Scotland spends £1.7 billion a year on unplanned hospital bed days for frail patients, and a new PhD project aims to fix the data tool that could cut that cost. Frailty affects millions of older people, making them vulnerable to falls, hospitalisations, and spiralling healthcare needs. The electronic Frailty Index (eFI) uses 35 routine GP data points to identify high-risk patients early, and pilot studies on 900,000 people in England showed it reduces unplanned admissions. But Scotland has not adopted it—visibility problems, GP workload, and doubts about real-world accuracy have blocked its use. This project will identify those bottlenecks, standardise Scottish GP data, and build a deprescribing model using regression and deep learning. The student will work with NHS data systems and a multidisciplinary team to develop a DELPHI study that harmonises eFI data across Scotland. If successful, the eFI could become a routine tool in Scottish primary care, helping pharmacists and GPs safely reduce unnecessary medications for frail patients. That would cut unplanned admissions, save the NHS money, and keep older people healthier at home—without requiring grip-strength tests or gait-speed measurements that primary care simply cannot deliver at scale.
View original technical description
Frailty is a crucial factor in global ageing, significantly impacting public health and healthcare costs. It is a multifactorial syndrome marked by increased vulnerability to stressors and a failure to maintain homeostasis, leading to adverse events like falls, hospitalizations, and heightened healthcare needs. In the UK, the NHS incurs an additional £5.6 billion annually due to frail individuals, with Scotland alone spending £1.7 billion on unplanned bed days for the frail population in 2019. Early identification of high-risk individuals is essential to preserve their wellbeing and inform appropriate healthcare interventions. Despite the availability of various frailty assessment tools, their application in primary care is limited due to the resource-intensive nature of physical assessments. Tools like the Fried Frailty Phenotype require measurements of grip strength and gait speed, posing challenges for widespread implementation. To address this, the electronic Frailty Index (eFI) was developed, utilizing 35 items of routine primary care data. Pilot studies on 900,000 individuals in England have shown the eFI's effectiveness in identifying frail patients and reducing unplanned admissions. However, its adoption in Scotland has been hindered by visibility issues, constraints in general practice, and concerns about validity outside academic settings. Emerging primary care pharmacists could support eFI usage to optimize prescribing, particularly for patients with polypharmacy. This project aims to identify implementation bottlenecks of the eFI in Scotland, standardize data, and explore the impact of pharmacological use on long-term outcomes. The student will work with a multidisciplinary team to develop a DELPHI study, harmonize eFI data, and create a deprescribing model using regression and deep learning tools. Training will include NHS data systems, eFI modeling, and coding in R and Python.
Plain English summaries and category classifications on this site are generated by AI and may not perfectly reflect the original research.
Is something wrong? Let us know