rhythm of cortisol and neighborhood characteristics in a population-based
sample: The Multi-Ethnic Study of Atherosclerosis. Health Place 2011, 17, 625632.
12. Mitchell, R.; Popham, F. Effect of exposure to natural environment on health inequalities:
An observational population study. The Lancet 2008, 372, 16551660.
13. De Vries, S.; Verheij, R.A.; Groenewegen, P.P.; Spreeuwenberg, P. Natural environments
healthy environments? An exploratory analysis of the relationship between greenspace and health.
Environ. Plan. A 2003, 35, 17171731.
14. Richardson, E.A.; Mitchell, R. Gender differences in relationships between urban green space and
health in the United Kingdom. Soc. Sci. Med. 2010, 71, 568575.
15. Ward Thompson, C.; Aspinall, P.; Bell, S.; Findlay, C. “It gets you away from everyday life”:
Local woodlands and community useWhat makes a difference? Landsc. Res. 2005, 30, 109146.
16. Foster, C.; Hillsdon, M.; Thorogood, M. Environmental perceptions and walking in English adults.
J. Epidemiol. Community Health 2004, 58, 924928.
17. Kumari, M.; Badrick, E.; Sacker, A.; Kirschbaum, C.; Marmot, M.; Chandola, T. Identifying
patterns in cortisol secretion in an older population. Findings from the Whitehall II study.
Psychoneuroendocrinology 2010, 35, 10911099.
18. Bremmer, M.A.; Deeg, D.J.H.; Beekman, A.T.F.; Penninx, B.W.J.H.; Lips, P.;
Hoogendijk, W.J.G. Major depression in late life is associated with both hypo- and
hypercortisolemia. Biol. Psychiatry 2007, 62, 479486.
19. Witteveen, A.B.; Huizink, A.C.; Slottje, P.; Bramsen, I.; Smid, T.; van der Ploeg, H.M.
Associations of cortisol with posttraumatic stresssymptoms and negative life events: A study of
police officers and firefighters. Psychoneuroendocrinology 2010, 35, 11131118.
20. Gerritsen, L.; Geerlings, M.I.; Beekman, A.T.F.; Deeg, D.J.H.; Penninx, B.W.J.H.; Comijs, H.C.
Early and late life events and salivary cortisol in older persons. Psychol. Med. 2010, 40,
15691578.
21. Power, C.; Thomas, C. Changes in BMI, duration of overweight and obesity, and glucose
metabolism: 45 years of follow-up of a birth cohort. Diabetes Care 2011, 34, 19861991.
22. Ward Thompson, C.; Roe, J.; Aspinall, P.; Mitchell, R.; Clow, A.; Miller, D. More green space is
linked to less stress in deprived communities: Evidence from salivary cortisol patterns.
Landsc. Urban Plan. 2012, 105, 221229.
23. Mitchell, R.; Astell-Burt, T.; Richardson, R. A comparison of green space indicators for
epidemiological research. J. Epidemiol. Community Health 2011, 65, 853858.
24. Carstairs, V.; Morris, R. Deprivation and Health in Scotland; Aberdeen University Press:
Aberdeen, Scotland, 1991.
25. Cohen, S.; Kamarck, T.; Mermelstein, R. A global measure of perceived stress. J. Health Soc.
Behav. 1983, 24, 385396.
Int. J. Environ. Res. Public Health 2013, 10 4102
26. Stewart-Brown, S.; Tennant, A.; Tennant, R.; Platt, S.; Parkinson, J.; Weich, S. Internal construct
validity of the Warwick-Edinburgh Mental Well-being Scale (WEMWBS): A Rasch analysis
using data from the Scottish Health Education Population Survey. Health Qual. Life Outcomes
2009, 7, 15, doi:10.1186/1477-7525-7-15.
27. Milton, K.; Bull, F.C.; Bauman, A. Reliability and validity testing of a single-item physical
activity measure. Br. J. Sports Med. 2011, 45, 203208.
28. The Centre for Research on Environment Society and Health (CRESH). Availble online:
www.cresh.org.uk (accessed on 30 August 2013).
