To compare the accuracy of the modified Fried Index (mFI) and the Clinical Frailty
Scale (CFS) to predict death or patient-reported new disability
90 days after major elective surgery
The association of frailty
with patient-reported outcomes
, and comparisons between preoperative frailty
instruments are poorly described.
This was a prospective multicenter cohort study. We determined frailty
status in individuals ≥65 years having elective noncardiac surgery
using the mFI and CFS. Outcomes
included death or patient-reported new disability
(primary); safety incidents, length of stay (LOS), and institutional discharge (secondary); ease of use, usefulness, benefit, clinical importance, and feasibility (tertiary). We measured the adjusted association of frailty
using regression analysis and compared true positive and false positive rates (TPR/FPR).
Of 702 participants, 645 had complete follow up. The CFS identified 297 (42.3%) with frailty
, the mFI 257 (36.6%); 72 (11.1%) died or experienced a new disability
was significantly associated with the primary outcome (CFS adjusted odds ratio, OR, 2.51, 95% confidence interval, CI, 1.50–4.21; mFI adjusted-OR 2.60, 95% CI 1.57–4.31). TPR and FPR were not significantly different between instruments. Frailty
was the only significant predictor of death or new disability
in a multivariable analysis. Need for institutional discharge, costs and LOS were significantly increased in individuals with frailty
. The CFS was easier to use, required less time and had less missing data.
Older people with frailty
are significantly more likely to die or experience a new patient-reported disability
. Clinicians performing frailty
assessments before surgery
should consider the CFS over the mFI as accuracy was similar, but ease of use and feasibility were higher.