Gentrification and childhood obesity: Evidence from New York City public school students in public housing

Eric G. Zhou, Amy Ellen Schwartz, Brian Elbel

Research output: Contribution to journalArticlepeer-review

Abstract

Objective: The study objective was to determine the effect of gentrification on the weight outcomes of New York City public school students living in public housing. Methods: In a prospective cohort of 19,022 New York City public school students in public housing followed during 2009–2017, weight outcomes of students living in public housing buildings in gentrified neighborhoods were compared to those living in consistently low-socioeconomic-status neighborhoods; assignment was quasi-random in each borough. Results: Among the 42,182 student-year observations, gentrification did not increase weight outcomes significantly, for BMI z scores (0.037; 95% CI: −0.012 to 0.086), obesity (0.6 percentage points [pp]; 95% CI: −0.9 to 2.1), or overweight (1.3 pp; 95% CI: −0.7 to 3.2). However, heterogeneous effects by borough were found, where the gentrification in Manhattan increased students' BMI z scores by 0.19 (95% CI: 0.09–0.29), obesity by 3.4 pp (95% CI: 0.03–6.5), and overweight by 9.2 pp (95% CI: 6.3–12.1). No heterogeneity by race and ethnicity, gender, or age was found. Conclusions: With strong internal validity, this study shows that neighborhood gentrification differentially influenced children's health through obesity, based on borough of residence. Such findings could inform policies or interventions focused on subpopulations and geographies.

Original languageEnglish (US)
Pages (from-to)390-397
Number of pages8
JournalObesity
Volume32
Issue number2
DOIs
StatePublished - Feb 2024

ASJC Scopus subject areas

  • Medicine (miscellaneous)
  • Endocrinology, Diabetes and Metabolism
  • Endocrinology
  • Nutrition and Dietetics

Fingerprint

Dive into the research topics of 'Gentrification and childhood obesity: Evidence from New York City public school students in public housing'. Together they form a unique fingerprint.

Cite this