The 130th Annual Meeting of APHA

3237.0: Monday, November 11, 2002 - 2:30 PM

Abstract #45435

Two syndromic surveillance methods for bioterrorism, the New York City experience

Don Weiss, MD, MPH1, Jennifer Leng, MD2, Rick Heffernan, MPH1, Debjani Das, MPH1, Katie Bornschlegel, MPH1, Bryan Cherry, VMD, PHD1, Joel Ackelsberg, MD, MPH3, Pauline Thomas, MD2, Marcelle Layton, MD1, Sharon Balter, MD, MFA1, Farzad Mostashari, MD, MSPH3, and Adam Karpati, MD, MPH3. (1) Communicable Disease Program, New York City Department of Health, 125 Worth St, Box 22a, New York, NY 10013, (212) 941-2265, dweiss@health.nyc.gov, (2) Integrated Surveillance, New York City Department of Health, 125 Worth St, Box 22A, New York, NY 10013, (3) Communicable Diseases, New York City Department of Health, 125 Worth St, 22A, New York, NY 10013

Immediately following the attack on the World Trade Center the NYC Department of Health and CDC implemented emergency department (ED) syndromic surveillance for agents of biological terrorism. The initial system involved 15 hospitals utilizing 18-24 hour staffing by Epidemic Intelligence Service Officers (EISO) who monitored hospital staff syndrome coding. This was replaced by a 30 hospital electronic system based on chief complaint logs (ELEC). Both systems operated seven days a week and analyzed for temporal and spatial clustering. Timeliness was similar with the daily analysis available for interpretation within12-15 hours of collection. Reporting completeness of EISO was 75% while for ELEC it is 95%. The EISO system was implemented in 3 days and required a large number of staff to operate. The ELEC system evolved over two weeks, required fewer staff but necessitated involvement of hospital information systems personnel. Alarm frequencies for the fever, respiratory and gastrointestinal syndromes were compared. During the 24 days of EISO surveillance there were 12 spatial (hospital or zip code) alarms (0.5/day) with a mean number of excess cases=9. For the ELEC system the number of alarms for a thirty-one day sample period was 6 (0.2/day) with a mean number of excess cases=22. None of the alarms were determined to be a biologic agent release. Syndromic surveillance has the ability to detect small clusters of disease, which may prove useful to state and local health departments as an early warning system for identification of an intentional biologic agent release.

Learning Objectives:

Keywords: Bioterrorism, Surveillance

Presenting author's disclosure statement:
I do not have any significant financial interest/arrangement or affiliation with any organization/institution whose products or services are being discussed in this session.

Bioterrorism and Epidemiology: Questions, Methods and Outcomes

The 130th Annual Meeting of APHA