Review

J Epidemiol Community Health: first published as 10.1136/jech-2018-210436 on 27 June 2018. Downloaded from http://jech.bmj.com/ on November 15, 2024 by guest. Protected by
Mental health benefits of interactions with nature in
children and teenagers: a systematic review
Suzanne Tillmann,1 Danielle Tobin,1 William Avison,2,3,4,5 Jason Gilliland1,3,4,5,6

►► Additional material is Abstract interpret. Therefore, there is an overwhelming need

published online only. To view Background It is commonly believed that nature has to critically review and synthesise what evidence
please visit the journal online
(http://d​ x.​doi.o​ rg/​10.​1136/​ positive impacts on children’s health, including physical, currently exists to make appropriate recommenda-
jech-​2018-​210436). mental and social dimensions. This review focuses on tions that can effectively support future research,
how accessibility to, exposure to and engagement policy and practice. Previous reviews on the rela-
1
Department of Geography, with nature affects the mental health of children and tionship between natural environments and mental
Western University, London, health have tended to lump in papers on children
teenagers.
Ontario, Canada
2
Department of Sociology, Methods Ten academic databases were used to with papers on adults.10–14 Due to the particular
Western University, London, systematically search and identify primary research objectives and/or inclusion/exclusion criteria of
Ontario, Canada papers in English or French from 1990 to 1 March these previous reviews, many relevant (and newer)
3
Department of Epidemiology 2017. Papers were included for review based on their studies with important findings for children have
and Biostatistics, Western
University, London, Ontario, incorporation of nature, children and teenagers (0–18 been overlooked. Likewise, there have been excel-
Canada years), quantitative results and focus on mental health. lent reviews focused on mental health issues among
4
Children’s Health Research Results Of the 35 papers included in the review, the unique subpopulations of youth in specific envi-
Institute, Lawson Health majority focused on emotional well-being and attention ronments (eg, indigenous youth, Arctic) that have
Research Institute, London, limited generalisability.15 Other reviews that deal
deficit disorder/hyperactivity disorder. Other outcome
Ontario, Canada
5
Department of Paediatrics, measures included overall mental health, self-esteem, with the benefits of nature for children’s health
Western University, London, stress, resilience, depression and health-related quality focus on a variety of other outcomes, such as phys-
Ontario, Canada of life. About half of all reported findings revealed ical health, rather than mental health.16–18 The
6
School of Health Studies, statistically significant positive relationships between specific objective of this review is to examine the
Western University, London,
Ontario, Canada nature and mental health outcomes and almost half evidence for all children and teenagers (birth to 18
reported no statistical significance. years) to determine how interacting with different

copyright.
Correspondence to Conclusions Findings support the contention that types of nature may benefit the mental health of
Dr Jason Gilliland, Department nature positively influences mental health; however, children and teenagers.
of Geography, Western in most cases, additional research with more rigorous
University, London, ON N6A study designs and objective measures of both nature
5K6, Canada; ​jgillila@​uwo.​ca Methods
and mental health outcomes are needed to confirm
The systematic review began with a scoping review
Received 4 January 2018 statistically significant relationships. Existing evidence is
to determine appropriate search terms related to
Revised 23 May 2018 limited by the cross-sectional nature of most papers.
Accepted 24 May 2018 nature, mental health and children and teenagers.19
Published Online First Search terms were identified by the authors and
27 June 2018 finalised by an advisory panel of subject experts
(see table 1). We used 10 bibliographic databases:
Introduction PubMed, Scopus, PsycINFO, Geobase, ProQuest,
Children’s mental health includes their emotional,
SPORTDiscus, Sociological Abstracts, Leisure and
psychological and social well-being and affects how
Tourism Database, Physical Education Index and
they reach developmental milestones, learn healthy
EMBASE. Within each database, we screened all
social skills, develop sound family and peer rela-
English and French papers published 1 January
tionships, develop a sense of identity and positive
1990 to 1 March 2017. This period represents
self-esteem and learn resilience and coping with
approximately one generation in the literature.
stress.1–7 Mental health issues developed at a young
age have the potential to persist into adulthood,
continuing the burden on the individual, family, Review process
friends and the healthcare system.8 9 While studies The review process was divided into three major
commonly examine individual-level factors (eg, steps: title screening, abstract screening and docu-
biological, socioeconomic) associated with chil- ment screening. Title screening involved reviewing
dren’s mental health outcomes, researchers are the outputs from each database search and down-
increasingly recognising the importance of external loading all titles that appeared relevant into a cita-
influences on children’s mental health, such as char- tion manager (Mendeley V.1.17.10). Of the 227 153
acteristics of their home, school and neighbourhood titles screened, 1731 documents were downloaded
environments. for further review. Abstracts of all 1731 were
This systematic review considered various then screened and 253 documents were retained
To cite: Tillmann S, forms of children’s and teenagers’ interactions which appeared to meet inclusion criteria: Popula-
Tobin D, Avison W, et al. J with nature. The evidence to support the connec- tion included children and teenagers 18 years and
Epidemiol Community Health tion between nature and children’s mental health under, Intervention incorporated an element of
2018;72:958–966. is extremely diverse, dispersed and difficult to nature, Outcome variable included a component of
958 Tillmann S, et al. J Epidemiol Community Health 2018;72:958–966. doi:10.1136/jech-2018-210436
 Review

