Resources

• Annual Report
• Disaster Preparedness
• Links
• Managing Conflict
• Medical Alerts
• Provider Directory
• Regulatory Updates
• Trends-Incidence
• Trends-National
• Trends-Prevalence
• Trends-Transplant
• Zip Code Data

Administration

• About Us
• Annual Meeting
• Coalition
• Email List
• ESRD Networks
• Privacy Policy
• Site Accessibility

Consumer Services

• Access to Success
• Common Concerns
• Consumer Committee
• Depression
• Exercise
• Network 11 Posters
• Grievance Process
• Living Well Video
• Patient Concerns
• Patient Rights

Data Management

• 2728 Form
• 2746 Form
• Address Change
• CROWNWeb
• Data Compliance
• Data Policy/Procedure
• Newsletter
• Patient Activity Report

Quality Improvement

• Bone Disease
• Elab Project
• Fistula First
• MRC Treatment Goals
• Patient Safety
• QI Information

Improving Hemodialysis Adequacy

ESRD Network 11 - Final Project Report on Improving Hemodialysis Adequacy in Network 11

Diane Carlson and Jan Deane, Network 11, St. Paul, MN; Bruce Lindgren and Cynthia Davey, Biostatistics Consulting Lab, University of Minnesota, Minneapolis, MN, Steve Helgerson, Epidemiology for Action, Seattle, WA

Abstract

The 2000 National ESRD Clinical Performance Measures (CPM) project (4th quarter 1999 data) showed that, based on a 5% random sample, 74% of adult in-center hemodialysis patients in Network 11 received the minimum dose of dialysis as measured by urea reduction ratio (URR) of 65%. The goal of this quality improvement project is to increase the number of hemodialysis patients in Network 11 receiving adequate dialysis (URR of 65%) to at least 80%. Facilities with < 80% of patients with URR of 65% were selected for intervention (39 facilities/3060 patients). The intervention consisted of regional workshops. Those facilities in the intervention group were required to attend one of four workshops. At the workshop, continuous quality improvement (CQI) principles and barriers to achieving dialysis adequacy were discussed. The participants were expected to complete a facility-specific root cause analysis and a quality improvement plan to be implemented at their facility. Follow-up data were requested for the 3rd quarter of 2001. For the 4th quarter of 1999, these 39 facilities had 62.6% (1777/2839) of patients with URR of 65%, while the non-intervention facilities had 88.1% (5804/6587) of patients with URR of 65%. In the 3rd quarter of 2001, these 39 facilities had 76.5% (2340/3060) of patients with URR of 65%. At the same time, a randomly selected 5% sample of the nonintervention facilities (n = 15) had 86.9% (747/860) of patients with URR 65%. The 13.9 percentage point improvement in dialysis adequacy in the intervention facilities is clinically and statistically significant. The 2001 National CPM assessment (4th quarter 2000) found that 81% of patients in Network 11 were receiving dialysis with a URR of 65%.

Introduction and Background

As of December 31, 1999, there were 14,069 patients receiving in-center hemodialysis therapy in 270 facilities in Network 11 (Michigan, Wisconsin, Minnesota, North Dakota, and South Dakota). The 2000 National ESRD Clinical Performance Measures Project (4th quarter 1999 data) showed that, based on a 5% random sample, 74% of adult in-center hemodialysis patients in Network 11 received the minimum dose of dialysis as measured by URR of 65%. Despite a steady increase in the proportion of patients receiving adequate dialysis from 1993 - 1999, important opportunities remain for improvement of hemodialysis adequacy in Network 11 facilities.

The goal of this project was to increase the percent of in-center hemodialysis patients in Network 11 receiving adequate dialysis (URR of 65%) to at least 80%.
To create the highest probability for change and to maximize the improvement, it was decided to conduct interventions with those facilities having the lowest percent of adequately dialyzed patients.

Methods

A. Quality Indicators

The quality indicator used in this study was the urea reduction ratio (URR). The URR was calculated from the first monthly pre/post BUN lab draw for each of the collection periods (October, November, and December 1999, July, August, and September 2001).

• Baseline: Percent of patients in the intervention facilities with URR 65%
  Numerator: All adult in-center hemodialysis patients in intervention facilities who have been on dialysis at least 6 months and have achieved a mean URR of 65% for the time period of 10/1/1999 - 12/31/1999.
  Denominator: All adult in-center hemodialysis patients in intervention facilities who have been on dialysis at least 6 months prior to 10/1/1999.
• Remeasurement: Percent of patients in intervention facilities with URR 65%
  Numerator: All adult in-center hemodialysis patients in intervention facilities who have been on dialysis at least 6 months prior to 7/1/2001 and have achieved a mean URR of 65% for the time period of 7/1/2001 - 9/30/2001.
  Denominator: All adult in-center hemodialysis patients in intervention facilities who have been on dialysis at least 6 months prior to 7/1/2001.

B. Data Collection

Data for the baseline measurement were collected as part of the Elab demonstration project, with data being collected directly from the national and local labs. Data were collected for the remeasurement by using a data collection form (attachment A) that has been used in previous Network 11 collection efforts. Facilities were asked to complete pre and post BUN values for their patient sample for the 3 months of July, August, and September 2001. The accuracy of the data collected using the data collection tool has been validated in past Network 11 projects.

