Background: Prolonged mechanical ventilation increases the risk of mortality and morbidity. Optimising sedation and early testing for possible liberation from invasive mechanical ventilation (IMV) has been shown to reduce time on the ventilator. Alongside a multicentre trial of sedation and ventilation weaning, we conducted a mixed method process evaluation to understand how the intervention content and delivery was linked to trial outcomes.
Methods: 10,495 children admitted to 18 paediatric intensive care units (ICUs) in the United Kingdom participated in a stepped-wedge, cluster randomised controlled trial, with 1955 clinical staff trained to deliver the intervention. The intervention comprised assessment and optimisation of sedation levels, and bedside screening of respiratory parameters to indicate readiness for a spontaneous breathing trial prior to liberation from ventilation. 193 clinical staff were interviewed towards the end of the trial. Interview data were thematically analysed, and quantitative adherence data were analysed using descriptive statistics.
Results: The intervention led to a reduced duration of IMV (adjusted median difference- 7.1 hours, 95% CI -9.6 to -5.3, p = 0.01). Overall intervention adherence was 75% (range 59-85%). Ease and flexibility of the intervention promoted it use; designated responsibilities, explicit pathways of decision-making and a shared language for communication fostered proactivity and consistency towards extubation. Delivery of the intervention was hindered by established hospital and unit organisational and patient care routines, clinician preference and absence of clinical leadership.
Conclusions: The SANDWICH trial showed a significant, although small, reduction in duration of IMV. Findings suggest that greater direction in decision-making pathways, robust embedment of new practice in unit routine, and capitalising on the skills of Advanced Nurse Practitioners and physiotherapists would have contributed to greater intervention effect.
Trial registration: isrctn.org Identifier: ISRCTN16998143.
Copyright: © 2023 Jordan et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.