SatCare: Remote Support for Ambulance Clinicians in Medical Emergencies

Overview

SatCare is a randomised controlled trial involving rapid standardised ultrasound assessment of patients with shock, major trauma, abdominal pain, chest pain or breathlessness in emergency ambulances. The scans will take less than 5 minutes and be transmitted to a hospital-based expert for review, providing support and instructions for optimal prehospital care. Five Highland Scottish Ambulance Service ambulances covering areas more than 30 minutes from Raigmore Hospital, Inverness, UK, will be equipped with an ultrasound machine (M-Turbo, FujiFilm Sonosite) and satellite transmission system plus webcam, and will be deployed in real emergency situations. When dispatched to a potentially eligible patient, the attending paramedic will contact Raigmore Hospital's emergency department to check the availability of an emergency medicine specialist and obtain study group allocation (ultrasound with enhanced telecommunications plus usual care versus usual care alone). Following verbal consent from the patient, trained paramedics will perform the condition-specific scan protocol in the ambulance at the incident site, and transmit the recordings and patient video via satellite to the emergency department for specialist analysis. The consultant will give advice on patient management via standard ambulance communications systems while it is en route to the hospital. The remotely supported prehospital ultrasound implementation will be examined in terms of its delivery and functioning. An economic evaluation will compare its use with care as usual for eligible patients transported by ambulance, modelling the costs and benefits of this service expansion and determining optimum use. It is hoped that the results, anticipated to be available in 2019, will provide an evidence base for the use of prehospital ultrasound for emergency care.

Research objectives: * To assess the patient health impact of enhanced telecommunications and remotely-supported point-of-care ultrasound scanning (RS-POCUS) in prehospital situations * To assess the impact of introducing this intervention on current care systems, via interviews with various stakeholders involved in the trial (e.g., patients, paramedics, emergency medicine consultants, hospital management, etc.) and health economic modelling Research questions: * What is the health gain, as assessed by the EQ-5D-5L (EuroQol five dimension questionnaire, 5-level version) score at three months, associated with the use of enhanced telecommunications and RS-POCUS? * What is the estimated cost-per-QALY associated with the prehospital RS-POCUS? * Where in the care system and in which situations does the intervention produce the greatest impact? * How does the use of RS-POCUS affect time to definitive treatment in the studied patient groups? Patient Randomisation: Patients will be randomized to the intervention or control group prior to obtaining consent (Zelen randomisation), an approach considered optimal in trials where care is time-critical and patients may be highly stressed. This simplifies and shortens the in-ambulance consent process because the details of the random allocation and its associated uncertainty don't need to be explained until later. Randomisation is performed using randomisation envelopes with codes printed on them, located in the ambulances. Codes will be reported and recorded on the emergency department contact log when a potential patient call-out occurs. If on arrival at the emergency scene the patient meets the inclusion criteria, the paramedic will obtain the patient's study group allocation by opening the envelope. Consent: It is crucial that delays to patient transfer are reduced to an absolute minimum in the ultrasound scanning group. The process of scanning is rapid and offers virtually no risk to the patient, but delays due to requesting consent are potentially more problematic and need to be minimised as much as possible. It is therefore proposed to have a two-stage consent process: a brief verbal consent to scanning and data transmission in the ambulance, and on arrival at the hospital, full written consent to health record analysis and to complete a follow-up questionnaire three months later. If patients are assessed by the paramedic to lack the mental capacity and understanding to provide meaningful consent, an eligible proxy (relative, welfare attorney or guardian) accompanying the patient may provide consent on his or her behalf in line with the Adults with Incapacity Act, Scotland (2000). If a participant who is unable to consent shows any signs of unwillingness to take part or distress, they will be immediately withdrawn from the study. Patient follow-up: Patients will be asked to complete a standard EQ-5D-5L (health-related quality of life) questionnaire 3 months after recruitment to the trial, plus a short satisfaction/service use questionnaire. Ten patients from both the scanned and control groups will be approached at this time to ask if they would be willing to be interviewed about the use of ambulance-based ultrasound. If other specific interesting themes meriting further investigation arise from the interviews, these may be explored further with additional interviews in other participants. Patient involvement in the trial will end on completion of this 3-month follow-up. Analyses: The primary outcome is defined as the patient EQ-5D-5L (EuroQol five dimension questionnaire) score three months after the original ambulance episode. Analysis will be on an intention to treat basis with a per protocol analysis as a sensitivity. Pre-specified descriptive subgroup analyses will examine EQ-5D-5L results by age group, reason for scan and transport distance. The economic analysis will include both a within-trial economic evaluation and a model-based economic evaluation. The within-trial analyses will take the form of a cost-utility analysis from the perspective of the National Health Service. The model-based analysis will also take the perspective of the NHS, but try to take into account wider societal costs such as unpaid carer time and costs, using standard national figures to estimate these contributions. A Markov model will be used to extrapolate from the short-term trial outcomes into the longer term.