*Purpose: The Karnofsky Performance Status (KPS) Scale has been used as a proxy for frailty and as a predictor of transplant outcomes, however reliability of the instrument in national transplant data registries is unknown. We investigate the inter-rater reliability of KPS reporting among transplant centers to assess its utility as a measure of physical function in kidney transplant candidates.

*Methods: Patients were selected from the Scientific Registry of Transplant Recipients (SRTR) database between 2006-2020 with at least two KPS scores in a three-month period. The reliability of KPS was estimated from the intraclass correlation coefficient (ICC) in a multilevel model with patient and center level random effects. All analyses were adjusted for the time elapse between measurements, year reported, patient demographics and comorbid conditions.

*Results: Among the 8,197 patients observed, 5,348 (65%) had a KPS score in the normal range (80-100 points) and only 87 (1%) had scores in the disabled range (0-40). We found substantial variability and poor reliability in KPS reporting between centers. Roughly one third of candidates had scores that varied 20-80 points in difference (See Table: Range between the maximum and minimum scores.) The Interrater reliability between centers was estimated to be 23% (95% CI: 21%, 25%) agreement, where estimates below 50% are considered poor.

*Conclusions: *Conclusions: Poor reliability in KPS reporting raises concerns for its utility as a measure of physical function and proxy for frailty. Kidney transplant candidates are a unique population that require a less subjective and more precise instrument for risk assessment.

« Back to 2021 American Transplant Congress