Before AI tells patients what to eat, it needs to know where they live: a cross-sectional ecological study of diabetes and food access in California and Michigan census tracts, with a spotlight on Metro Detroit and Metro Los Angeles
- 1University of the Cumberlands
Abstract
Background: AI chatbots now give people with diabetes diet advice, but whether those foods can be bought near home is rarely checked. Methods: In a cross-sectional ecological study of 11,974 California and Michigan census tracts, I linked diabetes prevalence (CDC PLACES 2025) with two USDA food-access maps: all SNAP-authorized stores (2025) and large grocery stores only (2019). I identified double-burden tracts (high diabetes and low income/low access) and hidden food deserts (low access to large grocery stores, but not to any SNAP store). In a pilot, ChatGPT, Gemini, and Claude generated meal plans for the same person across 20 high-diabetes tracts in Metro Los Angeles and Metro Detroit; I scored whether advice was healthy and whether foods were sold at nearby store types. Results: Double-burden tracts were few (71 in California, 19 in Michigan), almost none in the two metros. Food-desert tracts had a 1.5 percentage-point higher diabetes prevalence (95% CI, 1.0-2.0). Counting all SNAP stores hid 567 Californian and 329 Michigan food deserts; in Metro Detroit, all 97 disappeared, including 51 high-diabetes tracts with about 19,500 adults with diabetes. Chatbot advice was healthy, but in hidden tracts 30% of recommended foods were not reliably sold nearby; the chatbots barely adjusted. Of 55 places they named, 13% had errors, and most were over a mile away. Conclusions: The current USDA map obscures food deserts, especially in Detroit, and AI diet advice barely adjusts for them. AI nutrition tools for diabetes should include a place-aware check built on public food-access data.
Related
- Pre-diabetes and diabetes among young first-degree relatives of patients with type 2 diabetes: prevalence and associated factors in a cross-sectional study in Dar-es-Salaam, Tanzania
- Factors associated with chronic kidney disease stage among patients with uncontrolled diabetes and/or hypertension in Northeastern Thailand: a cross-sectional study
- Association of salivary function with oral health-related quality of life in patients with type 2 diabetes and xerostomia: cross-sectional study
- Serum Osteopontin Is Associated with Cardio-Ankle Vascular Index-Defined Arterial Stiffness in Patients with Type 2 Diabetes Mellitus: A Cross-Sectional Study
- Cardiovascular risk profile and treatment patterns in patients with type 2 diabetes in specialized cardiology outpatient care in Colombia : A retrospective cross-sectional study
BibTeX
@misc{chitipolu2026before,
title = {{Before AI tells patients what to eat, it needs to know where they live: a cross-sectional ecological study of diabetes and food access in California and Michigan census tracts, with a spotlight on Metro Detroit and Metro Los Angeles}},
author = {S. S. Chitipolu},
howpublished = {medRxiv},
year = {2026},
month = nov,
doi = {10.64898/2026.11.03.26367208},
url = {https://www.medrxiv.org/content/10.64898/2026.11.03.26367208v1},
note = {CC BY},
}
RIS
TY - UNPB
TI - Before AI tells patients what to eat, it needs to know where they live: a cross-sectional ecological study of diabetes and food access in California and Michigan census tracts, with a spotlight on Metro Detroit and Metro Los Angeles
AU - S. S. Chitipolu
PY - 2026
DA - 2026/11/03
PB - medRxiv
DO - 10.64898/2026.11.03.26367208
UR - https://www.medrxiv.org/content/10.64898/2026.11.03.26367208v1
AB - Background: AI chatbots now give people with diabetes diet advice, but whether those foods can be bought near home is rarely checked. Methods: In a cross-sectional ecological study of 11,974 California and Michigan census tracts, I linked diabetes prevalence (CDC PLACES 2025) with two USDA food-access maps: all SNAP-authorized stores (2025) and large grocery stores only (2019). I identified double-burden tracts (high diabetes and low income/low access) and hidden food deserts (low access to large grocery stores, but not to any SNAP store). In a pilot, ChatGPT, Gemini, and Claude generated meal plans for the same person across 20 high-diabetes tracts in Metro Los Angeles and Metro Detroit; I scored whether advice was healthy and whether foods were sold at nearby store types. Results: Double-burden tracts were few (71 in California, 19 in Michigan), almost none in the two metros. Food-desert tracts had a 1.5 percentage-point higher diabetes prevalence (95% CI, 1.0-2.0). Counting all SNAP stores hid 567 Californian and 329 Michigan food deserts; in Metro Detroit, all 97 disappeared, including 51 high-diabetes tracts with about 19,500 adults with diabetes. Chatbot advice was healthy, but in hidden tracts 30% of recommended foods were not reliably sold nearby; the chatbots barely adjusted. Of 55 places they named, 13% had errors, and most were over a mile away. Conclusions: The current USDA map obscures food deserts, especially in Detroit, and AI diet advice barely adjusts for them. AI nutrition tools for diabetes should include a place-aware check built on public food-access data.
ER -