Appropriateness and determinants of antibiotic prescribing in an Australian emergency department

Antibiotics are the second most frequent drug prescribed in the ED.1 ED staff acknowledge probable overuse of antibiotics,2 but Australian data to support this are scarce. In 2013, 151 Australian hospitals participated in a point prevalence study of antibiotic prescribing (National Antimicrobial Prescribing Survey [NAPS] study) demonstrating 60% of antibiotic use to be compliant with guidelines;3 however, this study specifically excluded the ED. We prospectively studied the appropriateness and determinants of antibiotic prescribing at Royal Perth Hospital, a tertiary hospital whose ED assesses 54 000 patients per year. Patients being admitted to hospital and initiated on empiric antibiotics in the ED were randomly selected and antibiotic appropriateness determined using the NAPS protocol, which assesses adherence to local or national guidelines.4 The main determinant of antibiotic prescription and prescriber attitude towards the role of electronic decision support (EDS) were also surveyed. Approval by our Hospital Research Ethics Committee (A14.005) was obtained. Over 2 months in mid-2014, we assessed 93 prescriptions from 63 patients. Table 1 shows the infectious syndromes and guideline adherence, which was 58% overall. Most prescriptions were written by ED staff (72%) compared with inpatient teams. Trainee registrars wrote more prescriptions (39%) than junior medical officers (24%), consultants (22%) or non-trainee registrars (16%). In keeping with the results of the 2013 NAPS study at our hospital (Matthew Rawlins, personal communication), the most frequently prescribed antibiotics were piperacillin-tazobactam (32%), azithromycin (15%), ceftriaxone (10%) and cephazolin (10%). The most frequent reason for guideline non-adherence was incorrect choice of antibiotic (27/39, 69%) all except one of which was excessively broad spectrum. Thirty-four of 50 (68%) prescriptions were continued unchanged by the inpatient teams after 48 h. The most frequent determinant of antibiotic choice was prior experience with that antibiotic (33%), followed by adherence to national (20%) or local guidelines (18%) and observation of more senior colleagues (13%). Seventy-five of 85 respondents (92%) agreed or strongly agreed that EDS would assist with the appropriateness of their prescribing. The choice and appropriateness of empiric antibiotic use in our ED paralleled that observed on our inpatient wards, with piperacillin-tazobactam being most frequently prescribed and nearly half of prescriptions being non-adherent to guidelines. ED is an important determinant of hospital antibiotic prescribing patterns as most inpatients are admitted via the ED and the majority of patient's antibiotics prescriptions remained unchanged 48 h following hospital admission. This likely reflects the limitations of microbiological testing leading to prolonged empiric prescribing, combined with 'clinical interia',2 or reluctance by doctors to alter antibiotic decisions made by their colleagues. Thus, efforts to improve hospital antibiotic use should extend to the ED, which has historically not been the focus of antimicrobial stewardship interventions.2 However, time constraints,2 a large number of prescribers with rapid staff turnover,5 diagnostic uncertainty2 and pressure to initiate antibiotic therapy promptly in unwell patients are all potential barriers to improving antibiotic prescribing in the ED. In a complex and often chaotic environment, our study demonstrates that ED clinicians mainly rely on prior personal experience, rather than guidelines, when making antimicrobial decisions. A multifaceted approach towards antimicrobial stewardship in the ED is required;2 however, the key to success will be improving accessibility of guidelines. We demonstrated strong ED support for using EDS as a tool for this purpose, which has been demonstrated to be of clinical benefit in the Australian ED context.6 In conclusion, we demonstrated that suboptimal, usually excessively broad spectrum antibiotic prescribing is common in the ED, and this drives inappropriate inpatient antibiotic use. In the busy ED setting, clinicians most frequently prescribe antibiotics based on personal experience; however, EDS offers a means of improving guideline-based decision-making. We thank the NAPS Coordinating Team for use of their assessment tool. None declared.