Epidemics: STDS- How we Can Predict them and Prevent them
How Facebook Can Predict the Spread of STDs
Peter Leone of the University of North Carolina’s Center for Infectious Diseases recently found that risks of contracting HIV can come from immediate friend circles, rather than just significant others/sexual partners. The reasoning is that people in the same social circles usually act in the same “risk-taking patterns” and tend to sleep with the same people. Leone studied an outbreak of syphilis in North Carolina as an example. He was able to connect 80% of the cases in this outbreak after asking patients who they hung out with. Leone then asked patients recently diagnosed with an STD for a list of friends most likely to be at risk. After getting the patient’s permission, doctors notified people that someone they know has been diagnosed with an STD and explained why they may be at risk as well. Not your everyday Facebook message. Using Facebook can enable researchers and doctors “to make connections that may not be obvious at first.”
There are many principles we’ve discussed in class in action here. First of all, when we look at the outbreak in North Carolina, notions of the small world phenomenon are brought up. The six degrees of separation- the idea that a person is separated from anyone else on this planet by just six other people at maximum- plays in here. It makes sense that Leone was able to like 80% of seemingly disparate cases together. If he looked further, he could link all of them, most likely, with not more than one or two degrees of separation. Facebook and its mutual friends feature a great way in seeing connections that are not so obvious. A threshold cascade model can also be brought up here. If we could figure out what the threshold is for a certain social circle/network, we could introduce a disease preventing technology (i.e. condoms) to the right amount of friends and to the friends with the most influence (usually the ones at “the center of the circle”- ones with the most connections) to eventually get the whole network to use this technology. For example, if we find that the threshold q=.3; this means that a node will switch to A if at least 30% of its neighbors use A. If we played this right, we could stop the spread of disease within a network to a large extent.