Biosurveillance systems have only recently begun to tap into call center data; these projects focus on 911 data in particular. Its value is not yet clearly established. The types of applications fall into two classes: stand alone systems that analyze only 911 data (perhaps implemented as part of a 911 system), and more general biosurveillance systems that collect 911 data as one of several types of data.
FirstWatch Biosurveillance®, a software suite produced by Stout Solutions, LLC, is an example of a stand-alone system used in Kansas City MO, Las Vegas, Plano TX, Tulsa OK, and Des Moines IA, and is being deployed in other areas of the Midwest. The vendor claims that its software will provide early warning during flu season by monitoring 911 data for spikes in respiratory and abdominal symptoms (Stout Solutions, LLC, 2005b). These systems' utility is under evaluation. Governmental public health spokespersons have noted that the FirstWatch alarm system does indeed become busier during flu season, but they question its true value; it has not justified any change in response strategy and its false positive rate is far too high for the system to be of any value, by itself, to detecting an unusual epidemic or bioterrorism event (Archer, 2005). A study conducted by New York City's Department of Health and Mental Hygiene (DOHMH) found that an EMS data collection alarm could be tuned to fairly high sensitivity for a community-wide respiratory outbreak (Mostashari F et al., 2003), but another found that the use of EMS data by itself achieved 58% sensitivity and a positive predictive value of 22% for "influenza-like illness'' (Greenko et al., 2003). We note that any system that focuses on a single category of data collection arrives with the limitations inherent in its respective category. For example, an upper respiratory ailment is usually first seen by a primary care physician or nurse practitioner, not by an ambulance crew. Similarly, a seemingly innocuous skin rash that is caused by a bioterrorism incident is not likely to be reported through an emergency call center.
These results and the paucity of peer-reviewed research regarding use of emergency call center data suggest that we will need to study this subject more closely before we know how to best use these kinds of systems for biosurveillance purposes.
Was this article helpful?