How to give patients greater transparency to improve the A&E experience
An aging population, winter pressures, staff shortages and financial strains are putting the future of our healthcare service at risk but emerging technologies are already lowering medical costs, improving patient care but one area of the NHS under strain are A&E departments.
The NHS has recorded its worst A&E waiting times in England since current targets began in 2004. On average patients are now expected to wait 5 hours 12 minutes before being seen. NHS Trusts across the UK are now making it a priority to reduce the average waiting time, but solutions to one of the country’s biggest healthcare management issues isn’t going to come overnight.
Hospitals are doing their best to manually predict approximate waiting times for patients, using current and historical A&E data but this can be inaccurate and often patients are left with long periods without visibility of their patient journey. In an NHS survey of 31,168 A&E patients, 59% of respondents said they were not told how long they would wait before being examined and 35% of patients either couldn’t find somewhere to get a suitable meal or weren’t even aware they were allowed to eat or drink. It therefore comes as no surprise that patients are left feeling extremely frustrated and distressed at a time when they’re already in a very vulnerable position.
So, what’s the solution?
Reducing A&E waiting times is the ultimate goal, but while we strive to get there, we need to focus on creating a better patient experience and one way to do this is to provide transparency of the patient journey.
Using a data centric platform would enable Trusts to more accurately predict waiting times and display real time data on screens in the A&E waiting area as well as a patient facing application to transform the patient and user experience.
To see how this would work in practice, let me introduce Joe. Joe has come into A&E with an injured leg after playing football. When Joe arrives, he is seen by the A&E administrator and is diagnosed with a suspected leg fracture. He is then assigned a patient number, password and, is given a QR code to download the NHS app. He’s also told to keep his eye on the screens in reception for updates.
Joe downloads the app and logs in with his patient number. He is presented with a timeline displaying his unique patient journey. The app is already showing the first step he completed when he saw the A&E administrator at 09:36 and he now has an estimated 32 minute wait before his assessment with the Triage Nurse.
When Joe is ready to be discharged, due to nature of his incident, the app will prompt him to either call a friend to pick him or, or to book a taxi.
So, how does the solution work?
It’s no secret that the NHS has multiple disparate sources of information so the first place to start is with a data cleansing and processing exercise. Using a data warehouse, NHS organisations will be able to join together sources of information held across different areas of the organisations to create new, meaningful datasets. These sources of data will consist of specific parameters such as time series data, patient identification data, varying patient ailment data and organisational and department data.
Once the data warehouse platform is set up, it will begin to collate all the information over time from the A&E department. This information would then be used to create further new datasets which would be used to run models and algorithms to detect relationships and correlations within the data. For instance, this could be understanding how long it will take before a patient, who has a suspected leg fracture, can be seen to. The solution would take all the necessary information from the warehouse, create the specific subset of data which will be used to train the model on. The model may well be a combination or singular model depending on the accuracy for each necessary success criteria. For example, an Elastic Net Regression model would enable us to understand how many factors of patient specific data, as well as A&E departmental data on wait times, injury/ailment response history, staff availability, data and time factors are correlated. Training the model with known data from previous cases during the exact time a new patient has arrived will give you an estimate on how long each stage of the A&E journey for the patient will entail.
The application also has the ability to link into external sources. In Joe’s case, he is facing a number of lengthy waiting periods, giving him the opportunity to visit a café for refreshments. The application can be designed to notify and encourage the patient to visit the café to give them an opportunity to get something to eat or drink. It is also designed to encourage patients to move to a more relaxing and comfortable waiting area, while prevent overcrowding in the hospital’s main A&E waiting area.
How to get started
Knowing where to get started can be overwhelming, especially when you’re collecting more patient, staff and organisational data than ever, making it increasingly difficult to sift through and pull out what really matters. But what if you could uncover and act on the trends hidden within your data?
Whether you’re looking for a solution to provide patient transparency, or just looking to innovate, our Data Platform Accelerator will help you leverage data to make intelligent decisions, faster. Check it out here.