A Retrospective, Cross-Sectional Study of Geographic Food Environment and Diabetes in Pregnancy
Recommended Citation
McClain S, Clinton T, Joseph J, Speak A, Goodson M, Dickson T, Khangura R, Kim SK, and Pitts D. A Retrospective, Cross-Sectional Study of Geographic Food Environment and Diabetes in Pregnancy. Am J Perinatol 2025.
Document Type
Article
Publication Date
11-24-2025
Publication Title
American journal of perinatology
Abstract
Insufficient access to healthy food has been linked to poor health outcomes in under-resourced communities. The relationship between neighborhood-level food insecurity and diabetes in pregnancy remains understudied, with previous studies reporting inconsistent results. This study examined the association between living in a low-income, low access (LILA) census tract and the prevalence of pregestational type 2 diabetes (T2D) and gestational diabetes (GDM) among pregnant individuals. This cross-sectional study included patients who delivered a singleton pregnancy at ≥20 weeks' gestation at Henry Ford Hospital between January 2014 and December 2019 and resided within Detroit city limits at the time of delivery. Residence in a LILA census tract, as designated by the USDA Food Access Research Atlas, was the exposure, and prevalences of pregestational T2D and GDM were the outcomes, which were collected retrospectively from patient records. A total of 117 census tracts were designated as LILA. Covariates that were adjusted for included maternal age at delivery, race, body mass index (BMI), insurance status, and substance use during pregnancy (drug, alcohol, and tobacco). Multivariate logistic regression models were used to analyze the data. Of the 3,897 patients included in this study, 1,377 (35.3%) resided in LILA tracts and 2,520 (64.7%) resided in non-LILA tracts. When individuals residing in LILA and non-LILA tracts were compared, there were no significant differences in the prevalences of pregestational T2D (4.8 vs. 4.6%, adjusted prevalence odds ratio [aPOR] = 1.00, 95% CI: 0.72-1.38, p = 0.99) and GDM (11.3 vs. 13.7%, aPOR = 0.96, 95% CI: 0.78-1.20, p = 0.74). Maternal age at delivery, maternal BMI, race, and insurance status were all significantly associated with the prevalences of GDM and pregestational T2D. Our results suggest that a LILA tract is not significantly associated with the prevalences of T2D and GDM during pregnancy. Living in a LILA tract was not linked with GDM or T2D in pregnancy. Age and BMI were significantly associated with T2D and GDM. Race and insurance status were significantly associated with T2D and GDM.
PubMed ID
41232568
ePublication
ePub ahead of print
