Feature: a data-driven future for healthcare?

The NHS, academics and industry professionals have been relying on data more than ever, to help shape COVID-19 responses, research studies, predict patient demand, identify stranded patients, develop new modernised pathways and improve patient outcomes.

Here, we take a look at one of the solutions available to healthcare providers – RW Health’s Data Science Platform (DSP) – and focus on specific cases of use and pathways.

The healthcare analytics company launched the platform to extract insights from the vast amounts of data healthcare providers hold. Built around the idea that, by combining data science and machine learning and applying it in the right way, such a platform can enable health services and academics to reduce costs, maximise returns, optimise operations and, most crucially of all, improve patient outcomes.

Honing-in on common challenges, the company notes ‘expertise, time or tech’ as the main three to make best use of data repositories and assist existing analytics teams. The automated machine learning platform aims to reduce these barriers to data-driven healthcare.

Described as ‘an advanced collection’ of ‘carefully curated’ libraries, it uses embedded AI to improve productivity and offer predictive analytics – acting, essentially, as a forecasting feature that can be used to develop ‘highly accurate predictive models’.

The company said to HTN: “Prior programming knowledge to introduce and use the tool is not required, due to the simplicity of integration and maintaining models, it can ease the burden on resources in a landscape where data and IT expertise is currently highly sought-after.”

As well as a tool for utilising data to create forecasts through ‘actionable, clinical insights’, the DSP also has other practical uses to assist with and enhance day-to-day processes, including simplifying workflows to help staff manage and distribute workloads – thereby aiming to reduce stress and burnout at a time when resources are precious and stretched.

Developing a new mental health care pathway

The first use case we focus on a mental health trust and how data was used to modernise pathways and deliver cost savings. The trust provides services across 6 boroughs, located from a Victorian-era hospital building that was, at the time, facing pressure to modernise.

The objective was to create a new, modernised mental health care pathway that would ultimately reduce pressure on local emergency departments and any unnecessary use of acute beds – potentially freeing up staff time and space for patients, and saving funds.

The DSP, focused on enhancing patient triage in regards to both admitting and sending home patients more efficiently, generating cost savings, and providing better visibility on future bed needs.

The pilot resulted in the trust being able to close its old facility, due to better utilisation of other sites. There were multiple benefits gained from trialling the system, aptly reflected in the data collected, which shows a regular drop below the mean in the average length of stay from April 2018 through to January 2020, as well as an overall decrease, before the start of the COVID-19 pandemic. The overall admission rate also embarked on a significant downward trend over the course of that period, while out of area bed usage was eliminated during the same time frame, too.

When it came to savings, the trust was also able to generate cost savings worth £5.5 million per annum, while the initial pilot has created an analytics and transformation process that’s now worth significant value to patients

Improving patient outcomes

In the next use case, we look at a large alliance supporting cancer care across a region, and how its utilising the DSP to improve outcomes for cancer patients in the midst of the current backlog.

5 NHS trusts combined to form the Cancer Alliance to drive visibility of the region’s needs, help to accelerate individual treatments and improve patient outcomes.

The ICS is leveraging patient data across multiple sites, at a system level, to identify where the highest risks are and where to focus resources. This enabled greater visibility of the demands of the wider area, enabling more specialisation for different types of cancer treatment.