J Epidemiol Community Health: first published as 10.1136/jech-2018-210436 on 27 June 2018. Downloaded from http://jech.bmj.com/ on November 15, 2024 by guest. Protected by
was not possible due to the heterogeneity of the methods and
Table 1 Database search terms
principal summary measures reported in the papers.
Population child* OR youth OR adolescen* OR teen* OR babies
OR infant* OR toddler* OR preschooler*
Intervention natur* OR green* OR “green space” OR greenspace OR “natural Assessing bias
environment” OR “b lue space” OR “open space” OR tree* OR Article quality was assessed using study quality assessment tools
outdoor* OR outside OR park* OR forest* OR wildlife* OR wilderness developed by the National Heart, Lung, and Blood Institute
OR wood* OR plant* OR garden* OR vegetation OR landscape OR (NHLBI) of the U.S. National Institutes of Health (NIH) Quality
playground Assessment Tools. To assess risk of bias due to study design or
Outcome “mental health” OR stress OR well-being OR “psychological well- implementation, reviewers used one of five NHLBI tools depen-
being” OR emotion* OR coping OR anxiety OR anxious OR sleep dent on study design. Two reviewers separately rated each study
OR mood OR “mood disorder” OR ADD OR ADHD OR “attention on a range of items in each tool and then considered the poten-
deficit disorder” OR autism OR depression OR schizo* OR tourettes
tial flaws (ie, item responses of ‘no’, ‘cannot determine’ and
OR “obsessive compulsive disorder” OR bipolar OR “depressive
symptoms” OR “psychological distress” OR flourishing OR ‘not reported’) to create a scale then used to judge each study
languishing OR behaviour OR behaviorbehaviour OR resiliency OR and assign an overall ranking of ‘good’, ‘fair’ or ‘poor’ quality.
self-esteem OR self-confidence Whenever there was a disagreement between two reviewers, a
third reviewer completed the rating and the majority rating was
taken. In general, a ‘good’ study has a low risk of bias and results
mental health and Study design was quantitative. Finally, the full are deemed to be valid, whereas a ‘fair’ study has weaknesses
text of all 253 retained documents was critically assessed using making it susceptible to some bias deemed not sufficient to inval-
the same inclusion/exclusion criteria as the abstract screening, idate its results. A ‘poor’ rating indicates significant risk of bias,
leaving 35 papers to be included in the systematic review. Finally, meaning results should be interpreted with caution or excluded
reference lists of all 35 papers were inspected for additional rele- from the body of evidence.
vant citations; however, this search found no new papers (see
figure 1). The review protocol was registered with PROSPERO Results
(CRD42016046085) and findings reported following PRISMA Of the 35 papers meeting eligibility criteria, 11 were conducted
guidelines. in the USA, 8 in the UK, 2 in Canada and the remaining 14 in
other countries. All papers focused on children and teens ranging
Data extraction from 9 months to 18 years of age, with early adolescence being
Relevant data from the 35 full-text articles were identified and the most commonly studied age group (see table 2).
compiled into a data extraction table. This information was All of the outcomes studied in the 35 papers were assigned

copyright.
used to create a summary of the key characteristics, outcome to 1 of 8 categories: emotional well-being (15 papers), atten-
measurement tools and findings of each study. A meta-analysis tion deficit disorder/hyperactivity disorder (ADD/ADHD) (10

Figure 1 Selection process of articles.

Tillmann S, et al. J Epidemiol Community Health 2018;72:958–966. doi:10.1136/jech-2018-210436 959
Review

J Epidemiol Community Health: first published as 10.1136/jech-2018-210436 on 27 June 2018. Downloaded from http://jech.bmj.com/ on November 15, 2024 by guest. Protected by
Table 2 Study characteristics and quality assessment of papers considering nature and mental health of children and teenagers
Ref# Ages Nature Study design
Country (years) N Element(s) of nature interaction(s) Outcome(s) (quality)
25 7–10 2111 Green Space Accessibility Emotional WB Cross-sectional*
Spain Blue Space Exposure Hyperactivity NC (G)
Greenness Inattention
Mental Health
20 4–6 1468 City Parks Accessibility Emotional WB Cross-sectional*
Lithuania Greenness Hyperactivity NC (F)
Mental Health
26 8–9 52 Nature Engagement Self-esteem Intervention*
England Orienteering NC (F)
27 13–16 53 Outdoor Engagement Mental Health Intervention*
Australia Adventure NC (P)
28 12–18 36 Wilderness Engagement Emotional WB Intervention*
Australia Therapy Mental Health NC (F)
Self-esteem
Depression
Resilience
29 12–18† 50 Horticulture Programme Engagement Self-esteem Intervention‡
U.S. C (F)
30 13–18 100 Wilderness Engagement Mental Health Quasi Empirical*
USA Therapy NC (G)
31 12–15 68 Parks Accessibility Stress Cross-sectional*
USA NC (F)
32 0.75, 3, 6348 Green Space Accessibility Emotional WB Cross-sectional‡
England 5, 7 Exposure Hyperactivity Inattention NC (F)
33 16–18 120 Outdoors Exposure Emotional WB Intervention‡
UK Attention NC (P)
34 12–15 401 Greenery Exposure Depression Longitudinal*

