Abstract
Background
The use of bystander video livestreaming from scene in Emergency Medical Services (EMS) is becoming increasingly common to inform decisions about the resources and support required. Possible benefits include clinical and financial gains, but evidence is sparse. We aimed to investigate the feasibility of conducting a definitive randomised controlled trial (RCT) of its use in major trauma incidents.
Objectives:
(i) To obtain data required to design a subsequent RCT.
(ii) To test trial processes.
(iii) To embed a process evaluation.
Design
A feasibility RCT with embedded process and economic evaluations where working shifts (n=62) in six trial weeks were randomised 1:1 to video livestreaming or standard care only; and two observational sub-studies: (i) assessment of acceptability in a diverse inner-city EMS that routinely uses video livestreaming; and (ii) assessment of staff wellbeing in an EMS that does not use livestreaming (for comparison to the trial site). Qualitative data collection included observations (286 hours) and interviews with staff (n=25) and bystander callers (n=2).
Setting
A pre-hospital EMS in South-East England, with follow-up in associated major trauma centres and trauma units; Sub-studies in (i) London and (ii) East of England EMS.
Participants
(i) Patients involved in trauma incidents (n=269); (ii) bystander callers (n=11); and (iii) ambulance service staff (n=67).
Intervention
Video livestreaming using GoodSAM Instant-on-Scene.
Main outcome measures
Progression to a definitive RCT based on four pre-defined criteria and consideration of qualitative data: (1) ≥ 70% bystanders with smartphones agreeing and able to activate livestreaming; (2) ≥50% requests to activate livestreaming resulting in footage being viewed; (3) Helicopter Emergency Medical Services stand-down rate reducing by ≥10% due to livestreaming; (4) no evidence of psychological harm to bystanders or staff caused by livestreaming.
Results
Sixty-two shifts were randomised, contributing 240 eligible incidents (132 control; 108 intervention). In a further three shifts we randomised by individual call which contributed four eligible incidents (2 control; 2 intervention), thereby totaling 244 incidents involving 269 patients. Video livestreaming was successful in 53 incidents in the intervention arm. Patient recruitment (to access medical records to assess appropriateness of dispatch), and bystander recruitment (to measure potential harm) were both low (58/269, 22% of patients, 4/244, 2% of bystanders). Two progression criteria were met: (1) 86% of bystanders with smartphones agreed and were able to activate livestreaming; (2) 85% of requests to activate livestreaming resulted in viewed footage; and two were indeterminate due to insufficient data: (3) 2/6 (33%) stand down due to livestreaming; (4) no evidence of psychological harm from survey, observations, or interviews. In sub study (i), dispatch staff reported that non/limited English language and older age may present barriers to video livestreaming.
Limitations
Poor recruitment of patients and bystanders limited assessment of appropriateness of dispatch decisions and potential psychological harm.
Conclusions
Video livestreaming is feasible to implement, acceptable to both bystanders and dispatchers, and may aid dispatch decision-making, but further assessment of benefits and harm is required.
Future work
Findings support the design and conduct of a future multi-centre study taking account of different triage systems and dispatch personnel, potentially using an alternative to an RCT due to rapid uptake of video livestreaming in this setting.