Dissertations and Theses

Date of Degree

1-1-2026

Document Type

Dissertation

Degree Name

Doctor of Philosophy (Ph.D.)

Department

Environmental, Occupational, and Geospatial Health Sciences

Advisor(s)

Glen D. Johnson

Terry T.-K. Huang

Committee Members

Glen D. Johnson

Terry T.-K. Huang

Brian Pavilonis

Katarzyna E. Wyka

Subject Categories

Community Health and Preventive Medicine | Environmental Design | Environmental Public Health | Epidemiology | Mental and Social Health | Public Health | Urban, Community and Regional Planning

Keywords

park use, park satisfaction, perceived stress, mental health, physical activity, sedentary behavior

Abstract

Background

The rapid pace of urbanization in the 21st century has profoundly reshaped global demographic patterns, with projections indicating that over two-thirds of the world’s population will reside in urban areas by 2050. This shift has also been associated with shifting population health risks, particularly concerning mental health stressors and physical inactivity, both of which are strongly linked to chronic conditions such as obesity, diabetes, cardiovascular disease, anxiety, depression, and dementia. Globally, there is a need for large-scale, innovative built environment interventions in urban areas that address the behavioral and structural drivers of poor mental health and physical inactivity and ultimately reduce chronic disease burdens.

Urban parks and greenspaces are increasingly being recognized as vital public health infrastructure, as they may offer pathways for improved health and well-being through heat and air pollution mitigation, opportunities for social interaction and physical activity, and provision of restorative environments that can reduce stress. Park quality is thought to be an important factor, as parks that are perceived as well-maintained and safe are used more frequently and may create the conditions necessary for the health and social benefits of park use to emerge. Furthermore, growing evidence suggests that the dosage of park use also matters, where greater frequency and/or duration of park use may be associated with better health outcomes.

This dissertation sought to evaluate the public health impacts of the Community Parks Initiative (CPI), a major citywide policy effort in New York City (NYC) providing targeted infrastructure improvements to parks in low-income, densely populated communities with a history of disinvestment in public spaces. The overall objective was to evaluate the impact of CPI park quality improvements on longitudinal changes in behavioral and health outcomes among adult residents of low-income neighborhoods served by the parks using data from the Physical Activity and Redesigned Community Spaces (PARCS) study. Specifically, this dissertation research assessed the impact of CPI on three primary areas: neighborhood-level changes in park use patterns and park satisfaction (Chapter 2), individual-level changes in perceived stress (Chapter 3), and individual-level changes in physical activity and sedentary behavior (Chapter 4).

Methods

The PARCS study was a quasi-experimental prospective cohort study conducted between 2016 and 2022. A Difference-in-Differences (DID) analytical approach was used to estimate the causal effects of CPI renovations by comparing changes over time in health and behavioral outcomes between intervention and control groups. The sample included adult residents (aged ≥18 years) residing within a 0.3-mile radius of 31 intervention parks scheduled for CPI renovations and 21 sociodemographically matched control parks. Both intervention and control parks were located in low-income neighborhoods that met specific criteria related to poverty, density, recent population growth, and lack of prior capital investment. The CPI intervention involved comprehensive redesigns and renovations implemented by NYC Parks, with a median capital investment of U.S. $3.9 million per park site. The intervention improved overall park quality, amenities, aesthetics, and accessibility, and included substantial upgrades to greenery, seating areas, walking paths, ball courts, playgrounds, and adult fitness equipment.

Data collection occurred in two waves: pre- and post-renovation. Park use and park satisfaction outcomes were derived from repeated cross-sectional survey data (n = 1,220 unique residents), measuring self-reported past-month park use frequency, typical visit duration in minutes, and satisfaction with overall park quality and facilities. Perceived stress was measured longitudinally using the validated 14-item Perceived Stress Scale (PSS) (n = 313 participants), collected pre- and post-renovation. Physical activity and sedentary behavior measures were collected longitudinally using ActiGraph accelerometer devices worn for seven consecutive days pre- and post-renovation (n = 344 participants), yielding total volume of physical activity (vector magnitude per minute, or VMPM), and minutes/day of sedentary behavior, light physical activity, and moderate-vigorous physical activity. Linear generalized estimating equations and mixed-effects regression models were used to obtain DID estimates while accounting for correlations within park sites and individuals. Exploratory models examined three-way interactions involving intervention status, time, and park use patterns to assess whether the magnitude of health or behavioral change was dependent on the dose of exposure to the renovated park environment.

