The Sydney Triage to Admission Risk Tool (START) study: using a data analytics tool to drive early senior decision making and improve patient outcomes in Emergency Departments

Using NSW State-wide linked ED data, we have previously derived and internally validated a prediction tool for ED disposition called the Sydney Triage Admission Risk Tool (START). This tool generates a risk score that correlates with a patient’s likelihood of admission or discharge. It can also be used to describe the risk of re-admission within 30 days of ED presentation. Pilot feasibility studies of the START tool indicate a potential Emergency Treatment Performance (ETP) improvement of around 10%. START will identify patients who are likely to require in-patient admission and streamed directly to a Medical Admissions Unit, or those who can be safely discharged and streamed to a Fast Track Unit or a co-located GP clinic depending on existing models of care at a given hospital. For instance, patients with chronic conditions such as cardiovascular disease often present with complex and co-existing problems that impede timely disposition decisions despite the patient understanding from past experiences, that an admission is required. The START score will be calculated by a designated trained ED Nurse in the ED Triage room at each of the three study institutions using a paper based form and supported by an online calculator version of START located at http://www.slhd.nsw.gov.au/rpa/start/default.html The score is a weighted sum of categorical variables (age, triage category, presenting problem, mode of arrival, in-patient admission within previous 30 days and time of day) with weightings derived using logistic regression. For patients allocated to the intervention group a copy of the paper START risk scoring checklist will be attached to the clinical notes and also noted in the comments section on the ED electronic patient tracking list (Cerner Firstnet) and will used by the Nurse Unit Manager or patient flow Navigator in ED (NUM1 or equivalent existing senior nursing role in ED), together with the treating clinician, to assist with disposition decisions and patient streaming to various clinical locations including fast track (minor injuries unit) and Hospital in the Home. Details of the scoring system and clinical recommendations are detailed in the above link and summarised below; Very likely in-patient admission – Bed management to locate appropriate ward bed and senior clinician to notify admitting team of likely admission prior to full assessment and final disposition. This represents around 10% of all presentations Likely Admission – Bed management to locate appropriate ward bed and await further advice from senior ED medical officer Around 20% of all presentations Unclear – Await further assessment in ED – around 40% of all patient presentations Likely discharge – if extended workup likely, prepare admission to short stay unit Around 20% of all patient presentations Very likely discharge – stream to ambulatory care, fast track unit, hospital in the home or alternative primary care provider. Around 10% of all patient presentations A two month education and implementation program will occur prior to the trial, educating clinical and bed management staff about the tool, and integration into clinical practice and bed management strategies across the hospitals. Education will involve ED Doctors, Navigators, Nurse Managers, Bed management and nursing administration.