409 - What about the control group? Provision of a lower-intensity intervention to control sites post-trial to improve bronchiolitis management in Australasian hospitals: A multi-method PREDICT study

Publication Number: 4400.409

Victoria Ramsden, Murdoch Children's Research Institute, randwick, New South Wales, Australia; Franz E. Babl, The Royal Children's Hospital, Melbourne, Victoria, Australia; Libby Haskell, The University of Auckland, Auckland, Auckland, New Zealand; Sandy Middleton, Australian Catholic University, Sydney, New South Wales, Australia; Nicola Straiton, Australia Catholic University, Sydney, New South Wales, Australia; Ed Oakley, The Royal Children's Hospital, Melbourne, Victoria, Australia; Anna Lithgow, The Royal Darwin Hospital, Tiwi, Northern Territory, Australia; Alexandra Wallace, Waikato Hospital, Health New Zealand, Hamilton, Waikato, New Zealand; Mark H. Corden, Northern Health, Ascot Vale, Victoria, Australia; Sonja Crone, Te Whatu Ora Lakes, Health New Zealand, Rotorua, Bay of Plenty, New Zealand; Kate Curtis, University of Sydney, Sydney, New South Wales, Australia; Shane George, Griffith University and Gold Coast Health, Gold Coast, Queensland, Australia; Lambros Halkidis, Cairns Hospital, Cairns, Queensland, Australia; Kirsten Hrabar, ROYAL DARWIN HOSPITAL, Darwin, Northern Territory, Australia; Amit Kochar, Womens And Childrens Hospital, ADELAIDE, South Australia, Australia; Parag Tambe, Southland District Health Board, Invercargill, Southland, New Zealand; Peter H. Wilson, Australian Catholic University, Melbourne, Victoria, Australia; Stuart Dalziel, University of Auckland, Auckland, Auckland, New Zealand; Emma Tavender, Murdoch Children's Research Institute, Melbourne, Victoria, Australia

Background: In 2017, the PREDICT Bronchiolitis Knowledge Translation (KT) Study, a cluster randomised controlled trial (cRCT) was conducted in 26 Australian and New Zealand paediatric hospitals. The cRCT demonstrated a 14.1% reduction at intervention sites in the use of five low-value bronchiolitis practices (albuterol, glucocorticoids, chest radiography, antibiotics, epinephrine) in the first 24-hours of hospitalization, following use of targeted, theory-informed de-implementation interventions. One-year post-trial, control sites received a lower-intensity version of the PREDICT Bronchiolitis KT interventions.

Objective: To determine whether lower-intensity interventions delivered to control hospitals one-year post-trial (2018), improved practice one-year post-receipt (2019) (i.e. two-years post-trial) in terms of improvements in care and clinician views.

Design/Methods: A retrospective medical record audit of bronchiolitis management was conducted in emergency departments (EDs) and paediatric wards that participated in the PREDICT Bronchiolitis KT cRCT. Data from infants with bronchiolitis admitted in 2019 were compared with the 2017 cRCT data. Qualitative interviews were conducted with a purposive sample of clinical staff from ED and inpatient paediatric units at three high- and three low-performing hospitals, examining views on facilitators and barriers that influenced improvement or deterioration in adherence to guideline recommendations. Data were analysed thematically and mapped to the Consolidated Framework for Implementation Research (CFIR).

Results: Audits were conducted at 10 control sites (n=1529 infants). Adherence to guideline recommendations, improved one-year post-receipt of lower-intensity interventions from 74.2%. to 80.8% (adjusted predicted adherence 80.9%, 95%CI: 77.5-84.4). Eleven clinicians were interviewed from 6 sites. Four main themes emerged representing facilitators to improving bronchiolitis management and included 1) a culture of evidence-based practice, 2) team-based autonomy, 3) the presence of nursing champions and 4) senior clinical leadership. The themes were mapped to two CFIR domains of Inner Setting and Individual Characteristics.

Conclusion(s): This study provides evidence of the effectiveness of providing lower-intensity interventions to control sites post an implementation trial to de-implement low-value practices in bronchiolitis management. It identified important facilitators and barriers that influence the de-implementation of low-value care and demonstrates the value in examining the ‘dose’ of a proven de-implementation intervention when scaling improvement efforts.