Kidney Week

Abstract: FR-OR149

Interventions to Reduce Early Dialysis Initiation in Canada: A Cluster Randomized Trial

Session Information

• High-Impact Clinical Trials
  October 26, 2018 | Location: 20A, San Diego Convention Center
  Abstract Time: 12:15 PM - 12:30 PM

Category: Dialysis

• No subcategory defined

Authors

• Tangri, Navdeep, Seven Oaks General Hospital, Winnipeg, Manitoba, Canada

Navdeep Tangri, MD, PhD

• Ferguson, Thomas W., Seven Oaks General Hospital, Winnipeg, Manitoba, Canada

Thomas W. Ferguson,

• Garg, Amit X., London Health Sciences Centre, London, Ontario, Canada

Amit X. Garg,

• Manns, Braden J., Foothills Medical Center, Calgary, Alberta, Canada

Braden J. Manns,

Background

There are higher healthcare costs and poorer outcomes when patients initiate dialysis earlier at a higher level of kidney function compared to when they initiate dialysis later. Despite the publication of The Initiating Dialysis Early and Late (IDEAL) study and concurrent clinical practice guidelines recommending an intent-to-defer dialysis initiation strategy, the estimated glomerular filtration rate (eGFR) at dialysis initiation varies widely.

Methods

We conducted a national cluster randomized controlled trial of a multifaceted knowledge translation intervention in 55 multi-disciplinary advanced chronic kidney disease (CKD) clinics in Canada. The intervention included academic detailing, audit and feedback, and provision of visual aids for patients and providers promoting the intent-to-defer strategy. The primary efficacy outcome was the proportion of patients who started dialysis early (eGFR > 10.5 ml/min/1.73m²) and the primary safety outcome was the proportion of dialysis starts occurring in the acute inpatient setting. Absolute risk differences were estimated using generalized-estimating-equations analysis of clustered binary data using the difference-in-differences approach.

Results

There were 6,836 patients that started dialysis during the study period, 3,521 people who started in the 1 year before the intervention and 3,314 people who started in the 1 year following the intervention. Prior to the intervention, the proportion of early dialysis starts was 28.1% in the intervention group and 31.1% in the control group, and following the intervention was 30.6% and 31.0% in the intervention and control groups respectively. In unadjusted and analyses adjusted for patient-level characteristics no differences in early initiation were observed (adjusted absolute risk difference 3.7%; 95% confidence interval [CI] -0.07% to 8.1%; p = 0.10). Similarly, no differences in acute inpatient dialysis starts between the intervention and control group were detected (adjusted relative risk 1.11; 95% CI 0.96-1.29; p = 0.17).

Conclusion

A multifaceted knowledge translation intervention did not result in further declines in early dialysis initiation in Canada. Secular trends in dialysis initiation from uptake of the IDEAL trial findings may have limited the impact of this intervention.

Funding

• Government Support - Non-U.S.