A growing body of evidence suggests that contact with nature, often referred to as green or blue space (ie, vegetated areas or water bodies), is associated with multiple health benefits. Recent reviews identify various mechanisms that may underlie the observed associations.1 In addition to encouraging healthy behaviors (eg, physical activity), nature exposure may have an effect on biological mechanisms by acting on the autonomous nervous system and reducing chronic stress,2 which can prevent systemic inflammation, a common cause of many noncommunicable diseases and related mortality. Nature exposure may also have direct structural and functional effects on the brain, resulting in reduced symptoms of depression (eg, rumination).3
Despite the mounting evidence, many associative studies suffer from the limitation of self-selection bias, which creates the possibility of reverse causality. For instance, property values are positively associated with the amount of green space in a neighborhood.4 Populations of high socioeconomic status, which may already have healthier lifestyles and more social advantages than less privileged groups, may agglomerate near green areas, possibly explaining the association of health with nature exposure. This bias challenges the validity of the results.
To address this problem, researchers have turned to evaluating natural experiments, where measurements of health are undertaken before and after a change in green space exposure.5 These quasi-experimental designs often include a sociodemographically matched comparison group that does not experience any landscape change. The hypothesis is usually that the intervention group who experience an increase in green space will achieve health benefits that will not occur in the comparison group. A 2016 review,6 however, raised questions about the quality of these designs, noting that many of the studies suffer from bias. Moreover, many quasi-experimental designs rely on strong a priori assumptions, such as parallel trends in the health outcome of interest between the intervention and the control group, something that is not required in studies in which individuals are randomized to different interventions. Other studies have used residential mobility to address the self-selection problem, showing some health benefits from moving from less to more green areas,7 although such studies are rare owing to the challenge of collecting accurate mobility data. Thus, even these quasi-experimental designs have several limitations compared with randomized controlled trials (RCTs).
South et al8 implement, to our knowledge, the first cluster randomized trial in general populations to assess whether a greening intervention near the residence improves mental health. By working with a nongovernmental organization focused on greening blighted vacant lots in Philadelphia, Pennsylvania, South and colleagues were able to capitalize on the timing of greening interventions so they could conduct an RCT. The trial had 3 study groups: (1) areas that experienced an increase in greenness near the home when the vacant lots were cleaned and converted to small green spaces, surrounded by a short fence; (2) areas that experienced litter cleanup but no greening; and (3) control areas that experienced neither greening nor cleaning. The researchers enrolled 442 participants in total, with roughly similar sizes of each study group. Participants were surveyed for mental health using the Kessler-6 Psychological Distress Scale both prior to the interventions and 18 months afterward.
Using intention-to-treat analyses, South and colleagues found that compared with the control group, the greening intervention group had a significant reductions in feeling depressed (−41.5%; 95% CI, −63.6% to −5.9%; P = .03) and feeling worthless (−50.9%; 95% CI, −74.7% to −4.7%; P = .04). The effects were stronger in poorer neighborhoods than in wealthier areas. They combined the 2 intervention groups in the analysis and compared them with the control group resulted in similar reductions but nonsignificant results.
The study is potentially important for several reasons. It is likely the first RCT to investigate mental health benefits of green space. In so doing, South et al8 avoid the self-selection bias limitation, which strengthens the evidence around the positive effect of green space on mental health. The greening intervention in this study specifically limited active use of the green space, as all greened lots had fences placed around them to restrict entry. This suggests an effect of simply seeing nature as opposed to interacting with nature through, for instance, physical activity. This improves the evidence that merely being visually exposed to nature may evoke critical biological responses. Because greening existing landscapes is usually less expensive than, say, constructing new urban parks, the finding may imply a cost-efficient way of improving mental health, especially in deprived neighborhoods. Additionally, the study design could be replicated in numerous cities around the world where various types of greening interventions occur.
