Using Electronic Health Records to Identify Frailty in Inpatient Settings

Lead Supervisor
Dr Julie Whitney
Lecturer in Physiotherapy
School of Population Health and Environmental Sciences, King’s College London
julie.whitney@kcl.ac.uk

Co-supervisor
Dr James Teo (King’s College Hospital), Dr Fiona Gaughran (King’s College London)

Project Details

Frailty is a distinctive health state related to the ageing process in which multiple body systems gradually lose their in-built reserves. Frailty is associated with increased risk of hospital admission related complications such as falls, immobility, malnutrition, delirium and medication error/adverse drug reactions. Accurate identification of frailty on admission to hospital allows for the provision of appropriate care pathways in order to prevent these complications

Screening tools have been developed to identify older people with frailty. However, there are differing approaches to the understanding of frailty which underpin these tools. The recent NHS long term plan requires all hospitals with a type 1 emergency department (ED) to provide specialist pathways for frail patients who present for emergency care.

There is currently no linkage between primary and secondary care and hospital frailty pathways rely on ED staff to carry out screening. Even when this process is optimised, the proportion of older people screened doesnt exceed 80%. Additionally, the accuracy of this screening method is unknown. As a result of these limitations, we currently lack an accurate figure on the prevalence of frailty, meaning it is difficult to effectively plan and commission services. 

This aim of this project is to use electronic health records (EHRs) to develop a model that accurately identifies frail hospital inpatients. A tool with greater accuracy than scoring at ED triage could replace this system. It would also allow identification of frail patients anywhere within a hospital to enable specialist teams to efficiently locate patients. Finally, it would allow for better data on the prevalence of frailty to support planning.

The candidate will:

  1. Work with KCH and SLAM electronic health records research systems like Cogstack to develop NLP models that can detect frail older people and indicate the severity of frailty.
  2. To validate the frailty detection models against expert assessment and clinical outcomes.

Terms associated with known phenotypic frailty markers (such as functional impairment, weakness and confusion) will form the unstructured data that informs the model. ICD10 codes for multi-morbidity associated with frailty (such as hip fracture, falls, incontinence, heart failure) will provide structured data to be used. 

Characteristics in EHRs associated with frailty will then be used to build a model to identify frailty. This model would need to provide an indication of the severity of frailty, as clinical interventions differ depending on this. 

The model will be validated by comparing it to expert clinician judgement as well evaluating the impact of frailty on outcomes such as length of hospital stay, readmissions and 30-day mortality.    

The student will be exposed to and expected to learn:

  • Core Data science skills including querying of both structured and unstructured data; and natural language processing (NLP);
  • Machine learning and classical statistical techniques;
  • Systems thinking for understanding healthcare processes
  • General understanding of interoperable Healthcare ontology

The supervision for this project will include a supervisor with a background in providing clinical frailty services. The second supervisor will bring expertise in data science, natural language processing and machine learning to this project and experience of working on similar projects. 

Datasets

The principle ethical and information governance issues affecting this project are that it will use the electronic health records of patients. The models will be developed at King‚Äôs College Hospital (KCH), which is at the forefront of the development of this technology. To that effect, KCH has recently set up the KERAI group which in conjunction with the Health Research Authority (HRA) and local research and innovation departments, have developed a protocol for reviewing all applications to use EHR data. This project will have to go through this committee as part of the approvals process. 

 The project will be undertaken in compliance with the code of conduct for data driven health technology (https://www.gov.uk/government/news/new-code-of-conduct-for-artificial-intelligence-ai-systems-used-by-the-nhs) and NHSX (https://www.nhsx.nhs.uk/key-information-and-tools).

Keywords

frailty, multi-morbidity, prevalence