Traditionally, the indicators of heart disease have included income, education, tobacco use, rates of obesity as well as rates of diabetes. Now it seems we need to add Twitter to the list. Researchers from Penn recently published a study that found a correlation between the number of negative tweets and the prevalence of heart disease. Specifically, researchers used geotagged tweets that were negative in nature (ie expressed anger, hostility) and then also examined the data on deaths due to heart disease. Now although the tweeters aren’t the ones dying of heart disease, the tweets served as a fairly reliable predictor of feeling in the community. The idea is that if your neighbors are angry, then you are more likely to die of heart disease.
The implications and applications of this finding are great and far reaching. One benefit that comes to mind is that this data could be a quick and inexpensive way to better target and direct funds communities in need. Additionally, this data can be used to supplement data that has already been gathered from more traditional means (paper surveys and phone surveys) and can help that data be more representative and inclusive. Next, the researchers plan to look into if positive tweets can protect people from heart disease.
Link to Washington Post Article: Tweets can better predict heart disease rates than income, smoking and diabetes, study finds
Link to Original Study: Psychological Language on Twitter Predicts County-Level Heart Disease Mortality (the full text of the article should be available for free if access from a Penn connected computer)