copyright.
Netherlands NC (F)
35 13–18 221 Wilderness Engagement Emotional WB Longitudinal
USA Therapy Mental Health Case Study*
NC (F)
36 12–15 25 Woodland Education Engagement Self-esteem Exploratory*
UK NC (F)
37 11–16 17 249 Natural Space Accessibility Emotional WB Cross-sectional*
Canada Green Space NC (F)
Blue Space
38 13–15 133 Schoolyard Exposure Emotional WB Pre-Post, Quasi-
Austria Experimental*
C (F)
39 9–11 92 Natural Accessibility HRQOL Cross-sectional‡
USA Environments NC (F)
40 5–18 452 Green Outdoor Exposure ADHD Cross-sectional‡
USA NC (P)
41 9.4–11.7 1932 Green Space Accessibility Emotional WB Cross-sectional*
Germany Hyperactivity NC (F)
Inattention
42 8–11 276 Green Space Accessibility Emotional WB Cross-sectional*
Scotland Exposure Self-esteem NC (G)
HRQOL
43 14 12 Outdoor Engagement Mental Health Intervention
Germany Adventure Stress Pilot Study*
NC (P)
44 Grade 76§ Outdoor Engagement Emotional WB Pre-Post
South 10 Adventure Education Stress Experimental*
Africa NC (G)
6 11–12 75 Park Exposure Self-esteem Counterbalanced
UK Randomised Cross Over*
C (P)
7 12–18 73 Outdoor Adventure Engagement Mental Health Intervention‡
Canada Leadership Experience Self-esteem C (G)
Resilience

Continued
960 Tillmann S, et al. J Epidemiol Community Health 2018;72:958–966. doi:10.1136/jech-2018-210436
 Review

J Epidemiol Community Health: first published as 10.1136/jech-2018-210436 on 27 June 2018. Downloaded from http://jech.bmj.com/ on November 15, 2024 by guest. Protected by
Table 2 Continued
Ref# Ages Nature Study design
Country (years) N Element(s) of nature interaction(s) Outcome(s) (quality)
45 11 18 Forest Schools Engagement Emotional WB Intervention*
UK NC (F)
46 15–18 94 Wilderness Engagement Self-esteem Intervention*
Israel Therapy C (F)
47 3–5.9 169 Outdoor Preschool Accessibility Emotional WB Cross-sectional*
Sweden Environment Exposure NC (P)
48 7–12 17 Park Exposure ADHD Single Blind
USA Control Trial*
C (G)
49 5–18 421 Grass Exposure ADD/ADHD Cross-sectional‡
USA Trees NC (P)
50 7–12 96 Greenness Trees Accessibility ADD/ADHD Cross-sectional‡
USA Grass Exposure NC (P)
51 9–17 12 Natural (Wooded) Setting Engagement Emotional WB Intervention‡
Netherlands ADHD NC (G)
52 12–18 8500 Garden Engagement Mental Health Cross-sectional*
New Depression NC (F)
Zealand
53 11–14 108 Green Space Exposure Emotional WB Cross-sectional‡
New NC (G)
Zealand
54 Grades 337 Outdoor Accessibility Stress Cross-sectional‡
USA 3–5 Yard NC (F)
21 10–15 87¶ Outdoor Engagement Resilience Intervention*
USA Adventure NC (G)
55 8–9 25 School Field Exposure Self-esteem Counterbalanced
UK Randomised Cross-over*

copyright.
NC (F)
*Study design originally mentioned in paper.
†Survey used was designed for children ages 12–18, but participant age not specified.
‡Study design assigned by reviewer.
§Males only.
¶Females only.
Study Quality Assessment: (G), Good; (F), Fair; (P), Poor.
ADD, attention deficit disorder; ADHD, attention deficit hyperactivity disorder; HRQOL, health-related quality of life; C, study design included control group; NC, no control group; WB, well-
being.

papers), overall mental health (9 papers), self-esteem (9 papers), accessibility refers to the ease of reaching destinations. Accessi-
stress (4 papers), resilience (3 papers), depression (3 papers) bility influences the likelihood a child will encounter or interact
and health-related quality of life (HRQOL) (2 papers). Online with nature, but does not necessarily equate to direct contact or
supplementary table 1 summarises the measurement tools used interaction. For example, in most studies reviewed here, acces-
in each paper, whereas online supplementary table 2 sorts the 35 sibility measures are passive and opportunity-based and tend to
papers by outcome, with the eight outcomes appearing in order be operationalised in terms of distance/proximity to one or more
according to number of papers studying that outcome. As some elements of nature or density/coverage of one or more nature
papers examined more than one outcome, the total entries in elements within an area around home. On the other hand, expo-
online supplementary table 2 is more than 35. sure can be defined as the condition of being presented to view,
Within the 35 papers, there was substantial diversity with having contact with or being subjected to some effect or influ-
respect to the specific elements of nature under consideration. ence. Exposure, therefore, implies that the child has a direct
The elements under study included green space (six papers), encounter with nature, rather than mere opportunity. Neverthe-
water/‘blue’ space (two papers), greenness/greenery (four less, in most studies reviewed here, exposure is a measure of
papers), vegetation (ie, grass, trees) (two papers), gardens (one incidental contact and is operationalised in terms of ‘time spent
paper), parks (four papers), outdoor programmes/education in/near’ or simply ‘use of ’, a natural area such as a park. Engage-
(eight papers), wilderness therapy (four papers), forest schools ment refers to involvement or participation in an activity and
(one paper) and various outdoor/natural settings (ie, school- differs from the other two categories in that it implies an interac-
yards, green outdoor settings) (nine papers). tion with nature which is more direct, intentional and sustained.
There was also considerable variation among the methods For example, the most popular form of engagement described
researchers used to assess children’s interaction with nature. in the studies reviewed here was participation in a wilderness
Despite the heterogeneity, a closer examination of study methods therapy programme for days/weeks.
allowed us to group each study into one of three broad catego- As displayed in table 3 (and online supplementary table 2),
ries we define as ‘accessibility’, ‘exposure’ and ‘engagement’ (see the 35 papers reported a total of 100 individual findings on the
table 3 for results based on nature interaction). In simple terms, relationship between children’s and teenagers’ mental health and
Tillmann S, et al. J Epidemiol Community Health 2018;72:958–966. doi:10.1136/jech-2018-210436 961
 Review

