Abstract: FR-PO860

A New, “Lighter” Operational Definition of Frailty in ESKD

Session Information

Category: Dialysis

  • 607 Dialysis: Epidemiology, Outcomes, Clinical Trials - Non-Cardiovascular


  • Moorthi, Ranjani N., Indiana University-Indianapolis, Indianapolis, Indiana, United States
  • Avin, Keith, Indiana University-Indianapolis, Indianapolis, Indiana, United States
  • Thadhani, Ravi I., Massachusetts General Hospital , Boston, Massachusetts, United States
  • Moe, Sharon M., Indiana University-Indianapolis, Indianapolis, Indiana, United States

Frailty detection in incident patients on hemodialysis enables prevention strategies for disability, falls, hospitalizations and death. Traditional operational definitions of frailty developed with community-dwelling older adults includes self-report of “involuntary weight loss of >4.5 kg in the past year”. Weight loss in ESKD is difficult to ascertain given volumetric flux during dialysis. We hypothesize that omitting weight criteria from the operational definition of frailty will reclassify incident dialysis patients to capture comorbidities, physical function and disability.


This is a cross-sectional analysis of subjects in the Indiana cohort in LUCID, a longitudinal study of incident dialysis patients. The 5 traditional frailty criteria are weight loss, exhaustion, low physical activity, slowness and weakness, with a sum score of >3 termed “frail”. Our “Light” (weight free) definition excludes weight loss and categorizes a sum > 2 as “new-frail”. SF36 scores, physical function (gait speed and grip strength), demographics as well as co-morbidities were compared between those “frail” and “new-frail”. We also determined the % of subjects of identified as “disabled” in the frail and new-frail groups, using published FiND criteria (which are all self-report).


Mean (SD) age of the 146 subjects was 54 (13) years; 54% male, 71% black, 53% had DM, and median (IQR) dialysis vintage was 90(65) days. New-frail criteria (4 elements) increased the number of participants identified as frail using the traditional 5 element definition from 43 to 90. The new frail compared to the traditional definition increased the odds of being identified as frail 1) in diabetics, (OR 1.56 to 2.31) and 2) for those with inability to walk several blocks (OR 3.032 to 7.792) (both p<0.001). (OR 1.381 to 3.047, p=0.003). The new definition also identified more subjects who are frail, but not yet disabled (n=37% vs traditional criteria 16.5% p<0.001).


Use of weight loss in traditional frailty definitions is complicated by its fluctuations during dialysis. The omission of weight and thus changing to a 4 element index demonstrated higher ORs for disease and function. This tool may be better at identifying frailty risk in women and those frail but not yet disabled. Longitudinal outcomes associated with this definition are currently being studied.


  • NIDDK Support