Nuffield Trust works on PARR30

  • 15 June 2012
Nuffield Trust works on PARR30
The Nuffield Trust has released a report - Delivering the Benefits of Digital Healthcare

Researchers at the Nuffield Trust are working on a new predictive risk tool for the NHS.

The latest version of the Patients at Risk of Re-hospitalisation, or PARR, tool will focus on patients at risk of being readmitted within 30 days.

The government announced last year that commissioners will be able to withhold payment from hospitals for emergency readmissions that occur within 30 days of a patient being discharged.

However, Ian Blunt, a senior fellow at the Nuffield Trust, told a conference earlier this week that there were few tools available to hospitals to spot patients at risk.

The existing PARR and PARR++ tools that were first developed in 2005 are designed to identify patients who may be at risk of re-admission within a year. But they use data that may not be available while the patient is still in hospital.

“For PARR30, we decided to use data that was available at the point of discharge, either from the hospital patient administration system or from the patient,” Blunt said.

This includes the hospital in which the patient is being treated, their age, some indicators of deprivation, their previous history of emergency admissions and some elements of their medical history.

Using these fields, Blunt said the researchers had been able to build a tool that had a ‘positive predictive value’ of around 59% – meaning that it is able to pick up a good number of those patients who will go on to be re-admitted without intervention.

On the other hand, he said the tool was not particularly sensitive – it tends to give low risk scores to patients, even to those that do go on to be re-admitted within 30 days.

The tool has been trialled at two hospitals – Chelsea and Westminster and Royal Berkshire. The first trust carried out a minimal amount of development work to build the tool into its data warehouse, while the second ran a spreadsheet version.

Blunt said that in both cases staff found it took less than 100 seconds to fill in the fields. The Nuffield Trust is looking for more organisations to test the tool, which it plans to make freely available to the NHS.

Read more about the Nuffield Trust’s latest conference on predictive risk in the Insight section.

Subscribe To Our Newsletters

Subscribe to our newsletter

Subscribe To Our Newsletter

Related News

Digital Health Coffee Time Briefing β˜•

Digital Health Coffee Time Briefing β˜•

Today's briefing includes news of a cardiovascular disease screening initiative and a fellowship to explore the application of AI in genomics.
Digital Health Coffee Time Briefing β˜•

Digital Health Coffee Time Briefing β˜•

Today's briefing includes an Epic deployment in Germany, and investment in a data platform to transform heart rhythm disorder.
Digital Health Coffee Time Briefing β˜•

Digital Health Coffee Time Briefing β˜•

Today's briefing has news on a significant milestone for home care firm Cera and a report showing how GPs use AI tools.