Creating And Validating A Claim-based Indicator Of Frailty

• PI:              Jodi Segal, M.D., M.P.H.
• Funder:     National Institute on Aging
• Status:      Ongoing

Linda Fried et al, described a frailty phenotype based upon observation of the cohort of individuals enrolled in the Cardiovascular Health Study. This phenotype is manifest when 3 of the 5 phenotypic criteria indicating compromised energetics, namely, low grip strength, low energy, slowed waking speed, low physical activity, and unintentional weight loss are present. Later work by her collaborators conceptualized frailty as a medical syndrome.  We propose to identify individuals with frailty using only administrative claims data.  Given the ubiquity of claims data, an indicator of frailty with good sensitivity and specificity would be widely used.

Frailty identified with administrative data may be used as an exposure variable to understand the health outcomes of individuals with frailty. It may be used as an outcome variable for evaluating the impact of interventions either designed to prevent or delay frailty or to assess frailty as an unwanted outcome of interventions such as hospitalization or surgery.  A claims-based indicator may also be used as an effect modifier in models of treatment effectiveness where there is expected to be differences in treatment-related response due to frailty.  We aim to develop a tool, which we will call the Claims-based Frailty Indicator (CFI) that can be generated with only claims data. This will be developed by translating existing frailty scales into a claims-based tool with appropriate validation.  We will explore the value of the CFI by demonstrating heterogeneity of treatment effect and benefit from use of interventions for treatment of aortic stenosis (transcatheter procedures and open surgical procedures).