29. Thorn, L.; Evans, P.; Cannon, A.; Hucklebridge, F.; Evans, P.; Clow, A. Seasonal differences in
the diurnal pattern of cortisol secretion in healthy participants and those with self-assessed
seasonal affective disorder. Psychoneuroendocrinology 2011, 36, 816823
30. Power, C.; Thomas, C.; Li, L.; Hertzman, C. Childhood psychosocial adversity and adult cortisol
patterns. Br. J. Psychiatry 2012, 201, 199206.
31. Nader, N.; Chrousos, G.P.; Kino, T. Interactions of the circadian CLOCK system and the HPA
axis. Trends Endocrinol. Metab. 2010, 21, 277286.
32. De Kloet, C.S.; Vermettena, E.; Heijnen, C.J.; Geuze, E.; Lentjes, E.G.W.M.; Westenberg, H.G.M.
Enhanced cortisol suppression in response to dexamethasone administration in traumatized
veterans with and without posttraumatic stress disorder. Psychoneuroendocrinology 2007, 32,
215226.
33. Giese-Davis, J.; Sephton, S.E.; Abercrombie, H.C.; Duran, R.E.F.; Spiegel, D. Repression and
high anxiety are associated with aberrant diurnal cortisol rhythms in women with metastatic breast
cancer. Health Psychol. 2004, 23, 645650.
34. Jerjes, W.K.; Cleare, A.J.; Wessely, S.; Wood, P.J.; Taylor, N.F. Diurnal patterns of salivary
cortisol and cortisone output in chronic fatigue syndrome. J. Affect. Disord. 2005, 87, 299304.
35. Deuschle, M.; Gotthardt, U.; Schweiger, U.; Weber, B.; Korner, A.; Schmider, J.; Standhardt, H.;
Lammers, C.-H.; Heuser, I. With aging in humans the activity of the hypothalamus-pituitary-
adrenal system increases and its diurnal amplitude flattens. Life Sci. 1997, 61, 22392246.
36. Weber, B.; Lewicka, S.; Deuschle, M.; Colla, M.; Vecsei, P.; Heuser, I. Increased diurnal plasma
concentrations of cortisone in depressed patients. J. Clin. Endocrinol. Metab. 2000, 85,
11331136.
37. Lederbogen, F.; Kirsch, P.; Haddad, L.; Streit, F.; Tost, H.; Schuch, P.; West, S.; Pruessner, J.C.;
Rietschel, M.; Deuschle, M.; Meyer-Lindenberg, A. City living and urban upbringing affect
neural social stress processing in humans. Nature 2011, 474, 498501.
38. Anderson, D.E.; Chesney, M.A. Gender Differences in Role of Stress and Emotion in
Cardiovascular Function and Disease, In Principles of Gender-Specific Medicine; Lagato, M.J.,
Ed.; Elsevier Academic Press: London, U.K., 2004; Volume 1, Chapter 26, pp. 263274
39. Koskinen, S.; Martelin, T. Why are socioeconomic differences smaller among women than among
men? Soc. Sci. Med. 1994, 38, 13851396.
40. Mackenbach, J.P.; Kunst, A.E.; Groenhof, F.; Borgan, J.K.; Costa, G.; Faggiano, F.; zan, P.;
Leinsalu, M.; Martikainen, P.; Rychtarikova, J.; et al. Socioeconomic inequalities in mortality
among women and among men: An international study. Am. J. Public Health 1999, 89,
18001806.
Int. J. Environ. Res. Public Health 2013, 10 4103
41. NHS Tayside Population Profile for Inequalities Strategy (2004), Directorate of Change and
Innovation. Available online http://www.nhstayside.scot.nhs.uk/about_nhstay/publications/
health_ineq/pop_profile.pdf (accessed on 27 August 2013).
42. Smyth, N.; Clow, A.; Hucklebridge, F.; Thorn, L.; Evans, P. Delays of 515 minutes between
awakening and the start of saliva sampling matter in assessment of the cortisol awakening
response. Psychoneuroendocrinology 2013, pii: S0306-4530(12)00426-X, doi: 10.1016/
j.psyneuen.2012.12.013.
© 2013 by the authors; licensee MDPI, Basel, Switzerland. This article is an open access article
distributed under the terms and conditions of the Creative Commons Attribution license
(http://creativecommons.org/licenses/by/3.0/).
