Citywide park renovations and changes in perceived stress: a quasi-experimental study among low-income communities in New York City

Abstract

Background: Quality parks have the potential to promote well-being and health equity in urban communities through reduced stress, yet high-quality epidemiological evidence is limited. This quasi-experimental study measured associations between park renovation and changes in perceived stress among low-income adults in New York City.

Methods: Pre- and post-renovation data on the Perceived Stress Scale and park use from 162 adults living near (< 0.3 miles) 31 renovated parks and 151 adults living near 21 sociodemographically matched control parks were analyzed. Linear mixed-effects difference-in-difference (DID) regression measured the association between park renovation and change in perceived stress (post-pre) in the overall sample and stratified by baseline sociodemographics. Additional models explored the interaction of post-renovation park use frequency [high (≥ once/week), low (< once/week)] and intervention status on changes in perceived stress.

Results: Overall, changes in perceived stress were similar between intervention and control groups [DID = 0.28 (95% CI -1.48, 2.03)]. However, park renovation was associated with a significant decrease in perceived stress among divorced/separated/widowed participants [DID = -4.22 (95% CI -7.92, -0.53)] and middle-aged participants (35-49y) with high park use [DID = -4.46 (95% CI -8.28, -0.64)]. Among intervention but not control participants, those with high park use experienced a significantly larger decrease in perceived stress compared to those with low park use [DID = -2.92 (95% CI -5.36, -0.47)].

Conclusions: In one of the first and largest studies on park quality improvement and mental health, park renovation near one's home was associated with decreased perceived stress among divorced/separated/widowed adults and middle-aged frequent park users. Frequent users of renovated parks experienced a larger drop in perceived stress than infrequent users, suggesting that high-quality parks may be an important pre-condition to the benefits of frequent park use on stress reduction.

Keywords: Built environment; Green space; Health equity; Mental health; Park renovation; Stress.

Fig. 1

Mean PSS scores among adult PARCS study participants before and after CPI park renovation intervention in (a) overall sample (b) stratified by age at baseline and (c) stratified by marital status at baseline. Means by intervention group (green square = intervention, grey circle = control) and 95% confidence intervals (vertical bars) were estimated using a linear mixed effects regression models adjusted for education, public housing, and marital status at baseline. Solid lines indicate significant (p < 0.05) change over time in mean PSS score within a given group, while dashed lines indicate non-significant change over time. The asterisk indicates a significant DID estimate among divorced, separated or widowed participants. Abbreviations– PSS: Perceived Stress Scale; PARCS: Physical Activity and Redesigned Community Spaces; CPI: Community Parks Initiative; DID: Difference-in-Difference

Fig. 2

Mean PSS scores among adult PARCS study participants before and after CPI park renovation (a) in intervention vs. control groups stratified by frequency of study park use at follow up, and (b) for participants with high park use (≥ once per week) vs. low park use (< once per week) stratified by intervention status. Means by park use frequency at follow-up and intervention group and 95% confidence intervals (vertical bars) were estimated using a linear mixed effects regression model adjusted for education, public housing, and marital status at baseline. In (a), the results are stratified by park use frequency at follow-up, with intervention group means represented by green squares, and control group means represented by grey circles. In (b), the same results are stratified by intervention status, with mean PSS for high park use represented by blue squares, and mean PSS for low park use represented by orange circles. Solid lines indicate significant (p < 0.05) change over time in mean PSS score within a given group, while dashed lines indicate non-significant change over time. The asterisk symbol indicates a significantly larger decrease in PSS over time among participants with high park use compared to those with low park use within the intervention group. Abbreviations– PSS: Perceived Stress Scale; PARCS: Physical Activity and Redesigned Community Spaces; CPI: Community Parks Initiative

Fig. 3

Mean PSS scores among adult PARCS study participants before and after CPI park renovation intervention stratified by age at baseline and study park use at follow-up. Means by intervention group (green square = intervention, grey circle = control) and 95% confidence intervals (vertical bars) were estimated using a linear mixed effects regression models adjusted for education, public housing, and marital status at baseline. Solid lines indicate significant (p < 0.05) change over time in mean PSS score within a given group, while dashed lines indicate non-significant change over time. The asterisks indicate significant differences between intervention and control groups in change in PSS within a given subgroup. Abbreviations– PSS: Perceived Stress Scale; PARCS: Physical Activity and Redesigned Community Spaces; CPI: Community Parks Initiative