Improving antibiotic prescribing for cellulitis: a collaboration between three rural health services.

Abstract

Background: An analysis of the National Antimicrobial Prescribing Survey (NAPS) data indicated that antibiotics for cellulitis were more often inappropriately prescribed in regional and remote hospitals compared to major-city hospitals (25.7% v 19.0%, p=<0.001). Given that antibiotic therapy for cellulitis is well-established, this represented an evidence–practice gap. Additionally, deficiencies existed in the information provided to patients about cellulitis.

Objective: To improve the appropriateness of antibiotic prescribing for cellulitis by implementing a cellulitis management plan developed through a collaboration between three Victorian regional or rural (rural) health services.

Methods: An adult lower limb cellulitis management plan, incorporating advice on antibiotic prescribing and a patient leaflet, was co-designed by three rural Victorian health services. Adults with ICD-10-AM codes for lower limb cellulitis or erysipelas admitted as inpatients of the three hospitals between 1 May and 30 November 2019 (baseline) and 1 March and 31 October 2020 (post-implementation) were included. Patients were excluded if they were admitted to ICU during their admission. Antibiotic prescriptions on Day 1 were assessed using the NAPS appropriateness definitions. Patient satisfaction with the information provided to them about cellulitis while in hospital was captured via a phone call after discharge.

Results: Overall, 29% (37/127) of patients with lower limb cellulitis were commenced on the cellulitis management plan. The overall appropriateness of antibiotic prescribing at Day 1 was similar in the baseline and post-implementation groups (79% and 82% respectively). In the post-implementation group, there was a non-statistically significant increase in antibiotic appropriateness when the cellulitis plan was initiated (88% v 79%, 95% CI -5.6% to 19.8%, p=0.20). The proportion of patients who were satisfied with the information provided to them about cellulitis was 100% (22/22) for those who received the cellulitis leaflet and 78% (46/58) in those who did not receive the leaflet (95% CI 4.8% to 34%, p=0.02).

Implications for practice: Cross-organisational collaborations present unique challenges, particularly in rural health services where resources and staffing structures can vary considerably. This study showed that a collaboration between three independent rural health services can conceptualise, design and implement shared resources. The provision of patient-centred information on cellulitis remains important to optimise care and increase patient satisfaction.