Frailty is an ageing-related health state in which multiple body systems gradually lose their in-built reserves. As a result, frail patients are less able to response to stressors such as trauma and acute illness. Frailty is associated with increased risk of worse outcome in people who present to hospital with a heart attack (acute coronary syndrome). As the population ages, the numbers of people who present with an acute coronary syndrome who also have multiple chronic conditions that are associated with frailty will increase. These people are also often underrepresented in clinical trials of interventions in acute coronary syndrome. This means that there is very little evidence to guide treatment stratified by frailty, which commonly results in variation in care. Furthermore, frailty assessment is not standard care in people presenting acute coronary syndrome. A number of frailty tools are available, but they are not routinely adopted for clinical care in part due to a lack of time and resources for assessment as these tools require some form of physical assessment.
Our research aims to develop a risk prediction score for frailty that can be easily implemented in clinical care. This will enable the development of effective stratified interventions to will address unmet needs specific to the frail population.
There is a growing interest in measuring frailty using routinely collected health data. Much of the work in the UK has been focused on developing frailty risk score for general acute care, with the predictors selected based on clinical coding in hospital admission1 or primary care databases2. In this project, we aim to examine the utility of the existing frail risk scores in cardiovascular patients, and develop a new frailty risk score using linked hospital and primary care data. As exploratory work, we will apply the newly developed frailty score, and investigate whether frailty is associated with variation in care and its impact on clinical outcomes. Our plan of investigation is as follows:
1. We will apply the existing Hospital Frailty Risk Score1 to admitted patients presented with an acute coronary syndrome and examine its performance to predict clinical outcomes.
2. To develop the new frailty score, we will first perform cluster analysis to identify frailty phenogroups characterised on the basis of clinical coding, and healthcare resource use.
3. Penalised regression will then be used to develop a risk scoring model which predicts frailty risk group membership based on diagnosis and procedure codes.
4. The frailty risk score will be validated against its performance to predict mortality and clinical outcomes.
5. External validation of the new frailty risk score using a local cohort of cardiovascular patients who have clinical measures of frailty and linked HES data will also be performed.
6. We will apply the new frailty risk score to stratify patients presented with an acute coronary syndrome. We will assess whether frailty explains the geographical variation in treatment decisions (stents vs bypass surgery), and how frailty affects survival in both operated and non-operated groups. We will model the surgical treatment rates controlling for frailty and case mix using logistic regression. Effect of frailty on patients’ survival will be analysed using competing risk analysis.
1. Gilbert T, Neuburger J, Kraindler J, Keeble E, Smith P, Ariti C, et al. Development and validation of a Hospital Frailty Risk Score focusing on older people in acute care settings using electronic hospital records: an observational study. Lancet 2018;391:1775-82. https://www.thelancet.com/journals/lancet/article/PIIS0140-6736(18)30668-8/fulltext
2. Clegg A, Bates C, Young J, Ryan R, Nichols L, Ann Teale E, et al. Development and validation of an electronic frailty index using routine primary care electronic health record data. Age Ageing 2016;45:353-60. https://pubmed.ncbi.nlm.nih.gov/26944937/