J Epidemiol Community Health: first published as 10.1136/jech-2018-210436 on 27 June 2018. Downloaded from http://jech.bmj.com/ on November 15, 2024 by guest. Protected by
Table 3 Findings by outcome, type of nature interaction and study quality
Accessibility Exposure Engagement Total
Outcome PR NS NR PR NS PR NS PR NS NR
Quality Good
 Emotional Well-being 1 3 – 2 2 2 – 5 5 –
 ADD/ADHD 2 2 – 1 2 – 1 3 5 –
 Overall Mental Health 1 – – 2 – 1 1 4 1 –
 Self–esteem – 1 – 1 – – 1 1 2 –
 Stress – – – – – 1 1 1 1 –
 Depression – – – – – – – – – –
 Resilience – – – – – 2 2 2 2 –
 HRQOL – 1 – 1 – – – 1 1 –
 Total Good 4 7 – 7 4 6 6 17 17 –
Quality Fair
 Emotional Well-being 2 5 – 1 1 2 3 5 9 –
 ADD/ADHD 3 1 – 1 – 0 0 4 1 –
 Overall Mental Health 1 – 1 – – 3 2 4 2 1
 Self-esteem – – – – 2 2 6 2 8 –
 Stress 3 – – – – – – 3 0 –
 Depression – – – – 1 2 3 2 4 –
 Resilience – – – – – 1 – 1 – –
 HRQOL 3 – – – – – – 3 – –
 Total Fair 12 6 1 2 4 10 14 24 24 1
Quality Poor
 Emotional Well–being – – – 2 1 – – 2 1 –
 ADD/ADHD 1 – – 5 – – – 6 – –

copyright.
 Overall Mental Health – – – – – 3 3 3 3 –
 Self-esteem – – – – – – – – – –
 Stress – – – – – 1 1 1 1 –
 Depression – – – – – – – – – –
 Resilience – – – – – – – – – –
 HRQOL – – – – – – – – – –
 Total Poor 1 – – 7 1 4 4 12 5 –
Total 17 13 1 16 9 20 24 53 46 1
ADD, attention deficit disorder; ADHD, attention deficit hyperactivity disorder; HRQOL, health-related quality of life; NR, nature has significant negative impact on outcome; NS, non-
significant finding; PR, nature has significant positive benefit on outcome.

nature. Over half of the findings (53 of 100) confirmed statis- of 19 findings, with 13 exhibiting statistically significant posi-
tically significant positive relationships (ie, positive benefits of tive results. Nevertheless, six findings came from studies rated
nature) (PR), whereas the remaining findings were non-signifi- poor quality. After removing poor studies, seven out of 13 find-
cant (NS). Only one paper reported a single finding suggesting ings identified a statistically significant positive relationship.
nature had negative effects on children’s mental health (NR). Increased accessibility to nature (6/9 significant positive find-
ings) and increased exposure (7/9 significant positive findings)
Emotional well-being to nature were associated with improvements in ADD/ADHD
Fifteen papers included emotional well-being as a dependent symptoms.
variable. Emotional well-being was captured through variables
such as emotional health, emotional symptoms, emotional intel- Overall mental health
ligence, mood and emotional problems. Within the 15 papers, Nine papers looked at an overall measure of children’s mental
12 findings demonstrated a significant positive relationship health. Six papers focused on how engagement with nature,
between nature and emotional well-being, whereas 15 findings through wilderness and adventure programming, can affect
were deemed non-significant. After removing two papers rated overall mental health in teenage children. Overall mental health
as poor quality, only 10 out of 23 findings identified a significant was assessed through several measures, for example psycholog-
positive relationship between nature and emotional well-being. ical well-being, psychological distress or overall mental health.
Eleven out of 18 findings within the 9 papers identified a signif-
Attention deficit disorder/hyperactivity disorder (ADD/ADHD) icant positive relationship with nature. After removing poor
Ten papers assessed nature interactions and ADD, ADHD or findings, 8 out of 12 findings identified a significant positive
symptoms related to these two disorders (hyperactivity, inatten- relationship. One study found a negative association between
tion and attention). Within the 10 papers, there were a total residential surrounding greenness and overall mental health.20
962 Tillmann S, et al. J Epidemiol Community Health 2018;72:958–966. doi:10.1136/jech-2018-210436
 Review

