Welcome, dear reader, to the first (of what we hope to be many) post in our new Digital Disease Detection column. My name is Elaine Nsoesie. I am a data-loving public health enthusiast with an interest in tracking and forecasting disease outbreaks. As a post-doc at HealthMap, I’m finding new data sources that could serve as early indicators of social disruption. And that includes public health events.
I’m trying to determine whether or not we can learn about events of social disruption before they happens. For my work, an event of social disruption is anything that changes a society’s routines. This includes public health events like outbreaks. Specifically, I am looking for new digital data sources that might hint that a socially-disruptive public health event is on its way.
I’m not alone in this endeavor. If you’ve been paying attention to the news lately, you might have come across stories covering the use of non-traditional or novel data sources such as Twitter and Facebook to study diseases. In case you’ve missed them, a few examples are “Facebook Likes Predict Obesity,” “Twitter Could Tell You Where Flu Is Ramping Up, Study Suggests,” and “Can Twitter Predict Where You'll Get Food Poisoning?”
Today, people share almost everything online – which can be good or bad. Sharing disease-related information is definitely good for researchers, like me, who use digital resources to discover who is sick, where and when. For example, a fellow HealthMapper showed that cholera-related tweets tracked the early stages of the 2010 cholera outbreak in Haiti. More recently, social media was used in sharing and tracking disease information during the 2013 H7N9 influenza outbreak in China. This rapidly-growing and exciting field of using digital resources to study diseases is usually referred to as digital epidemiology or digital disease detection.
To put it simply, digital disease detection is the use of media sources – like Facebook, Twitter, web searches, news reports, chat rooms, and blog posts – and other digital technologies to detect early indications of a disease outbreak, and track the spread of diseases. Several systems and apps have been developed to monitor news reports on disease outbreaks, rapidly spread information on outbreaks, monitor travel-related deaths etc. A major advantage of these systems is the ability to obtain and make disease data publicly available earlier than traditional surveillance systems. Traditional systems typically rely on data from physician visits and laboratory confirmations of disease—accurate but slow to collect. Ideally, if we detect disease outbreaks early, we can better respond to them and limit their impact on our communities.
In this column, we hope to post twice a month on recent advances in digital disease detection and expert opinions on the field, and to share our excitement on the potential of using digital resources to monitor, track and, ideally, forecast infectious diseases. As with any research field, there are ethical concerns and limitations and we’ll discuss those too. This column will also cover ethical issues regarding the use of personal information and tracking individuals online, and challenges to using web-based data sources. Ultimately, we hope to learn more about this exciting field and explore its potential to complement traditional disease surveillance.