Risk stratification tools

In 2009, NHS Somerset was nominated to run a national pilot project to improve the ways in which GP’s patient data was used for commissioning purposes. This included investigating new information management tools that would increase the involvement of GPs in collecting and using patient data and generate greater confidence in its quality.

A focus of this pilot was risk stratification. NHS Somerset was concerned that their GP’s existing risk stratification tools (RST) were difficult to use and did not return adequate identification or analysis of patients at risk of hospital admission.

As a result, we identified what was ideally required from a new RST (in terms of output and ease-of-use) and put in place a superior system.

After assessment, we identified a risk stratification tool for GP practices that exactly met the needs of all stakeholders. It was simple, easy to use, and returned reliable, comprehensive information from multiple sources. Features included:

• Access with just one click to a full list of patients, organised according to their likelihood of an unplanned chronic admission in the next 12 months

• An individual patient’s clinical data, including GP information, hospital activity, prescriptions, diagnoses and conditions, accessible in just two clicks

• Enough information to help a clinician understand why the patient is at risk

• Predictive models that suggest which care pathways or interventions could prevent a hospital admission 

• Easy-to-use features and reliable data

The previous RST was difficult to use, did not collect data from multiple sources and had limited capacity to inform clinicians why the patient was high-risk. We learnt that what GPs wanted was an information management tool that is simple to learn, quick to use, and capable of returning high-quality, reliable and useful data.

Using the new tool, GPs, clinicians and community matrons can now understand their patient population better and identify those at risk of hospital admission in just one click of a button. Once identified, enough patient information is available for the clinician to plan appropriate care pathways and monitor progress. So this system not only improves patient outcomes, it helps reduce the significant costs associated with unplanned hospital admission.

The new RST also supports Somerset CCG’s QIPP initiatives. For example, the tool was used to identify high-risk patients based on previous hospital admission who might benefit from Tele-health units.  The pilot was a success, reducing these patients’ incidence of hospital admission by 30%.

Around 98% of GPs in Somerset are using risk stratification regularly to identify and monitor high-risk patients. Risk stratification is also identifying incorrect use of secondary care services and supporting practice management. Our work has given Somerset a strong foundation for understanding integrated health and social care data.

Subsequently, we worked with Sir John Oldham, in his role as national clinical lead for QIPP, to demonstrate the potential for risk stratification within the health and social care system at events around the country. We also worked with United Health on the development of new risk stratification algorithms.