J Epidemiol Community Health: first published as 10.1136/jech-2018-210436 on 27 June 2018. Downloaded from http://jech.bmj.com/ on November 15, 2024 by guest. Protected by
Self-esteem studies in Europe (PR 23:NS 27:NR 1), Australia/New Zealand
Compared with other outcomes, self-esteem exhibited the most (PR 9: NS 8), Africa (PR 3: NS 1) and Asia (PR 0: NS 2).
non-significant findings compared with significant positive find-
ings. Nine papers measured the relationship between self-esteem
Risk of bias
and nature, with most focusing on nature through engagement.
The quality assessment process revealed 9 good, 19 fair and 7
Ten out of 13 findings supported a non-significant relationship;
poor papers. The majority of the findings fell within the papers
no studies were rated poor quality.
representing a fair quality assessment. After removing find-
ings from papers rated as ‘poor’, the evidence base contained
Stress 41 positive significant findings (previously 52), 41 non-signifi-
Accessibility and engagement to nature were both measured in cant results (previously 45), and 1 negative significant finding.
relationship to stress in four papers. Five out of seven findings This decrease resulted in an equal significant positive results to
found interacting with nature to be significantly positively asso- non-significant results ratio, creating an inconclusive set of find-
ciated with reduced stress. After removing all poor findings, four ings. Removing poor studies gives a more accurate picture of the
out of five findings identified a significant positive relationship. relationship between nature and the mental health of children
and teens. The majority of findings (12/17) that were removed
Depression came from studies researching the association between ADD/
The majority of findings in the three papers focusing on depres- ADHD or overall mental health and nature.
sion were non-significant, with four of six findings showing no
significant relationship with nature, and no studies were rated Discussion
poor. All three studies measured depressive symptoms through This review showed significant positive findings on the bene-
various scales. fits of nature for all mental health outcomes; however, ADD/
ADHD, overall mental health, stress, resilience and HRQOL
Resilience were the only outcomes that demonstrated more positive signif-
All three studies measuring resilience used a form of outdoor icant findings over non-significant findings. Several outcomes
programming, or engagement, to assess the relationship to (emotional well-being, self-esteem, depression) were associated
nature. Resilience was subdivided into measures of sense of with a greater number of non-significant findings than positive
mastery, relatedness and emotional reactivity. It was found significant findings, supporting the inconclusive nature of the
that adventure programmes resulted in an increase in mastery evidence reported in previous reviews.11 12 Furthermore, among
(improved self-efficacy and coping skills) and relatedness (more all studies only one finding reported a significant negative impact
comfortable interacting with others) and decrease in emotional of greenness on a subgroup of children.20 Clearly additional

copyright.
reactivity (ability to manage emotions when upset).21 Three out research is needed, with more rigorous study designs, to confirm
of five findings were found to show significant positive associa- the benefits of nature interactions and mental health outcomes.
tions between resilience and nature, no studies were rated poor. Framing the types of nature interactions in terms of ‘accessi-
bility’, ‘exposure’ and ‘engagement’ highlighted the distribution
Health-related quality of life (HRQOL) of significant positive findings. Among exposure studies, the
Two papers used HRQOL as a measure of mental health. Of larger ratio between positive-significant and non-significant find-
the five findings taken from these papers, four showed a signifi- ings suggests this type of interaction is the most beneficial and
cant positive association with nature, no studies were rated poor. may be the most effective approach for intervention strategies.
HRQOL takes into account factors influencing mental health Among accessibility studies, there was a smaller gap between
including physical, emotional, social, school, family, friends and positive-significant and non-significant findings, potentially due
self-esteem functioning. to the fact that accessibility to a particular environment does not
equate to use of that environment. Among engagement studies,
more findings were non-significant than positive-significant,
Accessibility, exposure, engagement indicating an inconclusive association between nature engage-
Engagement was the most commonly used interaction to measure ment and the mental health of children and teens; however, it
the relationship between children’s and teenagers’ mental health is noteworthy that the majority of these studies focus on less
and nature (15 papers); however, there were fewer positive healthy, ‘at risk’ populations participating in wilderness therapy
significant findings than non-significant findings for this type or outdoor adventure programmes. Further investigation needs
of nature interaction (20:24). Fourteen papers measured nature to examine how those programmes may benefit general healthy
through exposure and 11 through accessibility. The largest gap populations.
in the ratio between positive significant and non-significant find- Studies of emotional well-being, although the most studied
ings was for exposure to nature (16:9), with accessibility falling outcome, exhibited more non-significant findings than signifi-
between engagement and exposure (17:13). cant-positive findings (14:11). This calls for more rigorous inves-
tigations, as emotional well-being is critical for mental health.
Geographic differences Findings clearly demonstrate the benefits of nature interaction for
There are few geographical patterns among the findings; decreasing ADD/ADHD symptoms. This has important implica-
however, five out of the six studies set outside of North America tions for teachers implementing strategies to help children focus
and Europe dealt with children’s and teenagers’ engagement in the classroom. The holistic measure of overall mental health
with nature, whereas the type of interactions studied in North was also most commonly assessed through engagement, finding
America and Europe were more mixed among accessibility, an overall significant positive relationship; however, there was
exposure and engagement. Additionally, studies set in North considerable variation among the tools used by researchers to
America were more likely to report positive significant findings assess overall mental health. A more universal measure of overall
than non-significant findings (18:7), distinguishing them from mental health applied to accessibility, exposure and engagement
Tillmann S, et al. J Epidemiol Community Health 2018;72:958–966. doi:10.1136/jech-2018-210436 963
 Review