... Indeed, the median female score was in the 'strongly agree' range, whilst the median male score was in the 'agree' range. This is corroborated by Roe et al. (2013), who showed that female stress levels (measured via cortisol sampling) were significantly higher in areas of low green space compared to males', suggesting that there may be a gender difference in perceiving nature as a coping (or stress ameliorating) mechanism [56]. However, our results could also be the facet of the demographic limitations of this study, i.e., increasing the number of male respondents could potentially increase the median scores for males. ...
... Indeed, the median female score was in the 'strongly agree' range, whilst the median male score was in the 'agree' range. This is corroborated by Roe et al. (2013), who showed that female stress levels (measured via cortisol sampling) were significantly higher in areas of low green space compared to males', suggesting that there may be a gender difference in perceiving nature as a coping (or stress ameliorating) mechanism [56]. However, our results could also be the facet of the demographic limitations of this study, i.e., increasing the number of male respondents could potentially increase the median scores for males. ...
Article
Full-text available
The COVID-19 pandemic has brought about unprecedented changes to human lifestyles across the world. The virus and associated social restriction measures have been linked to an increase in mental health conditions. A considerable body of evidence shows that spending time in and engaging with nature can improve human health and wellbeing. Our study explores nature’s role in supporting health during the COVID-19 pandemic. We created web-based questionnaires with validated health instruments and conducted spatial analyses in a geographic information system (GIS). We collected data (n = 1184) on people’s patterns of nature exposure, associated health and wellbeing responses, and potential socioecological drivers such as relative deprivation, access to greenspaces, and land-cover greenness. The majority of responses came from England, UK (n = 993). We applied a range of statistical analyses including bootstrap-resampled correlations and binomial regression models, adjusting for several potential confounding factors. We found that respondents significantly changed their patterns of visiting nature as a result of the COVID-19 pandemic. People spent more time in nature and visited nature more often during the pandemic. People generally visited nature for a health and wellbeing benefit and felt that nature helped them cope during the pandemic. Greater land-cover greenness within a 250 m radius around a respondent’s postcode was important in predicting higher levels of mental wellbeing. There were significantly more food-growing allotments within 100 and 250 m around respondents with high mental wellbeing scores. The need for a mutually-advantageous relationship between humans and the wider biotic community has never been more important. We must conserve, restore and design nature-centric environments to maintain resilient societies and promote planetary health.
... The most recurrent method was calculating quantiles of the green space (Bixby et al., 2015;Lachowycz and Jones, 2014;Mitchell and Popham, 2008;Mytton et al., 2012;Wu et al., 2015). Other examples were the use of binary measurementsevaluating areas which had less or more than 43% of green spaces (Roe et al., 2013) -, and the assessment of areas which had 10% more green spaces than the average (Maas et al., 2009b). Apart from that, this review noticed that the classification of green spaces was not equal among the reviewed articles. ...
... In the same way, some researchers specified parkland categories as they were interested in open green spaces for public recreation (Astell-Burt et al., 2013b). Other authors did not consider private gardens because these might compensate for the absence of public green areas (de Vries et al., 2003) or might be seen as an additional socioeconomic indicator (Mytton et al., 2012;Roe et al., 2013;Ward Thompson et al., 2012). ...
