[Skip to Navigation]
Comment & Response
July 2017

Regarding Twitter as a Potential Data Source for Cardiovascular Disease Research

Author Affiliations
  • 1Department of Cardiovascular Diseases, Mayo Clinic, Rochester, Minnesota
JAMA Cardiol. 2017;2(7):819-820. doi:10.1001/jamacardio.2017.0194

To the Editor First, congratulations are in order for Sinnenberg et al1 for their wide-reaching investigation and insights into using Twitter for cardiovascular disease (CVD) research. This novel publication joins an emerging literature base attempting to demonstrate value for CVD professionals using Twitter.2 While innovative and important for the fields of CVD and health care social media (#hcsm), portions of their findings are troubling, specifically their assertions regarding the relative age of those Tweeting about CVD and comparison of terms used in CVD Tweets. The mean age of those Tweeting about CVD was older than the average age of Twitter users (28.7 years vs 25 years); whether this imparts any true value to the effect of #hcsm on the field of CVD is uncertain. Likewise, seeing that the term myocardial infarction was Tweeted nearly 30-fold more than the term heart failure does not fit with our own experience.3 Is it possible there was more noise than signal in some of these terms or perhaps that these were not the best terms to study and analyze?

Add or change institution