Electronic Surveillance for CAUTI: Meeting Compliance for Regulatory Standards
Automated electronic surveillance systems used to identify potential hospital-acquired infections (HAIs) help streamline the work of the Infection Preventionist (IP), reduce human error in applying complicated definitions and enable the IP to spend time on developing interventions to reduce HAIs . Our large healthcare system uses Surveillance Assistant (SA), an internally developed software program, to identify candidate infections for each IP to review. Most facilities currently perform catheter associated urinary tract infection (CAUTI) surveillance only for intensive care unit (ICU) patients, however, mandatory CAUTI reporting in 2015 will expand to include non-ICU locations. This prompted an interest in developing an enhanced electronic surveillance system for CAUTI.
Currently, SA identifies candidate CAUTI - patients with urinary catheters (UC) and positive urine cultures - on specific inpatient units set by the IP. The enhanced algorithm pulled all hospitalized patients from January - June 2013 with a UC, positive urine culture, and documentation of fever (> 38°C) on dates that the UC was in place, including 2 days post removal of UC. A single IP evaluated each patient in the sample data set using January 2013 National Healthcare Safety Network (NHSN) CAUTI definitions. We compared the yield of each surveillance method. UC days were captured electronically.
See Table 1. The current SA algorithm identified 1399 candidate CAUTI from target units (mostly ICU) with a total of 34,671 UC days (1 candidate CAUTI/25 UC days). The refined algorithm applied to all units system wide identified 343 candidate infections with a total of 73,492 UC days (1 candidate CAUTI/214 UC days). Hospital IPs identified 101 CAUTI (7.2% yield, 101/1399), while 171 CAUTI were confirmed from the enhanced algorithm (49.8% yield, 171/343).
Adding fever to the current surveillance method reduced the number of patients requiring review by almost 75% and allowed surveillance on more than twice as many UC days. We plan to evaluate the sensitivity and specificity of the enhanced algorithm before implementation as well as further refine the criteria by adding urinalysis results.
H. M. Babcock, None
R. R. Khoury, None
S. Mccormick, None
See more of: Oral Abstract Session
Findings in the abstracts are embargoed until 12:01 a.m. EDT, Oct. 8th with the exception of research findings presented at the IDWeek press conferences.