Effect of renal center characteristics on mortality and technique failure on peritoneal dialysis

BACKGROUND: Recent studies report decreased mortality in patients on peritoneal dialysis (PD) over time, suggesting that advances in PD have resulted in improved patient outcomes. Our investigation sought to assess the effect of renal center characteristics on mortality and technique failure (TF) rates. METHODS: Covariates of interest included center-specific cumulative number of PD patients treated, percentage of patients who initiated dialysis on PD, and academic status. Using data obtained from the Canadian Organ Replacement Register, the 17,900 patients who received PD during the 1981 to 1997 period were studied. Mortality and TF rate ratios (RR) were estimated using Poisson regression, adjusting for age, gender, race, primary renal diagnosis, province, follow-up time, and type of PD. RESULTS: As the cumulative number of PD patients treated increased, covariate-adjusted mortality significantly decreased (P < 0.05); a weaker yet significant association was observed between number of PD patients treated and TF. As the percentage of patients initiating dialysis on PD increased, TF rates decreased significantly. No association was observed between center academic status and PD mortality or TF rates. CONCLUSIONS: These results imply that a center's experience with and degree of specialization toward PD impact strongly on PD outcomes. One hypothesis is that a center's propensity to exploit technical and non-technical advances in PD increases directly with these variables. It is also possible that, through experience, centers become more adept at identifying appropriate patients to receive PD. More detailed research is required to evaluate these hypotheses.