Community greening interventions have a positive impact on community mental health: a summary of the first ever city-wide RCT

by Jacob King, UD/MH Associate and junior doctor practising in the UK

Published in the prestigious Journal of the American Medical Association last week at long last those of us interested in green space and mental health have a city-wide experimental study, and it’s good news! (South, Hohl, Kondo, MacDonald, & Branas, 2018)

KEY FINDINGS

• Creating urban green spaces reduced feelings of depression and worthlessness
• The effect was most pronounced in people with low income
• The effect did not involve active use of the green spaces
• The effect of simply tidying but not greening the spaces had minimal impact
• The effect was achieved with small-scale greening interventions

Until now, green spaces in one’s urban environment have been shown to confer a range of mental health benefits to their local populations only in observational epidemiological studies. In the most common type of these studies, snapshots of a populations’ access to green space and their mental health are measured at the same moment in time. These methodologies are of course hindered by problems for inferring causality.

The relationship between green spaces and mental health is a hugely complex one. There are a long list of ways in which the benefits are explained. The most well-evidenced mechanisms to date are: promoting exercise and socialisation, reducing exposure to air and noise pollution, reducing stress and restoring attention, and building senses of community and place attachment . These mechanisms, among many others, are hugely complex and very difficult to adjust for in observational studies despite best efforts. A common criticism which therefore arises from observational studies is whether the effect could be caused by any one of a thousand factors associated with green spaces, which could be good for mental health. Furthermore, the observational studies so far have reported widely variable results. Some have demonstrated impressive reductions in anxiety (de Vries et al., 2016) and depressive symptomatology (Triguero-Mas et al., 2015). While others have shown virtually zero impact at all (Houlden, Weich, & Jarvis, 2017). These variable results are likely in a large part due to the many confounding factors. In response, study after study, commentary after commentary, has been crying out for experimental style studies - natural experiments or randomized control trials (RCTs) – the benefits of which allow for the single issue of interest to be studied in isolation from the disruptive noise of the complex co-factors in the relationship. In RCTs of sound methodology we can be quite confident that the results we see are due to the factor we are interested in.

Step forward Eugenia South and her colleagues from the University of Pennsylvania and their RCT set in Philadelphia.
 
Green fingered Philly
 
Initially concerned with the high burden of mental health conditions, and in light of the then fledgling evidence for green space benefit, researchers extended their work which had previously shown reductions in crime rates following neighbourhood improvement projects to consider mental health outcomes (Kondo, Hohl, Han, & Branas, 2016). By early 2013 city officials in Philadelphia had identified nearly 45,000 lots of unused vacant, often derelict brownfield land across the city. Given such an opportunity authors designed their methodology to include three study arms into which randomly selected plots, grouped together into local clusters of a 0.25 mile radius, would be allocated. The first arm would be left as they were at present. The vacant plots in the second arm would be tidied up, and the third would be “greened”. The researchers would then be able to differentiate whether the “greenness” of the spruced-up space itself contributed anything to outcomes.

Random plots were selected from the master list, and random plots also from the list and within a 0.25 mile radius were included in the cluster. To be eligible for the study, lots were to be less than 5500sqft, deemed to be abandoned, and stricken with ‘blight’, for example that there was evidence of fly-tipping (dumping), abandoned cars, or numerous police reports concerning crimes associated with the lot. In total 110 clusters were formed, containing 541 lots. Over a period of two months gardeners from the Pennsylvania Horticultural Society transformed the third of these vacant plots allocated to the “greening” arm, and tidied the third in the second arm, they will continue to maintain these lots monthly for the foreseeable future.

Before and after examples of greening the vacant lots. Used without permission of copyright holder for educational purposes. License held by JAMA network and authors.

Gardeners were instructed to follow a strict, replicable, modification process of grading the land, cleaning debris, planting grass and a small number of trees, and enclosing the space with a wooden fence with openings, in the aim of avoiding future dumping.
 
What impact on locals’ mental health?
 
In their study researchers randomly selected individuals living within the catchment areas of clusters and administered questionnaires before the intervention, and again after the intervention. There was a 77.4% success rate at interviewing the same people post-intervention at 18 months, achieving a final sample size of 342 subjects used in analysis. Questionnaires primarily consisted of key demographic information, financial status and a measure for mental health status. The short form “Kessler-6 scale” is a quick screening tool widely used for assessing poor mental health. Each question concerns a key symptom of psychological distress: nervousness, hopelessness, restlessness, depressive feelings, worthlessness, the feeling that everything is an effort, and a summary result which gives a good approximation of overall mental health and psychological distress. An annoying limitation of the short form Kessler scale is that we cannot make clinical judgments about the results: we can only identify the presence of depressive symptomatology, rather than making a diagnosis of clinical depression. However the two are of course highly related.

On to the results. Between individuals living in clusters which were greened versus those which were not, authors demonstrated significant reductions in two of the sub-categories of the K-6: depressive feelings were reduced by 41.5% and feelings of worthlessness by 50.9%(!) All other components had major drops in prevalence too. The combined figure showed impressive community wide reductions of psychological distress by 62.8% (95% CI, −86.2 to 0.4; P = 0.051). In the second arm of the study, tidying up the lots compared to no intervention produced weaker result than greening did, and while the prevalence of all psychological categories decreased, non came close to a real significance (a strong likelihood of true difference); overall psychological distress for example was reduced by 30.1% (95% CI, −74.7 to 93.2; P = 0.49).

