Response to the Letter From Xue and Varadhan: 'What Is Missing in the Validation of Frailty Instruments?'

摘要

There are defenses to be made for each argument (frailty as a specific phenotype vs frailty as deficit accumulation), but that is to what they amount. The issue is resolved, one way or the other, as soon as the premises of the argument are established. Even so, and despite considerable progress, the specific biological premises of the frailty phenotype are as yet elusive and occur only at a broad level of description. The approach taken with the frailty index (FI) makes no claim as to which processes give rise to deficits; in any case, these are likely to be several.3 The FI employs an approach that does not sacrifice individually informative pieces of data, as do "informative on average" approaches, such as searching for latent constructs, to use one of the examples suggested. For this reason, I am comfortable counting a whole range of problems, knowing that some will be more common than others, but that uncommon problems remain important to the people who experience them. That the FI itself has characteristic behaviors4'5 that allow highly generalizable description6 is an important explanatory variable across a range of important problems,7'8 and also informs clinical management4 (eg, in recognizing the increase in risk associated with increasing degrees of frailty) are reasons enough to continue with it.