SNAP: Supportive Non-Invasive Ventilation for Acute Chest Syndrome Prevention in Hospitalized Children with Sickle Cell Disease: Facilitators, Barriers, and Contextual Factors Related to Implementation

Background: Acute chest syndrome (ACS) causes significant morbidity and mortality for patients with sickle cell disease (SCD), and often develops during a hospitalization for acute SCD pain due to chest wall splinting, hypoventilation, and atelectasis due to prior rib infarcts, pain and opioid use. While incentive spirometry is an important component of ACS prevention and management, additional strategies are needed to prevent atelectasis during sleep. Bi-level positive airway pressure ventilation (BiPAP) provides positive pressure breaths through a mask to support ventilation, and is most commonly used to treat acute and chronic respiratory failure and obstructive sleep apnea. In 2017, we began utilizing BiPAP during sleep as supportive care on the general pediatric inpatient unit for hospitalized children with SCD and vasoocclusive pain to prevent ACS and/or respiratory decompensation among children with mild-moderate ACS. We previously demonstrated that novel use of BiPAP on a general pediatric unit was safe, feasible and tolerated in 40/53 (75%) hospitalizations. A subsequent qualitative study of patients, parents, and health care team members at 3 sites found strong agreement that BiPAP is perceived as effective at preventing adverse respiratory outcomes; it is appropriate on a general pediatric unit and does not require transfer to the intensive care unit (ICU); and that the patient experience is the biggest challenge to BiPAP use. The goals of the current study were to understand contextual factors, facilitators and barriers to widespread implementation of “supportive non-invasive ventilation for ACS prevention” (SNAP) in general pediatric inpatient units across a variety of institutions.

Methods: We conducted semi-structured interviews with health care team members at 3 sites with varying levels of SNAP implementation: Boston Medical Center (BMC, extensive SNAP use), Children’s Medical Center Dallas (CMCD, limited SNAP use), and Children’s Hospital Pittsburgh (CHP, no SNAP use) about their experiences with and/or perceptions of SNAP. Interviews and a priori codes for the preliminary codebook were guided by the Promoting Action Research on Implementation in Health Services (PARiHS) framework using the Evidence, Context, and Facilitation constructs. Interview transcripts were double coded by three team members using NVivo 12. The codebook was refined until consensus was reached. Team members independently reviewed data to identify preliminary themes and subsequently consolidated themes as a group. Here we present unit- and institutional-level contextual factors, facilitators, and barriers to implementation of SNAP in hospitalized patients with SCD.

Results: Interviews were completed with 29 participants (BMC, n=11, CMCD, n=6, CHP, n=12) until thematic saturation was reached, including: physicians (hospitalists, hematologists, and pulmonologists), nurses, respiratory therapists (RT), child life specialists, psychologists, and clinical leadership. Themes included: 1) Communication among the different clinical staff involved is critical for successful implementation of SNAP. 2) Nurses are key to implementation success, including daytime nurses because many patients with SCD will sleep during the day, and this can help with BiPAP acclimatization. 3) Multidisciplinary support to facilitate daytime BiPAP trials would alleviate some of the burden on overnight nurses, and aid both implementation and patient acceptability. Integration of RT and Child Life is helpful in this process. 4) Specific institutional facilitators and barriers impact implementation, including unit size, whether patients with SCD are admitted to a specific unit, staffing levels, equipment availability, buy-in from leadership, and culture of innovation.

Conclusions/Future directions: Members of health care teams at 3 sites perceive that optimizing communication among clinical team members, improving integration of RT and child life, and incorporating day shift team members into the implementation of SNAP would improve success. Each institution has its own specific and distinct facilitators and barriers to implementation, and strategies will need to adapt to those unique features. These data offer important insights as we develop a protocol for a multi-center hybrid effectiveness/implementation trial of SNAP for hospitalized children with SCD.