Surveillance, Community Needs Assessment, Pedagogy
Internet and Social Media Based Surveillance Systems for Detecting Influenza Activity in Maryland
(School of Public Health (UMD) Epidemiology and Biostatistics Master's Student)
Influenza continues to be a public health problem. Seasonal influenza is associated with 3,000-49,000 deaths per year. While pandemic strains, such as the “Spanish Flu” in 1918, killed more than 21 million people worldwide. Influenza surveillance is essential to detecting and managing outbreaks. The Maryland Department of Health and Mental Hygiene (DHMH) currently reports the number of emergency room visits for influenza-like-illness (ILI) to track flu activity. Recently, internet and social media based surveillance methods have emerged as useful in detecting outbreaks. This study aims to determine if internet based surveillance methods are useful in monitoring ILI in Maryland through assessing how Google Flu Trend data compares to DHMH’s formal reporting system and determining if Twitter messages provide more timely information on influenza outbreaks. Previous studies have found that Google Flu Trends, and Twitter data are correlated with the Centers for Disease Control and Prevention (CDC) ILI data. However, these studies have focused on keywords “influenza” and “flu” when searching Twitter. Publically available data from DHMH Maryland Influenza Surveillance Reports and Google Flu Trends will be used to determine correlation between current and past flu seasons by calculating the Pearson’s correlation coefficient. Twitter’s streaming Application Programming Interface was used to gather tweets consistent with the case definition of ILI, fever (cough OR sore throat). Twitter data will be plotted and compared to Google Flu Trends and DHMH data using Pearson’s correlation coefficient. Favorable results have important implications for emergency preparedness and planning procedures.