[Skip to Content]
Sign In
Individual Sign In
Create an Account
Institutional Sign In
OpenAthens Shibboleth
[Skip to Content Landing]
Figure 1.
Distribution of Study Vacant Lots Across Philadelphia, Pennsylvania
Distribution of Study Vacant Lots Across Philadelphia, Pennsylvania

This map shows the distribution of randomly selected study vacant lots across 3 groups of the trial: the greening intervention, the trash cleanup intervention, and no intervention. The distribution of vacant lots shown is representative of those in the study, although for the purposes of confidentiality are not the locations of actual study lots.

Figure 2.
Vacant Lot Main Greening Intervention
Vacant Lot Main Greening Intervention

Images show blighted preperiod conditions and remediated postperiod restorations. A, The image shows the grass seeding method used to rapidly complete the treatment process. B, The after image shows the low wooden perimeter fence. Vacant lots shown here are representative of those in the study, although for purposes of confidentiality are not actual study lots.

Figure 3.
Flowchart of Vacant Lots and Participants Through Vacant Lot Greening Trial
Flowchart of Vacant Lots and Participants Through Vacant Lot Greening Trial

aVacant lots were classified as blighted if they (1) had existing violations signaling blight, including illegal dumping, abandoned cars, and/or unmanaged vegetation growth; and (2) had been abandoned, as confirmed through contact with the owner of record who, within a 10-day period, either authorized the intervention or did not reply. Those excluded as having insufficient blight or not confirmed as abandoned did not meet these conditions.

