A clinical decision process model for evaluating vancomycin use with modified HICPAC guidelines. Hospital Infection...

The objective of this study was to evaluate a clinical decision process model for the appropriateness of vancomycin use, using modified Hospital Infection Control Practice Advisory Committee (HICPAC) guidelines.

All nondialysis vancomycin use was reviewed using the retrospective chart review method. The HICPAC guidelines were modified to distinguish between documented and suspected infections and appropriateness of vancomycin use initially and after 3 days of therapy. Data were collected on both vancomycin-use orders and vancomycin-use days.

446-bed health maintenance organization teaching hospital.

758 uses of vancomycin from 1993 through 1995 were evaluated using the modified HICPAC guidelines. Initial use was appropriate in 71% of the cases, with 26% used for documented infections and 74% for suspected infections. Of the 536 orders of initial appropriate use, 176 courses of treatment with vancomycin were discontinued appropriately within 3 days. Ongoing use evaluation after 3 days revealed appropriate use in 45%, inappropriate ongoing use in 25%, and empirical ongoing use in 30% of the cases. There were adequate clinical or laboratory data available in 70% of cases after 3 days to discontinue vancomycin or to reclassify from suspected to documented infections or indications. Vancomycin-use evaluation solely after 3 days would not have disclosed 537 initial inappropriate vancomycin-use days, which were 44% of the total inappropriate use days.

Comprehensive evaluation of vancomycin use with HICPAC guidelines should include a modification to encompass initial and 3-day reevaluation, because most initial use is for suspected, and not documented, infections. HICPAC guidelines do not address the issues of differentiating suspected from documented infection indications or ongoing empirical use. The clinical decision process model is a framework for documentation and data collection for use evaluation and addresses issues not covered in HICPAC vancomycin guidelines. This model could be used by other medical centers for evaluation of vancomycin or other antibiotics.