Background In the end stage renal failure (ESRF) population frailty is associated with early mortality, increased hospitalisations, and significant symptom burden. After identifying a high degree of frailty in patients withdrawing from dialysis in our population we examined the use of formal frailty scoring and its use in identifying deteriorating patients on renal replacement therapy (RRT).

Methods The Rockwood Clinical Frailty Scale (CFS) is a 9-point scale which enables measurement of frailty it has high inter-rater reliability and correlates well with objective measures of frailty and has been validated in the ESRF population. We introduced routine recording of the CFS for all ESRF in our region for patients at three monthly intervals and for low clearance patients at the time of RRT education.

Results A total of 1663 scores (range 1–9) have been recorded in 798 patients. Mean age 63.9 years. Of those patients currently undergoing haemodialysis the median CFS score was 4 (n=533). The median score prior to death was 5.5. Evidence of deterioration in CFS score (last score greater than the first) was present in 50% (14/28) of those who were deceased at follow-up, whereas a deteriorating score was only present in 22.3% (97/435) of those who remained alive, p=0.001. A documented score of 6 or greater was present in 51.4% (38/74) of those who subsequently died vs. 21.7% (158/727) of those who remained alive, p<0.001.

Conclusion Deterioration in frailty score is associated with death at follow-up. Furthermore, a score 6 or greater at any point is predictive of death at follow-up. Routine monitoring of frailty using the CFS provides a simple tool to identify patients who are deteriorating and at risk of death. High or deteriorating CFS score should trigger clinical review and anticipatory care planning where appropriate.