Leading the world in frailty identification and management

Doctors frequently see older patients with health issues, but frailty is rarely identified in these interactions. Frailty develops as we get older because our bodies change and lose their inbuilt reserves. The condition is so common in old age that it accounts for around £6 billion of annual NHS expenditure, because patients can experience sudden, dramatic changes in health when they have an illness. 

“The problem is that frailty is invisible, yet it surrounds us in healthcare,” says Professor Andrew Clegg, lead of the Older People with Frailty theme at the NIHR Applied Research Collaboration Yorkshire and Humber (ARC YH). “The majority of people I see have frailty and typically present to hospital in crisis. If we could identify them earlier, we could drive a shift away from a reactionary, crisis-driven response and provide interventions that could reduce injury and hospitalisation.”

International guidelines recommend that frailty should be identified routinely. Until recently, identification of frailty relied on additional assessment over and above what a doctor may do in practice, such as measuring walking speed or grip strength, or carrying out a questionnaire. As such, patients often remain undiagnosed until they have an accident or are admitted to hospital in crisis. 

There is however a growing evidence base for a range of interventions. The best evidence is on exercise programmes aimed at muscle strengthening for relatively simple mobility tasks, and for falls prevention interventions. Patients can be supported to improve their self-monitoring skills, and checks can be done on medications and trying to reduce harm through inappropriate or multiple prescriptions.

Professor Clegg has led on the development and national implementation of an electronic frailty index (eFI) that is revolutionising the way patients with frailty are identified and supported in the UK and beyond. Professor Clegg and his colleagues realised that by using data that is already captured by general practitioners (GPs), they could create a cumulative frailty model, relying on the simple premise that the more little things that are wrong with an individual, the more likely they are to have frailty.

Around the same time, a database of anonymised patient data was made available to researchers. ResearchOne was set up by the key provider of a primary healthcare electronic records tool, SystmOne, which is used by about a third of the UK’s GPs.

Professor Clegg designed a research study using ResearchOne as a proxy for the real world setting, collecting anonymised data from 931,000 patients aged 65-95 to develop and validate an eFI. His study was funded by ARC YH in its previous form as the NIHR Collaboration for Leadership in Applied Health Research (CLAHRC) Yorkshire and Humber and NIHR CLAHRC West Midlands.

“We saw that we could use routinely available primary care data, and identified 36 equally weighted clinical health deficits across signs, symptoms, diseases, impairments and disability,” says Professor Clegg. “Using anonymised data, we could calculate a frailty index score as a proportion of the number of deficits identified in a patient record.”

The eFI successfully identified older people with mild, moderate and severe frailty, with the results validated against outcomes of nursing home admissions, hospital admissions and mortality. The study demonstrated the validity of identifying frailty using routine health data, and indicated that routine implementation of the eFI could enable delivery of evidence-based interventions to improve outcomes for this vulnerable group.

The results were striking enough that implementation followed quickly in the live GP electronic health record system, SystmOne, and in other electronic health records systems used in primary health care.

Professor Clegg describes the impact on real clinical practice: “Over the course of about a year we made the eFI available to every practice in England. Fortunately there are only four providers of GP electronic health records in the country so it’s available to about 95% of practices across the UK.

“GP contracts were updated in 2017/18 to include frailty identification and management as a new contractual requirement, as a result of the work that we did to provide the infrastructure to enable practices to consider frailty care. This is a huge impact story!”

For patients this means interventions can begin before they reach a crisis. “Since the eFI has been implemented there have been around 1.2 million people assessed for the presence of frailty, and a clinical confirmation made before any decisions on interventions,” says Professor Clegg. “There have been around 200,000 medication reviews, and 30,000 patients have been referred for falls prevention services since the eFI was rolled out in 2017/18”. 

The work has been recognised by a Royal College of Physicians Excellence in Patient Care Award, and was cited in the NHS Long Term Plan. It was selected as an example of research impact to support the 2018 Government Spending Review. 

The team now has further funding through the NIHR Health Technology Assessment programme to develop improvements to the eFI, such as improving the performance of the tool and the accuracy of frailty categories. This is a result of feedback since the national rollout and to be expected following such groundbreaking work. 

Professor Clegg concludes: “We knew before we started that frailty was a huge issue that needed addressing, but it is astonishing how much impact the eFI has had in just four years. This NIHR-funded work has enabled the UK to be the first country in the world to implement a national programme of frailty care, and be at the vanguard of frailty work internationally.”