91Ö±²¥

Background

With Emergency Departments becoming increasingly crowded, it is vital paramedics are supported to make complex decisions on-scene to ensure the patients get to the right place; first time. This project aims to develop and validate a risk prediction model to help Paramedics decide if their patient will have a clear benefit if they got transported to hospital.

This research is supporting paramedics to make more appropriate and effective decisions for patients who may not require the level of care provided by a hospital. It is important as it is aiming to navigate care decisions that will safely provide patients with the right care, first time. If a paramedic can see the likelihood that their patient may have an avoidable attendance, it opens up an opportunity to explore community options. It also empowers the patient to be an active partner in developing a self-care plan

---

Aim

The aim of this project is to design a tool that can be automatically calculated for paramedics to use on scene, which will inform them their patient might not need the Emergency Department. 

Primary objective

To build classification models deriving risk predictions using prehospital clinical data as input variables, and ED experience as the output variable.

Secondary objectives

1. Internally validate the model and apply to a retrospective cohort of non-conveyed patients.
2. Compare the different classification models for most accurate and feasible to embed in practice.

---

Design

Phase 1 will start by creating a dataset of all patients who called an ambulance and got transported to the Emergency Department in Yorkshire. Each episode in the dataset will contain clinical information from both the ambulance service and ED. 

Different mathematical models will be applied to the dataset to try and predict an avoidable attendance at ED (primary outcome measure). Phase 1 will end when the models have been created.

Phase 2 will use statistical methods to internally validate the models.  The models will also be applied to random samples of patients who were not conveyed to the ED. Phase 2 will end when the most accurate model has been identified and selected.

---

Funding

This project is funded by the National Institute of Health Research and Health Education England as part of a Clinical Doctoral Research Fellowship. 

 

---

Team

Job title	Name	Organisation	Email address
Project Lead	Jamie Miles	Yorkshire Ambulance Service and University of 91Ö±²¥	j.miles@sheffield.ac.uk
Primary Supervision	Suzanne Mason	University of 91Ö±²¥	s.mason@sheffield.ac.uk
Statistical supervision	Richard Jacques	University of 91Ö±²¥	r.jacques@sheffield.ac.uk
Ambulance policy and design supervision	Janette Turner	University of 91Ö±²¥	j.turner@sheffield.ac.uk
Clinical supervision	Julia Williams	South East Coast Ambulance Service	Julia.Williams@secamb.nhs.uk

---

Downloads

Privacy Notice

SINEPOST Protocol