According to new research, a new process that tracked infected patients by address simplified the detection of flu outbreaks at long-term care facilities.
Researchers at the New York City Dept of Health and Mental Hygiene (DOHMH) obtained a list of 175 nursing homes in the New York metropolitan area at the start of the 2013-2014 influenza season and used geographical coding techniques to assign each facility a building identification number (BIN).
The same technique was used to code the addresses of patients with laboratory-confirmed influenza after they were logged into the New York State Electronic Clinical Laboratory Reporting System (NYSECLRS).
The automated program was operated daily from Sept 2013 through May 2014 — roughly spanning the flu season — and DOHMH received 15,876 reports for 13,508 cases of influenza, of which 99% were received through the NYSECLRS. Of these, 12,681 were coded by address, and 426 either matched the BIN codes of long-term care facilities (55%) or were identified by approximate text match (44%) or keyword search (1%).
Upon manual review, 249 of the 426 matches were found to be valid; the rest were deemed faulty for various reasons including inaccurate text matching and incorrect address reporting. Valid matches corresponded to 109 influenza outbreaks at 70 long-term care facilities (LTCFs), more than 90% of the 119 outbreaks reported at 75 long-term care facilities in the New York metro area during the 2013-2014 flu season, the researchers wrote. Nearly half of the outbreaks were detected by building analysis before other methods of detection.