Article
Background: In recent years, research has been increasingly devoted to understanding the complex human health-environment relationship. Nevertheless, many different measurements have been applied to characterize the environment. Among them, the application of Land Use and Land Cover (LULC) data is becoming more noticeable over time. Aims: This research aims to analyse the reliability of Land Use and Land Cover data (LULC) data as a suitable describer of the environment in studies relating human health to the environment. With a specific focus on the methodologies using LULC data, we also examine the study designs and analytical methods that have been commonly performed so far. Materials and Methods: We gathered studies relating human health outcomes to Land Use and Land Cover (LULC) data. A Boolean search limited to reviews was conducted in February 2019 using Web of Science Core Collection search engines. Five reviews were selected as our preliminary starting set of literature and from those, two backward snowballing searches were conducted. The first backward snowballing search used the reference lists of the first 5 reviews and revealed 17 articles. From these, the second search gathered 24 new articles also fulfilling the inclusion criteria established. In total, 41 articles were examined. Results: Our main results reported that Land Use and Land Cover (LULC) data national level data was preferred over LULC international level data. However, this tendency seems to be strongly related to the specific aims of the articles. They essentially defined the living environment either through buffer zones, using the administrative boundaries wherein the individuals reside, or using the specific location of the individuals assessed. As for the characterization of the environment, authors performed 4 principal methodologies: extracting the percentage of green space, computing the "Land Use mix", recording the type of land cover, and using the percentage of tree canopy. Besides, all the articles included measurements in urban contexts and most of them evaluated the accessibility of individuals to their surroundings. Furthermore, it was clearly stated that the complexity of the topic and the challenging data leads authors to carry out advanced statistical methods and mostly cross-sectional designs with no causal relations. Discussion and Conclusions: Land Use and Land Cover (LULC) data has been demonstrated to be a versatile tool supporting both local-focused studies with few individuals involved and broad territorial-scoped studies with huge populations. Promising synergy has been highlighted between Electronic Health Records (EHR) and LULC data in studies dealing with massive information and broader scopes with regards to the assessment of territorial realities. As this emerging topic matures, investigators should (1) elucidate subjects of ongoing debate such as the 2 measurement of the living environment and its characterization; (2) explore the whole potential of LULC data, using methodologies that encompass both their biophysical and socioeconomic information; (3) perform innovative designs that are able to establish causal relationships among the studied variables (for example, Cellular Automata models), and (4) expand the current set of studied health outcomes leveraging comprehensive and trustworthy health data sources such as EHR.
... Green space offers citizens more opportunities for social contact and stress relief -whether impromptu or planned [48]. Studies show UGS should be of critical importance to public mental health, especially from an urban planning perspective [21,40]. ...
Preprint
Full-text available
Continued population growth and urbanization is shifting research to consider the quality of urban green space over the quantity of these parks, woods, and wetlands. The quality of urban green space has been hitherto measured by expert assessments, including in-situ observations, surveys, and remote sensing analyses. Location data platforms, such as TripAdvisor, can provide people's opinion on many destinations and experiences, including UGS. This paper leverages Artificial Intelligence techniques for opinion mining and text classification using such platform's reviews as a novel approach to urban green space quality assessments. Natural Language Processing is used to analyze contextual information given supervised scores of words by implementing computational analysis. Such an application can support local authorities and stakeholders in their understanding of and justification for future investments in urban green space.
... A significant part of the world's population lives in urban areas, where there are many stressors [1]. Studies show that the less green areas there are in a neighborhood, the higher level of cortisol, the "stress hormone", in the blood of its residents [2]. The health benefits of contact with nature have been widely researched. ...
Article
Full-text available
The category of green areas which intensify feelings using a small area includes sensory gardens. They also have major potential for eliminating stress symptoms and stimulating sensory feelings among people with disabilities. Sensory gardens, as an element of universal recreational development in urban and suburban conditions, can perfectly fulfill social, educational, and therapeutic functions. Their distinguishing feature is a strong influence on non-visual senses. Spaces that are adapted to the needs of people with disabilities give a sense of security, provide greater independence, can be used for passive and active therapy, and constitute a zone of social inclusion. Sensory gardens can be considered as an example of a comprehensive development of green areas that meets the requirements of universal design. The presented study is a review. The authors show that providing urban communities with a form of therapeutic contact with nature through the use of urban forests is very important of improving their quality of life. Special attention was paid to the potential of sensory gardens in the designing of forms of recreational development, which may be an opportunity to reduce the process of social exclusion of people with disabilities in the local environment.
... Green space offers citizens more opportunities for social contact and stress reliefwhether impromptu or planned [48]. Studies show UGS should be of critical importance to public mental health, especially from an urban planning perspective [21,40]. ...
Article
Full-text available
Continued population growth and urbanization is shifting research to consider the quality of urban green space over the quantity of these parks, woods, and wetlands. The quality of urban green space has been hitherto measured by expert assessments, including in-situ observations, surveys, and remote sensing analyses. Location data platforms, such as TripAdvisor, can provide people’s opinion on many destinations and experiences, including UGS. This paper leverages Artificial Intelligence techniques for opinion mining and text classification using such platform’s reviews as a novel approach to urban green space quality assessments. Natural Language Processing is used to analyze contextual information given supervised scores of words by implementing computational analysis. Such an application can support local authorities and stakeholders in their understanding of–and justification for–future investments in urban green space.