To assist with the evaluation of the QIP results, remeasurement data were requested from a 5% sample (15 facilities, 860 patients) of the nonintervention facilities. Data were collected in the same manner as with the intervention facilities.

C. Intervention

The purpose of the intervention was to identify root cause(s) of inadequate dialysis, and to develop and implement a quality improvement plan. The intervention was divided into three phases as follows.

Phase I: Distribute a Hemodialysis Adequacy Toolbox to all intervention facilities. This toolbox contained the following resources:

• Hemodialysis adequacy information, including journal articles regarding barriers to achieving adequate dialysis.
• DOQI guidelines for hemodialysis adequacy.
• Root cause analysis worksheet (to be completed before and brought to the workshop).

Phase II: Required attendance of Medical Director and Nurse Manager at one of four regional workshops. The workshops were one day in length. The workshops covered the following topics:

1. Facility-specific CQI education

a. CQI tools
b. Case study practice
c. Facility-specific identification of root cause and facility barriers

2. Breakout sessions addressing specific process indicators

a. Prescription issues (including determination of adequate prescription, staff adherence to prescription, and strategies for maximizing dialysis)
b. Vascular access (including early placement issues, monitoring, and timely interventions)
c. Patient adherence to prescription (including identification of causes, strategies for behavior modification, and the Network 11 Special Needs Manual)

Phase III: Follow-up reporting

• Facilities were required to complete a patient-specific follow-up form and submit it to the Network office.
• Network staff provided technical assistance as needed to assist the facilities to implement their plans.

Results

Chart 1 shows the change in percent of adequately dialyzed patients for nonintervention facilities, intervention facilities, and both groups combined from baseline to remeasurement. In 1999, the total percent of patients with URR 65% for the nonintervention and intervention groups was 75.4%. At remeasurement (using a 5% random sample of nonintervention facilities) the total (intervention and nonintervention) percent of patients with URR 65% was 81.5%. This indicates that the goal of the project, to increase the percent of in-center hemodialysis patients in Network 11 to at least 80%, has been met. This compares with the 2000 CPM data collection (4th quarter 2000) results, which show that 81% of patients in Network 11 had URR 65%.

For purposes of comparison, the baseline percent of adequately dialyzed patients in nonintervention facilities was limited to the 15 randomly selected facilities. The percent of patients with URR 65% in this group at baseline was 86%, and 86.9% at remeasurement. This results in an estimated 0.9% improvement in dialysis adequacy in the nonintervention facilities. The observed improvement in the 39 intervention facilities was 13.9 percentage points (62.6% at baseline, 76.5% at remeasurement). Since facilities with the highest percent of inadequately dialyzed patients were selected for intervention, it can be reasonably argued that some of the observed 13.9 percentage point improvement is attributed to regression towards the mean. An unplanned outcome of this project provided data for 17 facilities that were eligible for intervention (<80% of patients with URR 65%) but did not participate in the intervention workshops. The observed improvement in dialysis adequacy of 7.8 percentage points between 4th quarter 1999 and 4th quarter 2000 in these 17 facilities provides an estimate of the regression towards the mean effect (although data collection times are different, this provides the best available estimate). Subtracting the estimated 7.8 percentage point regression towards the mean effect from 13.9 percentage points results in an improvement in dialysis adequacy of 6.1 percentage points in the intervention group.

The difference from baseline to remeasurement in percent of adequately dialyzed patients is approximately a normally distributed variable with a mean (SD) of 0.061 (0.0115) for the intervention group and 0.009 (0.0165) for the nonintervention group. The two-sample t-test was used to test the null hypothesis that these mean differences are equal (p < 0.0001). Results of this test indicate that, even after adjusting for regression towards the mean effect, the 6.1 percentage point improvement in dialysis adequacy observed in the intervention group was significantly greater than the 0.9% improvement in the nonintervention group.

Conclusions

Dialysis facilities in Network 11 have consistently demonstrated improvement in hemodialysis adequacy since the inception of the Core Indicators project in 1993 (chart 2). As previously mentioned, the 2000 CPM results from 4th quarter of 2000 show that 81% of patients in Network 11 have URR 65%. In spite of this improvement, there continue to be facilities that have been unable to meet that goal. This project enabled Network 11 to maximize the use of resources by focusing on those facilities with the most opportunities for improvement.

This was the first time Network 11 used the mandatory workshops as an intervention for quality improvement. There were several positive outcomes of this type of intervention.

• Increased awareness of accountability for improving outcomes. The exercise of having to develop a root cause analysis and identify barriers forced participants to look beyond the obvious towards a more comprehensive review of hemodialysis adequacy.
• The synergy of working together with other medical directors and nurse managers to identify similar problems and brainstorm solutions.
• Ample time was allowed to focus on process change (prescription, access, and non-adherence) as opposed to just focusing on outcomes. Again, the sharing of processes among facilities was very helpful.

The workshops focused only on the poorly performing facilities. One change that might be made to future workshops would be to include top performers who could share best practices. It is always helpful to see what types of strategies work in similar situations. This type of intervention (regional workshops) appears to be an effective method for improving care since the improvement seen in the intervention facilities was significant. It was a cost effective method for intervening in a large number of facilities (4 workshops for 39 facilities vs. 39 on-site focused reviews). This intervention method should be considered for use in future projects and can be modified as needed for other types of quality improvement projects where process change is within the control of the dialysis facility, e.g., anemia management.