Background The fast turnover of patients make the Emergency Department Observation Units (EDOU) a complex setting for antimicrobial stewardship interventions. To identify modifiable factors to improve inappropriate antimicrobial prescriptions (AT) can help in the design of targeted interventions.

Purpose Our objective was to identify modifiable factors related to inadequate AT in the EDOU by performing repeated point prevalence surveys (PPS).

Material and methods PPS of all antimicrobial prescriptions for non-trauma patients admitted to the EDOU were performed daily for 5 consecutive weeks starting in February 2015. The main outcome variable was the rate of inadequate ATs, when any of the following criteria were not optimal according to local guidelines. Data included demographics, clinical assessment performed by the prescriptor (syndrome, source, severity at onset, type of acquisition), microbiological samples taken and antimicrobial prescriptions including the drug, dose and route of administration, if empirical or targeted, and mono or combination. Multivariate analysis was performed using logistic regression.

Results Overall, 406 ATs were analysed. The most frequent syndromes were pneumonia (24%), urinary tract infections (22%) and non-pneumonic lower respiratory tract infections (22%); 51.5% (n=209) AT were inadequate (26% of them: drug with a reasonable spectrum was prescribed despite not being recommended as first line, 45% antibiotic not needed, 25% ‘inadequate spectrum’ and 4% others). In multivariable analysis, microbiological samples before AT (OR: 1.9; 95% CI: 1.2 to 2.8; p=0.004), specification of the source of infection in patient’s charts (OR: 2.0; 95% CI: 1.1 to 4.2; p=0.05) and severe sepsis or shock (OR: 1.9; 95% CI: 1.2 to 2.9; p=0.003) were independent predictors of adequate AT.

Conclusion Half of the prescriptions were inadequate using very strict criteria. Interventions aiming at improving antibiotic use in this Unit should include education and promotion of optimal clinical procedures for antibiotic prescribing. Quality indicators such as taken micrrobiological samples and the description of source of infection in the medical chart were predictors of better AT.

References and/or acknowledgements No conflict of interest.