Predicting older people’s frailty helps doctors intervene earlier

UNDER EMBARGO UNTIL 00.01 AM BST ON TUESDAY 1 APRIL

• The new Electronic Frailty Index 2 (eFI2) is now available to 60% of England’s GPs thanks to research funded by the National Institute of Health and Care Research (NIHR) and conducted by researchers at the University of Leeds and UCL 

• GP data on 36 health problems such as dementia, falls and fractures will help medical professionals to more accurately identify older people’s frailty and intervene earlier 

• Interventions may include a holistic assessment and treatment plan, falls prevention, targeted medicines review, and resistance exercise training 

• Older people living with frailty account for around £6 billion of annual NHS spending - it is hoped that early intervention will maximise independence, prevent falls and reduce these costs. 

GPs can more accurately identify older people’s frailty and intervene earlier as a result of NIHR-funded research led by the University of Leeds. 

Researchers have successfully improved the Electronic Frailty Index (eFI) – a tool that uses data to predict older patients’ risks of living with frailty – so medical professionals can provide holistic care, help to prevent falls, reduce burdensome medications and provide targeted exercise programmes to maximise independence.  

The groundbreaking eFI was first developed by Leeds academics and introduced in 2016 across the UK. In just one year of use by NHS England, more than 25,000 people with frailty were referred to a falls service, with an estimated prevention of around 2,300 future falls. Researchers estimate that in 2018 alone, these interventions saved the NHS nearly £7m. The world-first eFI system also influenced similar approaches in the US, Canada, Spain and Australia. 

Now, a new eFI2 system will improve the accuracy of the service by integrating data on 36 health problems including dementia, falls and fractures, weight loss and the number of regular prescriptions people have.  

A paper published today (Tuesday 1 April) in academic journal Age and Ageing by researchers at Leeds and University College London (UCL) confirms that the eFI2 can more accurately predict older people’s need for home care, risk of falls, care home admission or death.  

The authors hope that the eFI2, which is now available to 3 in 5 GPs in England through Optum (formerly known as EMIS) software, will help more older people stay independent for longer.  

Andrew Clegg, who led the study, is NIHR Research Professor and Head of Ageing and Stroke Research at the University of Leeds School of Medicine, and Honorary Consultant Geriatrician at Bradford Royal Infirmary. Professor Clegg said: “This landmark health data study, funded by the National Institute for Health and Care Research (NIHR), is a major step forward in transforming health and social care services for older people with frailty.  

“The eFI2 is a significant improvement on the original eFI and will be extremely valuable for helping GPs identify older people living with frailty so that they can be provided with personalized treatments to prevent costly loss of independence and falls in older age. We are delighted that the eFI2 has already been made available to 60% of GPs and is an exemplar of the planned NHS ‘analogue to digital’ shift.” 

Professor Marian Knight, Scientific Director for NIHR Infrastructure, said: "The eFI has already proven that it can improve patient outcomes and save the NHS millions of pounds. This evolution of the tool is extremely exciting, enabling people to receive personalised treatments from their GPs and maintain their independence for longer, bringing crucial cost savings to the health system." 

Frailty is identified when older people have a high risk of a range of adverse outcomes such as requirement for home care services, falls and admission to a hospital or care home. It is estimated that frailty costs the NHS £6bn every year. 

The eFI2 algorithm is based on routine data from Connected Bradford and the Welsh Secure Anonymised Information Linkage dataset, drawing on 750,000 linked records across medical, community and social care data to assign categories of frailty to older people.   

It uses 36 variables, including dementia, falls and fractures, weight loss and the number of regular prescriptions people have to predict which groups of people are more likely to be living with frailty. GPs are then encouraged to use their clinical judgement to apply a personalised approach to each patient. The accuracy of the eFI2 has significantly improved from the first model. 

Kate Walters, Professor of Primary Care & Epidemiology at UCL, a GP and one of the paper authors, said: "The eFI2 has great potential as a simple tool to support GPs in identifying people living with frailty who may benefit from further support to help them stay independent."

This work was funded by the National Institute for Health and Care Research Applied Research Collaboration Yorkshire & Humber (NIHR ARC YH) and supported by funding from the NIHR ARC North Thames. Professor Clegg is funded by the NIHR Research Professorship Award, and supported by the NIHR Leeds Biomedical Research Centre and Health Data Research UK, an initiative funded by UK Research and Innovation Councils, NIHR and the UK devolved administrations and leading medical research charities.