Article Text

Download PDFPDF

2245 Mixed methods study exploring factors influencing ambulance clinician decisions to pre-alert emergency departments (EDs) of a patient’s arrival
  1. Fiona Sampson1,
  2. Richard Pilbery2,
  3. Esther Herbert1,
  4. Jaqui Long1,
  5. Joanne Coster3,
  6. Rachel O’Hara3,
  7. Fiona Bell2,
  8. Steve Goodacre1,
  9. Andy Rosser4,
  10. Rob Spaight4,
  11. Mark Millins4
  1. 1University of Sheffield
  2. 2Yorkshire Ambulance Service
  3. 3ScHARR, University of Sheffield
  4. 4West Midlands Ambulance Service


Aims and Objectives Ambulance pre-alert calls can lead to improved treatment of time-critical patients by enabling Emergency Departments (EDs) to prepare for their arrival but need to be used judiciously to optimise patient care. Despite their importance, there is a lack of research understanding how pre-alert decisions are made. We aimed to understand factors influencing ambulance clinician pre-alert decision-making.

Method and Design Using a convergent parallel mixed-methods design we integrated quantitative and qualitative data from three Ambulance Services and six Emergency Departments using: 1) linked routine dataset of 12 months’ (2020/21) electronic patient records (3 Ambulance Services), clinician information and routine hospital statistics 2) semi-structured interviews with 35 ambulance clinicians and 40 ED staff and 156 hours non-participation observation of pre-alerts across six EDs. Lasso regression to identify candidate variables for multivariate logistic regression was undertaken in R(™) to explain variation in pre-alert rates in terms of patient (NEWS2 score, working diagnosis, age, sex), ambulance clinician (experience, role, sex, time to end of shift) and hospital factors (journey time,% ambulances waiting >30 mins). Qualitative data was analysed using thematic analysis in NVivo(™). Findings were integrated using a triangulation protocol.

Results and Conclusion Variation in pre-alert practice was not fully explained by casemix. Overall 142,795/1,363,274 conveyances were pre-alerted. Highest overall odds ratios (ORs) for pre-alert were associated with patient factors (working diagnosis OR:4.16,CI:4.05-4.26, NEWS2 OR:1.4,CI:1.39-1.4) but thresholds for pre-alerting varied between ambulance clinicians. Pre-alerts were more likely when there were longer turnaround times at EDs (OR:1.83,CI:1.69-1.98), potentially due to ambulance clinicians’ concerns about their ability to effectively manage deteriorating patients where long handovers were anticipated. There was a significant difference in pre-alert rates between EDs (figure 1) that was not explained by type of hospital (e.g. Major Trauma Centres). Anticipated ED response to pre-alerts had a significant impact on pre-alert decisions due to variation in ED protocols and expectations.

Abstract 2245 Figure 1

Odds ratio of pre-alert being made stratified by receiving hospital

Statistics from

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.