Table 1.  
Baseline Characteristics Demonstrating Balance Across Study Groupsa
Baseline Characteristics Demonstrating Balance Across Study Groupsa
Table 2.  
Intention-to-Treat Analyses of Vacant Lot Interventions and Self-reported Mental Health Outcomes
Intention-to-Treat Analyses of Vacant Lot Interventions and Self-reported Mental Health Outcomes
1.
Murray  CJL, Atkinson  C, Bhalla  K,  et al; US Burden of Disease Collaborators.  The state of US health, 1990-2010: burden of diseases, injuries, and risk factors.  JAMA. 2013;310(6):591-608. doi:10.1001/jama.2013.13805PubMedGoogle ScholarCrossref
2.
US Center for Behavioral Health Statistics and Quality. Key substance use and mental health indicators in the United States: results from the 2015 National Survey on Drug Use and Health. http://www.samhsa.gov/data/. Accessed November 13, 2017.
3.
Kessler  RC, Berglund  P, Demler  O, Jin  R, Merikangas  KR, Walters  EE.  Lifetime prevalence and age-of-onset distributions of DSM-IV disorders in the National Comorbidity Survey Replication.  Arch Gen Psychiatry. 2005;62(6):593-602. doi:10.1001/archpsyc.62.6.593PubMedGoogle ScholarCrossref
4.
Kamal  R, Cox  C, Rousseau  D; Kaiser Family Foundation.  Costs and outcomes of mental health and substance use disorders in the US.  JAMA. 2017;318(5):415. doi:10.1001/jama.2017.8558PubMedGoogle ScholarCrossref
5.
Mair  C, Diez Roux  AV, Morenoff  JD.  Neighborhood stressors and social support as predictors of depressive symptoms in the Chicago Community Adult Health Study.  Health Place. 2010;16(5):811-819. doi:10.1016/j.healthplace.2010.04.006PubMedGoogle ScholarCrossref
6.
Latkin  CA, Curry  AD.  Stressful neighborhoods and depression: a prospective study of the impact of neighborhood disorder.  J Health Soc Behav. 2003;44(1):34-44. doi:10.2307/1519814PubMedGoogle ScholarCrossref
7.
Hill  TD, Ross  CE, Angel  RJ.  Neighborhood disorder, psychophysiological distress, and health.  J Health Soc Behav. 2005;46(2):170-186. doi:10.1177/002214650504600204PubMedGoogle ScholarCrossref
8.
Mair  C, Diez Roux  AV, Galea  S.  Are neighbourhood characteristics associated with depressive symptoms? a review of evidence.  J Epidemiol Community Health. 2008;62(11):940-946.PubMedGoogle Scholar
9.
Mair  C, Diez Roux  AV, Shen  M,  et al.  Cross-sectional and longitudinal associations of neighborhood cohesion and stressors with depressive symptoms in the Multiethnic Study of Atherosclerosis.  Ann Epidemiol. 2009;19(1):49-57. doi:10.1016/j.annepidem.2008.10.002PubMedGoogle ScholarCrossref
10.
Hoebel  J, Maske  UE, Zeeb  H, Lampert  T.  Social inequalities and depressive symptoms in adults: the role of objective and subjective socioeconomic status.  PLoS One. 2017;12(1):e0169764. doi:10.1371/journal.pone.0169764PubMedGoogle ScholarCrossref
11.
Garvin  E, Branas  C, Keddem  S, Sellman  J, Cannuscio  C.  More than just an eyesore: local insights and solutions on vacant land and urban health.  J Urban Health. 2013;90(3):412-426. doi:10.1007/s11524-012-9782-7PubMedGoogle ScholarCrossref
12.
South  EC, Kondo  MC, Cheney  RA, Branas  CC.  Neighborhood blight, stress, and health: a walking trial of urban greening and ambulatory heart rate.  Am J Public Health. 2015;105(5):909-913. doi:10.2105/AJPH.2014.302526PubMedGoogle ScholarCrossref
13.
James  P, Banay  RF, Hart  JE, Laden  F.  A review of the health benefits of greenness.  Curr Epidemiol Rep. 2015;2(2):131-142. doi:10.1007/s40471-015-0043-7PubMedGoogle ScholarCrossref
14.
Seymour  V.  The human-nature relationship and its impact on health: a critical review.  Front Public Health. 2016;4:260. doi:10.3389/fpubh.2016.00260PubMedGoogle ScholarCrossref
15.
Beyer  KM, Kaltenbach  A, Szabo  A, Bogar  S, Nieto  FJ, Malecki  KM.  Exposure to neighborhood green space and mental health: evidence from the Survey of the Health of Wisconsin.  Int J Environ Res Public Health. 2014;11(3):3453-3472. doi:10.3390/ijerph110303453PubMedGoogle ScholarCrossref
16.
McEachan  RRC, Prady  SL, Smith  G,  et al The association between green space and depressive symptoms in pregnant women: moderating roles of socioeconomic status and physical activity.  J Epidemiol Community Health. 2016:70(3):253-259. doi:10.1136/jech-2015-205954PubMedGoogle ScholarCrossref
17.
Wu  Y-T, Prina  AM, Jones  A, Matthews  FE, Brayne  C; The Medical Research Council Cognitive Function and Ageing Studies.  Older people, the natural environment and common mental disorders: cross-sectional results from the Cognitive Function and Ageing Study.  BMJ Open. 2015;5(9):e007936. doi:10.1136/bmjopen-2015-007936PubMedGoogle ScholarCrossref
18.
Roe  JJ, Thompson  CW, Aspinall  PA,  et al.  Green space and stress: evidence from cortisol measures in deprived urban communities.  Int J Environ Res Public Health. 2013;10(9):4086-4103. doi:10.3390/ijerph10094086PubMedGoogle ScholarCrossref
19.
Morita  E, Fukuda  S, Nagano  J,  et al.  Psychological effects of forest environments on healthy adults: Shinrin-yoku (forest-air bathing, walking) as a possible method of stress reduction.  Public Health. 2007;121(1):54-63. doi:10.1016/j.puhe.2006.05.024PubMedGoogle ScholarCrossref
20.
Cox  DTC, Shanahan  DF, Hudson  HL,  et al.  Doses of nearby nature simultaneously associated with multiple health benefits.  Int J Environ Res Public Health. 2017;14(2):172. doi:10.3390/ijerph14020172PubMedGoogle ScholarCrossref
21.
Shanahan  DF, Bush  R, Gaston  KJ,  et al.  Health benefits from nature experiences depend on dose.  Sci Rep. 2016;6:28551. doi:10.1038/srep28551PubMedGoogle ScholarCrossref
22.
Mitchell  R, Popham  F.  Effect of exposure to natural environment on health inequalities: an observational population study.  Lancet. 2008;372(9650):1655-1660. doi:10.1016/S0140-6736(08)61689-XPubMedGoogle ScholarCrossref
23.
Hartig  T.  Green space, psychological restoration, and health inequality.  Lancet. 2008;372(9650):1614-1615. doi:10.1016/S0140-6736(08)61669-4PubMedGoogle ScholarCrossref
24.
Mitchell  RJ, Richardson  EA, Shortt  NK, Pearce  JR.  Neighborhood environments and socioeconomic inequalities in mental well-being.  Am J Prev Med. 2015;49(1):80-84. doi:10.1016/j.amepre.2015.01.017PubMedGoogle ScholarCrossref
25.
Frieden  TR.  A framework for public health action: the health impact pyramid.  Am J Public Health. 2010;100(4):590-595. doi:10.2105/AJPH.2009.185652PubMedGoogle ScholarCrossref
26.
Gong  Y, Palmer  S, Gallacher  J, Marsden  T, Fone  D.  A systematic review of the relationship between objective measurements of the urban environment and psychological distress.  Environ Int. 2016;96:48-57. doi:10.1016/j.envint.2016.08.019PubMedGoogle ScholarCrossref
27.
Branas  CC, Cheney  RA, MacDonald  JM, Tam  VW, Jackson  TD, Ten Have  TR.  A difference-in-differences analysis of health, safety, and greening vacant urban space.  Am J Epidemiol. 2011;174(11):1296-1306. doi:10.1093/aje/kwr273PubMedGoogle ScholarCrossref
28.
Garvin  EC, Cannuscio  CC, Branas  CC.  Greening vacant lots to reduce violent crime: a randomised controlled trial.  Inj Prev. 2013;19(3):198-203. doi:10.1136/injuryprev-2012-040439PubMedGoogle ScholarCrossref
29.
