# Beyond supermarkets: ethnic stores, food environments, and the limits of the Food Access Research Atlas

## TL;DR

This study shows that the Food Access Research Atlas may not accurately reflect food environments in minority neighborhoods by overlooking the role of ethnic stores.

## Contribution

The study introduces a Google Maps-based measure and a multi-ethnic nutritional survey to evaluate food store healthfulness.

## Key findings

- Low-access census tracts do not consistently align with store count or density.
- Ethnic and large stores offer healthier, more affordable food than conventional and small stores.
- FARA's assumption that supermarkets are the gold standard is challenged by the findings.

## Abstract

The Food Access Research Atlas (FARA) is a nationally used measure for community food environment that informs resource allocation to improve food access and population health. However, because FARA flags census tracts (CTs) as low-access solely based on CT population’s proximity to supermarkets, it assumes supermarkets as the gold standard of food stores and may not adequately capture the food environments in racial minority neighborhoods, where ethnic stores can play a critical role.

To examine the accuracy of FARA and its underlying assumption by comparing FARA with our novel Google Maps-based Measure and evaluating the healthfulness of diverse food store types with our multi-ethnic compilation of Nutritional Environment Measures Survey (NEMS).

This cross-sectional study in Durham, North Carolina, leveraged Google Maps to develop three CT-level variables for food store access (intensity, per capita count, and density) and compared them between low-access and not-low-access CTs classified by FARA. This study then developed the first multi-ethnic NEMS compilation and conducted it among small, large, conventional, and ethnic food stores in Durham to evaluate their respective ability to provide healthy, affordable, and quality food.

The geographic distribution of low-access CTs was not consistent with that of CT-level store count. The intensity, per capita count, and density of large stores and ethnic stores did not significantly differ between low-access and not-low-access CTs. From NEMS, ethnic and large food stores could provide healthier, more affordable, and higher quality food than conventional and small food stores.

By highlighting FARA’s limitations in measuring community food environment and casting doubt on FARA’s underlying assumption, this study highlights the need to shift the discourse away from the binary narrative that a lack of supermarkets equals a food desert, and instead examine the food access provided by existing networks of grocery stores, particularly ethnic stores in minority neighborhoods.