GPS, user data and other factors can give health researchers and officials a new channel for tracking epidemics. Location can be used to monitor disease movement as well as other public health concerns.
Twitter has been shown to be an interesting tool for tracking the spread of disease, and researchers are currently mining its users’ posts and location data to determine how to best use the social network for this purpose. The Journal of Medical Research recently published the article “Right Time, Right Place” on the value of this content in monitoring health communication - how there are “several location indicators in Twitter, and when taken together, they offer a sizable sample of individuals whose location can be accurately inferred.” The authors predict that this could influence how health organizations survey and mine information and “right time, right place” health communication.
Twitter research found to be nearly as accurate as traditional surveys
The large population of Twitter users may only make available a small percentage of reliable GPS coordinates (2.7 percent), but the data set is still useful as this still corresponds to hundreds of thousands of tweets during a 2-week period, the time taken for this study. Between GPS information and user-supplied location information, reliable geo-data is available for between roughly 15 and 17 percent of users. To verify the usefulness of their sample size, the authors compared their results to a Pew survey and achieved statistically similar results. But with Twitter, researchers are able to get minimally less accurate rates with the added benefit of speed and other factors over more traditional methodology.
This social health data stream can illuminate other issues and democratize epidemiology
Besides the potential for tracking or predicting epidemics, there are various public health concerns that can be better understood. Among the issues that have been the subject of recent studies, as referenced in “Right Time, Right Place” are tobacco-related issues, problem drinking, dental pain, and breastfeeding. Observations on real-time, user-generated health content have created a new, high-volume health data stream populated with text, images and video. Due to Twitter’s diverse user demographic, the authors suggest that health disparities may be reduced thanks to more awareness and reach to underrepresented groups and low-income populations.