Enhanced Algorithm for Predicting Frailty in Seniors Aims to Facilitate Early Medical Interventions

Researchers at the University of Leeds have introduced an enhanced version of the Electronic Frailty Index (eFI), a predictive tool designed to assess the risks of frailty in older adults. This improved algorithm is expected to enable healthcare providers to offer more comprehensive care, prevent falls, minimize unnecessary medications, and implement tailored exercise programs, thus promoting greater independence among seniors.

The original eFI was developed by Leeds researchers and rolled out in 2016 across the United Kingdom. Within its first year, the tool led to referrals for over 25,000 individuals with frailty to fall prevention services, resulting in an estimated 2,300 avoided falls. In 2018, these interventions were believed to have saved the National Health Service (NHS) approximately £7 million. The eFI has since inspired similar initiatives in countries such as the United States, Canada, Spain, and Australia.

The latest iteration, referred to as eFI2, enhances predictive capabilities by integrating comprehensive data on 36 health conditions, including dementia, falls, fractures, weight loss, and the number of regular medications patients are prescribed. A recent study published in the journal Age and Ageing by researchers from Leeds and University College London (UCL) confirms that eFI2 demonstrates improved accuracy in predicting older adults' needs for home care, risk of falls, potential care home admissions, or mortality.

With eFI2 now accessible to approximately 60% of general practitioners (GPs) in England via Optum software, the researchers hope it will assist a larger number of older adults in maintaining their independence for a more extended period.

Andrew Clegg, who spearheaded the study, emphasized that this innovative health data project represents a significant advancement in the transformation of health and social care services for frail elderly citizens. Clegg highlighted that eFI2 is a substantial enhancement over its predecessor and will be invaluable for GPs in identifying frail patients, enabling them to provide personalized treatment plans that help prevent costly losses of independence and decrease the incidence of falls.

Professor Marian Knight, Scientific Director for the National Institute for Health Research (NIHR) Infrastructure, reiterated the positive impact of the original eFI on patient outcomes and its cost-saving potential for the NHS. She expressed excitement about the evolution of the tool, which allows for personalized treatment from GPs and supports seniors in sustaining their independence, ultimately benefiting the healthcare system financially.

Frailty is generally characterized by a heightened risk of adverse events, including the need for home care, increased likelihood of falls, and hospital or care home admissions. The financial burden of frailty on the NHS is estimated to be around £6 billion annually.

The eFI2 algorithm utilizes routine data sourced from Connected Bradford and the Welsh Secure Anonymized Information Linkage dataset, aggregating 750,000 linked records from medical, community, and social care data to categorize frailty levels among older individuals. By employing 36 variables, including dementia, falls, fractures, weight loss, and regular prescription counts, the algorithm identifies populations at higher risk of frailty. GPs are advised to leverage their clinical judgment when determining personalized care strategies for each patient, resulting in improved accuracy compared to the original model.

Kate Walters, a Professor of Primary Care & Epidemiology at UCL and co-author of the study, noted the eFI2's potential as an effective tool for GPs to identify seniors living with frailty who may require additional support to preserve their independence.