Branas  CC, Kondo  MC, Murphy  SM, South  EC, Polsky  D, MacDonald  JM.  Urban blight remediation as a cost-beneficial solution to firearm violence.  Am J Public Health. 2016;106(12):2158-2164. doi:10.2105/AJPH.2016.303434PubMedGoogle ScholarCrossref
30.
Branas  CC, South  E, Kondo  MC,  et al.  Citywide cluster randomized trial to restore blighted vacant land and its effects on violence, crime, and fear.  Proc Natl Acad Sci U S A. 2018;115(12):2946-2951. doi:10.1073/pnas.1718503115PubMedGoogle ScholarCrossref
31.
Campbell  MK, Piaggio  G, Elbourne  DR, Altman  DG; CONSORT Group.  CONSORT 2010 statement: extension to cluster randomised trials.  BMJ. 2012;345:e5661. doi:10.1136/bmj.e5661PubMedGoogle ScholarCrossref
32.
Boruch  R, May  H, Turner  H,  et al.  Estimating the effects of interventions that are deployed in many places: place-randomized trials.  Am Behav Sci. 2004;47(5):608-633. doi:10.1177/0002764203259291Google ScholarCrossref
33.
Schulz  KF, Grimes  DA.  Generation of allocation sequences in randomised trials: chance, not choice.  Lancet. 2002;359(9305):515-519. doi:10.1016/S0140-6736(02)07683-3PubMedGoogle ScholarCrossref
34.
American Association for Public Opinion Research. AAPOR response rate calculator. https://www.aapor.org/Education-Resources/For-Researchers/Poll-Survey-FAQ/Response-Rates-An-Overview.aspx. Accessed November 27, 2017.
35.
Groves  RM.  Nonresponse rates and nonresponse bias in household surveys.  Public Opin Q. 2006;70(5):646-675. doi:10.1093/poq/nfl033Google ScholarCrossref
36.
Galea  S, Tracy  M.  Participation rates in epidemiologic studies.  Ann Epidemiol. 2007;17(9):643-653. doi:10.1016/j.annepidem.2007.03.013PubMedGoogle ScholarCrossref
37.
Keeter  S, Kennedy  C, Dimock  M, Best  J, Craighill  P.  Gauging the impact of growing nonresponse on estimates from a national RDD telephone survey.  Public Opin Q. 2006;70(5):759-779. doi:10.1093/poq/nfl035Google ScholarCrossref
38.
Kessler  RC, Andrews  G, Colp  LJ,  et al Short screening scales to monitor population prevalences and trends in non-specific psychological distress.  Psychol Med. 2002;32(6):959-976. doi:10.1017/S0033291702006074PubMedGoogle ScholarCrossref
39.
Kessler  RC, Barker  PR, Colpe  LJ,  et al.  Screening for serious mental illness in the general population.  Arch Gen Psychiatry. 2003;60(2):184-189. doi:10.1001/archpsyc.60.2.184PubMedGoogle ScholarCrossref
40.
Cohen  J.  Statistical Power for the Behavioral Sciences. 2nd ed. Hillsdale, NJ: Lawrence Erlbaum Associates; 1988.
41.
Meyer  BD.  Natural and quasi-experiments in economics.  J Bus Econ Stat. 1995;13(2):151-161. doi:10.1080/07350015.1995.10524589Google Scholar
42.
US Department of Health and Human Services, Assistant Secretary for Planning and Evaluation. Poverty guidelines. https://aspe.hhs.gov/2013-poverty-guidelines#thresholds. Accessed April 20, 2018.
43.
Branas  CC, Macdonald  JM.  A simple strategy to transform health, all over the place.  J Public Health Manag Pract. 2014;20(2):157-159. doi:10.1097/PHH.0000000000000051PubMedGoogle ScholarCrossref
44.
Kondo  MC, South  EC, Branas  CC.  Nature-based strategies for improving urban health and safety.  J Urban Health. 2015;92(5):800-814. doi:10.1007/s11524-015-9983-yPubMedGoogle ScholarCrossref
45.
Woolf  SH, Braveman  P.  Where health disparities begin: the role of social and economic determinants—and why current policies may make matters worse.  Health Aff (Millwood). 2011;30(10):1852-1859. doi:10.1377/hlthaff.2011.0685PubMedGoogle ScholarCrossref
46.
Kuo  FE, Sullivan  WC.  Aggression and violence in the inner city: effects of environment via mental fatigue.  Sage Journals. 2001;33(4):543-571. doi:10.1177/00139160121973124Google Scholar
47.
Kaplan  S.  The restorative benefits of nature: toward an integrative framework.  J Environ Psychol. 1995;15(3):169-182. doi:10.1016/0272-4944(95)90001-2Google ScholarCrossref
48.
Hartig  T, Evans  GW, Jamner  LD, Davis  DS, Gärling  T.  Tracking restoration in natural and urban field settings.  J Environ Psychol. 2003;23(2):109-123. doi:10.1016/S0272-4944(02)00109-3Google ScholarCrossref
49.
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(3):221-229. doi:10.1016/j.landurbplan.2011.12.015Google ScholarCrossref
50.
Cohen  DA, Inagami  S, Finch  B.  The built environment and collective efficacy.  Health Place. 2008;14(2):198-208. doi:10.1016/j.healthplace.2007.06.001PubMedGoogle ScholarCrossref
51.
Maas  J, van Dillen  SME, Verheij  RA, Groenewegen  PP.  Social contacts as a possible mechanism behind the relation between green space and health.  Health Place. 2009;15(2):586-595. doi:10.1016/j.healthplace.2008.09.006PubMedGoogle ScholarCrossref
52.
Kweon  B-S, Sullivan  WC, Wiley  AR.  Green common spaces and the social integration of inner-city older adults.  Sage Journals. 1998;30(6):832-858. doi:10.1177/001391659803000605Google Scholar
53.
Kuo  F, Sullivan  W, Coley  R, Brunson  L.  Fertile ground for community: inner-city neighborhood common spaces.  Am J Community Psychol. 1998;26(6):823-851. doi:10.1023/A:1022294028903Google ScholarCrossref
54.
Sullivan  WC.  The fruit of urban nature: vital neighborhood spaces.  Sage Journals. 2004;36(5):678-700. doi:10.1177/0193841X04264945Google Scholar
55.
Fone  D, White  J, Farewell  D,  et al.  Effect of neighbourhood deprivation and social cohesion on mental health inequality: a multilevel population-based longitudinal study.  Psychol Med. 2014;44(11):2449-2460. doi:10.1017/S0033291713003255PubMedGoogle ScholarCrossref
56.
Kruger  DJ, Reischl  TM, Gee  GC.  Neighborhood social conditions mediate the association between physical deterioration and mental health.  Am J Community Psychol. 2007;40(3-4):261-271. doi:10.1007/s10464-007-9139-7PubMedGoogle ScholarCrossref
57.
Fone  D, Dunstan  F, Lloyd  K, Williams  G, Watkins  J, Palmer  S.  Does social cohesion modify the association between area income deprivation and mental health? a multilevel analysis.  Int J Epidemiol. 2007;36(2):338-345. doi:10.1093/ije/dym004PubMedGoogle ScholarCrossref
58.
de Vries  S, van Dillen  SME, Groenewegen  PP, Spreeuwenberg  P.  Streetscape greenery and health: stress, social cohesion and physical activity as mediators.  Soc Sci Med. 2013;94:26-33. doi:10.1016/j.socscimed.2013.06.030PubMedGoogle ScholarCrossref
59.
Triguero-Mas  M, Dadvand  P, Cirach  M,  et al.  Natural outdoor environments and mental and physical health: relationships and mechanisms.  Environ Int. 2015;77:35-41. doi:10.1016/j.envint.2015.01.012PubMedGoogle ScholarCrossref
60.
Sugiyama  T, Leslie  E, Giles-Corti  B, Owen  N.  Associations of neighbourhood greenness with physical and mental health: do walking, social coherence and local social interaction explain the relationships?  J Epidemiol Community Health. 2008;62(5):e9. doi:10.1136/jech.2007.064287PubMedGoogle ScholarCrossref
61.
Stafford  M, Chandola  T, Marmot  M.  Association between fear of crime and mental health and physical functioning.  Am J Public Health. 2007;97(11):2076-2081. doi:10.2105/AJPH.2006.097154PubMedGoogle ScholarCrossref
62.
Heckert  M, Kondo  M. Can “cleaned and greened” lots take on the role of public greenspace?  J Plan Educ Res. 2017;38(2):211-221. doi:10.1177/0739456X16688766.Google ScholarCrossref
Limit 200 characters
Limit 25 characters
Conflicts of Interest Disclosure