Adapted from South et al., 2018. A table showing pre/post intervention differences in those clusters which were greened (arm 3) and those which were not altered (arm 1).

Furthermore, and importantly, in line with other studies of green spaces in local communities and mental health outcomes (Roe, Aspinall, & Ward Thompson, 2016), evidence from this study suggests that these benefits are even more pronounced for those individuals with low incomes (in this study judged to be household income under $25,000 pa). Notably, feelings of depression dropped in this sub-group by 68.7%, (−86.5 to −27.5; P < 0.01). All other aspects dropped by large amounts, but with wide confidence intervals and without strong evidence of a true difference.

In short, authors, and readers, can conclude from this presented data, that the greening interventions conducted by these gardeners notably reduced the overall number of citizens with poor mental health, and has been especially good for reducing the number of people with depressed feelings, particularly for those with low incomes.

A call to arms for communities

The evidence presented by South and her colleagues marks an important point for green space / mental health research. For the first time, this is large scale, experimental data, which provides key, and long-needed reassurance that the work of observational studies to date is replicable when the complex web of confounding factors are evaporated away.

Furthermore this study offers much to the way in which to think about green space within urban design. Especially in conjunction with this team’s previous work on the reductions in crime rates in ‘greened’ neighbourhoods, this paper adds to the conversation about the mechanisms of action of the now-undeniable benefits of green neighbourhoods to the mental health of their residents. Recent emphasis in the debate had been placed on active use of green spaces, but this study may now shift thought back towards passive or indirect observable functions of green spaces (such as attention restoration, stress reduction and protection from nuisance environmental exposures), and promotes greenery as a key facet of improving the quality of neighbourhoods, given tidying the area alone produced only marginal benefit.

Next, we must ask environmental psychologists to consider why South’s interventions delivered improvement to rather specific facets of psychiatric symptomatology (depressive feelings and worthlessness specifically: the authors propose a renewed sense of local authorities caring about their communities as a possible explanation). Hence whether specific mechanisms of green space produce specific mental health symptom benefits? In this sense, facilitating other mechanisms with other flavours of green space interventions, perhaps larger green spaces for promoting recreation, and as community foci, other facets of psychiatric symptomatology will be addressed for an overall multifaceted tackling of community psychiatric burden.

Other important areas for consideration now should be the replication of these results across other cities, with larger sample sizes, and more rigorous, clinically validated assessments.

More than ever, we should feel renewed in a community focused approach to urban (re)design. That efforts in renewing small scale (and very small scale) blighted vacant lots in our communities (some clusters only renovating 5 lots to produce such improvements in mental health) is to be of benefit. It is highly likely that these small projects are achievable for many communities. Authors further report that these initiatives are  affordable too: in their previous work, greening improvements of this kind cost on average US$1,597, plus US$180 in yearly maintenance. Local government structures can now add “improving the mental health of my community” to the long list of reasons for revitalising derelict land that is perhaps too small and financially unappealing to property developers. Otherwise, in the spirit of work which has suggested community involvement and directorship of a community’s spaces is of multifaceted benefit through building a sense of community, of place, and of stewardship, councils might look to devolve authority of these small projects to community groups themselves. When all is said and done improving mental health is not the only outcome of improving the quality of local communities, but it is a major player in an interconnected web of community, environment and health, which the work presented here by South and colleagues could more reliably inform and encourage local and national decision makers to take a little more seriously.

READ THE STUDY HERE

References

de Vries, S., ten Have, M., van Dorsselaer, S., van Wezep, M., Hermans, T., & de Graaf, R. (2016). Local availability of green and blue space and prevalence of common mental disorders in the Netherlands. British Journal of Psychiatry Open, 2(6), 366-372. doi:10.1192/bjpo.bp.115.002469

Houlden, V., Weich, S., & Jarvis, S. (2017). A cross-sectional analysis of green space prevalence and mental wellbeing in England. BMC Public Health, 17(1), 460. doi:10.1186/s12889-017-4401-x

Kondo, M., Hohl, B., Han, S., & Branas, C. (2016). Effects of greening and community reuse of vacant lots on crime. Urban Stud, 53(15), 3279-3295. doi:10.1177/0042098015608058

Roe, J., Aspinall, P. A., & Ward Thompson, C. (2016). Understanding Relationships between Health, Ethnicity, Place and the Role of Urban Green Space in Deprived Urban Communities. Int J Environ Res Public Health, 13(7). doi:10.3390/ijerph13070681

South, E. C., Hohl, B. C., Kondo, M. C., MacDonald, J. M., & Branas, C. C. (2018). Effect of greening vacant land on mental health of community-dwelling adults: A cluster randomized trial. JAMA Network Open, 1(3), e180298. doi:10.1001/jamanetworkopen.2018.0298

Triguero-Mas, M., ., Dadvand, P., Cirach, M., Martínez, D., Medina, A., Mompart, A., . . . Nieuwenhuijsen, M. J. (2015). Natural outdoor environments and mental and physical health: Relationships and mechanisms. Environment International, 77, 35-41. doi:http://dx.doi.org/10.1016/j.envint.2015.01.012

About the Author

Jacob King is a UD/MH Associate and junior doctor practising in UK. His main interest concerns the association between green space exposure and mental health, and how we can design interventions to promote this relationship.