... Par exemple, la présence d'espaces verts publics à proximité d'un lieu de résidence a été associée à une meilleure santé métabolique et cardiaque des habitants (Paquet, Coffee, Haren, Howard, Adams, Taylor & Daniel, 2014) ou encore à une amélioration de la prévention contre le diabète, le surpoids et l'obésité chez les enfants et les adolescents (Bell, Wilson & Liu, 2008). De nombreuses études rapportent les bénéfices de la nature et de la biodiversité pour la gestion du stress et une plus grande capacité à faire face aux problèmes et aux grandes difficultés de la vie (Roe, Thompson, Aspinall, Brewer, Duff, Miller, Mitchell & Clow, 2013 ;Kuo & Sullivan, 2001). ...
Technical Report
Full-text available
La journée d’étude 2019 du laboratoire ESPI Réflexions et Recherches (ESPI2R) est première à double titre : elle inaugure la série – que l’on souhaite longue – des évènements réunissant chercheurs et spécialistes de tous horizons pour partager leurs réflexions sur les thèmes liés à l’immobilier, à la ville et aux territoires.
... A mounting body of evidence shows that exposure to elements of these natural environments (e.g., trees, vegetation, lakes, and oceans) contributes to human health and well-being [5][6][7][8]. For example, exposure to greenness (i.e., trees, vegetation) has been associated with better sleep [9][10][11][12][13], cognitive function [14][15][16], and immune function [17][18][19], as well as reduced stress [15,[20][21][22][23], and depressive symptoms [24][25][26][27]. Similarly, exposure to blueness (i.e., lakes, rivers, sea) has been linked to mental health and well-being [28][29][30][31][32][33][34]. ...
Article
Full-text available
Evolutionary psychology theories propose that contact with green, natural environments may benefit physical health, but little comparable evidence exists for brown, natural environments, such as the desert. In this study, we examined the association between “brownness” and “greenness” with fasting glucose among young residents of El Paso, Texas. We defined brownness as the surface not covered by vegetation or impervious land within Euclidian buffers around participants’ homes. Fasting glucose along with demographic and behavioral data were obtained from the Nurse Engagement and Wellness Study (n = 517). We found that residential proximity to brownness was not associated with fasting glucose when modeled independently. In contrast, we found that residential greenness was associated with decreased levels of fasting glucose, despite the relatively low levels of greenness within the predominantly desert environment of El Paso. A difference between the top and bottom greenness exposure quartiles within a 250 m buffer was associated with a 3.5 mg/dL decrease in fasting glucose levels (95% confidence interval: −6.2, −0.8). Our results suggest that within the understudied context of the desert, green vegetation may be health promoting to a degree that is similar to other, non-desert locations in the world that have higher baselines levels of green.
... Urban agriculture proposes to be a solution by bringing fresh produce into urban neighborhoods with low access (Horst et al. 2017). In these same neighborhoods, green space is often lacking, so urban agriculture promises increased access to green space and the health benefits that accompany it: mitigating stress, anxiety, and depression (Roe et al. 2013;Beyer et al. 2014); increasing physical activity (Branas et al. 2011); providing space for social interaction (Kuo et al. 1998;Holtan et al. 2015); and improving public safety (Kuo et al. 1998;Kondo et al. 2015). ...
Article
Full-text available
Critical food scholars have argued that activists’ political ideologies and environmental values are important influences on their food justice projects. However, this body of work has given little attention to religion and spirituality even though religious studies scholars maintain that religious values affect environmental and social action. Bringing together these perspectives considers the way religious values and meaning making intersect with actions toward food justice outside of traditionally religious spaces. This paper draws on qualitative research, including a dozen interviews and 11 months of participant observation, at Franklinton Farms, a nonprofit urban farm in Columbus, Ohio. I demonstrate that Franklinton Farms team members reference diverse religious values and practices when explaining the meaning and significance of their farming. In addition, I argue that they renegotiate their religious values in light of the injustices they see in the food system. By examining religion and spirituality within a secular food space, this paper sheds light on an underexplored influence on whether and how alternative food spaces realize food justice.
