Background and Goal of Study:Frailty assessment may be of value to identify high-risk surgical patients, highlighting additional post-operative care needs. A number of frailty assessment tools exist but there is uncertainty as to which is best in surgical patients. This study investigated the agreement between 9 widely accepted frailty assessments.
Materials and Methods:Elective colorectal surgical patients were assessed using 9 frailty tools, as part of the FAST study. The clinical frailty scale (CFS); 36-point accumulative deficit (AD); frailty phenotype (FP); frailty, non-disabled tool (FiND); fatigue, resistance, ambulatory, illness and loss of weight tool (FRAIL); Edmonton frailty scale (EFS); Gerontopole frailty screening tool (GFST) polypharmacy as defined by 5+ medications (Poly); and PRISMA 7 frailty score (PRISMA) were used. Cohen’s kappa statistic was used to assess agreement between tools. Ethical approval has been granted for the FAST studies by a UK National Health Service Research Ethics Committee.
Results and Discussion:95 patients were recruited into FAST, mean age 74 (SD 6.2) years with 7 over85. 66/95 patients had metastatic disease. Frailty was identified by the different measurements (Table 1).
There was significant agreement between the different frailty tools ranging from K=0.096 to K=0.590 (figure 1).
Conclusion(s):Frailty tools have limited agreement in a surgical population. The adoption of a frailty assessment into a surgical pathway requires scientific and clinical rigour to ensure the optimal assessment is used, providing clinically significant information. We are currently exploring, as part of the FAST study, which assessment has the strongest validity in relation with post-operative adverse outcomes.

09 / FAST (Frailty And Sarcopenia Trials): Poor agreement between commonly-used frailty assessments in elective colorectal surgical patients.