Results

The CPI park renovations led to significant neighborhood-level increases in park use duration and park satisfaction among residents living near renovated parks when compared to control neighborhoods. Specifically, the intervention led to a significantly larger increase in the usual length of time residents reported spending at the study park on weekdays [DID = 30.0 minutes (95% CI 10.3, 49.7)] and the total minutes spent at the park in the past month [DID = 466.3 minutes (95% CI 63.0, 869.6)]. Substantial increases were consistently reported across all park satisfaction metrics, with the largest improvements observed in satisfaction with overall park quality [DID = 38.4% (95% CI 25.2, 51.6)] and the maintenance of grounds and facilities [DID = 40.9% (95% CI 27.7, 54.1)].

In the assessment of individual-level health outcomes, CPI was not associated with an overall change in perceived stress in the full longitudinal sample [DID = 0.28 (95% CI -1.47, 2.03)]. However, a significant decrease in perceived stress attributable to the renovations was found for the divorced, separated, or widowed participant subgroup [DID = -4.22 (95% CI -7.92, -0.53)] and middle-aged participants (35-49y) with frequent park use (once/week or more) [DID = -4.46 (95% CI -8.28, -0.64)]. Furthermore, among intervention but not control participants, those with frequent park use (once/week or more) experienced a significantly larger decrease in perceived stress compared to those with infrequent park use (less than once/week) [difference in change = -2.92 (95% CI -5.36, -0.47)].

Similarly, the park renovations alone were not associated with overall changes in accelerometry-measured physical activity and sedentary behavior. Nonetheless, significant dose-response relationships between park use duration and changes in physical activity and sedentary behavior emerged exclusively within the renovated park neighborhoods, suggesting that longer typical visit durations to renovated parks were associated with increased physical activity and decreased sedentary behavior over time. Intervention participants who reported visiting their study park for 30 minutes or more on a weekend day experienced a 140.0-unit/day greater increase in VMPM (95% CI 52.7, 227.2) and a 61.1-minute/day larger decrease in sedentary time (95% CI -108.8, -13.3) compared to those with shorter visit durations. Furthermore, in the intervention group, each additional 60 minutes spent at a study park on weekdays was associated with 18.0 more minutes/day of light physical activity (95% CI 4.7, 31.3), and each additional 60 minutes spent at the study park on weekend days corresponded to 19.0 more minutes/day of light physical activity (95% CI 8.1, 29.9) and a 54.1-unit increase in average VMPM/day (95% CI 7.5, 100.7). No associations between park use duration and changes in physical activity or sedentary behavior were observed among residents living near control parks.

Conclusions

In one of the largest citywide park redesign and renovation projects to-date, CPI achieved considerable success in enhancing park perceptions and increasing the amount of time spent at parks among residents in low-income neighborhoods. However, physical renovations alone were not sufficient to drive widespread community-level improvements in physical activity or perceived stress. The primary and novel finding of this research is that high-quality, renovated park environments may serve as a necessary precondition for translating higher park use into measurable health benefits, including both stress reduction and physical activity improvements. This is supported by the consistent observation that the dose-response relationship between park use and health outcomes was only present among renovated park sites, indicating that the quality of the physical space may be essential prerequisite to the health-benefitting potential of park use. To maximize the public health return on investments in urban greenspaces, future built environment interventions should adopt multidimensional approaches that strategically couple infrastructure upgrades with community-tailored programming and social supports designed to promote the increased park use necessary to fully realize the health benefits of improved parks.

Download

Available for download on Saturday, December 18, 2027

Share

COinS