J Epidemiol Community Health: first published as 10.1136/jech-2018-210436 on 27 June 2018. Downloaded from http://jech.bmj.com/ on November 15, 2024 by guest. Protected by
with nature could help to clarify this relationship. The majority potential persistence of effects. The majority of the findings
of findings related to nature and self-esteem were non-signif- presented here illustrate that nature benefits children’s and teen-
icant; however, most of these studies focused on engagement, agers’ mental health. Some contradictory findings, however,
suggesting further research should investigate how other types highlight the need for greater attention on how nature’s effects
of nature interactions may impact self-esteem. Findings indicate differ between populations (ie, toddlers, adolescents). Further-
that interacting with nature can help reduce children’s and teen- more, very few studies assessed childhood depression and no
agers’ stress levels; however, these findings are based on a small studies assessed anxiety, which have more recently come to
number of studies and additional research could help confirm the the attention of public health professionals. Therefore, more
benefits. Finally, for those outcomes with few findings (depres- research on nature’s connection to these health issues is strongly
sion, resilience and HRQOL), it is difficult to interpret a rela- encouraged. The majority of studies focusing on nature engage-
tionship one way or the other. Therefore, more research needs ment target vulnerable or ‘at risk’ populations, limiting gener-
to be conducted to build on potential findings discovered here. alisability of findings; future studies should focus on healthy
All of the findings here suggest that more universal tools should populations to inform change in policy and practice more gener-
be used to measure both outcomes of mental health as well as ally. Likewise, more research is needed in regions outside North
nature interactions, in order to more confidently conclude a American and Europe, especially in less developed nations, to
relationship between the mental health of children and teenagers improve generalisability of findings. More rigorous tools are
and nature. required for measuring nature, nature interactions and mental
This review supports the application of these findings in health outcomes. Indeed, more rigorous measures would allow
various forms of policy, including municipal planning, public researchers to more robustly identify associations and causal
health and school board policies. The findings can support relationships and to better understand the potential pathways
policymakers in designing future plans as well as strengthening linking nature and positive outcomes for children’s and teen-
current policies that take into consideration the importance of agers’ mental health.22–24
natural environments. Furthermore, school boards can use these
findings to prioritise school outdoor spaces which are benefi- Conclusion
cial to the students and to the whole community. By prioritising The primary purpose of this review was to compile and eval-
investments of natural spaces at all levels of government as uate the existing evidence linking nature and the mental health
well as within school districts, children have a better chance of of children and teenagers. The results demonstrate that inter-
receiving the benefits of interacting with nature. acting with nature is positively associated with the mental health
of children and teenagers. The findings, although somewhat
Strengths and limitations

copyright.
This systematic review was comprehensive, searching 10 data-
bases resulting in 227 153 titles screened.19 Having multiple What is already known on this subject?
researchers assess abstracts and extract data added methodolog-
ical rigour. Providing a quality assessment for each paper allows ►► Nature has a significant impact on health. Previous reviews
for a more accurate assessment of the weight of the evidence. have identified the overall health effects of nature on a
Additionally, the review focused on children in general rather variety of health outcomes including physical, mental, social
than a special subgroup of children, allowing the findings to and cognitive health.
be applicable to a wider population. Finally, conceptualising ►► These reviews have largely highlighted the impact of nature
interactions with nature in terms of accessibility, exposure and on adult populations, reporting positive effects as well as
engagement was a significant advancement over previous reviews many inconclusive results.
and provides a deeper understanding as to what type, dose and ►► The current review helps to close gaps in the literature
duration of nature is required to influence change in the mental related to the impact of nature on children’s and teenagers’
health of children and teenagers. mental health.
One limitation relates to the difficulty of scoring study quality.
The subjective nature of observational studies does not allow for
a clear ‘yes’ or ‘no’ answer to some questions designed for inter-
vention based studies. Furthermore, papers based on qualitative What this study adds
methods were not included in this review due to the difficulties
of comparing findings among studies. Their inclusion may have ►► This study critically examines current literature focusing
provided for a more fulsome understanding of the benefits of on how nature influences children’s and teenagers’ mental
nature for the mental health of children and teenagers. We were health.
unable to complete a meta-analysis with the 35 studies collected ►► It presents a framework for facilitating comparisons among
due to the heterogeneity of the measures used in each study. The the heterogeneous body of literature by categorising papers
majority of the studies had fairly small sample sizes and were into one of three groups based on type of nature interaction:
from Europe, North America and other developed regions which accessibility, exposure and engagement.
can also limit the generalisability of the findings. ►► The study highlights the need for more rigorous tools to
measure nature interactions.
►► Additionally, it highlights the growth of research focusing on
Future directions
child populations in the last 5 years.
This review calls for more longitudinal studies to assess the
►► This study concludes that although the findings vary based
long-term effects that interactions with nature have on mental
on mental health outcome and type of nature interaction, it
health, as changes in mental health outcomes cannot always be
can be argued that nature does have a beneficial influence
assessed over a short period. Longitudinal studies would support
on children’s and teenagers’ mental health.
the assessment of the effects of different doses of nature and
964 Tillmann S, et al. J Epidemiol Community Health 2018;72:958–966. doi:10.1136/jech-2018-210436
 Review

