Preventing Early Dialysis Starts

Knowledge Translation Interventions to Prevent the Early Initiation of Dialysis: A Cluster Randomized Trial

Manitoba Health Research Council, University of Calgary, University of Western Ontario, Canada, University of Ottawa, University of Toronto, Dalhousie University, McGill University, University of British Columbia

Chronic kidney disease (CKD) and its end stage of kidney failure requiring dialysis are important contributors to morbidity, mortality and health care costs. Over the last two decades, there has been a strong secular trend in the earlier initiation of dialysis for treatment of kidney failure from progressive CKD. These trends have occurred in spite of evidence showing harms with early dialysis initiation and increased health care costs. Recently, investigators from the Canadian Society of Nephrology, including study co-investigators, have proposed clinical practice guidelines to recommend an "intent-to-defer" approach for dialysis initiation. Whether these guidelines require an active knowledge translation strategy or they will simply translate through passive dissemination is unknown. In the investigators' proposed national cluster parallel group randomized clinical trial, we will randomize CKD clinics across Canada to an active knowledge translation strategy to defer dialysis initiation or passive dissemination of guidelines (current practice). The unit of observation will be the patient (i.e., outcomes will be measured at the level of an individual patient), and the unit of randomization will be at the level of the multidisciplinary CKD clinic. The investigators will then evaluate the kidney function (estimated glomerular filtration rate - eGFR) at dialysis initiation for all dialysis starts originating from these clinics to examine whether our KT strategy is safe and effective at delaying dialysis initiation. Our active KT strategy, if effective, will have a significant impact on patient morbidity and health care costs. The investigators' hypothesis and specific aims are as follows: Hypothesis: The investigators hypothesize that the clinics randomized to the active KT strategy will start a greater proportion of patients on dialysis later (eGFR below 10.5 ml/min/1.73m2) compared to the control. Aim 1 - Efficacy: To compare the impact of an active KT intervention with passive guideline release on the proportion of patients followed by a Nephrologist ( > 3 months) who start dialysis with an eGFR >10.5ml/min/1.73 m2 across the randomized CKD clinics (clusters) in Canada. Aim 2 - Safety: To compare the impact of an active KT intervention with passive guideline release on safe dialysis initiation (acute unplanned dialysis starts) across the randomized CKD clinics in Canada.

Background: End Stage Renal Disease (ESRD) requiring chronic dialysis treatment is associated with poor health outcomes and high costs. Recent data shows that early initiation of dialysis, defined as starting dialysis with an estimated glomerular filtration rate >10.5 ml/min/1.73m2 (eGFR; the measure of kidney function used in practice), has risen rapidly in the last two decades. In 2010, a large randomized trial was published that evaluated the effect of early vs. late initiation, noting no health benefits but higher costs. Despite this, in a recent national cohort study, the investigators noted substantial practice pattern variation in the timing dialysis initiation in Canada, noting that > 40% of all patients started dialysis "early", ranging from 10% to 57% across regions. The Canadian Society of Nephrology has recently released clinical practice guidelines on the timing of dialysis initiation, recommending an "intent-to-defer" over an "intent-to-start early" approach for the initiation of chronic dialysis. Since simply releasing guidelines does not ensure that the evidence practice gap is bridged, the Canadian Kidney Knowledge Translation and Generation Network (CANN-NET), a national network of clinicians, researchers and knowledge users, was established to ensure best practices for patients with chronic kidney disease (CKD). On behalf of CANN-NET, we propose a cluster randomized controlled trial (RCT) of a knowledge translation (KT) strategy to reduce early initiation of dialysis in patients with severe CKD. Informed by careful survey work, the knowledge translation intervention will consist of patient- and provider-directed educational tools based on the recent published clinical practice guidelines, and will include compelling visual aids (infographic and whiteboard video), audit and feedback, and in-person medical detailing. The control group will have access to the published clinical practice guidelines, consistent with current clinical practice. The investigators' hypothesis and specific aims are as follows: Hypothesis: The investigators hypothesize that the clinics randomized to the active KT strategy will start a greater proportion of patients on dialysis later (eGFR below 10.5 ml/min/1.73m2) compared to the control. Aim 1 - Efficacy: To compare the impact of an active KT intervention with passive guideline release on the proportion of patients followed by a Nephrologist ( > 3 months) who start dialysis with an eGFR >10.5ml/min/1.73 m2 across the randomized CKD clinics (clusters) in Canada. Aim 2 - Safety: To compare the impact of an active KT intervention with passive guideline release on safe dialysis initiation (acute unplanned dialysis starts) across the randomized CKD clinics in Canada. Study Design: A cluster randomized trial of CKD clinics across Canada comparing the efficacy and safety of a KT intervention targeting early initiation of dialysis in patients with advanced CKD. The unit of observation will be the patient (i.e., outcomes will be measured at the level of an individual patient), and the unit of randomization will be at the level of the multidisciplinary CKD clinic. Team: The investigators' study team includes experts in the clinical epidemiology of CKD and kidney failure, local opinion leaders from every province/region, as well experts in knowledge translation and cluster randomized design. As such, the team is well positioned to carry out the proposed study. Research Significance: Early initiation of dialysis leads to uncertain benefit and potential harm to patients with CKD, with an increase in health care costs. This topic was deemed the highest priority area for knowledge translation intervention by regional and provincial kidney health administrators across Canada in a 2010 survey. If successful, the investigators' intervention will reduce the practice pattern variation in dialysis initiation, provide a successful framework for future KT interventions, and could have significant health and economic benefits.

