Social Media Tracking Illnesses
Greetings! Happy 2013! This is the first full week after the holidays and I thought I would start off the new year right with a blog post.
Hopefully you got your flu shot and weren’t hit with the flu bug that seems to have hit hard and early this year. We got a small stomach bug in the Krafty household. Thankfully it was a 24 hr thing that was gone as quick as it came.
Over the holidays while dealing with the stomach bug, I noticed a reference to a site that tracks illness via social media posts. Sickweather.com, tracks self reported illnesses using social media.
“Just as Doppler radar scans the skies for indicators of bad weather, Sickweather scans social networks for indicators of illness, allowing you to check for the chance of sickness as easily as you can check for the chance of rain.”
Sickweather trolls the social media sites like Facebook and Twitter looking for when people post that they are sick. Using the location based information from the social networks, they are able to map the illness. Their system supposedly is smart enough to differentiate between somebody saying, “I’m sick” and “I’m sick of the Browns losing.” (Interesting little fun fact: According to All Things D, “The company has found it must filter out messages with any mention of the word “fever” that also include the word “Bieber.” Hee hee)
There is one fairly BIG caveat. The social information has to be publicly available. Most people I know lock down their Facebook accounts which means their posts usually aren’t publicly available. (I say usually because Facebook likes to change settings and some people like myself might find themselves posting publicly for a bit thinking they were posting privately.)
Sickweather is an interesting concept, it isn’t the first time people have used the Internet to track illnesses. In August of 2009 FluPortal (seems to be no longer active) used the Internet to “collect and curate content from across public media as well as from trusted governement sources.” They were using the reports from the news media and organizations like the CDC and WHO to build their outbreak maps. It appears that Sickweather is the first company to use social media to track the spread of diseases. (Others like Salathe and Christianini & Lampos for example, have studied the use of social media to track illnesses.)
Despite only having access to public social media updates, Sickweather claims to have declared the start of the flu season 6 weeks earlier than the CDC and may have dectected two whooping cough outbreaks. Clearly there are enough people out there with public accounts (for example my Twitter feed is public but my Facebook isn’t) mentioning their illnesses to make for some interesting results.
Several people have mentioned that you probably don’t want to rely on Sickweather if you have serious health concerns. Duh. However, what I find interesting is the data. There is a boat load of data out there ripe for the plucking and Sickweather is just another example of somebody finding and using the data. There is already an overwhelming amount of medical data out there, and I’m not just talking about “I’m sick” tweets. Data management is big right now. Why? Because as I mentioned there is a ton of data out there. Do a quick search on data management and librarian. The whole first page on Google retrieves pretty relevant results on data management and librarians, including information ARL’s Guide for Research Libraries: The NSF Data Sharing Policy (top result), MIT Libraries support of management & curation of the MIT community research data (second result), and a position description for a Data Management Librarian at Oregon State University Libraries (fourth result, the third result was a slide presentation). Data management has been mentioned several times during #medlibs chat sessions as an emerging role for medical librarians. Margaret Henderson (@mehlibrarian) even stated “Data is the new book. That is where we need to go.”
So how are medical librarians positioned to deal with the onslaught of data? How are library schools teaching librarians or information professionals to work with and manage data? Or are we still playing around with our cataloging systems and copying the journals tables of contents and routing them?
Just seconds after this post went live I saw Kevin the Librarian’s post “A Data Management and Data Sharing Bibliography for Librarians” where he compiled a list of all the literature on data management a librarian would find useful. He and his colleague @fsayre hope to have ”Mendeley group where more librarians can join and share their experiences and ideas about working with data management.” So if you are interested in learning more about data management, go to his site.