Benchmarking inappropriate empirical antibiotic treatment

Area of Science:

• Infectious Diseases
• Pharmacology
• Public Health

Background:

• Inappropriate empirical antibiotic treatment for severe infections is linked to increased mortality.
• Superfluous antibiotic use contributes to the induction of antimicrobial resistance.

Purpose of the Study:

• To define acceptable rates of inappropriate empirical antibiotic treatment for severe infections.
• To establish benchmarks for evaluating local treatment performance.

Main Methods:

• Systematic review and meta-analysis of prospective cohort studies (1975-2009).
• Inclusion of microbiologically documented infections from 87 studies (27,628 patients).
• Pooled estimate calculation using a beta-binomial model and control limits derived from overdispersion factor technique with winsorized data.

Main Results:

• The pooled rate of inappropriate empirical antibiotic treatment was 28.6% (95% CI 25.4-31.8).
• Benchmarking using control limits indicated 86% of studies fell within acceptable parameters after applying specific statistical methods.
• Significant heterogeneity was observed, not explained by clinical or methodological factors.

Conclusions:

• A pooled rate of 28.6% for inappropriate empirical antibiotic treatment highlights a significant clinical challenge.
• The developed funnel plot and control limits provide a tool for benchmarking local performance.
• Evaluating at least 500 patients is recommended before establishing local treatment rates.