A dollar-aware food-environment index and a 27-year trajectory typology: a measurement foundation for diet and childhood-obesity research in Mississippi, 1997-2024
- 1Brown University School of Public Health
Abstract
Background. The retail food environment is a widely used exposure in behavioural-nutrition and obesity research. Grocery stores have declined across rural America while limited-assortment discount ("dollar") stores have proliferated, yet standard food-environment indices typically exclude dollar stores. We developed a dollar-aware index, assessed how counting dollar stores changes the measured exposure, and derived a trajectory typology. Methods. Using annual establishment data from Data Axle for all 878 Mississippi census tracts (1997-2024), we classified food retailers with code-based rules and computed the modified Retail Food Environment Index (mRFEI) in standard and dollar-aware forms. We fitted latent class trajectory models suited to the bounded index, tested their stability in seven sensitivity analyses, and related class membership to the Social Vulnerability Index (SVI) and urbanicity with multinomial regression. Results. Grocery stores fell from 1,616 to 716 while dollar stores rose from 315 to 1,005, crossing in 2018. The indices ranked tracts almost identically (within-year Spearman [≥]0.99), but where dollar stores and healthy outlets coexisted, counting dollar stores lowered the index by 4.5 points in 1997 and 9.1 in 2024, and tracts with only dollar-store retail, which the standard index cannot score, rose from 1 to 32. Six trajectory classes emerged: stable adequate (7.8% of tracts), steady decline (15.3%), early collapse (12.9%), late collapse (9.7%), persistently constrained (34.8%) and chronically low (19.6%); only 7.5% of children in classified tracts lived in stable-adequate tracts. Classes were stable across random starts and pandemic-year exclusion; collapse was robust, but its early-late timing split depended on tracts with few retailers. Chronically low and persistently-constrained tracts were predominantly urban (relative risk ratios 8.3 and 4.5), consistent with food swamps; persistently-constrained and steady-decline membership rose modestly with vulnerability (1.17 and 1.20 per 0.1 SVI), an association that was weaker with 2010 SVI; collapse classes had no cross-sectional social signature. Conclusions. Counting dollar stores changes the level of a widely used food-environment index, and whether a growing set of tracts can be scored, rather than how tracts rank. A 27-year typology separates urban food swamps from tracts whose food environments collapsed and provides a time-varying exposure for diet and childhood-obesity research.
Related
- Childhood-Oncology Intelligence for Predicting Cancer Emergence, Optimizing Precision Therapies, and Simulating Lifelong Outcomes in Pediatric Malignancies Using Autonomous Foundation Models
- Shared reading and independent book engagement in early childhood: associations with the home literacy environment in a Turkish pediatric outpatient population
- Relationship between BMI-SDS trajectories and childhood asthma: a prospective cohort study from the Japan Environment and Children’s Study.
- Beyond Categorical Tiers: Improving Measurement of Medical Urgency as the Foundation for Heart Allocation Reform
- Early childhood changes in BMI z-score and serum IGF-1 levels at age 6 years: the Japan environment and children study.
BibTeX
@misc{mandalapu2026dollaraware,
title = {{A dollar-aware food-environment index and a 27-year trajectory typology: a measurement foundation for diet and childhood-obesity research in Mississippi, 1997-2024}},
author = {S. V. Mandalapu and S. Lefebvre and E. D. Walker},
howpublished = {medRxiv},
year = {2026},
month = aug,
doi = {10.64898/2026.08.20.26360912},
url = {https://www.medrxiv.org/content/10.64898/2026.08.20.26360912v2},
note = {CC BY},
}
RIS
TY - UNPB
TI - A dollar-aware food-environment index and a 27-year trajectory typology: a measurement foundation for diet and childhood-obesity research in Mississippi, 1997-2024
AU - S. V. Mandalapu
AU - S. Lefebvre
AU - E. D. Walker
PY - 2026
DA - 2026/08/20
PB - medRxiv
DO - 10.64898/2026.08.20.26360912
UR - https://www.medrxiv.org/content/10.64898/2026.08.20.26360912v2
AB - Background. The retail food environment is a widely used exposure in behavioural-nutrition and obesity research. Grocery stores have declined across rural America while limited-assortment discount ("dollar") stores have proliferated, yet standard food-environment indices typically exclude dollar stores. We developed a dollar-aware index, assessed how counting dollar stores changes the measured exposure, and derived a trajectory typology. Methods. Using annual establishment data from Data Axle for all 878 Mississippi census tracts (1997-2024), we classified food retailers with code-based rules and computed the modified Retail Food Environment Index (mRFEI) in standard and dollar-aware forms. We fitted latent class trajectory models suited to the bounded index, tested their stability in seven sensitivity analyses, and related class membership to the Social Vulnerability Index (SVI) and urbanicity with multinomial regression. Results. Grocery stores fell from 1,616 to 716 while dollar stores rose from 315 to 1,005, crossing in 2018. The indices ranked tracts almost identically (within-year Spearman [≥]0.99), but where dollar stores and healthy outlets coexisted, counting dollar stores lowered the index by 4.5 points in 1997 and 9.1 in 2024, and tracts with only dollar-store retail, which the standard index cannot score, rose from 1 to 32. Six trajectory classes emerged: stable adequate (7.8% of tracts), steady decline (15.3%), early collapse (12.9%), late collapse (9.7%), persistently constrained (34.8%) and chronically low (19.6%); only 7.5% of children in classified tracts lived in stable-adequate tracts. Classes were stable across random starts and pandemic-year exclusion; collapse was robust, but its early-late timing split depended on tracts with few retailers. Chronically low and persistently-constrained tracts were predominantly urban (relative risk ratios 8.3 and 4.5), consistent with food swamps; persistently-constrained and steady-decline membership rose modestly with vulnerability (1.17 and 1.20 per 0.1 SVI), an association that was weaker with 2010 SVI; collapse classes had no cross-sectional social signature. Conclusions. Counting dollar stores changes the level of a widely used food-environment index, and whether a growing set of tracts can be scored, rather than how tracts rank. A 27-year typology separates urban food swamps from tracts whose food environments collapsed and provides a time-varying exposure for diet and childhood-obesity research.
ER -