Article
Full-text available
The therapeutic advantages of seeing plants have gained increasing consideration in stressful modern societies, however, evidence-based studies on how physiological and emotional states of individuals from different nationalities change when seeing different foliage colors are limited. The study was conducted to explore the physiological and psychological advantages of foliage colors as visual stimuli. The experiment included 40 men from two nations (age: 21.34 ± 3.50 years) and was carried out using five foliage colors including green, light green, green-yellow, green-red and green-white. Participants were exposed to each color for 2 min, when seeing the foliage colors, eye movements and oxy-Hb concentrations were continuously measured. Subjective evaluations of emotions were performed utilizing a semantic differential questionnaire. A significant decrease in oxy-Hb concentration in the frontal lobe was associated with the viewing of green and green-white plants by the Japanese participants and with viewing light green and green-yellow by the Egyptian participants. Participants spent higher fixation numbers and longer durations on these colors. The findings indicate that viewing of these plant colors was positively associated with physiological relaxation. Furthermore, these colors were associated with more positive feelings, such as calmness, comfort and naturalness. Therefore, the presence of these colors in spaces may have positive impacts on relaxation and emotional status.
Article
Full-text available
Access to green schoolyards (schoolyards designed with greenery and natural elements to create a park-like environment, as opposed to asphalt-based playgrounds) are associated with many benefits for students, including improvements in physical and mental health. While many studies examining these associations are cross-sectional, some feature experimental designs that offer the possibility of causal inference. In this review, we looked at experimental studies that examine the impact of schoolyard greening on measures of physical activity and socioemotional health in children. Four electronic databases (Ovid Medline, PsycINFO, Scopus and Greenfile) were searched, and from 1843 articles retrieved, 6 articles met the inclusion criteria. Examination of the eligible studies revealed a general consensus on the positive impact of schoolyard greening on both physical activity and socioemotional health outcomes for students, suggesting that schoolyard greening is a viable intervention in reducing the health equity gaps and improving children’s health regardless of their racial or ethnic backgrounds or residential neighborhood socioeconomic status. Further experimental research on this topic should elucidate how educators, administrators, policy makers, and other stakeholders can harness the benefits of schoolyard greening to improve the health and well-being of children in their communities.
Article
Full-text available
Green space has been associated with a wide range of health benefits, including stress reduction, but much pertinent evidence has relied on self-reported health indicators or experiments in artificially controlled environmental conditions. Little research has been reported using ecologically valid objective measures with participants in their everyday, residential settings. This paper describes the results of an exploratory study (n = 25) to establish whether salivary cortisol can act as a biomarker for variation in stress levels which may be associated with varying levels of exposure to green spaces, and whether recruitment and adherence to the required, unsupervised, salivary cortisol sampling protocol within the domestic setting could be achieved in a highly deprived urban population. Self-reported measures of stress and general wellbeing were also captured, allowing exploration of relationships between cortisol, wellbeing and exposure to green space close to home. Results indicate significant relationships between self-reported stress (P < 0.01), diurnal patterns of cortisol secretion (P < 0.05), and quantity of green space in the living environment. Regression analysis indicates percentage of green space in the living environment is a significant (P < 0.05) and independent predictor of the circadian cortisol cycle, in addition to self-reported physical activity (P < 0.02). Results also show that compliance with the study protocol was good. We conclude that salivary cortisol measurement offers considerable potential for exploring relationships between wellbeing and green space and discuss how this ecologically valid methodology can be developed to confirm and extend findings in deprived city areas to illuminate why provision of green space close to home might enhance health.
Article
Full-text available
Cortisol levels may be altered in childhood in association with maltreatment (neglect, abuse and witnessing abuse) and other adversities, yet little is known about whether effects on cortisol persist into later life. To establish whether childhood psychosocial adversities predict cortisol levels in mid-adulthood. Childhood psychosocial adversities were ascertained in the 1958 British birth cohort and cortisol was measured in two saliva samples, one 45 min after awaking (T(1)) and the other 3 h later the same day (T(2)), from 6524 participants aged 45 years. No association was seen for abuse or household dysfunction in childhood and adult cortisol levels. In women but not men, T(1) cortisol was lowered by 7.9% per unit increase in childhood neglect score (range 0-3); T(1) to T(2) cortisol decline was less steep. High levels of maltreatment (abuse, neglect, witnessed abuse) were associated with >25% lower T(1) cortisol in both men and women, and 24% higher T(2) cortisol for men after adjustment for concurrent depressive/anxiety symptoms. In a non-clinical population, cumulative maltreatments in childhood were associated with flattened morning cortisol secretion in mid-adult life.