The study design also demonstrates the reciprocal benefits when academics collaborate with governmental and nongovernmental organizations. South et al8 appear to have worked with a nongovernmental organization to stage the greening intervention in a way that they could conduct an RCT, which ultimately provides better evidence for optimizing policies and planning. For the most part, such cooperation between bodies implementing changes to the built environment and researchers remains a rarity.
The study reinforces findings from previous quasi-experimental studies and RCTs of the same data set, which show that the greening intervention also appears to reduce crime in the neighborhood.9 Whether the crime reduction results from improvements in mental health or vice versa remains an open question. Much work is still required for understanding the pathways from green space to health and the likely biological mechanisms. Such explanatory questions are perhaps answered better in structural equation models or in biomonitoring studies than in RCTs, but a crucial first step is establishing the possible causal relationship.
While the study has made an important advance, a certain limitation deserves mention. South et al8 did not compare the litter cleanup group with the control group, which raises questions about whether the effect arose from greening or a general decrease in the disarray associated with the blighted vacant lots. Photographs of the vacant lots shown in the article amplify this concern because the lots were green before but weedy and littered; hence, the greening intervention may not have increased green as much as it increased order with trash removal and a less blighted, unkempt look. We recommend that future studies should compare other control intervention study arms that do not include greening, such as the litter cleanup intervention, as this would more clearly delineate whether the greening effect was the primary reason for improvements in mental health. Different types of comparator interventions with possible effects on mental health should also be tested, which is important as the trial participants cannot be blinded to the treatment.
Notwithstanding this limitation, South et al8 have made a contribution to improving the evidence on mental health benefits of nature exposure. Beyond the direct benefits of improved mental health, greening cities holds many other cobenefits. Green spaces have been shown to reduce heat-related morbidity and mortality in urban areas,1 and when properly designed, they may also reduce local noise levels and air pollution concentrations and may improve stormwater management.10 As evidence accumulates on health benefits of greening from this study and others, investments in greening cities will become better justified, possibly providing cost-efficient solutions for dealing with current and future challenges to the health of ecosystems and people.
Published: July 20, 2018. doi:10.1001/jamanetworkopen.2018.0299
Open Access: This is an open access article distributed under the terms of the CC-BY License. © 2018 Jerrett M et al. JAMA Network Open.
Corresponding Author: Michael Jerrett, PhD, Department of Environmental Health Sciences, Center for Occupational and Environmental Health Fielding School of Public Health, University of California, Los Angeles, 650 Charles E. Young Dr S, Room 56-070 CHS, MC 177220, Los Angeles, CA 90095 (firstname.lastname@example.org).
Conflict of Interest Disclosures: Drs Jerrett and van den Bosch were supported by project grant 156152 from the Canadian Institutes of Health Research. Dr Jerrett also received support from 1R01ES027027-01A1 from the US National Institutes of Environmental Health Science.
Identify all potential conflicts of interest that might be relevant to your comment.
Conflicts of interest comprise financial interests, activities, and relationships within the past 3 years including but not limited to employment, affiliation, grants or funding, consultancies, honoraria or payment, speaker's bureaus, stock ownership or options, expert testimony, royalties, donation of medical equipment, or patents planned, pending, or issued.
Err on the side of full disclosure.
If you have no conflicts of interest, check "No potential conflicts of interest" in the box below. The information will be posted with your response.
Not all submitted comments are published. Please see our commenting policy for details.
The study design included a means to assess possible changes in neighborhood Well-Being from a Hawthorne Effect. Most interesting would be the use of the Kessler-6 Psychological Distress Scale 1-2 years after the 1 year intervention. These are very difficult to do, since Social Capital changes occurring within neighborhoods or a community have generational dimensions. As a side note, the concept of a Cluster will likely become a more useful concept within population studies. All in all, a trailhead study!
Jerrett M, van den Bosch M. Nature Exposure Gets a Boost From a Cluster Randomized Trial on the Mental Health Benefits of Greening Vacant Lots. JAMA Netw Open. 2018;1(3):e180299. doi:10.1001/jamanetworkopen.2018.0299
Customize your JAMA Network experience by selecting one or more topics from the list below.
Create a personal account or sign in to: