Frailty Trait Scale-Short Form: A Frailty Instrument for Clinical Practice.

Objectives: To develop short versions of the Frailty Trait Scale (FTS) for use in clinical settings.

Design: Prospective population-based cohort study.

Setting And Participants: Data from 1634 participants from the Toledo Study for Healthy Aging.

Methods: The 12-item Frailty Trait Scale (FTS) reduction was performed based on an area under the curve (AUC) analysis adjusted by age, sex, and comorbidity. Items that maximized prognostic information for adverse events were selected. Each item score was done at the same time as the reduction, identifying the score that maximized the predictive ability for adverse events. For each short version of the FTS, cutoffs that optimized the prognostic information (sensitivity and specificity) were chosen, and their predictive value was later compared with a surrogate gold standard for frailty (the Fried Phenotype).

Results: Two short forms, the 5-item (FTS) (range 0-50) and 3-item (FTS) (range 0-30), were identified, both with AUCs for health adverse events similar to the 12-item FTS. The identified cutoffs were >25 for the FTS scale and >15 for the FTS. The frailty prevalence with these cutoffs was 24% and 20% for the FTS and FTS, respectively, whereas frailty according to Fried Phenotype (FP) reached 8% and prefrailty reached 41%. In general, the FTS showed better prognostic performance than the FP, especially with prefrail individuals, in whom the FTS form identified 65% of participants with an almost basal risk and 35% with a very high risk for mortality (OR: 4) and frailty (OR: 6.6-8.7), a high risk for hospitalization (OR: 1.9-2.1), and a moderate risk for disability (OR: 1.7) who could be considered frail. The FTS form had worse performance than the FTS, showing 31% of false negatives between frail participants identified by FP with a high risk of adverse events.

Conclusions And Implications: The FTS is a short scale that is easy to administer and has a similar performance to the FTS, and it can be used in clinical settings for frailty diagnosis and evolution.