Improving the Adequacy of Hemodialysis
A Quality Improvement Project Conducted Jointly

By The Mid-Atlantic Renal Coalition & Delmarva Foundation

Executive Summary

Background: The literature contains compelling evidence of a correlation between the delivered dose of hemodialysis and patient mortality and morbidity, suggesting that mortality and morbidity among patients with end stage renal disease (ESRD) decreases as sufficient hemodialysis treatments are delivered. Clinical practice guidelines, generated by the National Kidney Foundation's Dialysis Outcome Quality Initiative (DOQI), established a Kt/V ³ 1.2 as the minimum acceptable delivered dose.

The 1997 National ESRD Core Indicator Project revealed that Network 5 had the lowest average Kt/V of all 18 Networks. The Network's Medical Review Board determined that a Quality Improvement Project should be conducted to improve this important parameter of care. A collaborative effort was undertaken between the Mid-Atlantic Renal Coalition and Delmarva Foundation, the PRO contractor for Maryland and Washington, D.C. The PRO contributed resources and expertise in the areas of project design, statistical analysis, and electronic data collection. The Network furnished expertise in the care of patients with ESRD, through its 20-member Medical Review Board, and conducted all intervention activities.

Project Description: The goal of the project was to improve the adequacy of dialysis delivered to outpatient hemodialysis patients. Three indicators were developed to measure the proportion of patients receiving adequate dialysis: the proportion of patients receiving adequate dialysis, measured by Kt/V; the proportion of prescriptions written to achieve adequate dialysis; and the proportion of treatments delivered as prescribed. The latter was further divided into sub-indicators measuring five reasons a dialysis treatment may not be delivered as prescribed (time, dialyzer, needle, blood flow rate, and dialysate flow rate).

The project employed an equivalent control group design with pre- and post-measurements. Based on inclusion and exclusion criteria, twelve (12) units in Maryland were selected and assigned to either a control group (6 units) or an intervention group (6 units). The two groups were shown to be statistically equivalent at the start of the project.

Thirty (30) patients were randomly selected from each unit and data from six treatments during the observation periods were gathered for each patient. Baseline data were collected from January-March 1998. Re-measurement data were taken from treatments delivered January-March 1999. PRO reviewers abstracted data on-site at each facility using a customized MedQuest tool.

Interventions: The intervention plan was based on the philosophy that the root cause of inadequate dialysis differs between facilities. Using baseline data, intervention facilities received a feedback report displaying their performance on each indicator. A Pareto chart was included to show the probable root causes, based on the indicator with five sub-parts. An educational workshop, addressing both dialysis adequacy and QI tools, was conducted for the 6 intervention units. Patient and staff educational materials were provided. MRB teams acted as facilitators, and assisted each intervention unit in embracing the project and in identifying and correcting factors unique to that facility. Each team consisted of a nephrologist, a nephrology nurse, a renal social worker, and a renal dietitian. A team visited each of the intervention facilities and assisted in launching a unit-specific improvement project.

Measurement Results: The first step in achieving adequate dialysis is a prescription sufficient to deliver adequate dialysis treatments. Notably, the most profound improvement was seen in this indicator. The proportion of prescriptions sufficient to achieve adequate dialysis improved significantly in the intervention group over the control group and to a substantively important degree from 48% to 60% in the intervention group vs. from 56% to 50% in the control group. Furthermore, estimates of the proportion of patients receiving adequate dialysis increased. The project is available to other Network 5 dialysis facilities using the "project-in-a-box" approach.