Is automated electronic surveillance for healthcare-associated infections accurate in the burn unit?

As monitoring requirements for healthcare-acquired infection increase, an efficient and accurate method for surveillance has been sought. The authors evaluated the accuracy of electronic surveillance in multiple intensive care unit settings. Data from 500 intensive care unit patients were reviewed to determine the presence of central line-associated blood stream infection (CLABSI) and catheter-associated urinary tract infection (CAUTI). An electronic surveillance report was obtained to determine whether patients had a blood-line nosocomial infection marker or a urine nosocomial infection marker. Manual review was based on Centers for Disease Control and Prevention criteria. An infection preventionist then reviewed all discrepant cases and made a final determination, which was used as the gold standard. Sensitivity, specificity, false-positive rate, and false-negative rate were then calculated for electronic surveillance. In the burn population the sensitivity of electronic surveillance for CAUTI was 66.66%, specificity 96.5%, false-positive rate 3.44%, false-negative rate 33%; and for CLABSI the sensitivity was 100%, specificity 95%, false-positive rate 4.96%, false-negative rate 0%. In the nonburn population the sensitivity for CAUTI was 50%, specificity 97.9%, false-positive rate 2%, and false-negative rate 30%; and for CLABSI sensitivity was 60%, specificity 98.8%, false-positive rate 1%, and false-negative rate 60%. Burn centers may experience a higher false-positive rate for electronic surveillance of CLABSI and CAUTI than other critical care units.