Comparison of prior authorization and prospective audit with feedback for antimicrobial stewardship.

OBJECTIVE: Although prior authorization and prospective audit with feedback are both effective antimicrobial stewardship program (ASP) strategies, the relative impact of these approaches remains unclear. We compared these core ASP strategies at an academic medical center. DESIGN: Quasi-experimental study. METHODS: We compared antimicrobial use during the 24 months before and after implementation of an ASP strategy change. The ASP used prior authorization alone during the preintervention period, June 2007 through May 2009. In June 2009, many antimicrobials were unrestricted and prospective audit was implemented for cefepime, piperacillin/tazobactam, and vancomycin, marking the start of the postintervention period, July 2009 through June 2011. All adult inpatients who received more than or equal to 1 dose of an antimicrobial were included. The primary end point was antimicrobial consumption in days of therapy per 1,000 patient-days (DOT/1,000-PD). Secondary end points included length of stay (LOS). RESULTS: In total, 55,336 patients were included (29,660 preintervention and 25,676 postintervention). During the preintervention period, both total systemic antimicrobial use (-9.75 DOT/1,000-PD per month) and broad-spectrum anti-gram-negative antimicrobial use (-4.00 DOT/1,000-PD) declined. After the introduction of prospective audit with feedback, however, both total antimicrobial use (+9.65 DOT/1,000-PD per month; P < .001) and broad-spectrum anti-gram-negative antimicrobial use (+4.80 DOT/1,000-PD per month; P < .001) increased significantly. Use of cefepime and piperacillin/tazobactam both significantly increased after the intervention (P = .03). Hospital LOS and LOS after first antimicrobial dose also significantly increased after the intervention (P = .016 and .004, respectively). CONCLUSIONS: Significant increases in antimicrobial consumption and LOS were observed after the change in ASP strategy.

A retrospective analysis of the effect of patient-specific factors on voriconazole concentrations in oncology patients.

OBJECTIVE: To identify patient-specific factors significantly associated with voriconazole exposure. DESIGN SETTING, AND PARTICIPANTS: Retrospective, single center at an academic medical center. Consecutive, adult oncology inpatients who received voriconazole by mouth and had at least one voriconazole level over a 14-month period. Voriconazole was ordered for 292 patients during the study period, a level was obtained in 41 patients. Nine patients were excluded; the study population was comprised of 32 patients. METHODS AND RESULTS: Univariate and multivariable regression analyses were utilized to predict the patient-specific factors significantly associated with level. A total of 36 levels meeting inclusion/exclusion criteria were performed in 32 patients. Sixty percent of levels (22/36) were between 1 and 5.5 µg/mL. Data were available to calculate a Child Pugh score for 66% (21/32) of patients. Using multivariable linear regression, three covariates were found to be statistically significant: age, international normalized ratio (INR), and alkaline phosphatase (ALP). Three outliers were notable in the ALP category, when removing the three individuals from the model, only an increasing INR remains significantly associated with increasing voriconazole level (p = 0.0093). No correlation was found with trough level and Child Pugh score. CONCLUSIONS: Sixty percent of voriconazole trough levels were between 1 and 5.5 µg/mL (range 0.1-7.4 µg/mL), only increasing INR was significantly associated with rising voriconazole level. Increasing Child Pugh score was not associated with increasing level. More investigation is warranted into the usefulness of the Child Pugh score for guidance of dose modifications in non-cirrhotic patients with acute hepatic injury.