Chapter
This chapter is structured around a conceptual framework based on a research on emotion and environment-person interactions in relation to cardiovascular function and disease. It summarizes what is known about gender differences in the cardiovascular effects of specific behavioral stresses presumed to elicit emotional states. The chapter also describes gender differences in basal cardiovascular function and the effects of conditions that activate the sympathetic nervous system, including personality and social factors. There is discussion on the role of emotional experience in cardiovascular disease and survey on the putative roles of anger and hostility, depression, hopelessness, and anxiety in hypertension as well as CVD. The chapter also discusses the gender differences in the role of macro-environmental factors such as social class, employment, job strain, and social support in CVD pathogenesis and progression. Complex interactions and social systems of humans are seen to provide the conditions for eliciting or avoiding the experiential and physiologic components of emotional mechanisms that can mediate cardiovascular disease.
Article
This paper explores who uses woodlands near their homes, why they visit, what benefits they believe they obtain and what makes the difference between them choosing to visit or not. In the research, supported by the Forestry Commission, a multi-method, user-led approach was used, based on focus groups, questionnaire surveys and on-site observation in relation to five different communities in the central belt of Scotland. The conclusions demonstrate the overriding importance of childhood woodland visits as predictors of adult patterns of use. Proximity of woodlands is important for regular woodland users and freedom from rubbish is the physical quality people care most about. The physical qualities that make a difference as to whether people visit woodlands or not include directional signs, good information boards, variety of trees and tidiness of appearance. Perceptions of woodlands differ according to age and sex but are predominantly positive across all groups sampled: most people feel at peace in a woodland.
Article
We compared psychophysiological stress recovery and directed attention restoration in natural and urban field settings using repeated measures of ambulatory blood pressure, emotion, and attention collected from 112 randomly assigned young adults. To vary restoration needs, we had half of the subjects begin the environmental treatment directly after driving to the field site. The other half completed attentionally demanding tasks just before the treatment. After the drive or the tasks, sitting in a room with tree views promoted more rapid decline in diastolic blood pressure than sitting in a viewless room. Subsequently walking in a nature reserve initially fostered blood pressure change that indicated greater stress reduction than afforded by walking in the urban surroundings. Performance on an attentional test improved slightly from the pretest to the midpoint of the walk in the nature reserve, while it declined in the urban setting. This opened a performance gap that persisted after the walk. Positive affect increased and anger decreased in the nature reserve by the end of the walk; the opposite pattern emerged in the urban environment. The task manipulation affected emotional self-reports. We discuss implications of the results for theories about restorative environments and environmental health promotion measures.
Article
There is compelling evidence for feedback disturbances in the hypothalamus-pituitary-adrenal system associated with human aging as assessed by challenge tests. However, reports about age-related changes in human basal activity are ambiguous and to date little is known about changes in the pulsatile features of the HPA system. To investigate these changes we studied twenty-two healthy male and eleven healthy female subjects ranging from 23 to 85 and 24 to 81 years respectively. 24-hour blood sampling with 30 minute sampling intervals was performed. From 18.00 to 24.00 hours blood was sampled every 10 minutes for analysis of pulsatile features of HPA activity. Statistical analysis revealed that age in particular had major effects upon basal HPA-system activity: there was a significant age-associated increase in minimal (p < 0.0001) and mean (p < 0.0002) cortisol plasma concentrations, but no alteration in pulsatile features. We found no age-cortisol correlation during daytime, but were able to demonstrate a strong impact of age upon cortisol plasma levels from 20.00 to 1.30 hours. The diurnal amplitude of cortisol (p < 0.0005) and ACTH (p < 0.0006), relative to the 24-hour mean of the hormones, showed an age-associated decline. Additionally, the evening cortisol quiescent period (p < 0.0001) was shortened in the elderly, suggesting increasingly impaired circadian function in aging. Our results suggest an increased basal activity and a flattened diurnal amplitude of the HPA system in the elderly.