Monday, September 30, 2013
Twitter, Public Health and Supply Chains
This week’s topic is about the role of technology and the web on global supply chains. Last semester I took a population health class, where we discussed how researchers have begun to use social media data to, “…model public health at a population scale.” For example, Michael Paul and Mark Dredze, researchers from John Hopkins University, analyzed data from Twitter to see what public health information was out there but more specifically to track influenza. The technique used is known as syndromic surveillance, where they tracked trends in medical conditions over time. The focus on influenza was because the infection is episodic and widespread. They were able to find patterns with influenza outbreaks within a specific location. Their hope is to eventually link up the geography data with the known public health data to have a valid model to predict disease outbreaks.
Another article stated that they were able to quantify the impact of social status, exposure to pollution, interpersonal interactions, travel patterns, and other important lifestyle factors on health by analyzing Twitter posts. They have also found that models developed for infectious disease can be applied to study mental health disorders.
I think that data mining social media to identify trends within public health is a great tool. The models created not only can predict where the next outbreak could be globally, but could also contribute to predicting the amount of vaccinations and medications needed. For example, using the influenza outbreak model could help the pharmaceutical industry produce the predicted demand for vaccinations and know where to ship it, in a timely manner. This additional information could be applied to FEMA’s or the American Red Cross’s supply chain system. They can be better prepared in terms of having the appropriate amount of supplies and staff to control infectious disease outbreaks.
Do you think social media data is a valid source to identify public health patterns and predict future trends?