A nurse-driven penicillin allergy risk score in the preoperative setting was associated with increased cefazolin use perioperatively.

STUDY OBJECTIVE: To characterize and assess the effects of a preoperative, nurse-driven penicillin allergy risk stratification tool on rates of perioperative cefazolin and second-line antibiotic use. DESIGN: Quasi-experimental quality improvement study of penicillin-allergic surgical patients undergoing procedures for which cefazolin is indicated. SETTING: Outpatient Perioperative Care Clinic (PCC) for preoperative surgical patients at a tertiary care center. PATIENTS: 670 and 1371 adult penicillin-allergic PCC attendants and non-attendants, respectively. INTERVENTION: A paper penicillin allergy risk stratification questionnaire was administered during the PCC visit. Nurses were educated on its use. MEASUREMENTS: Antibiotic (cefazolin, clindamycin, vancomycin) use rates in the 24 months before and 17 months after intervention implementation in November 2020 (November 2018 - April 2022) were assessed in penicillin-allergic PCC attendants with statistical process control charts. Multivariable logistic regression assessed antibiotic use rates pre- and post-intervention adjusting for age, sex, surgical specialty and penicillin allergy history severity. Similar analyses were done in penicillin-allergic PCC non-attendants. MAIN RESULTS: Of 670 penicillin-allergic PCC attendants, 451 (median [IQR] age, 66 (Sousa-Pinto et al., 2021 [14])) were analyzed pre-intervention and 219 (median [IQR] age, 66 (Mine et al., 1970 [13])) post-intervention. One month after implementation, process measures demonstrated an upward shift in cefazolin use for PCC attendants versus no shift or other special cause variation for PCC non-attendants. There were increased odds of cefazolin use (aOR 1.67, 95% CI [1.09-2.57], P = 0.019), decreased odds of clindamycin use (aOR 0.61, 95% CI [0.42-0.89], P = 0.010) and decreased odds of vancomycin use (aOR 0.56, 95% CI [0.35-0.88], P = 0.013) in PCC attendants post-intervention. This effect did not occur in PCC non-attendants. There was no increase in perioperative anaphylaxis post-intervention. CONCLUSIONS: A simple penicillin allergy risk stratification tool implemented in the preoperative setting was associated with increased use of cefazolin and decreased rates of second-line agents post implementation.