Frailty is a clinical condition that is associated with an increased risk of falls, damaging events, institutional care and disability. There are several tools available for assessing frailty, including the Clincial Frailty Scale and the Freid Phenotype Model. Understanding a patient's level of frailty can be the critical factor between receiving or withholding care. This is especially true in light of COVID-19 and the allocation of limited resources.
Explore frailty with Scott Buxton
The Clinical Frailty Scale (CFS) is the recommended instrument for assessing the frailty of a patient in acute situations. However, in most places outside of geriatric / frailty services, few clinicians receive formal frailty assessment training.
The interrater reliability of the CFS is generally very good, but a number of biases can play a role in the assessment, especially for clinicians who are inexperienced with the scale. With this in mind, a new classification tree has been created to improve the CFS assessment by inexperienced assessors.
Methods
The aim of the study was to evaluate the intra-rater reliability between a classification tree for the clinical vulnerability scale and the assessment of the clinical vulnerability scale by beginners and experts in 115 older adults.
This was achieved through a prospective design study in an emergency room. Assessment of frailty was done after patients were referred to a specialty and a comprehensive geriatric examination was completed.
After the assessment, an experienced frailty clinician assigned the patient a clinical frailty value. In 104 of the patients, a frail beginner rated the patient independently of the expert. In 11 patients, they were only rated by the expert as they were assessed outside of the hospital setting.
After completion of the judgment-based assessment after the assessment, the vulnerability scale was assessed separately using the classification tree (see below). The classification tree is essentially a step-by-step guide to assessing a patient's independence, multimorbidity, fatigue, and activity.
The nice thing about the tool is that are based on the questions at each level of the clinical vulnerability scale and link each level to a simple question that is asked during routine exams / assessments. Therefore, no specialist training is required.
The scores of the beginners and experts were compared using intraclass correlation coefficients (ICCs) and 95% confidence intervals (CIs). ICC is a method of assessing the reliability of results and describes how similar units in the same group are. For this type of tool, a score of 0.70 or more indicates good interrater reliability.
Results and clinical implications
The overall agreement for the CFS classification tree was good with an ICC value of over 0.800 compared to the expert and classification tree and for the assessment of beginners and classification trees.
Most of the classification tree evaluations were either the same or differed by one level compared to the other evaluators. This is acceptable in most cases when small differences in rating could determine whether or not someone receives stricter measures of care.
This discrepancy probably arises from the complexity of patients with frailty and the inability to translate this complexity into decision trees. It is difficult to get relevant details from these complex cases at the time of an acute illness and this can affect the classification.
Quick summary of the classification tree for clinical frailty scales
The evaluation of frailty by inexperienced evaluators may not always be the same as the evaluation by experts
Using a classification tree for the clinical vulnerability scale has good reliability between experts and inexperienced clinicians
Beginners judge more easily frail people than moderately frail
The clinical frailty classification tree can aid routine frailty scoring in clinical practice
This classification tree does not replace the CGA but is a useful tool for inexperienced frail clinicians to assess frailty in clinical practice. Understanding the degree of frailty of a patient is important to prognosis, and any tool used to aid clinician frailty assessment should definitely be considered.
Apart from that it is a single-site study with a small sample size with only a few evaluators . Although the study was sufficiently powerful to determine the intrarateral reliability with the classification tree, further validation must be sought, especially across locations.
