pexels-sarowar-hussain-12769937

Mental Health and Accessibility to Green Space

Due to environmental stressors that are often associated with areas that have high population densities, such as higher rates of criminality, mortality, social isolation, air pollution, or noise, it is often assumed that higher rates of mental illness correspond with urban areas¹. While negative characteristics of an urban neighborhood may affect the mental health of its residents, it is also useful to consider the effects of positive characteristics of urban neighborhoods on mental health.

More specifically, it has been shown that the presence of green spaces within urban neighborhoods are “health-promoting characteristics” linked to stress reduction and better overall health². While previous literature has examined the role of green spaces and mental wellness and urban areas, few studies have examined this relationship spatially within New York City. This study seeks to examine the relationship between vegetation, green spaces, and history of depression among residents in New York City.

Collaborators: Advised by Dr. Hongmian Gong
Client: Completed at Hunter College
Industry: Urban Design, Urban Planning, GIS Analysis
Skills: Quantitative Research and Analysis, GIS Analysis, Research Design
Timeline: Completed January 2022-May 2022
Tools: ArcGIS Pro, Stata

DOWNLOAD FULL PAPER HERE

INTRODUCTION

Research Question

Do urban neighborhoods in New York City with greater access to green features, such as parks and street trees, correlate with lower individual histories of depression?

My Role

I downloaded publicly available datasets to create dependent, explanatory, and control variables within the study. I utilized Excel and ArcGIS Pro to clean and manipulate the data so that it would be useable for spatial analysis tests, such as geographically weighted regression, to answer my research question. After cleaning the data and running analyses, I prepared tables and maps for presentation in addition to writing up my findings. 

LITERATURE REVIEW

The literature review revealed two main takeaways: (1) population density and income affect mental health in urban areas, and (2) higher vegetation/access to green spaces is associated with better mental wellness in urban environments

Mental Health in Urban Areas

1. There is a strong positive correlation between rate of inpatient treatment for mental illness and population density³

2. Mood and anxiety disorders are higher in urban areas compared to rural areas¹

3. In a study on mental health in China, population density is a predictor of depression

4. In New York City, high-income neighborhoods show increased resilience to urban stressors when compared to low-income neighborhoods

Main Takeaway: Population density and income affect mental health in urban areas.

Jump to references.

A 2014 study found that the difference in depressive symptoms between an individual living in an environment with  100% tree canopy and an environment with no tree canopy is larger than the difference in symptoms between someone who has health insurance and someone who does not.²

Mental Health in Urban Areas and Nature

1. Residents living in neighborhoods with more than 20% vegetation cover showed reduced symptoms of depression by up to 11%

2. Improvement in urban parks in Seoul is positively associated with residents’ subjective well-being

3. Well-designed urban landscapes can be associated with physical, cognitive, and emotional restoration

Main Takeaway: Higher vegetation/access to green spaces is associated with better mental wellness in urban environments.

Jump to references.

VARIABLES

Dependent Variable: History of Depression

New York City Department of Health and Mental Hygiene’s 2009 Community Survey was a survey of ~10,000 randomly selected New Yorkers. Depression data was provided as the percentage of respondents with a history of depression. Spatial data is coded into the UHF 34 Neighborhood Index, which is depicted below.

Depression Data Source

 

uhf-zip-information-01
uhf-zip-information-02

Explanatory Variable: Street Tree Count

The 2015 Street Tree Census collected tree count as point data, which I spatially joined to the UHF 34 Neighborhood data to be more easily compared with the history of depression variable. 

https://data.cityofnewyork.us/Environment/2015-Street-Tree-Census-Tree-Data/pi5s-9
p35

 

Picture5

Explanatory Variable: Walk-to-Park Ratio

 

Walk-to-a-Park Service Area (2017) data from the Department of Parks and Recreation Depicts areas within ½ mile of an urban park. I clipped this data to UHF 34 Neighborhoods, and divided the walk- to-park area by total area of the neighborhood to get the percentage of the neighborhood that was a walkable distance from an urban park.

Walk-to-Park Ratio Data Source

Screenshot-2023-03-10-082757

Control Variables: Median Income and Population Density

These variables were pulled from 2020 U.S. Census Data by Zip Code. I used the GIS “Summarize Within” tool to combine it with UHF 34 Neighborhood data.

