JMIR Publications

ABSTRACT

Background:

The COVID-19 pandemic has inflicted global devastation, infecting over 750 million and causing 6 million deaths. In an effort to control the spread of the virus, governments around the world implemented a variety of measures, including stay-at-home orders, school closures, and mask mandates. These measures had a significant impact on dietary behavior with individuals discussing more home-cooked meals and snacking on social media.

Objective:

The study explores pandemic-induced dietary behavior changes using Twitter images and text, particularly in relation to obesity, to inform interventions and understand societal influences on eating habits. Additionally, the study investigate the impact of COVID-19 on emotions and eating patterns.

Methods:

In this study, we collected 200,000 tweets related to food between May and July in 2019, 2020, and 2021. We employed transfer learning and a pre-trained ResNet-101 neural network to classify images into four health categories: definitely healthy, healthy, unhealthy, and definitely unhealthy. We then used the state obesity rates from The Behavioral Risk Factor Surveillance System (BRFSS) to assess the correlation between state obesity rates and dietary images on Twitter. The study further investigates the effects of COVID-19 on emotion changes and its relation to eating patterns via sentiment analysis. Furthermore, we illustrated how the popularity of meal terms and health categories changed over time, considering varying time zones by incorporating geolocation data.

Results:

A significant correlation was observed between state obesity rates and the percentages of definitely healthy and definitely unhealthy food images in 2019. However, no significant correlation was observed in 2020 and 2021, despite higher obesity rates. A significant increase in the percentage of healthy food consumption was observed during and after the shutdown, as compared to the pre-shutdown period. A Sentiment analysis from 2019, 2020, and 2021 revealed a significantly more positive sentiment associated with dietary posts from 2019. This was true regardless of the healthiness of the food mentioned in the tweet. Lastly, we found the average meal timing shifted later in the day and snack consumption increased during the pandemic. Overall, dietary behavior shifted towards healthier choices at the population level during and post the COVID-19 shutdown, with potential for long-term health consequences.

Conclusions:

This study used social media data to investigate the effects of the COVID-19 on dietary behaviors. Deep learning for image classification and text analysis were applied, revealing a decline in users' emotions and a change in dietary patterns and attitudes. The findings of this study suggest the need for further investigations into the factors that influence dietary behaviors and the pandemic's implications of these changes for long-term health outcomes.