Sustainability of evidence-based practices in the care for infants with bronchiolitis in hospitals involved in the Knowledge Translation (KT) in Australasian Paediatric Acute Care Settings Study

The 2017 PREDICT ‘Knowledge Translation in Australasian Paediatric Acute Care Settings: a multi-centred, cluster-randomised controlled trial comparing a tailored, theory informed Knowledge Translation intervention versus passive dissemination of a bronchiolitis guideline’ study (PREDICT Bronchiolitis Knowledge Translation (KT) Study) (ACTRN12616001567415), successfully implemented a targeted, theory-informed bronchiolitis intervention which reduced the use of five inappropriate therapies (chest X-ray, salbutamol, glucocorticoids, antibiotics, and adrenaline). The intervention included the following components: engagement of local nursing and medical clinical opinion leaders; a one-day train the trainer workshop; educational materials; and audit and feedback. Our study will undertake medical record audits at the 13 intervention trial hospitals to determine the sustainability of this intervention over time including any improvements. Medical record audits will be conducted on infants who presented to hospital between 1 January 2018 to 31 December 2019, with a diagnosis of bronchiolitis. Data will be extracted on patient demographics; chest x-ray; medication administration (salbutamol, antibiotic, glucocorticoid, adrenaline); respiratory support (non-invasive or invasive ventilation) and disposition from emergency department or inpatient paediatric area (i.e., discharged home or admitted to Intensive Care Unit). Semi-structured interviews (individual or group - depending on clinician availability, via teleconference or face-to-face) will be conducted at 13 intervention trial hospital sites to determine fidelity and adaptation of the intervention and explore factors which may have contributed to improvements in intervention group hospitals. The semi-structured interviews will be approximately 30-45 minutes in duration and will be conducted by two study researchers with a convenience sample of doctors and nurses who treat infants in the participating EDs and inpatient paediatric areas. There will be a maximum of three to five individuals per site (n=78-130).