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.

Limit 140 characters
Limit 3600 characters or approximately 600 words
    3 Comments for this article
    EXPAND ALL
    Good Local Example of Planetary Health
    Roberto Almeida, MD | UNILA, Federal University of Latin American Integration

    Planetary health deals with the effects of human species on nature (the Anthropocene Age) and the reverse impact of nature on human health. In the case of this article, the urbanization process in some cities generates degraded environments, and this affects the mental health that in turn produces more carelessness with the natural environment. This article shows a light in the implementation of planetary health.

    CONFLICT OF INTEREST: None Reported
    Another Example of the Effects of Greening
    Frederick Rivara, MD, MPH | University of Washington
    I love this study because it is another example of what greening of vacant lots in urban environments can accomplish. Prior work by this group has shown that greening of vacant lots can reduce violent crime. Local groups- get busy and do this!
    CONFLICT OF INTEREST: Editor in Chief, JAMA Network Open
    Tidiness Effect or Greening Effect?
    Carson Yap, M.D. | Chang Gung Memorial Hospital

    I hope the authors can comment on the question "Is this the tidiness effect or greening effect?", especially when the demo photo shows that the vacant lot before the intervention is pretty green to me, although much less tidy compared to after intervention.

    1. Effect of "greening intervention" should be compared to that of "cleanup intervention" in order to attribute the effect to solely "greening" effect.

    2. From the illustration shown, it makes me think that the difference between the two intervention is actually "the size of the already green area being cleanup" and not "greening or not greening" per se., i.e. the area of the green space could have been an effect modifier on cleanup, and not a independent variable. If this the case, policy makers should not only provide the "greening intervention" but more importantly to maintain the tidiness of those GREEN areas, in order to gain such effects.

    3. It would be better to show two sample picture of the same vacant area illustrating how they were different after underwent the two different intervention; albeit "hypothetical", it gives the readers clearer view on the difference of the two intervention. For example, I'm especially curious about how it would look like if "Figure 2A" was managed by the "cleanup" intervention, how different would it be from the "greening" intervention?

    Thank you

    CONFLICT OF INTEREST: None Reported
    READ MORE
    Original Investigation
    Public Health
    July 20, 2018

    Effect of Greening Vacant Land on Mental Health of Community-Dwelling AdultsA Cluster Randomized Trial

    Author Affiliations
    • 1Department of Emergency Medicine, Perelman School of Medicine, University of Pennsylvania, Philadelphia
    • 2Center for Emergency Care Policy and Research, Perelman School of Medicine, University of Pennsylvania, Philadelphia
    • 3Department of Epidemiology, School of Public Health, School of Criminal Justice, Rutgers University, Newark, New Jersey
    • 4Northern Research Station, Forest Service, US Department of Agriculture, Philadelphia, Pennsylvania
    • 5Department of Criminology, School of Arts and Sciences, University of Pennsylvania, Philadelphia
    • 6Department of Epidemiology, Mailman School of Public Health, Columbia University, New York, New York
    • 7Department of Biostatistics and Epidemiology, Perelman School of Medicine, University of Pennsylvania, Philadelphia
    JAMA Network Open. 2018;1(3):e180298. doi:10.1001/jamanetworkopen.2018.0298
    Key Points

    Question  Does the greening of vacant urban land reduce self-reported poor mental health in community-dwelling adults?

    Findings  In this cluster randomized trial of urban greening and mental health, 110 randomly sampled vacant lot clusters were randomly assigned to 3 study groups. Among 342 participants included in the analysis, feeling depressed significantly decreased by 41.5% and self-reported poor mental health showed a reduction of 62.8% for those living near greened vacant lots compared with control participants.

    Meaning  The remediation of vacant and dilapidated physical environments, particularly in resource-limited urban settings, can be an important tool for communities to address mental health problems, alongside other patient-level treatments.

    Abstract

    Importance  Neighborhood physical conditions have been associated with mental illness and may partially explain persistent socioeconomic disparities in the prevalence of poor mental health.

    Objective  To evaluate whether interventions to green vacant urban land can improve self-reported mental health.

    Design, Setting, and Participants  This citywide cluster randomized trial examined 442 community-dwelling sampled adults living in Philadelphia, Pennsylvania, within 110 vacant lot clusters randomly assigned to 3 study groups. Participants were followed up for 18 months preintervention and postintervention. This trial was conducted from October 1, 2011, to November 30, 2014. Data were analyzed from July 1, 2015, to April 16, 2017.

    Interventions  The greening intervention involved removing trash, grading the land, planting new grass and a small number of trees, installing a low wooden perimeter fence, and performing regular monthly maintenance. The trash cleanup intervention involved removal of trash, limited grass mowing where possible, and regular monthly maintenance. The control group received no intervention.

