An assessment of selective surveillance methods for detecting hospital-acquired infection.

Three selective surveillance methods were compared to a reference method in their ability to detect hospital-acquired infection (HAI) in patients occupying 122 beds of a district general hospital. The time for data collection was also assessed. The selective methods consisted of: (a) ward liaison surveillance (WLS), conferring with nursing staff twice weekly to determine patients with infections; (b) risk factor surveillance (RFS), the follow-up of patients with "clues" that indicated a risk of infection; and (c) laboratory-based ward liaison surveillance (LBWLS), the follow-up of positive microbiology reports by reviewing case records, in addition to conferring with nursing staff. The reference method consisted of total continuous clinical surveillance and the review of laboratory reports. During the 11-month period of the study, the reference method identified 306 HAI in 3,326 patients. LBWLS identified 71%, WLS 58%, and RFS 49% of HAI. The time for data collection (per week) was 7.75 hours for LBWLS, 4.3 hours for WLS, and 7.9 for RFS. In the United Kingdom, LBWLS was concluded to be an effective method of surveillance.