Identification of the Sick Patient in the ED

This is a single-centre, non-interventional, prospective observational cohort study collecting data from patients presenting to the ED of a large teaching hospital in the Northwest of England. The study is designed to gather information about the number of patients receiving life-saving intervention within 48 hours of presentation and the best of five proposed ways of identifying these patients when they present to the ED. This study is non-interventional and there will be no interference with treatment decisions or alteration in clinical practice.

A prospective observational approach is chosen as it will allow direct access to information and better capture of the systems being assessed in the ED. Information about each patient's perceived risk at presentation (in terms of the five proposed systems) will be collected at the time of presentation. Information will include rating on the five proposed systems and a number of outcome variables, all of which are recorded as part of routine clinical practice. All information will be captured directly into a specially designed database including a study number to preclude the need to record any personal identifiable information for the study.

Data will be collected for each patient on an anonymised database. Demographic details (age, gender, presenting condition) will be collected to identify if the sample is representative of a general ED population. Each patient's Manchester Triage category, NEWS at presentation and the novel score calculated from NEWS data will be included, as will the impression (yes/no) of the triaging nurse and the treating clinician as to whether the patient will need a lifesaving intervention. A number of outcome measures (death, admission to high dependency or intensive care, admission directly to the cardiac catheter laboratory or coronary care, emergency endoscopy, thrombolysis, life-saving surgery, intubation, emergency intravenous medication, CPR, DC cardioversion/external or internal cardiac pacing, massive transfusion for ongoing haemorrhage and non-invasive ventilatory support) will be extracted, having been defined a priori as being potentially life-saving. The demographic details and variables chosen are all recorded or calculated for each patient as part of routine practice on critical care. No additional tests or measurements will be required for this study.

Data will be anonymised by allocation of a study number and this figure will be cross-referenced to an electronic masterlist which will contain the patient name and hospital number. In this way the database can be linked back to the patient to continue data collection and entry. The masterlist will be an electronic spreadsheet which will be stored on a secure server at the hospital and accessed via password protected computers within the ED. The masterlist will be destroyed on completion of the study.

Patient information will be collected in a spreadsheet designed to collate the information and allow statistical analysis. No identifiable patient information will be inputted onto the spreadsheet