(Medinfo Panel) Is social media a good tool for disease surveillance?

Panel: New Trends in Health Social Media: Hype or Evidence-based medicine?

Panelist: Kerstin Denecke, http://www.linkedin.com/pub/kerstin-denecke/34/73b/a90

Threats to the health of individuals are of concern for all of us. When something happens around us, we are communicating this to our friends, colleagues or to the world often through the Web in blogs, twitter messages or forum postings. “I feel so sick” or “I have a high temperature, pain everywhere” – these are messages that can be found on Twitter. But to what extent are those messages referring to real disease outbreaks? In the last years, the idea came up of making use of this chatter for the early detection of disease outbreaks. Methods were developed that exploit content from informal sources, applying sophisticated event-detection techniques to identify potential threats and providing signals to the user in a personalized way [1]. 

However, many challenges still need to be addressed. The language used in social media, in particular in microblogging services such as Twitter is extremely difficult to analyze automatically due to the high variability of language constructions and the immense volume of irrelevant postings [2]. “Football fever” or “(Justin) Bieber fever” are not of interest for public health officials, but these or similar word constructs are coming up from time to time. Sophisticated filtering algorithms were developed, but their applicability and efficacy still need to be evaluated.

There are still open questions:

  • Does social media really provide us with the information necessary for early warning of disease outbreaks? 
  • To what extent is the information reliable we can get from social media?
  • Privacy versus the need to act?


[1] Kerstin Denecke , Peter Dolog , Pavel Smrz: Making use of social media data in public health. Alain Mille et al. (Eds.): Proceedings of the 21st World Wide Web Confer-ence, WWW 2012, Lyon, France, April 16-20, 2012, pp. 243-246

[2] Mustafa  Sofean , Avaré Stewart , Matthew Smith , Kerstin Denecke: Medical Case-Driven Classification of Microblogs: Characteristics and Annotation. ACM SIGHIT International Health Informatics Symposium (IHI 2012)


Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

%d bloggers like this: