Combining Crowd-Sourcing, Census Data, and Public Review Forums for Real-Time, High-Resolution Food Desert Estimation

Abstract

Objective The metrics for measuring the food environment’s health, called food desert indices, are primarily based on decadal census data, limiting their frequency and geographical resolution to that of the census. This research augments census data with other information sources to create a food desert index that is real-time and geographically more granular. Materials and Methods We started with a tract-level food desert index. We used demographic, geographic, and health data, in addition to real-time data from platforms such as Yelp, Google Maps, and crowd-sourced answers to questionnaires by the Amazon Mechanical Turks. We trained a model that learns the relationship between the baseline food desert index and the more granular or up-to-date information sources. Results We use this real-time, context-aware, and geographically refined index in a concept application that suggests alternative routes with similar ETAs between a source and destination in the Atlanta metropolitan area to expose a traveler to better food environments. The resulting model was sensitive to changes in the environment that occurred after the census data was collected. Discussion Our calculations of 248,000 routes to 28,000 food retailers show that the Atlanta food environment creates a strong bias towards eating out rather than preparing a meal at home when access to vehicles is limited. Conclusions Modern, real-time, and more diverse information sources provide opportunities for researchers to analyze varying food environments. In addition to informing community policymakers more accurately, this approach allows us to measure environmental effects on individuals and suggest personal behavioral changes.