    Main Outcomes and Measures  Self-reported mental health measured by the Kessler-6 Psychological Distress Scale and the components of this scale.

    Results  A total of 110 clusters containing 541 vacant lots were enrolled in the trial and randomly allocated to the following 1 of 3 study groups: the greening intervention (37 clusters [33.6%]), the trash cleanup intervention (36 clusters [32.7%]), or no intervention (37 clusters [33.6%]). Of the 442 participants, the mean (SD) age was 44.6 (15.1) years, 264 (59.7%) were female, and 194 (43.9%) had a family income less than $25 000. A total of 342 participants (77.4%) had follow-up data and were included in the analysis. Of these, 117 (34.2%) received the greening intervention, 107 (31.3%) the trash cleanup intervention, and 118 (34.5%) no intervention. Intention-to-treat analysis of the greening intervention compared with no intervention demonstrated a significant decrease in participants who were feeling depressed (−41.5%; 95% CI, −63.6% to −5.9%; P = .03) and worthless (−50.9%; 95% CI, −74.7% to −4.7%; P = .04), as well as a nonsignificant reduction in overall self-reported poor mental health (−62.8%; 95% CI, −86.2% to 0.4%; P = .051). For participants living in neighborhoods below the poverty line, the greening intervention demonstrated a significant decrease in feeling depressed (−68.7%; 95% CI, −86.5% to −27.5%; P = .007). Intention-to-treat analysis of those living near the trash cleanup intervention compared with no intervention showed no significant changes in self-reported poor mental health.

    Conclusions and Relevance  Among community-dwelling adults, self-reported feelings of depression and worthlessness were significantly decreased, and self-reported poor mental health was nonsignificantly reduced for those living near greened vacant land. The treatment of blighted physical environments, particularly in resource-limited urban settings, can be an important treatment for mental health problems alongside other patient-level treatments.

    Trial Registration  isrctn.org Identifier: ISRCTN92582209

    Introduction

    Almost 1 in 5 US adults report some form of mental illness. Depression is the second largest contributor to years lived with disability in the United States,1 with more than 16 million adults experiencing an episode annually.2,3 Yet patient mental health services only account for an estimated 5% of total medical care spending in the United States.4 A broadening of treatment options to improve mental health is necessary, including interventions that fundamentally change harmful environmental surroundings that may be key contributors to mental illness.

    Neighborhood physical conditions, including vacant or dilapidated spaces, trash, and lack of quality infrastructure such as sidewalks and parks, are associated with depression5-9 and are factors that may explain the persistent prevalence of mental illness in resource-limited communities.10 Vacant and dilapidated spaces are unavoidable neighborhood conditions that residents in low-resource communities encounter every day, making the very existence of these spaces a constant source of stress11,12 and possibly mental illness.

    However, neighborhood physical conditions can also positively influence mental health.13,14 Spending time and living near green spaces have been associated with various improved mental health outcomes, including less depression, anxiety, and stress.15-19 Several studies have demonstrated a dose-response relationship between more time spent in green spaces and lower depression rates.20,21 Therefore, green space may be a potential buffer between inequitable neighborhood conditions and poor mental health outcomes.22-24

    While patient-level therapies for mental illness will always be a vital aspect of treatment, changing the places where people live, work, and play may have broad population-level effects on mental health outcomes.25 There have been calls for the development of urban environmental interventions to improve mental health outcomes and well-being.1,26 In support of this, a number of observational studies have demonstrated the positive effect of vacant land greening interventions on urban health, crime, and stress.12,27-29 However, these prior studies have not been experimental and have not tested mental health outcomes. Given this, we evaluated data from, to our knowledge, the first citywide cluster randomized trial with the objective to test the effects of inexpensive, standardized, and reproducible vacant land remediation interventions—greening and trash cleanup—on health and safety. We report here on the mental health outcomes. Analysis of crime outcomes is reported elsewhere.30

    Methods
    Study Design

    This citywide cluster randomized trial of a standardized, reproducible vacant lot greening intervention and vacant lot trash cleanup intervention was conducted in Philadelphia, Pennsylvania. The University of Pennsylvania institutional review board approved this trial. All participants provided written informed consent. All sections of this article were written using the Consolidated Standards of Reporting Trials (CONSORT) reporting guideline.31The trial protocol can be found in the Supplement.

    Vacant Lot Random Sampling and Random Assignment

    A master list was compiled of all vacant lots citywide available from the city administrative records throughout January 2011. Vacant lots that were authorized by municipal ordinance as blighted and eligible for the intervention were randomly sampled for the trial. Eligible lots were included if they specifically (1) had existing violations signaling blight, including illegal dumping, abandoned cars, and/or unmanaged vegetation growth; and (2) had been abandoned, as confirmed through contact with the owner of record who, within a 10-day period, either authorized the intervention or did not reply. Owners included the city itself for publicly owned lots. We excluded lots that had insufficient blight or lack of abandonment, lots that were greater than 5500 sq ft, and lots that were fully paved parking lots.

    Vacant lot clusters served as the intervention unit for the study. To form these clusters, the master list of eligible vacant lots was ordered based on the assignment of random numbers within 4 sections of the city.32 In each section of the city, the first vacant lot in the randomly ordered list was chosen as an index lot and a 0.25-mile radius buffer was created around that lot. All other eligible vacant lots on the master list that fell within this radius were used to form a cluster grouping of geographically proximal vacant lots that summed between 4500 to 5500 total sq ft; these lots were then removed from consideration as future index lots. This process then cycled to the next randomly ordered index lot on the list that was at least 0.25 miles away from the edge of prior clusters until all clusters were formed. This process guaranteed that no clusters overlapped, reducing potential spillover and contamination effects across trial arms.

    Within each city section, clusters were randomly assigned to 1 of 3 study groups—the greening intervention, trash cleanup intervention, or no intervention (Figure 1). A repeated randomization procedure33 was used under a predetermined protocol that permitted repeated random allocation of the 3 study groups until a statistically significant balance was achieved with a set of potential confounding variables, including the total area and mean separating distance of the study vacant lots, the total vacant lots, resident population, and number of serious crimes (part I violent and property crimes), in each cluster.

    Vacant Lot Interventions and Control Group

    The vacant lot greening intervention involved the cleaning and greening of vacant lots via a standard, reproducible process of removing trash and debris, grading the land, planting new grass and a small number of trees, installing a low wooden perimeter fence with openings, and performing regular maintenance (Figure 2). The vacant lot trash cleanup intervention group involved removal of trash and debris, limited grass mowing on the lot where possible, and regular maintenance. The Pennsylvania Horticultural Society designed and carried out the interventions over a 2-month period, from April 1, 2013, to May 31, 2013, followed by monthly maintenance. At the end of the postintervention period, vacant lots assigned to the control condition were scheduled for cleaning and greening.

