The Florida Department of Health (DOH) utilizes the Electronic Surveillance System for the Early Notification of Community-Based Epidemics (ESSENCE-FL) as its syndromic surveillance system. Two of the data sources in ESSENCE-FL are emergency department/urgent care center (ED\UCC) visits and Florida Poison Information Center Network (FPICN) calls. In response to the 2014 Ebola outbreak occurring in West Africa, these two data sources were queried daily in an effort to detect any related visits or calls.
METHODS:
Free text queries containing relevant terms and countries were created and run against the ESSENCE-FL ED/UCC chief complaint and discharge diagnosis data during the period of October 9 – November 12, 2014. As new countries in West Africa reported cases of Ebola and others remained free of cases for two incubation periods, the query was modified to reflect these changes. Here is the last iteration of the query:
^africa^,or,(,^guinea^,andnot,^pig^,),or,^liberia^,or,^sierra^,or,(,^leone^,andnot,^pyleone^,),or,^congo^,or,mali ^,or,^ mali,or,^ mali ^,or,^Ebola^,or,(,^eboli^,andnot,^ebolism^,)
When Ebola could not be ruled out from information available in the query results themselves, follow up was conducted by the county health departments.
Specialists in Poison Information (SPIs) or those answering incoming calls to the FPICN, were instructed to enter one of two codes into their database when a potential Ebola-related call was received. Additionally, SPIs were asked to immediately notify DOH when they received these calls.
RESULTS:
During the period of October 9 – November 12, 2014, 27 visits were identified using the ED/UCC queries. The majority, 16/27 (59%), were patients who had no risk factors for Ebola. Four patients (15%) were part of a hospital Ebola drill, four (15%) had travel to other parts of Africa, two (7%) had other diagnoses, and one (4%) was flagged for a word in the chief complaint that was similar to Ebola. Six calls to FPICN were coded as being related to Ebola. Five (83%) were informational calls and one was incorrectly coded.
CONCLUSIONS:
Syndromic surveillance systems are extremely useful when there is concern about emerging pathogens and can provide crucial redundancy that can help detect cases. In this situation, no cases of Ebola were identified, but important situational awareness was provided, and the query was sensitive enough to identify unknown Ebola related drills. Better communication between hospitals and public health would be mutually beneficial; hospitals would better understand public health surveillance systems and county health departments would be aware of drills and events that may impact these systems.