Background

As infection with vancomycin-resistant enterococci (VRE) increases in hospitals, knowledge about VRE reservoirs and improved accuracy of epidemiologic measures are needed. Many assessments underestimate incidence by including prevalent carriers in at-risk populations. Routine surveillance cultures can substantially improve prevalence and incidence estimates, and assessing the range of improvement across diverse units is important.

Methods

We performed a retrospective cohort study using accurate at-risk populations to evaluate the range of benefit of admission and weekly surveillance cultures in detecting unrecognized VRE in 14 patient-care units.

Results

We assessed 165 unit-months. The admission prevalence of VRE was 2.2%-27.2%, with admission surveillance providing 2.2-17-fold increased detection. Medical units were significantly more likely to admit VRE carriers than were surgical units. Monthly incidence was 0.8%-9.7%, with weekly surveillance providing 3.3-15.4-fold increased detection. The common practice of reporting incidence using the total number of patients, rather than patients at risk, underestimated incidence by one-third. Overall, routine surveillance prevented the misclassification of 43.0% (unit range, 0%-85.7%) of "incident" carriers on the basis of clinical cultures alone and increased VRE precaution days by 2.4-fold (unit range, 2.0-2.6-fold).

Conclusions

Routine surveillance markedly increases the detection of VRE, despite variability across patient-care units. Correct denominators prevent the substantial underestimation of incidence.