    Random Sampling of Participants

    Two preintervention interview survey waves were conducted from October 1, 2011, to March 31, 2013, and 2 postintervention survey waves were conducted from June 1, 2013, to November 30, 2014, with a sample of residents from each cluster. All participants completed at least 1 preintervention survey and 1 postintervention survey. The outer-bounding polygon and its centroid were calculated for each grouping of vacant lots per cluster. This centroid represented the point location that was mathematically closest to all the study vacant lots in each cluster. The address of the closest building to this point location was then determined as the starting point for house-to-house random sampling and enrollment of survey participants. At each starting address, a 2-person survey team walked in a predetermined random direction on the corresponding city block followed by randomly chosen adjacent city blocks within the cluster until a total of 5 participants had been identified, consented, and were interviewed. Both the survey team and participants were blinded to cluster intervention. Participants were told the study was about improving our understanding of urban health. One participant per household was chosen; in households with multiple eligible participants, the individual with the most recent birthday was chosen. All baseline interviews and most follow-up interviews were conducted in person; a handful of follow-up interviews were conducted by telephone. Both English-speaking and Spanish-speaking individuals 18 years and older were administered the survey in the language of their choice; only 2 Spanish-language surveys were administered. Each participant was compensated $25 per interview, which took an average of 39.6 minutes to complete. Based on the American Association for Public Opinion Research response rate calculator, our survey response rate was 47.4%.34 Our response rate matched or exceeded that of other surveys and was high enough to produce a reasonably representative sample of our target population.35-37

    Outcome Measures

    At each interview, participants responded to questions about their perceptions of mental health, focusing on their experiences within the past 30 days to anchor responses in time relative to the intervention period and to avoid telescoping and overestimation. We used the validated short-form Kessler-6 Psychological Distress Scale (K6), a widely used community screening tool. The K6 was designed to evaluate the prevalence of serious mental illness in the community and does not make a clinical diagnosis of mental illness. Participants were asked to indicate how often they felt nervous, hopeless, restless, depressed, that everything was an effort, and worthless using the following scale: all of the time, most of the time, more than half of the time, less than half of the time, some of the time, or at no time. In keeping with the K6 order and scoring, the 2 middle categories were combined to create a score of 0 to 4 for each marker, which was then summed for a total score of 0 to 24. Using standard scoring guidelines, a score of 13 or greater indicated higher prevalence of serious mental illness or what we call self-reported poor mental health.38,39 Participants self-reported their race and/or ethnicity.

    Statistical Analysis

    Prior to the study, sample size was determined by taking into account anticipated intracluster correlation, participant response prevalence, number of crimes reported to the police in each area, effect size, and power. The minimally detectable effect size, given 80% power and 4 time points based on the group before vs after interaction test for any pairwise comparison among the randomly allocated groups of lots, was calculated.40 From this, and predicting a 25% loss-to-follow-up rate, we estimated that we would maintain more than 80% power if we randomly surveyed 3 people per cluster twice before and twice after the intervention.

    Intention-to-treat analyses of participants were conducted according to the randomly assigned vacant lot cluster intervention group in which they lived. Pairwise comparisons were completed for all study outcomes between the greening intervention group and the no intervention group as well as the trash cleanup intervention group and the no intervention group. These pairwise comparisons were tested for statistical significance (all tests were 2-sided and statistical significance was defined as P ≤ .05) using unadjusted random-effects, cross-sectional time series regressions that accounted for the cluster design of the trial. Random-effects regressions were chosen because we assumed that unobserved lot-specific effects were correlated over time at the cluster level. All statistical analyses were conducted using Stata, version 14.1 (StataCorp LLC).

    Difference-in-differences analyses were calculated as interaction terms of 1-0 intervention-control differences multiplied by 0-1 pre-post differences. These difference-in-differences interaction terms were the primary independent variables of interest interpreted as the true effect of the interventions on the outcomes studied. The estimates from the difference-in-differences analysis were then divided by the overall magnitude of occurrence for each outcome in the intervention group to obtain percentage reductions.27,29,41 Additional subset analyses were also completed by neighborhood poverty levels using the census tracts within which study participants lived. The poverty threshold for 2013 was determined to be $19 530 per the average size of persons per household in Philadelphia and the 2013 poverty guidelines from the US Census Bureau and the Department of Health and Human Services Office of the Assistant Secretary for Planning and Evaluation.42

    Results
    Vacant Lots and Clusters

    The master list included 44 768 vacant lots, 34 149 (76.3%) of which were deemed eligible for inclusion in the study. Ineligible lots were excluded owing to insufficient blight or not being abandoned (4284), being greater than 5500 sq ft (3755), and being existing private or commercial parking lots (2580). A total of 110 clusters containing 541 vacant lots were enrolled in the trial and randomly allocated to the following 1 of 3 study arms: the greening intervention (37 clusters [33.6%]), the trash cleanup intervention (36 clusters [32.7%]), or no intervention (37 clusters [33.6%]) (Figure 3). Of the clusters, 47 (42.7%) were included in neighborhood poverty subset analysis.

    Balance was evident at the cluster level between the 3 intervention conditions in terms of total number of study lots per study arm (range, 161-206 lots), the mean number of study lots per cluster (range, 4.5-5.4 lots), the total square footage of study lots per cluster (range, 4844-4935 sq ft), the mean number of residents per cluster (range, 285-297 people), and the mean number of serious crimes, as reported by the Philadelphia Police Department, occurring within each cluster during the 18-month baseline period (range, 16.5-18.3 crimes) (Table 1).

    Participant Baseline Characteristics

    Of the 442 participants, the mean (SD) age was 44.6 (15.1) years, 264 (59.7%) were female, and 194 (43.9%) had a family income less than $25 000. A total of 442 participants were interviewed during the preintervention period, and 342 (77.4%) of these original participants were interviewed during the postintervention period and are included in this analysis. This amounted to a 22.6% loss to follow-up; of the 100 lost participants, 78% could not be found in their original cluster, and 22% refused to participate in subsequent waves. Of the 442 participants, 149 (33.7%) were assigned to the greening intervention, 145 (32.8%) to the trash cleanup intervention, and 148 (33.5%) to no intervention. Of the 342 participants included in the analysis, 117 (34.2%) received the greening intervention, 107 (31.3%) the trash cleanup intervention, and 118 (34.5%) no intervention. A total of 139 people (40.6%) were included in the neighborhood poverty subset analyses, including 45 (32.4%) receiving the greening intervention, 51 (36.7%) the trash cleanup intervention, and 43 (30.9%) no intervention. Participant demographic characteristics were balanced between the 3 study arms, including mean tenure in the home (range, 12.0-13.7 years), mean age (range, 43.3-45.3 years), and percentage with family income less than $25 000 (range, 41.0%-46.9%) (Table 1).

    Participant-Reported Mental Health Outcomes

    Intention-to-treat analyses demonstrated significant changes in participant-reported mental health outcomes. Intention-to-treat analyses of the greening intervention compared with no intervention demonstrated a significant decrease 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). Analysis also demonstrated a nonsignificant reduction in overall self-reported poor mental health (−62.8%; 95% CI, −86.2% to 0.4%; P = .051), as calculated by the K6 (Table 2). When looking only at neighborhoods below the poverty line, feeling depressed significantly decreased (−68.7%; 95% CI, −86.5% to −27.5%; P = .007). There was no significant difference in self-reported poor mental health in neighborhoods below the poverty line.

    Intention-to-treat analyses of the trash cleanup intervention compared with no intervention did not show any statistically significant differences between self-reported poor mental health measured by the K6 (Table 2). There was also no difference between groups for the individual components of the K6. The analysis of neighborhoods below the poverty line also did not indicate any difference in self-reported mental health between the groups.

    Discussion

    In this citywide cluster randomized trial of 2 vacant land remediation interventions, greening was associated with a significant reduction in feeling depressed and worthless as well as a nonsignificant reduction in overall self-reported poor mental health for randomly sampled residents living nearby. The trash cleanup intervention was not associated with a reduction in feeling depressed or self-reported poor mental health.

    To our knowledge, this is the first citywide cluster randomized trial of actual place-based changes to urban spaces. These results add much needed experimental evidence to a growing body of literature calling for structural changes to neighborhoods as a method for improving health and safety.43,44 This study extends previous work showing a clear association between green space and mental illness,13-21 by demonstrating that adding green space to people’s neighborhood environment can improve the trajectory of their mental health. Additionally, vacant lot greening is a relatively low-cost intervention (approximately $1597 per vacant lot and $180 per year to maintain) that we have previously shown to be a cost-beneficial solution to firearm violence.29 For these reasons, vacant lot greening may be an extremely attractive intervention for policy makers seeking to address urban blight.