 

VARIABLE MAPS

Average of Median Incomes

1-01

Population Density

1-02

History of Depression

ALL_MAPS_Fig-1-1

Tree Count

ALL_MAPS_Fig-2

% of Neighborhood within 1/4 Mile of Park

ALL_MAPS_Fig-3

FINDINGS

Ordinary Least Squares (OLS) Regressions

OLS-table

Notes on Ordinary Least Squares Regressions (OLS)

  • I conducted several OLS regressions using history of depression as dependent variable and tree count/walk to a park ratio as explanatory variables, and median income/population density as control variables

  • Models 1-3 show that the tree count variable has statistical significance even with the addition of median income as a control variable
  • Walk-to-park ratio does not show statistical significance in any of the models

  • When population density is added as a control variable, tree count is no longer statistically significant

Notes on Geographically Weighted Regression (GWR)

  • All models showed evidence of spatial autocorrelation

  • Geographically weighted regressions were conducted where possible
  • Walk-to-park ratio variable was highly correlated with median income so it was dropped from GWR

  • AIC and R-squared values indicate that GWR was a better fit for all models where it was conducted

GWR on History of Depression, Tree Count, Walk-to-Park Ratio

ALL_MAPS_Fig-4

GWR on History of Depression, Tree Count, Income, Density

ALL_MAPS_Fig-9

GWR on Tree Count and Population Density

ALL_MAPS_Fig-10
table-2-white

Includes OLS regression with tree count as dependent variable and population density as explanatory variable shows statistical significance

LIMITATIONS & CONCLUSIONS

Limitations

  • No data on how “green” each park was

  • Depression data was provided as only a single percentage that indicated any history of depression at all, within the UHF 34 Neighborhood Index

  • Data was used from different years, where conditions for any of the variables may have changed throughout the years

Conclusions

  • Some evidence that supports hypothesis: there is a relationship between high tree counts and low history of depression

  • Population density may mediate the relationship between tree counts and history of depression

  • No relationship found between proximity to parks and history of depression

  • This study provides a starting point for examining the relationship between vegetation and history of depression in New York City

DOWNLOAD FULL PAPER HERE

References

1. Peen, J., R. A. Schoevers, A. T. Beekman, and J. Dekker. “The Current Status of Urban-Rural Differences in Psychiatric Disorders.” Acta Psychiatrica Scandinavica 121, no. 2 (February 2010): 84–93. https://doi.org/10.1111/j.1600-0447.2009.01438.x.

2. Beyer, Kirsten, Andrea Kaltenbach, Aniko Szabo, Sandra Bogar, F. Nieto, and Kristen Malecki. “Exposure to Neighborhood Green Space and Mental Health: Evidence from the Survey of the Health of Wisconsin.” International Journal of Environmental Research and Public Health 11, no. 3 (March 21, 2014): 3453–72. https://doi.org/10.3390/ijerph110303453.

3. Williams, David G. “Population Density and Mental Illness.” The Journal of Social Psychology 134, no. 4 (August 1994): 545–46. https://doi.org/10.1080/00224545.1994.9712205.

4. Chen, Juan, Shuo Chen, and Pierre Landry. “Urbanization and Mental Health in China: Linking the 2010 Population Census with a Cross-Sectional Survey.” International Journal of Environmental Research and Public Health 12, no. 8 (July 31, 2015): 9012–24. https://doi.org/10.3390/ijerph120809012.

5. Beard, John R., Magda Cerdá, Shannon Blaney, Jennifer Ahern, David Vlahov, and Sandro Galea. “Neighborhood Characteristics and Change in Depressive Symptoms Among Older Residents of New York City.” American Journal of Public Health 99, no. 7 (July 2009): 1308–14. https://doi.org/10.2105/AJPH.2007.125104.

6. Cox, Daniel T. C., Danielle F. Shanahan, Hannah L. Hudson, Kate E. Plummer, Gavin M. Siriwardena, Richard A. Fuller, Karen Anderson, Steven Hancock, and Kevin J. Gaston. “Doses of Neighborhood Nature: The Benefits for Mental Health of Living with Nature.” BioScience, January 25, 2017, biw173. https://doi.org/10.1093/biosci/biw173.

7. Kim, Danya, and Jangik Jin. “Does Happiness Data Say Urban Parks Are Worth It?” Landscape and Urban Planning 178 (October 2018): 1–11. https://doi.org/10.1016/j.landurbplan.2018.05.010.

8. Berto, Rita. “The Role of Nature in Coping with Psycho-Physiological Stress: A Literature Review on Restorativeness.” Behavioral Sciences 4, no. 4 (October 21, 2014): 394–409. https://doi.org/10.3390/bs4040394.