[Assessment of treatment quality by hierarchical models in the Italian dialysis and transplantation registry]

Given the operative difficulties and expenses inherent in a clinical audit, we explored whether the Italian Dialysis and Transplantation Registry (RIDT) might offer a valid alternative for quality analysis about two aspects of renal replacement therapy: how widespread is the practice of peritoneal dialysis, and how often is a temporary catheter used as first vascular access for dialysis. We analyzed the data of all patients recorded in the RIDT in 2007 with regard to the type of first treatment, age, gender and primary kidney disease. For peritoneal dialysis we compared all Italian regions having scattered data. With regard to the types of vascular access and the comorbidities at the start of treatment, we evaluated patients from Veneto only. The performance of regions and centers were evaluated using a random-effects multilevel logistic model. Only 65% of Italian regions were available in RIDT; 13.8% of patients began RRT with peritoneal dialysis, with only 3 regions exceeding 20%. Differences among regions were due more to differences in patient characteristics than to between center differences in treatment strategies. Data on vascular access were available for 83% of the patients. Almost 40% began RRT with a temporary catheter; in 12 of 22 centers this was less than 35%. Also in this case, differences were due to patient characteristics. Incomplete data limit the strength of the interpretation of our study results. At any rate, differences among regions or centers seem due more to patient characteristics than to treatment strategies.