Automating case extraction and screening for hospital-associated Staphylococcus aureus bacteremia surveillance in two Canadian hospitals.

OBJECTIVE

Outline our process in creating an efficient surveillance system through increased automation of detecting hospital-associated Staphylococcus aureus bacteremia (HA-SAB). To describe the rate and attributable source of HA-SAB over time at our institutions.

METHODS

We developed an algorithm for extracting cases that met the temporal case definition of hospital-onset Staphylococcus aureus bacteremia (HO-SAB) through our electronic medical record system. The expanded temporal definition included both instances of S. aureus bacteremia (SAB) if occurring>48 hours after admission (HO-SAB Current Admission) or if admitted ≤48 hours but had a previous admission within the past 4 weeks (HO-SAB Previous Admission). Following extraction, clinical review was performed to ascertain if the algorithm extracted HO-SAB cases were hospital-associated SAB (HA-SAB) versus community-associated SAB (CA-SAB) and source(s) of infection.

RESULTS

Over a 5-year period from 2020 to 2025, the rate of HA-SAB was 15.2-26.2 per 10,000 patient admissions and 1.5-2.7 per 10,000 patient days. Our automated process screened 5,146 total cases of SAB, and determined that 222 cases (4%) fit the temporal definition of HO-SAB. Of these 222 HO-SAB, we classified 189 (85%) cases as HA-SAB and 33 (15%) as CA-SAB. The most common sources of HA-SAB were central lines (23%) and peripheral IVs (22%).

CONCLUSION

An automated process is effective at screening cases of SAB. This can improve workflows to enable a sustainable HA-SAB surveillance program as a quality metric for institutions. Our expanded definition of HO-SAB allowed detection and evaluation of an additional 38 (20%) previous admission cases.