1. Development of an electronic Frailty Index (eFI)  

Development of an electronic Frailty Index (eFI)

As part of NIHR CLAHRC Yorkshire and Humber, Dr Andrew Clegg and Professor John Young, from the Academic Unit of Elderly Care & Rehabilitation, University of Leeds, have collaborated with Chris Bates and John Parry at TPP/ResearchOne to develop and validate an electronic frailty index (eFI) that uses existing electronic health record data to identify and severity grade frailty. The eFI has been developed in collaboration between the University of Leeds, TPP, the University of Bradford, the University of Birmingham and Bradford Teaching Hospitals NHS Foundation Trust and funding from the National Institute for Health Research Collaboration for Leadership in Applied Health Research and Care, Yorkshire and Humber (NIHR CLAHRC YH). It has undergone internal and external validation in a retrospective cohort study using data from around 1 million patients in the ResearchOne and THIN databases.

The eFI has been developed using the cumulative deficit model of frailty, whereby frailty is defined on the basis of the accumulation of a range of deficits, which are clinical signs (e.g. tremor), symptoms (e.g. breathlessness), diseases (e.g. hypertension) and disabilities. The eFI consists of 36 deficits which have been constructed using around 2,000 primary care clinical codes (Read codes). The eFI calculates a frailty score by dividing the number of deficits present by the total possible. For example, if a patient has 9 out of 36 deficits, the eFI score is 0.25. The score is a robust predictor of those who are at greater risk of adverse outcomes (e.g. care home admission and mortality).

The eFI is now available to GPs in in SystmOne and EMISweb care electronic health records, covering approximately 90% of the UK population. The eFI will enable treatments and services to be targeted to a person based on their frailty status rather than their chronological age thus providing a paradigm shift in care for older people living in the community.

The eFI will help improve care for older people with frailty by: directing better primary care pathways for older people by considering individual frailty rather than chronological age; identifying the top 2% most vulnerable patients for targeted care planning to reduce unplanned hospitalisations; structuring integrated care around frailty, to ensure that those who may benefit most are identified for integrated services; enabling targeted medication reviews for older people with frailty using evidence-based checklists (e.g. STOPP/START criteria; identifying the presence of frailty to guide more appropriate, shared decision making in secondary care, for example in cancer services for older people; identifying those with advanced frailty who may be entering the terminal phase of life for advance care planning discussions.


We have established a Healthy Ageing collaborative as part of the 
Yorkshire & Humber AHSN Improvement Academy (http://www.improvement-academy.co.uk/improving-services/healthy-ageing.html) 
to implement and evaluate the eFI. We have engaged locally, regionally and nationally with 55 CCGs to develop new models of frailty care using the widespread availability of the eFI to enable delivery of evidence-based interventions to potentially modify frailty trajectories and improve outcomes for older people.

The eFI helps primary care providers to identify older people with frailty and who face an increased risk of care home admission, hospitalisation and mortality by using information within a patient’s electronic health record. Examples of the eFI in action include: falls prevention interventions for people with moderate frailty; adding people with severe frailty to practice palliative/Gold Standards Framework registers and offering advance care planning interventions: offering self-management support to people with mild frailty; medication reviews for people with severe frailty and care home residents; adding people with severe frailty to a GP practice top two per cent of patients at risk of avoidable unplanned admissions register; nurse-led frailty assessments for people with mild, moderate and severe frailty; and identifying patients with moderate and severe frailty for geriatrician led Frailty Clinics or comprehensive geriatric assessment clinics.

The eFI has been included as a recommended tool within the recently published NICE Multimorbidity guideline. The eFI has won the Healthcare IT Product Innovation category at the EHI Live 2016 Awards.