This article is more than 5 years old. Information may no longer be current.
Emotions expressed on Twitter may predict CV risk in communities
A new study suggested that it may be feasible to capture community psychological characteristics through social media. Further, these characteristics appear to be strong markers of CV mortality at the community level, according to research published in Psychological Science.
“Hostility and chronic stress are known risk factors for heart disease, but they are costly to assess on a large scale. We used language expressed on Twitter to characterize community level psychological correlates of age-adjusted mortality from atherosclerotic heart disease,” researchers from the World Well-Being Project wrote.
Using a set of public tweets made from 2009 to 2010, the researchers used established emotional dictionaries and automatically generated clusters of words reflecting behaviors and attitudes to analyze a random sample of tweets from individuals who had made their locations available. The random sample included tweets from about 1,300 counties, which covered 88% of the US population, according to a press release.
Language patterns on Twitter reflective of negative social relationships, psychological disengagement, negative emotions including anger, and use of expletives and words like “hate” were identified as risk factors for atherosclerotic heart disease at the community level. Positive emotions and use of words such as “wonderful” or “friends” in tweets were identified as protective factors. These associations remained after the researchers controlled for variables including income and education.
Further analyses indicated that Twitter language was a better predictor of atherosclerotic heart disease than a combination of 10 common health, demographic and socioeconomic risk factors, including as diabetes, hypertension, obesity and smoking.
“Twitter seems to capture a lot of the same information that you get from health and demographic indicators, but it also adds something extra,” researcher Gregory Park, PhD, a postdoctoral fellow in the University of Pennsylvania School of Arts and Science’s department of psychology, said in the release. “So predictions from Twitter can actually be more accurate than using a set of traditional variables.”
These findings complement previous sociological research that suggested that the combined characteristics of communities may be more predictive of physical health than reports of any one individual, according to the release.
“We believe that we are picking up more long-term characteristics of communities. … We can’t predict the number of heart attacks a county will have in a given timeframe, but the language may reveal places to intervene,” researcher Lyle H. Ungar, PhD, from the departments of psychology and computer and information science at University of Pennsylvania, said in the release.
Disclosure: The study is supported by the Robert Wood Johnson Foundation and Templeton Religion Trust.