Earlier this month, Michelle posted about Sickweather, a Baltimore-based start-up that seeks to harness the power of social networks to report the prevalence of local illnesses. Sickweather has developed an algorithm to probe through publicly available information on social networks like Facebook and Twitter to predict the emergence of local illnesses, based on reported symptoms or diagnoses. Sickweather will organize the geographically-tagged information in a “weather map” of symptoms, allowing users to find information relevant to their local communities.
While Sickweather is the first site to utilize user-generated content on social networks to track emerging illnesses, other sites have developed illness trackers that compile information into maps of symptoms. Sickweather’s leading competitor is HealthMap, developed by a team of researchers, epidemiologists and software developers at Children’s Hospital Boston in 2006. This website (as well as the companion mobile app Outbreaks Near Me), monitors and filters infectious disease information from online news aggregators, eyewitness reports, expert-curated discussions, and validated official reports across 9 languages and organizes it into a map. Users can directly click on reports posted on the map, or peruse through a list of all reports. HealthMap has created a flu-specific map, Flu Near You, and a dengue-specific map DengueMap in collaboration with the Center for Disease Control. HealthMap has also collaborated with the US Department of Health and Human Services to build an interactive map for Flu.Gov, a website that provides government-wide information on seasonal and pandemic influenza. Users can directly access state news and state resources directly from the map. Other initiatives include MedWatcher, which compiles reports on new drug warnings and adverse events, and OutbreakMD, a platform for health professionals with limited internet access to track and report diseases.
Google has taken a different approach to tracking illnesses with its Flu Trends and Dengue Trends maps, which use aggregated Google search data to estimate disease activity throughout the world. While Google recognizes that not every individual who conducts searches related to flu or dengue are infected with the illness, comparisons of query counts with traditional illness trackers have shown a high correlation between search queries and disease prevalence. Google has developed an algorithm for estimating cases of the illnesses throughout the world based on search queries. A comparison of Google trends estimates for dengue and influenza and actual government-reported data demonstrate high concordance, legitimizing Google’s approach. Google’s flu and dengue trend maps provide real-time information, allowing for early illness detection and prevention.
Who is Sick? compiles user-reported illness information and symptoms and organizes it in a map. Users report their symptoms, such as runny nose, cough, or stomach ache, how long they have experienced those symptoms, zip code, sex, and age, and have the option to provide more details. They can then directly access other reports from the map, or search and filter all reports by symptoms, location, sex, age and keywords. Who is Sick? organizes illness information in analytics summaries, identifying trends and outbreaks. Discussion boards give users the opportunity to discuss illnesses. Users can also sign up to receive information by e-mail about local outbreaks.
HealthMap, Google and Who is Sick? already utilize different approaches to tracking emerging illnesses, from official data to search queries to user reports, and present this information in the form of a map. Sickweather’s methodology, which compiles publicly available information on social networks, takes advantage of illness information that users report to friends and family in real-time. Only time will tell whether social networks provide accurate information in regard to illness!