J Epidemiol Community Health: first published as 10.1136/jech-2018-210436 on 27 June 2018. Downloaded from http://jech.bmj.com/ on November 15, 2024 by guest. Protected by
inconsistent and often non-significant, demonstrate the need 13 Ack L, Maheswaran R. The health benefits of urban green spaces: A review of the
for more in depth and rigorous research. Creating a more stan- evidence. J Public Health 2011;33:212–22.
14 Thompson Coon J, Boddy K, Stein K, et al. Does participating in physical activity
dardised measure for operationalising nature is necessary to in outdoor natural environments have a greater effect on physical and mental
make these findings generalisable. Understanding why there are wellbeing than physical activity indoors? A systematic review. Environ Sci Technol
differences in the findings is critical to establishing evidence- 2011;45:1761–72.
based recommendations for policy makers and planners in 15 Lehti V, Niemelä S, Hoven C, et al. Mental health, substance use and suicidal
behaviour among young indigenous people in the Arctic: a systematic review. Soc Sci
designing neighbourhoods and cities. This review identified the Med 2009;69:1194–203.
importance in promoting nature interactions to children and 16 Audrey S, Batista-Ferrer H. Healthy urban environments for children and young
teenagers in supporting their mental health. people: A systematic review of intervention studies. Health Place 2015;36:97–117.
17 Brussoni M, Gibbons R, Gray C, et al. What is the Relationship between Risky Outdoor
Acknowledgements We would like to thank the Expert Advisory Panel and Play and Health in Children? A Systematic Review. Int J Environ Res Public Health
Systematic Review Team for their guidance and assistance in the completion of this 2015;12:6423–54.
systematic review. 18 Christian H, Zubrick SR, Foster S, et al. The influence of the neighborhood physical
environment on early child health and development: A review and call for research.
Contributors JG devised and supervised the project. All authors made Health Place 2015;33:25–36.
substantial contributions to study design, including development of systematic 19 Petticrew M, Roberts H. Systematic Reviews in the Social Sciences: A Practical Guide.
review procedures. ST and DT conducted the literature search and data extraction Oxford: Blackwell Publishing, 2006.
and independently assessed the methodological quality of included studies and 20 Balseviciene B, Sinkariova L, Grazuleviciene R, et al. Impact of residential greenness
conducted analysis. JG adjudicated in any disagreements in methodological on preschool children’s emotional and behavioral problems. Int J Environ Res Public
quality assessments and contributed to analysis. ST, DT and JG drafted the original Health 2014;11:6757–70.
manuscript and WA critically reviewed the manuscript for important intellectual 21 Whittington A, Aspelmeier JE, Budbill NW. Promoting resiliency in adolescent girls
content. All authors have given approval of this final version to be published and through adventure programming. Journal of Adventure Education and Outdoor
agree to be accountable for all aspects of the work. No others fulfil the criteria for Learning 2016;16:2–15.
authorship. 22 Markevych I, Schoierer J, Hartig T, et al. Exploring pathways linking greenspace to
Funding This systematic review was made possible through funding from The health: Theoretical and methodological guidance. Environ Res 2017;158:301–17.
Lawson Foundation. ST and DT received trainee support from the Children’s Health 23 Dzhambov A, Hartig T, Markevych I, et al. Urban residential greenspace and mental
Research Institute through funding from the Children’s Health Foundation. health in youth: Different approaches to testing multiple pathways yield different
conclusions. Environ Res 2018;160:47–59.
Disclaimer The funders were not involved in the design or completion of the 24 Triguero-Mas M, Donaire-Gonzalez D, Seto E, et al. Natural outdoor
review. environments and mental health: Stress as a possible mechanism. Environ Res
Competing interests None declared. 2017;159:629–38.
25 Amoly E, Dadvand P, Forns J, et al. Green and blue spaces and behavioral
Patient consent Not required. development in Barcelona schoolchildren: the BREATHE project. Environ Health
Provenance and peer review Not commissioned; externally peer reviewed. Perspect 2014;122:1351–8.
26 Barton J, Sandercock G, Pretty J, et al. The effect of playground- and nature-based
Open access This is an open access article distributed in accordance with the

copyright.
playtime interventions on physical activity and self-esteem in UK school children. Int J
Creative Commons Attribution Non Commercial (CC BY-NC 4.0) license, which
Environ Health Res 2015;25:196–206.
permits others to distribute, remix, adapt, build upon this work non-commercially,
27 Bowen DJ, Neill JT. Effects of the PCYC Catalyst outdoor adventure intervention
and license their derivative works on different terms, provided the original work is
program on youths’ life skills, mental health, and delinquent behaviour. Int J Adolesc
properly cited, appropriate credit is given, any changes made indicated, and the use Youth 2016;21:34–55.
is non-commercial. See:http://​creativecommons.​org/​licenses/​by-​nc/​4.​0/. 28 Bowen DJ, Neill JT, Crisp SJR. Wilderness adventure therapy effects on the mental
© Article author(s) (or their employer(s) unless otherwise stated in the text of the health of youth participants. Eval Program Plann 2016;58:49–59.
article) 2018. All rights reserved. No commercial use is permitted unless otherwise 29 Cammack C, Waliczek TM, Zajicek JM. The Green Brigade: The psychological effects of
expressly granted. a community-based horticultural program on the self-development characteristics of
juvenile offenders. Horttechnology 2002;12:82–6.
30 Clark JP, Marmol LM, Cooley R, et al. The Effects of Wilderness Therapy on the Clinical
Concerns (on Axes I, II, and IV) of Troubled Adolescents. Journal of Experiential
References Education 2004;27:213–32.
1 Waddell C, Hua JM, Garland OM, et al. Preventing mental disorders in children: a 31 Feda DM, Seelbinder A, Baek S, et al. Neighbourhood parks and reduction in stress
systematic review to inform policy-making. Can J Public Health 2007;98:166–73. among adolescents: Results from Buffalo, New York. Indoor Built Environment
2 World Health Organization. Mental health: strengthening our response. Geneva: 2015;24:631–9.
World Health Organization, 2016. 32 Flouri E, Midouhas E, Joshi H. The role of urban neighbourhood green space in
3 Avison WR. Incorporating children’s lives into a life course perspective on stress and children’s emotional and behavioural resilience. J Environ Psychol 2014;40:179–86.
mental health. J Health Soc Behav 2010;51:361–75. 33 Greenwood A, Gatersleben B. Let’s go outside! Environmental restoration amongst
4 Maller C, Townsend M. Children’s mental health and wellbeing and hands-on contact adolescents and the impact of friends and phones. J Environ Psychol 2016;48:131–9.
with nature. Int J Learn 2006;12:359–72. 34 Gubbels JS, Kremers SP, Droomers M, et al. The impact of greenery on physical activity
5 Mantler A, Logan AC. Natural environments and mental health. Adv Integr Med and mental health of adolescent and adult residents of deprived neighborhoods: A
2015;2:5–12. longitudinal study. Health Place 2016;40:153–60.
6 Reed K, Wood C, Barton J, et al. A repeated measures experiment of green exercise to 35 Harper NJ, Russell KC, Cooley R, et al. Catherine Freer Wilderness Therapy
improve self-esteem in UK school children. PLoS One 2013;8:e69176. Expeditions: An Exploratory Case Study of Adolescent Wilderness Therapy, Family
7 Ritchie SD, Wabano MJ, Russell K, et al. Promoting resilience and wellbeing through Functioning, and the Maintenance of Change. Child and Youth Care Forum
an outdoor intervention designed for Aboriginal adolescents. Rural Remote Health 2007;36:111–29.
2014;14:2523. 36 Hinds J. Woodland Adventure for Marginalized Adolescents: Environmental Attitudes,
8 Bardone AM, Moffitt TE, Caspi A, et al. Adult mental health and social outcomes Identity and Competence. Applied Environmental Education & Communication
of adolescent girls with depression and conduct disorder. Dev Psychopathol 2011;10:228–37.
1996;8:811–29. 37 Huynh Q, Craig W, Janssen I, et al. Exposure to public natural space as a protective
9 Eaton WW, Martins SS, Nestadt G, et al. The burden of mental disorders. Epidemiol factor for emotional well-being among young people in Canada. BMC Public Health
Rev 2008;30:1–14. 2013;13:1–14.
10 Annerstedt M, Währborg P. Nature-assisted therapy: systematic review of controlled 38 Kelz C, Evans GW, Röderer K. The Restorative Effects of Redesigning the Schoolyard:
and observational studies. Scand J Public Health 2011;39:371–88. A Multi-Methodological, Quasi-Experimental Study in Rural Austrian Middle Schools.
11 Clark C, Myron R, Stansfeld S, et al. A systematic review of the evidence on the Environ Behav 2015;47:119–39.
effect of the built and physical environment on mental health. J Public Ment Health 39 Kim JH, Lee C, Sohn W. Urban Natural Environments, Obesity, and Health-Related
2007;6:14–27. Quality of Life among Hispanic Children Living in Inner-City Neighborhoods. Int J
12 Gascon M, Triguero-Mas M, Martínez D, et al. Mental health benefits of long-term Environ Res Public Health 2016;13:121–15.
exposure to residential green and blue spaces: a systematic review. Int J Environ Res 40 Kuo FE, Taylor AF. A potential natural treatment for attention-deficit/hyperactivity
Public Health 2015;12:4354–79. disorder: evidence from a national study. Am J Public Health 2004;94:1580–6.

