Residents and Fellows Research in Progress
Tuesday, November 14, 2023
08:30 AM–10:00 AM
Abstract
Introduction: Previous studies suggest that second-dose delays for antibiotics in sepsis and septic shock are associated with worse outcomes. Initiatives to prevent antibiotic delays may cause problems, including redundant antibiotic administration and dose overlap.
Research Question or Hypothesis: What factors are associated with delays or overlap in antibiotic redosing time in patients with sepsis?
Study Design: Single-center, retrospective cohort
Methods: Adult inpatients (n=600) with sepsis or septic shock were identified for screening using SEP-1 compliance reporting criteria. Data were collected from electronic health records. Delay in redosing was defined as >125% of the recommended dosing interval for a single agent or different agents in the same class, accounting for both indication and renal function. Analysis of early redosing was limited to beta-lactams and defined as occurring sooner than 75% of the recommended dosing interval. Utilizing STATA 15.1, patient and process variables were analyzed with Chi-square, Fisher’s exact, and Wilcoxon rank sum as appropriate. Factors associated with delays or early redosing were identified by multivariable logistic regression analysis.
Results: Dosing delays occurred in 14% of patients (n=30/222) who met inclusion criteria (52% septic shock). Early beta-lactam redosing occurred in 23% of patients (n=50/216). In the multivariable analysis, delays were statistically significantly associated with ordering provider type (OR 3.23, 95%CI 1.13-9.30). For early redosing, change in beta-lactam after the first dose was a perfect predictor and excluded from the model. Process variables (e.g., verification time) were significant in univariable analyses, but only ordering provider type (OR 2.62, 95%CI 1.10-6.24), antibiotic redosing interval =12 hours (OR 6.94, 95%CI 2.91-16.56), and sex (OR 2.10, 95%CI 1.02-4.31) remained significant in the multivariable model.
Conclusion: Redosing delays were uncommon and associated with ordering provider type. Early redosing was associated with provider type, dosing interval, and patient sex. Future study is needed to determine the impact of protocols and process changes on antibiotic administration patterns.
Presenting Author
Hannah Stark Pharm.D.Texas Tech Health Science Center School of Pharmacy
Authors
Taryn Bainum Pharm.D., BCPS, BCCCP
Texas Tech University Health Sciences Center School of Pharmacy
Emily M. Buatois Pharm.D., BCPS
Texas Tech University Health Sciences Center School of Pharmacy
Casey Mantzke Pharm.D., BCPS, BCCCP
Northwest Texas Healthcare System
Krystal K. Haase Pharm.D., FCCP, BCPS, BCCCP
Texas Tech University Health Sciences Center School of Pharmacy
Ronald Hall II Pharm.D., MSCS
Texas Tech Health Sciences Center
Amanda N. Charlton Pharm.D.
Texas Tech University Health Sciences Center Jerry H. Hodge School of Pharmacy