Determinants of inappropriate empirical antibiotic treatment: systematic review and meta-analysis.

This systematic review assessed study-level determinants of non-covering (inappropriate) empirical antibiotic treatment (IEAT), focusing on the influence of study years and the prevalence of multidrug-resistant organisms (MDROs) in the study. Prospective and retrospective observational studies reporting on the association between IEAT and mortality in adult patients with microbiologically documented infections published between 2008-2016 were included. A meta-analysis of IEAT rates was conducted using a random-effects model. Subgroup analyses and mixed-effect single-covariate meta-regression were conducted to identify the association between clinical and methodological study-level covariates and IEAT rates. Heterogeneity was assessed using the I2 measure of inconsistency. Multi-covariate meta-regression was conducted including variables with a P-value of <0.1 on single-covariate analysis. A total of 191 studies were included assessing 73 595 patients, most commonly with bacteraemia. The pooled IEAT event rate was 32% [95% confidence interval (CI) 30-35%], with large heterogeneity (I2 = 97.7%). On multi-covariate analyses, the prevalence of any MDRO [odds ratio (OR) per 10% increase in prevalence = 1.11, 95% CI 1.07-1.15], Acinetobacter spp. specifically (OR = 1.99, 95% CI 1.22-3.25) and advancing study years were associated with IEAT rates (OR = 1.03, 95% CI 1.00-1.06 per year). MDRO rates were independently associated with mortality rates in the studies, adjusting to the rate of IEAT. The prevalence of MDROs, mainly multidrug-resistant Gram-negative bacteria, is significantly associated with the probability of prescribing IEAT and mortality rates in recent studies. We show how antibiotic resistance impacts patient management and outcomes.