Tillmann S, et al. J Epidemiol Community Health 2018;72:958–966. doi:10.1136/jech-2018-210436 965

 Review

J Epidemiol Community Health: first published as 10.1136/jech-2018-210436 on 27 June 2018. Downloaded from http://jech.bmj.com/ on November 15, 2024 by guest. Protected by
41 Markevych I, Tiesler CM, Fuertes E, et al. Access to urban green spaces and 48 Taylor AF, Kuo FE. Children with attention deficits concentrate better after walk in the
behavioural problems in children: Results from the GINIplus and LISAplus studies. park. J Atten Disord 2009;12:402–9.
Environ Int 2014;71:29–35. 49 Faber Taylor A, Kuo FEM. Could Exposure to Everyday Green Spaces Help Treat ADHD?
42 McCracken DS, Allen DA, Gow AJ. Associations between urban greenspace and Evidence from Children’s Play Settings. Appl Psychol 2011;3:281–303.
health-related quality of life in children. Prev Med Rep 2016;3:211–21. 50 Taylor AF, Kuo FE, Sullivan WC. Coping with ADD. The Surprising Connection to Green
43 Mutz M, Müller J. Mental health benefits of outdoor adventures: Results from two Play Settings. Environ Behav 2001;33:54–77.
pilot studies. J Adolesc 2016;49:105–14. 51 van den Berg AE, van den Berg CG. A comparison of children with ADHD in a natural
44 Opper B, Maree JG, Fletcher L, et al. Efficacy of outdoor adventure education in and built setting. Child Care Health Dev 2011;37:430–9.
developing emotional intelligence during adolescence. Journal of Psychology in Africa 52 van Lier LE, Utter J, Denny S, et al. Home Gardening and the Health and Well-Being of
2014;24:193–6. Adolescents. Health Promot Pract 2016;18:34–43.
45 Roe J, Aspinall P. The restorative outcomes of forest school and conventional 53 Ward JS, Duncan JS, Jarden A, et al. The impact of children’s exposure to greenspace
school in young people with good and poor behaviour. Urban For Urban Green on physical activity, cognitive development, emotional wellbeing, and ability to
2011;10:205–12. appraise risk. Health Place 2016;40:44–50.
46 Romi S, Kohan E. Wilderness Programs: Principles, Possibilities and Opportunities for 54 Wells NM, Evans GW. Nearby Nature: A Buffer of Life Stress among Rural Children.
Intervention with Dropout Adolescents. Child Youth Care Forum 2004;33:115–36. Environ Behav 2003;35:311–30.
47 Söderström M, Boldemann C, Sahlin U, et al. The quality of the outdoor environment 55 Wood C, Gladwell V, Barton J. A repeated measures experiment of school playing
influences childrens health -- A cross-sectional study of preschools. Acta Paediatr environment to increase physical activity and enhance self-esteem in UK school
2013;102:83–91. children. PLoS One 2014;9:e108701.

copyright.

966 Tillmann S, et al. J Epidemiol Community Health 2018;72:958–966. doi:10.1136/jech-2018-210436