Variation in access to kidney transplantation across dialysis facilities: using process of care measures for quality improvement

Abstract

Background: Kidney transplantation rates differ by patient and dialysis-facility characteristics, yet little is known about the sources of this variation or how access to transplantation can be improved. Examining specific steps in the transplantation process may guide quality improvement efforts that ultimately improve the equity and efficiency of transplantation.

Methods: We sought to examine variation across dialysis facilities in the completion of four specific steps (medical suitability, interest in transplantation, pretransplantation workup, and waiting list) after adjustment for patient characteristics. Using 11,674 patients on chronic hemodialysis therapy between January 1993 and December 1996 at 206 facilities in Indiana, Kentucky, and Ohio, we derived facility-specific: (1) actual completion rates at each step; (2) expected completion rates based on patient age, race, sex, and cause of renal failure; and (3) ratio of actual to expected completion rates.

Results: Actual completion rates varied greatly, with facilities at the 75th percentile having a twofold greater completion rate than facilities at the 25th percentile (eg, 59% versus 32% at the first step). Although patient characteristics were strongly associated with step completion, adjustment for these characteristics did not decrease variation across facilities. At each step, approximately 20% of facilities had a significantly greater than expected completion rate, whereas approximately 15% had a significantly lower than expected completion rate.

Conclusion: There is substantial variation across dialysis facilities in access to kidney transplantation, even after adjustment for patient characteristics. Identifying steps with less than expected completion rates may help facilities target such efforts as treatment of medical conditions, patient education, and early referral for pretransplantation workup and waiting list placement.