Clinical knowledge and ‘big data’ analysis are being combined to develop a frailty indicator based on routinely collected records data. With a growing elderly population it’s increasingly important to identify when patients are becoming frail so that the resources for appropriate care can be arranged.
There is national focus on how to be more pro-active about patient care in the community, as Jeremy Hunt highlighted in September. ResearchOne and researchers at Bradford Foundation Trust and the University of Leeds are developing an indicator of patient frailty that uses GP practice data. This could then be run by GPs on their clinical data to identify patient frailty and monitor how this is changing on a patient and practice-wide level.
Making frailty quantifiable, and identifiable from routinely collected data, is a huge step forward for consistent patient care. Professor John Young, the National Clinical Director for Integration and the Frail Elderly at NHS England, is working with ResearchOne on developing the frailty indicator. “Frailty is not seen as a formal diagnosis and yet it is progressive, increasingly prevelant, impacts on quality of life and is expensive. Having an index of frailty will help it to be seen and treated as a long-term condition, and can help with selecting an appropriate care pathway for each patient.” Read an interview with Professor Young.
A combination of clinical understanding and big data analysis is being used to develop the frailty indicator. Firstly, analysis of non-identifiable records data from ResearchOne unearths potential clusters of information that may relate to frailty. These are reviewed by gerontologists and researchers to identify clinically relevant indicators – deficits - that are used to build an index of frailty.
Dr Andrew Clegg, clinical senior lecturer at Leeds University, and honorary consultant geriatrician at Bradford Teaching Hospitals is a member of the research team who is also keen to use the frailty indicator in daily practice. “Deficits are things that indicate frailty, such as osteorporosis or an abnormal blood test result, and the index can quickly add up such instances in the record to give a simple frailty score.” He says “it would help with decisions around care planning”.