Census Tract Food Tweets and Chronic Disease Outcomes in the U.S., 2015⁻2018.

Yuru Huang,Quynh C. Nguyen,Dina Huang

doi:10.3390/ijerph16060975

Abstract

There is a growing recognition of social media data as being useful for understanding local area patterns. In this study, we sought to utilize geotagged tweets—specifically, the frequency and type of food mentions—to understand the neighborhood food environment and the social modeling of food behavior. Additionally, we examined associations between aggregated food-related tweet characteristics and prevalent chronic health outcomes at the census tract level. We used a Twitter streaming application programming interface (API) to continuously collect ~1% random sample of public tweets in the United States. A total of 4,785,104 geotagged food tweets from 71,844 census tracts were collected from April 2015 to May 2018. We obtained census tract chronic disease outcomes from the CDC 500 Cities Project. We investigated associations between Twitter-derived food variables and chronic outcomes (obesity, diabetes and high blood pressure) using the median regression. Census tracts with higher average calories per tweet, less frequent healthy food mentions, and a higher percentage of food tweets about fast food had higher obesity and hypertension prevalence. Twitter-derived food variables were not predictive of diabetes prevalence. Food-related tweets can be leveraged to help characterize the neighborhood social and food environment, which in turn are linked with community levels of obesity and hypertension.

Highlights

The food environment, including access to grocery stores and different types of restaurants, may shape an individual’s food choices as well as eating habits, which could influence chronic disease outcomes [1]
Twitter data have been used to show the geographical variation in diet choices and nutrition, which furthers the understanding of food environments in different neighborhoods and identifies “food deserts” in certain regions [10]
Chow test to assess the differences between the separate regression models, found there was test to assess the differences between the separate regression models, we we found thatthat there was no no statistically significant difference between the coefficient estimates for the Twitter-derived statistically significant difference between the coefficient estimates for the Twitter-derived variables variables in the male-specific obesity model in the female-specific obesity model,a in the male-specific obesity model versus thoseversus in thethose female-specific obesity model, suggesting suggesting a similar strength of associations between the Twitter food environment variables and the sex-specific obesity rates

Summary

Introduction

The food environment, including access to grocery stores and different types of restaurants, may shape an individual’s food choices as well as eating habits, which could influence chronic disease outcomes [1]. Researchers have extensively studied the neighborhood food environment in relation to obesity. Food environment indicators, such as a short distance from supermarkets, more favorable retail food environments [2], and less access to fast food restaurants [3], have all shown to be associated with lower obesity rates [4]. The prevalence of diabetes and hypertension are affected by local food environments [5,6]. Social media, such as Twitter and Facebook, provides a massive amount of user-generated data. Public health researchers have used social media data to track the spread of communicable disease outbreaks. The information retrieved from Twitter is a reflection of the Twitter user’s health behavior, as well as their utilization of neighborhood resources [11]

Methods

Results

Discussion

Conclusion