Whether you have had some good news or a bad day at the office, chances are you are likely to take to social media and share your thoughts and feelings. But you may be surprised to learn that these publicly displayed emotions could predict your risk of heart disease, according to a new study by researchers from the University of Pennsylvania.
Lead researcher Johannes Eichstaedt, of the Department of Psychology, and his team found that the social media platform Twitter can be used to assess the psychological well-being of the US population and predict rates of heart disease.
The researchers publish their findings in the journal Psychological Science.
Approximately 600,000 people in the US die from heart disease each year, making it the leading cause of death in the US.
According to Eichstaedt and his team, certain psychological factors – such as anxiety, stress and depression – are known risk factors for heart disease. Past research from the team has also shown that negative emotions portrayed through language – both speaking and writing – can trigger social behaviors that may lead to heart disease, such as a poor diet and alcohol use.
All of these risk factors, however, have proved challenging to monitor on a large scale, and previous attempts have involved the use of surveys.
“Getting this data through surveys is expensive and time-consuming, but more important, you’re limited by the questions included on the survey,” notes Eichstaedt. “You’ll never get the psychological richness that comes with the infinite variables of what language people choose to use.”
As such, the team turned to Twitter and analyzed the language used in an array of public tweets from over 1,300 counties in the US – accounting for around 88% of the population – made between 2009 and 2010.
Image credit: University of Pennsylvania.
The researchers also analyzed public health data from the Centers for Disease Control and Prevention (CDC) for each county, including information on deaths from heart disease and rates of smoking, obesity, hypertension and physical activity.
Overall, the researchers found that more positive emotional tweets (identified by displays of excitement or optimism through words such as “wonderful” or “friends”) correlated with lower risk of death from heart disease, while negative emotional tweets (established through the use of expletives or words such as “hate”) were associated with increased risk of heart disease death.
“The relationship between language and mortality is particularly surprising, since the people tweeting angry words and topics are in general not the ones dying of heart disease,” says study co-author Hansen Andrew Schwartz, of the Departments of Psychology and Computer and Information Science. “But that means if many of your neighbors are angry, you are more likely to die of heart disease.”
According to study co-author Lyle Ungar, also of the Departments of Psychology and Computer and Information Science, the findings support previous research suggesting that psychological factors at community level can be better predictors of health than such factors identified in a single person. He adds:
“We believe that we are picking up more long-term characteristics of communities. The language may represent the ‘drying out of the wood’ rather than the ‘spark’ that immediately leads to mortality. 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.”
The team says Twitter may be useful for identifying the drivers behind negative emotional states by county, which could help inform public health interventions.
“Twitter seems to capture a lot of the same information that you get from health and demographic indicators,” says study co-author Gregory Park of the Department of Psychology, “but it also adds something extra. So predictions from Twitter can actually be more accurate than using a set of traditional variables.”
Earlier this week, Medical News Today reported on a study claiming the use of social media does not directly contribute to increased levels of stress.