Evaluation of timeliness and models of transporting critically ill children for intensive care: The DEPICT mixed-methods study
Ramnarayan P., Seaton S., Evans R., Barber V., Hudson E., Kung E., Entwistle M., Pearce A., Davies P., Marriage W., Mouncey P., Polke E., Rajah F., Hudson N., Darnell R., Draper E., Wray J., Morris S., Pagel C.
Background: Centralisation of paediatric intensive care has increased the need for specialist critical care transport teams to transfer sick children from general hospitals to tertiary centres. National audit data show variation in how quickly transport teams reach the patient’s bedside and in the models of care provided during transport; however, the impact of this variation on clinical outcomes and the experience of patients, families and clinicians is unknown. Objectives: We aimed to understand if and how clinical outcomes and experience of children transported for intensive care are affected by timeliness of access to a transport team and different models of transport care. Methods: We used a mixed-methods approach with a convergent triangulation study design. There were four study workstreams: A retrospective analysis of linked national clinical audit data (2014–16) (workstream A), a prospective questionnaire study to collect experience data from parents of transported children and qualitative analysis of interviews with patients, families and clinicians (workstream B), health economic evaluation of paediatric transport services (workstream C) and mathematical modelling evaluating the potential impact of alternative service configurations (workstream D). Results: Transport data from over 9000 children were analysed in workstream A. Transport teams reached the patient bedside within 3 hours of accepting the referral in > 85% of transports, and there was no apparent association between time to bedside and 30-day mortality. Similarly, the grade of the transport team leader or stabilisation approach did not appear to affect mortality. Patient-related critical incidents were associated with higher mortality (adjusted odds ratio 3.07, 95% confidence interval 1.48 to 6.35). In workstream B, 2133 parents completed experience questionnaires pertaining to 2084 unique transports of 1998 children. Interviews were conducted with 30 parents and 48 staff. Regardless of the actual time to bedside, parent satisfaction was higher when parents were kept informed about the team’s arrival time and when their expectation matched the actual arrival time. Satisfaction was lower when parents were unsure who the team leader was or when they were not told who the team leader was. Staff confidence, rather than seniority, and the choice for parents to travel with their child in the ambulance were identified as key factors associated with a positive experience. The health economic evaluation found that team composition was variable between transport teams, but not significantly associated with cost and outcome measures. Modelling showed marginal benefit in changing current transport team locations, some benefit in reallocating existing teams and suggested where additional transport teams could be allocated in winter to cope with the expected surge in demand. Limitations: Our analysis plans were limited by the impact of the pandemic. Unmeasured confounding may have affected workstream A findings. Conclusions: There is no evidence that reducing the current 3-hour time-to-bedside target for transport teams will improve patient outcomes, although timeliness is an important consideration for parents and staff. Improving communication during transport and providing parents the choice to travel in the ambulance with their child are two key service changes to enhance patient/family experience. Future work: More research is needed to develop suitable risk-adjustment tools for paediatric transport and to validate the short patient-related experience measure developed in this study.