Diabetes
Neighborhood Typology and A1C Trajectories among Women with Gestational Diabetes in New York City: A Longitudinal Study Teresa Janevic* Bohao Wu Wu Wu Wu Wu Wu Department of Epidemiology, Columbia University Mailman School of Public Health, New York, NY, USA
Background: Neighborhood environments are associated with type 2 diabetes (T2D), yet prior studies have focused on single neighborhood determinants rather than multidimensional profiles. Women with gestational diabetes (GDM) experience heterogeneous postpartum glycemic trajectories. We used a data-driven approach to classify New York City (NYC) neighborhoods into distinct profiles and examined their associations with postpartum A1C trajectories among women with GDM.
Methods: Using 2009-2021 linked NYC birth, hospital discharge, and NYC A1c Registry data, we identified GDM cases in birth/hospital data. Women with pre-pregnancy diabetes were excluded. We applied latent profile analysis to 10 social and built environment characteristics in 2010 to classify neighborhoods. The profile with the most average neighborhood characteristics was used as the reference. Participants’ residential addresses were updated at each A1C test to assign neighborhood profiles as the time-varying exposure. We used linear mixed-effects models with random intercepts at the census block and individual levels to estimate adjusted differences in repeated A1C measurements across neighborhood profiles, adjusting for maternal age, race and ethnicity, educational attainment, nativity, and parity.
Results: Among 13,231 women with GDM at baseline, seven distinct neighborhood profiles were identified. Compared with the Profile 5 neighborhoods (the most average), residence in Profile 1 neighborhoods (“Green and Wealthy”, mean difference [β]: -0.11, 95% CI -0.21, -0.01), Profile 4 neighborhoods (“Unhealthy Built Environment and Wealthy”, β: -0.11, 95% CI -0.21, -0.02), and Profile 7 neighborhoods (“Affluent Urban”, β: -0.16, 95% CI -0.25, -0.07) was associated with lower A1C levels.
Conclusions: Considering multidimensional neighborhood environments revealed distinct A1C trajectories after GDM, underscoring the value of neighborhood-based glycemic control.