    Our findings indicate that the effect of vacant lot greening on feeling depressed was slightly stronger for those living in neighborhoods below the poverty line. Urban blight is an environmental condition that disproportionately affects low-resource neighborhoods, as evidenced by the fact that almost half of our participants had yearly family incomes less than $25 000. Making structural changes to the lowest-resource neighborhoods can make them healthier and may be an important mechanism to address persistent and entrenched socioeconomic health disparities.45

    There are several possible mechanisms through which the vacant lot greening intervention but not the trash cleanup intervention improved feeling depressed and self-reported poor mental health. One significant difference between the 2 interventions was the creation of new green space. Green space, particularly in urban environments more likely to have a dearth of vegetation, has been linked to recovery from mental fatigue,46 a state of inattentiveness and irritability resulting from the information-processing demands of daily life. Spending time in or near nature can combat mental fatigue because it allows engagement without paying explicit attention.46-48 A related concept is the association between spending time in or near green space and stress reduction,18,49 which may in turn reduce mental illness. For example, walking past green space has been associated with reduction in heart rate,12 one marker of acute physiological stress.

    Additionally, the presence of green space is associated with improved neighborhood social milieu, including the concepts of social cohesion, social capital, and collective efficacy.50-53 The presence of grass and trees is related to use of outdoor space and increased social activity that takes place in those outdoor spaces.54 Improved social conditions are, in turn, associated with better mental health.55,56 For example, living in a low-income neighborhood is associated with worse mental health indicators for people with low but not high social cohesion.57 Studies have found that social cohesion mediated a positive green space–mental health relationship.58-60 Additionally, previous studies have demonstrated an association of vacant lot greening with increased feelings of safety and decreased violent crime, both of which may work to improve mental illness.27,28 Fear of crime, for example, is associated with almost 2-fold higher likelihood of having depression.61

    The other significant difference between the greening and trash cleanup interventions was the presence of a simple wooden post and rail fence. The fence delineates the newly greened space as one that is cared for but does have openings to indicate that people can enter the space. The fence is also meant to deter illegal dumping. Previous qualitative work conducted by our team indicated that vacant land causes people to feel stigmatized and abandoned by their community and government.11 Countering this with clear signs of neighborhood investment, such as a clearly marked newly greened vacant lot, may contribute to the improvements seen in feeling depressed and self-reported poor mental health.

    Limitations

    There were several limitations to this study. We used the K6 to measure our outcome of interest and mental health. While this is a validated and widely used scale, it is still a single scale, and other mental illness screening and diagnosis tools and scales may produce different results. Furthermore, we did not conduct a Diagnostic and Statistical Manual of Mental Disorders, Fifth Edition–level diagnosis of mental illness but rather used a community screening tool. Another limitation is the duration of our study and loss to follow-up. We followed up people for 18 months following the blight remediation interventions and are unable to know if the effect of the interventions on mental health outcomes persisted past the study period. We also made every effort to minimize loss to follow-up of our study participants after they were first enrolled, although differential, nonrandom dropout in our 3 study arms and across all study waves could have affected our results. Finally, we did not specifically track if and how study participants used (or did not use) study vacant lots, although prior work has demonstrated signs of use, such as barbeques or chairs on similar vacant lots.62

    Conclusions

    Among community-dwelling adults, self-reported feelings of depression and worthlessness were significantly decreased and self-reported poor mental health was nonsignificantly reduced for those living near greened vacant lots compared with control lots. The treatment of dilapidated physical environments can be an important tool for communities to address persistent mental health problems. These findings provide support to health care clinicians concerned with positively transforming the often chaotic and harmful environments that affect their patients. Our findings also offer evidence to policy makers interested in increasing municipal investments in the remediation of blighted urban spaces as an inexpensive29 and scalable way to improve mental health, particularly in low-resource neighborhoods.

    Back to top
    Article Information

    Accepted for Publication: April 26, 2018.

    Published: July 20, 2018. doi:10.1001/jamanetworkopen.2018.0298

    Correction: This article was corrected on August 17, 2018, to fix an error in Figure 2B.

    Open Access: This is an open access article distributed under the terms of the CC-BY License. © 2018 South EC et al. JAMA Network Open.

    Corresponding Author: Eugenia C. South, MD, MS, Department of Emergency Medicine, Perelman School of Medicine, University of Pennsylvania, 423 Guardian Dr, Blockley Hall, Room 408, Philadelphia, PA 19104 (eugenia.south@uphs.upenn.edu).

    Author Contributions: Drs MacDonald and Branas had full access to all of the data in the study and take responsibility for the integrity of the data and the accuracy of the data analysis.

    Concept and design: South, Hohl, MacDonald, Branas.

    Acquisition, analysis, or interpretation of data: All authors.

    Drafting of the manuscript: South, Hohl, Branas.

    Critical revision of the manuscript for important intellectual content: Hohl, Kondo, MacDonald, Branas.

    Statistical analysis: MacDonald, Branas.

    Obtained funding: Branas.

    Administrative, technical, or material support: Hohl, Kondo, Branas.

    Supervision: MacDonald, Branas.

    Conflict of Interest Disclosures: Dr Hohl reported receiving grants from the National Institutes of Health during the conduct of the study. Dr MacDonald reported receiving grants from the National Institutes of Health and Centers for Disease Control and Prevention during the conduct of the study. No other disclosures were reported.

    Funding/Support: This study was funded in part by grants R01AA020331 and R01DA010164 from the National Institutes of Health and grant R49CE002474 from the Centers for Disease Control.

    Role of the Funder/Sponsor: The funders had no role in the design and conduct of the study; collection management, analysis, and interpretation of the data; preparation, review, or approval of the manuscript; or decision to submit the manuscript for publication.

    Additional Contributions: Philippe Bourgois, PhD (University of California, Los Angeles), contributed to the study design and execution; Keith Green (Pennsylvania Horticulture Society, Philadelphia), study intervention design and implementation; Jamillah Millner, BA (University of Pennsylvania, Philadelphia), participant recruitment, enrollment, and retention; Vicky Tam, MA (University of Pennsylvania, Philadelphia), geospatial planning and implementation; Douglas Wiebe, PhD (University of Pennsylvania, Philadelphia), study design and execution; and Jeremy Levenson, BA (University of California, Los Angeles), study execution. These individuals were affiliated with this project and key to its success. Drs Bourgois and Wiebe and Mss Millner and Tam received salary support for their contribution. Mr Green’s organization received funds to perform the intervention. Mr Levenson received compensation for his field work.

