Through an analysis of posts on X (formerly Twitter) relating to obesity published between April 2019 and December 2022, has revealed that most of these tweets convey negative sentiments about overweight people. The positions taken by influential politicians and celebrities play an important role in shaping public opinion on this issue, according to the study’s researchers from the Geneva University Hospitals (HUG) and the University of Geneva (UNIGE).
The study points the finger at the discrimination suffered by people who are overweight and the role of social networks in the propagation of these criticisms. Understanding public attitudes and perceptions toward obesity is essential for the development of effective health policies, prevention strategies and treatment approaches.
The study was carried out by Dr Jorge César Correia, senior resident in the Therapeutic Patient Education Unit at the HUG, under the supervision of Professor Zoltan Pataky, Staff Physician, Head off Therapeutic Patient Education Unit at the HUG and Associate Professor at the UNIGE Faculty of Medicine. It was performed in collaboration with researchers from the Universities of Liverpool and the Greater Manchester Mental Health NHS Foundation Trust.
The team conducted an algorithmic and manual analysis to acquire an in-depth understanding of the widespread attitudes and perceptions relating to obesity on Twitter. It used an artificial intelligence-based platform of text classification fine-tuned on about 198 million tweets for different applications, including sentiment analysis in eight different languages.
The study examined the sentiments of the general public, political personalities, celebrities and important organizations relating to obesity by analysing 53,414 tweets published in English on this issue between April 2019 and December 2022 in the midst of the COVID-19 pandemic. Analysis of these tweets revealed that they were mainly negative (69.36%) and on the rise, whereas the neutral (20.91%) and positive ones (9.73%) remained stable.
Tweets connected to obesity were also very often associated with racism, with less judicious life choices and with social ills such as the consumption of illicit substances and alcohol.
The study demonstrates that negative representations of obesity by influential politicians and celebrities contribute to negative sentiments among the general public and to the perpetuation of stereotypes and prejudices toward people who are obese or overweight. This stigmatisation may have impacts on the mental health and well-being of these individuals and have adverse effects on public health.
The researchers noted that spikes in criticism were linked to political events and comments by celebrities in the US and in the UK. For example, many negative messages were transmitted when overweight celebrities were hospitalized with COVID-19, when the President of the US announced he wanted to lose weight and when the British government launched its anti-obesity campaign.
The study notes that when prominent people publish negative comments or opinions on obesity, their subscribers are more likely to engage in similar conversations, thus perpetuating the negativity even more. It therefore demonstrates the role of influencers in shaping public opinion on health issues and underscores that public personalities should be aware of the potential consequences of their pronouncements in terms of public health.
The researchers noted that the results of this study have improved understanding of the role of social media in shaping the sentiments, attitudes and health behaviours of the general public on a range of issues, namely obesity in this case.
They will also be useful in the development of effective public health policies, prevention strategies and treatment approaches which today must take account of the unique challenges posed by social media and digital communications in terms of public health.
The findings were reported int eh paper, ‘Exploring Public Emotions on Obesity During the COVID-19 Pandemic Using Sentiment Analysis and Topic Modeling: Cross-Sectional Study’, published in the Journal of Medical Internet Research. To access this paper, please click here
Comments