Could a frailty index enhance primary care for frail older adults at the end of life?

Talk Code: 
Holly Standing
Daniel Stow, Catherine Exley, Louise Robinson, Steven Iliffe, Stephen Barclay, Andrew Clegg, Sarah De-Baise, Barbara Hanratty
Author institutions: 
Newcastle University, University College London, University of Cambridge, University of Leeds


As people are living longer, more and more people are dying with frailty. Estimates suggest that between one quarter and one half of all patients aged over 85 years are frail. It is important that health professionals recognise when a patient may be nearing the end of their lives, so that they can talk to them about the sort of care that would prefer, and where they would like to be treated during their final illness. It can be very difficult for a General Practitioner (GP) to know when a frail older person is likely to die, as their health may have been declining for some time. To help GPs identify and assess the severity of frailty amongst their patients, researchers have put together a simple tool in one of the GP computer systems. The electronic frailty index (eFI) defines frailty as an accumulation of deficits, and utilises information recorded in patients’ electronic health records. A score is calculated by expressing the number of deficits (signs, symptoms, diseases or disabilities) as a proportion of the total possible (36 deficits derived from 2000 Read codes).


In this study we are exploring the use of the eFI as a tool to identify older people who may benefit from advanced care planning and palliative care. Semi-structured interviews with up to 25 GPs from different geographical regions of England are exploring GPs’ understanding of frailty and experiences of managing frail patients at the end of life. Data are being collected on GP perceptions of the appropriateness, acceptability and interpretability of the eFI for routine use in end of life care. Thematic analysis of the data is ongoing.


Early interviews suggest that GPs view the eFI as a potentially useful tool, but there are concerns about workload, and the loss of clinical skills. To be successfully adopted into routine practice, the eFI will need to be integrated seamlessly into existing systems. The completed findings of this study will be available for presentation at SAPC.


The findings of this study will contribute to enhancing end of life care for a relatively neglected patient population.

Submitted by: 
Barbara Hanratty
Funding acknowledgement: 
This study is funded by NIHR School for Primary Care Research