    References
    1.
    Murray  CJL, Atkinson  C, Bhalla  K,  et al; US Burden of Disease Collaborators.  The state of US health, 1990-2010: burden of diseases, injuries, and risk factors.  JAMA. 2013;310(6):591-608. doi:10.1001/jama.2013.13805PubMedGoogle ScholarCrossref
    2.
    US Center for Behavioral Health Statistics and Quality. Key substance use and mental health indicators in the United States: results from the 2015 National Survey on Drug Use and Health. http://www.samhsa.gov/data/. Accessed November 13, 2017.
    3.
    Kessler  RC, Berglund  P, Demler  O, Jin  R, Merikangas  KR, Walters  EE.  Lifetime prevalence and age-of-onset distributions of DSM-IV disorders in the National Comorbidity Survey Replication.  Arch Gen Psychiatry. 2005;62(6):593-602. doi:10.1001/archpsyc.62.6.593PubMedGoogle ScholarCrossref
    4.
    Kamal  R, Cox  C, Rousseau  D; Kaiser Family Foundation.  Costs and outcomes of mental health and substance use disorders in the US.  JAMA. 2017;318(5):415. doi:10.1001/jama.2017.8558PubMedGoogle ScholarCrossref
    5.
    Mair  C, Diez Roux  AV, Morenoff  JD.  Neighborhood stressors and social support as predictors of depressive symptoms in the Chicago Community Adult Health Study.  Health Place. 2010;16(5):811-819. doi:10.1016/j.healthplace.2010.04.006PubMedGoogle ScholarCrossref
    6.
    Latkin  CA, Curry  AD.  Stressful neighborhoods and depression: a prospective study of the impact of neighborhood disorder.  J Health Soc Behav. 2003;44(1):34-44. doi:10.2307/1519814PubMedGoogle ScholarCrossref
    7.
    Hill  TD, Ross  CE, Angel  RJ.  Neighborhood disorder, psychophysiological distress, and health.  J Health Soc Behav. 2005;46(2):170-186. doi:10.1177/002214650504600204PubMedGoogle ScholarCrossref
    8.
    Mair  C, Diez Roux  AV, Galea  S.  Are neighbourhood characteristics associated with depressive symptoms? a review of evidence.  J Epidemiol Community Health. 2008;62(11):940-946.PubMedGoogle Scholar
    9.
    Mair  C, Diez Roux  AV, Shen  M,  et al.  Cross-sectional and longitudinal associations of neighborhood cohesion and stressors with depressive symptoms in the Multiethnic Study of Atherosclerosis.  Ann Epidemiol. 2009;19(1):49-57. doi:10.1016/j.annepidem.2008.10.002PubMedGoogle ScholarCrossref
    10.
    Hoebel  J, Maske  UE, Zeeb  H, Lampert  T.  Social inequalities and depressive symptoms in adults: the role of objective and subjective socioeconomic status.  PLoS One. 2017;12(1):e0169764. doi:10.1371/journal.pone.0169764PubMedGoogle ScholarCrossref
    11.
    Garvin  E, Branas  C, Keddem  S, Sellman  J, Cannuscio  C.  More than just an eyesore: local insights and solutions on vacant land and urban health.  J Urban Health. 2013;90(3):412-426. doi:10.1007/s11524-012-9782-7PubMedGoogle ScholarCrossref
    12.
    South  EC, Kondo  MC, Cheney  RA, Branas  CC.  Neighborhood blight, stress, and health: a walking trial of urban greening and ambulatory heart rate.  Am J Public Health. 2015;105(5):909-913. doi:10.2105/AJPH.2014.302526PubMedGoogle ScholarCrossref
    13.
    James  P, Banay  RF, Hart  JE, Laden  F.  A review of the health benefits of greenness.  Curr Epidemiol Rep. 2015;2(2):131-142. doi:10.1007/s40471-015-0043-7PubMedGoogle ScholarCrossref
    14.
    Seymour  V.  The human-nature relationship and its impact on health: a critical review.  Front Public Health. 2016;4:260. doi:10.3389/fpubh.2016.00260PubMedGoogle ScholarCrossref
    15.
    Beyer  KM, Kaltenbach  A, Szabo  A, Bogar  S, Nieto  FJ, Malecki  KM.  Exposure to neighborhood green space and mental health: evidence from the Survey of the Health of Wisconsin.  Int J Environ Res Public Health. 2014;11(3):3453-3472. doi:10.3390/ijerph110303453PubMedGoogle ScholarCrossref
    16.
    McEachan  RRC, Prady  SL, Smith  G,  et al The association between green space and depressive symptoms in pregnant women: moderating roles of socioeconomic status and physical activity.  J Epidemiol Community Health. 2016:70(3):253-259. doi:10.1136/jech-2015-205954PubMedGoogle ScholarCrossref
    17.
    Wu  Y-T, Prina  AM, Jones  A, Matthews  FE, Brayne  C; The Medical Research Council Cognitive Function and Ageing Studies.  Older people, the natural environment and common mental disorders: cross-sectional results from the Cognitive Function and Ageing Study.  BMJ Open. 2015;5(9):e007936. doi:10.1136/bmjopen-2015-007936PubMedGoogle ScholarCrossref
    18.
    Roe  JJ, Thompson  CW, Aspinall  PA,  et al.  Green space and stress: evidence from cortisol measures in deprived urban communities.  Int J Environ Res Public Health. 2013;10(9):4086-4103. doi:10.3390/ijerph10094086PubMedGoogle ScholarCrossref
    19.
    Morita  E, Fukuda  S, Nagano  J,  et al.  Psychological effects of forest environments on healthy adults: Shinrin-yoku (forest-air bathing, walking) as a possible method of stress reduction.  Public Health. 2007;121(1):54-63. doi:10.1016/j.puhe.2006.05.024PubMedGoogle ScholarCrossref
    20.
    Cox  DTC, Shanahan  DF, Hudson  HL,  et al.  Doses of nearby nature simultaneously associated with multiple health benefits.  Int J Environ Res Public Health. 2017;14(2):172. doi:10.3390/ijerph14020172PubMedGoogle ScholarCrossref
    21.
    Shanahan  DF, Bush  R, Gaston  KJ,  et al.  Health benefits from nature experiences depend on dose.  Sci Rep. 2016;6:28551. doi:10.1038/srep28551PubMedGoogle ScholarCrossref
    22.
    Mitchell  R, Popham  F.  Effect of exposure to natural environment on health inequalities: an observational population study.  Lancet. 2008;372(9650):1655-1660. doi:10.1016/S0140-6736(08)61689-XPubMedGoogle ScholarCrossref
    23.
    Hartig  T.  Green space, psychological restoration, and health inequality.  Lancet. 2008;372(9650):1614-1615. doi:10.1016/S0140-6736(08)61669-4PubMedGoogle ScholarCrossref
    24.
    Mitchell  RJ, Richardson  EA, Shortt  NK, Pearce  JR.  Neighborhood environments and socioeconomic inequalities in mental well-being.  Am J Prev Med. 2015;49(1):80-84. doi:10.1016/j.amepre.2015.01.017PubMedGoogle ScholarCrossref
    25.
    Frieden  TR.  A framework for public health action: the health impact pyramid.  Am J Public Health. 2010;100(4):590-595. doi:10.2105/AJPH.2009.185652PubMedGoogle ScholarCrossref
    26.
    Gong  Y, Palmer  S, Gallacher  J, Marsden  T, Fone  D.  A systematic review of the relationship between objective measurements of the urban environment and psychological distress.  Environ Int. 2016;96:48-57. doi:10.1016/j.envint.2016.08.019PubMedGoogle ScholarCrossref
    27.
    Branas  CC, Cheney  RA, MacDonald  JM, Tam  VW, Jackson  TD, Ten Have  TR.  A difference-in-differences analysis of health, safety, and greening vacant urban space.  Am J Epidemiol. 2011;174(11):1296-1306. doi:10.1093/aje/kwr273PubMedGoogle ScholarCrossref
    28.
    Garvin  EC, Cannuscio  CC, Branas  CC.  Greening vacant lots to reduce violent crime: a randomised controlled trial.  Inj Prev. 2013;19(3):198-203. doi:10.1136/injuryprev-2012-040439PubMedGoogle ScholarCrossref
    29.
    Branas  CC, Kondo  MC, Murphy  SM, South  EC, Polsky  D, MacDonald  JM.  Urban blight remediation as a cost-beneficial solution to firearm violence.  Am J Public Health. 2016;106(12):2158-2164. doi:10.2105/AJPH.2016.303434PubMedGoogle ScholarCrossref
    30.
    Branas  CC, South  E, Kondo  MC,  et al.  Citywide cluster randomized trial to restore blighted vacant land and its effects on violence, crime, and fear.  Proc Natl Acad Sci U S A. 2018;115(12):2946-2951. doi:10.1073/pnas.1718503115PubMedGoogle ScholarCrossref
    31.
    Campbell  MK, Piaggio  G, Elbourne  DR, Altman  DG; CONSORT Group.  CONSORT 2010 statement: extension to cluster randomised trials.  BMJ. 2012;345:e5661. doi:10.1136/bmj.e5661PubMedGoogle ScholarCrossref
    32.
    Boruch  R, May  H, Turner  H,  et al.  Estimating the effects of interventions that are deployed in many places: place-randomized trials.  Am Behav Sci. 2004;47(5):608-633. doi:10.1177/0002764203259291Google ScholarCrossref
    33.
    Schulz  KF, Grimes  DA.  Generation of allocation sequences in randomised trials: chance, not choice.  Lancet. 2002;359(9305):515-519. doi:10.1016/S0140-6736(02)07683-3PubMedGoogle ScholarCrossref
    34.
    American Association for Public Opinion Research. AAPOR response rate calculator. https://www.aapor.org/Education-Resources/For-Researchers/Poll-Survey-FAQ/Response-Rates-An-Overview.aspx. Accessed November 27, 2017.
    35.
    Groves  RM.  Nonresponse rates and nonresponse bias in household surveys.  Public Opin Q. 2006;70(5):646-675. doi:10.1093/poq/nfl033Google ScholarCrossref
    36.
    Galea  S, Tracy  M.  Participation rates in epidemiologic studies.  Ann Epidemiol. 2007;17(9):643-653. doi:10.1016/j.annepidem.2007.03.013PubMedGoogle ScholarCrossref
    37.
    Keeter  S, Kennedy  C, Dimock  M, Best  J, Craighill  P.  Gauging the impact of growing nonresponse on estimates from a national RDD telephone survey.  Public Opin Q. 2006;70(5):759-779. doi:10.1093/poq/nfl035Google ScholarCrossref
    38.
    Kessler  RC, Andrews  G, Colp  LJ,  et al Short screening scales to monitor population prevalences and trends in non-specific psychological distress.  Psychol Med. 2002;32(6):959-976. doi:10.1017/S0033291702006074PubMedGoogle ScholarCrossref
    39.
    Kessler  RC, Barker  PR, Colpe  LJ,  et al.  Screening for serious mental illness in the general population.  Arch Gen Psychiatry. 2003;60(2):184-189. doi:10.1001/archpsyc.60.2.184PubMedGoogle ScholarCrossref
    40.
    Cohen  J.  Statistical Power for the Behavioral Sciences. 2nd ed. Hillsdale, NJ: Lawrence Erlbaum Associates; 1988.
    41.
    Meyer  BD.  Natural and quasi-experiments in economics.  J Bus Econ Stat. 1995;13(2):151-161. doi:10.1080/07350015.1995.10524589Google Scholar
    42.
    US Department of Health and Human Services, Assistant Secretary for Planning and Evaluation. Poverty guidelines. https://aspe.hhs.gov/2013-poverty-guidelines#thresholds. Accessed April 20, 2018.
    43.
    Branas  CC, Macdonald  JM.  A simple strategy to transform health, all over the place.  J Public Health Manag Pract. 2014;20(2):157-159. doi:10.1097/PHH.0000000000000051PubMedGoogle ScholarCrossref
    44.
    Kondo  MC, South  EC, Branas  CC.  Nature-based strategies for improving urban health and safety.  J Urban Health. 2015;92(5):800-814. doi:10.1007/s11524-015-9983-yPubMedGoogle ScholarCrossref
    45.
    Woolf  SH, Braveman  P.  Where health disparities begin: the role of social and economic determinants—and why current policies may make matters worse.  Health Aff (Millwood). 2011;30(10):1852-1859. doi:10.1377/hlthaff.2011.0685PubMedGoogle ScholarCrossref
    46.
    Kuo  FE, Sullivan  WC.  Aggression and violence in the inner city: effects of environment via mental fatigue.  Sage Journals. 2001;33(4):543-571. doi:10.1177/00139160121973124Google Scholar
    47.
    Kaplan  S.  The restorative benefits of nature: toward an integrative framework.  J Environ Psychol. 1995;15(3):169-182. doi:10.1016/0272-4944(95)90001-2Google ScholarCrossref
    48.
    Hartig  T, Evans  GW, Jamner  LD, Davis  DS, Gärling  T.  Tracking restoration in natural and urban field settings.  J Environ Psychol. 2003;23(2):109-123. doi:10.1016/S0272-4944(02)00109-3Google ScholarCrossref
    49.
    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(3):221-229. doi:10.1016/j.landurbplan.2011.12.015Google ScholarCrossref
    50.
    Cohen  DA, Inagami  S, Finch  B.  The built environment and collective efficacy.  Health Place. 2008;14(2):198-208. doi:10.1016/j.healthplace.2007.06.001PubMedGoogle ScholarCrossref
    51.
    Maas  J, van Dillen  SME, Verheij  RA, Groenewegen  PP.  Social contacts as a possible mechanism behind the relation between green space and health.  Health Place. 2009;15(2):586-595. doi:10.1016/j.healthplace.2008.09.006PubMedGoogle ScholarCrossref
    52.
    Kweon  B-S, Sullivan  WC, Wiley  AR.  Green common spaces and the social integration of inner-city older adults.  Sage Journals. 1998;30(6):832-858. doi:10.1177/001391659803000605Google Scholar
    53.
    Kuo  F, Sullivan  W, Coley  R, Brunson  L.  Fertile ground for community: inner-city neighborhood common spaces.  Am J Community Psychol. 1998;26(6):823-851. doi:10.1023/A:1022294028903Google ScholarCrossref
    54.
    Sullivan  WC.  The fruit of urban nature: vital neighborhood spaces.  Sage Journals. 2004;36(5):678-700. doi:10.1177/0193841X04264945Google Scholar
    55.
    Fone  D, White  J, Farewell  D,  et al.  Effect of neighbourhood deprivation and social cohesion on mental health inequality: a multilevel population-based longitudinal study.  Psychol Med. 2014;44(11):2449-2460. doi:10.1017/S0033291713003255PubMedGoogle ScholarCrossref
    56.
    Kruger  DJ, Reischl  TM, Gee  GC.  Neighborhood social conditions mediate the association between physical deterioration and mental health.  Am J Community Psychol. 2007;40(3-4):261-271. doi:10.1007/s10464-007-9139-7PubMedGoogle ScholarCrossref
    57.
    Fone  D, Dunstan  F, Lloyd  K, Williams  G, Watkins  J, Palmer  S.  Does social cohesion modify the association between area income deprivation and mental health? a multilevel analysis.  Int J Epidemiol. 2007;36(2):338-345. doi:10.1093/ije/dym004PubMedGoogle ScholarCrossref
    58.
    de Vries  S, van Dillen  SME, Groenewegen  PP, Spreeuwenberg  P.  Streetscape greenery and health: stress, social cohesion and physical activity as mediators.  Soc Sci Med. 2013;94:26-33. doi:10.1016/j.socscimed.2013.06.030PubMedGoogle ScholarCrossref
    59.
    Triguero-Mas  M, Dadvand  P, Cirach  M,  et al.  Natural outdoor environments and mental and physical health: relationships and mechanisms.  Environ Int. 2015;77:35-41. doi:10.1016/j.envint.2015.01.012PubMedGoogle ScholarCrossref
    60.
    Sugiyama  T, Leslie  E, Giles-Corti  B, Owen  N.  Associations of neighbourhood greenness with physical and mental health: do walking, social coherence and local social interaction explain the relationships?  J Epidemiol Community Health. 2008;62(5):e9. doi:10.1136/jech.2007.064287PubMedGoogle ScholarCrossref
    61.
    Stafford  M, Chandola  T, Marmot  M.  Association between fear of crime and mental health and physical functioning.  Am J Public Health. 2007;97(11):2076-2081. doi:10.2105/AJPH.2006.097154PubMedGoogle ScholarCrossref
    62.
    Heckert  M, Kondo  M. Can “cleaned and greened” lots take on the role of public greenspace?  J Plan Educ Res. 2017;38(2):211-221. doi:10.1177/0739456X16688766.Google ScholarCrossref
    ×