Study: Complaining on Twitter Correlates with Heart Disease Risks

This Past Week, a Study was Released by Researchers at the University of Pennsylvania that Found a Surprising Correlation When Studying 2 Kinds of Maps: those that Mapped the County-Level Frequency of Cardiac Disease and those that Mapped the Emotional State of an Area's Twitter Posts. In All, Researchers Sifted through over 826 Million Tweets, Made Available by Twitter's Research-Friendly "Garden Hose" Server Access, then Narrowed those Down to Roughly 146 Million Tweets that had been Posted with Geolocation Data from over 1,300 Counties (Each County Needed to Have at least 50,000 Tweets to Sift through to Qualify). The Team then Measured an Individual County's Expected "Health" Level Based on Frequency of Certain Phrases, Using Dictionaries that had been Put through Scrutiny over their Application to Emotional States. Negative Statements about Health, Jobs and Attractiveness — Along with a Bump in Curse Words — Would Put a County in the "Risk" Camp, while Words like "Opportunities", "Overcome" and "Weekend" Added More Points to a County's "Protective" Rating.


Not Only did this Measure Correlate Strongly with Age-Adjusted Heart Disease Rate Data, it Turned Out to be a More Efficient Predictor of Higher or Lower Disease Likelihood than "Ten Classical Predictors" Combined, including Education, Obesity and Smoking. Twitter Beat that Data by a Rate of 42% to 36%. The Study Acknowledged the Age Issues with Such Data — Namely because the Average Twitter User's Age When the Study was Conducted (Between June 2009 and March 2010) was about 31 Years Old, which is Quite a Few Years Younger than the Average Sufferer of Heart Disease. “The people tweeting are not the people dying", the Study's Authors Wrote. "However, the tweets of younger adults may disclose characteristics of their community, reflecting a shared economic, physical and psychological environment.” Ultimately, the Study's Authors Stated their Data Reinforces the Long-Assumed Health Benefits of Being an Active Participant in your Community, Whether that Happens at Community Centers or among your Best Twitter Friends. "The combined psychological character of the community is more informative for predicting risk than are the self-reports of any one individual", they Added.