Knowledge Translation, Chronic Kidney Disease, Dialysis, Intent-to-defer, Early dialysis, Cluster randomized trial

Clinics will have access to the Canadian Society of Nephrology (CSN) guidelines on the optimal timing of dialysis initiation (current practice). These guidelines have been published in the Canadian Medical Association Journal (CMAJ) and have been recently presented at the annual meeting of the Canadian Society of Nephrology.

Access to CSN guidelines, & provider- & patient-directed infographics recommending an intent-to-defer dialysis initiation strategy will be displayed in prominent clinic wall space & disseminated to patients. Educational whiteboard video will be made available as a resource for clinic staff & patients. Each clinic will receive reports from the Canadian Organ Replacement Register (CORR) outlining the proportion of patients followed by a Nephrologist (>3 months) starting dialysis early (eGFR >10.5 ml/min), for all incident dialysis patients from the clinic, with provincial & national average comparisons. These reports, & the CSN guideline on timing of dialysis initiation recommendation, will be delivered to the medical lead for each CKD clinic. Each clinic will receive an in-person visit from one of the study investigators/collaborators highlighting the clinical practice guidelines & evidence supporting an intent-to-defer strategy, & will receive follow-up.

Primary Efficacy Outcome: Proportion of patients followed by a Nephrologist ( > 3 months) who start dialysis with an eGFR > 10.5 ml/min

Proportion of patients followed by a Nephrologist ( > 3 months) who start dialysis with an eGFR > 10.5 ml/min in the follow-up period. eGFR at dialysis initiation will be ascertained from the clinic clusters and confirmed by linkage with the Canadian Organ Replacement Register (CORR).

Primary Safety Outcome: Proportion of patients starting dialysis as inpatients or in an emergency room

Proportion of all incident dialysis patients originating from the randomized clinic clusters that start dialysis in a hospital or in an emergency room in the follow-up period. Patient location at dialysis initiation will be ascertained from the Canadian Organ Replacement Register (CORR) via linkage with the Canadian Institute for Health Information (CIHI)-Discharge Abstract Database.

The rate of change in early dialysis starts will be analyzed to assess whether the effect of the active knowledge translation intervention dissipates over time, and for non-linear effects.

Secondary Outcome: Outcomes of all patients followed in the nephrology clinics using provincial data linkages, wherever available (presently Ontario, Manitoba and Alberta)

Examine the outcomes of all patients followed in the nephrology clinics using provincial data linkages, wherever available (presently Ontario, Manitoba and Alberta) to examine rates of hospitalizations, deaths, and cost of pre-dialysis care in both study arms.

Secondary Outcome: Quarterly proportion of new starts from each clinic, and the differences in this proportion between the two study arms.

Quarterly proportion of new starts from each clinic (new starts/total number of patients followed in the clinic), and the differences in this proportion between the two study arms.

Acceptability of the knowledge translation materials provided to the clinic measured using semi-structured interviews and surveys in the follow-up period.

Inclusion Criteria: Multidisciplinary clinics in Canada that provide care coordinated by a Nephrologist to patients with chronic kidney disease (CKD) These clinics have already been identified in a previous survey

Nesrallah GE, Mustafa RA, Clark WF, Bass A, Barnieh L, Hemmelgarn BR, Klarenbach S, Quinn RR, Hiremath S, Ravani P, Sood MM, Moist LM; Canadian Society of Nephrology. Canadian Society of Nephrology 2014 clinical practice guideline for timing the initiation of chronic dialysis. CMAJ. 2014 Feb 4;186(2):112-7. doi: 10.1503/cmaj.130363. No abstract available.

Ferguson TW, Garg AX, Sood MM, Rigatto C, Chau E, Komenda P, Naimark D, Nesrallah GE, Soroka SD, Beaulieu M, Alam A, Kim SJ, Dixon S, Manns B, Tangri N. Association Between the Publication of the Initiating Dialysis Early and Late Trial and the Timing of Dialysis Initiation in Canada. JAMA Intern Med. 2019 Jul 1;179(7):934-941. doi: 10.1001/jamainternmed.2019.0489.

Chau EM, Manns BJ, Garg AX, Sood MM, Kim SJ, Naimark D, Nesrallah GE, Soroka SD, Beaulieu M, Dixon S, Alam A, Tangri N; Canadian Kidney Knowledge Translation and Generation Network (CANN-NET). Knowledge Translation Interventions to Improve the Timing of Dialysis Initiation: Protocol for a Cluster Randomized Trial. Can J Kidney Health Dis. 2016 Sep 14;3:2054358116665257. doi: 10.1177/2054358116665257. eCollection 2016.

Tangri N, Garg AX, Ferguson TW, Dixon S, Rigatto C, Allu S, Chau E, Komenda P, Naimark D, Nesrallah GE, Soroka SD, Beaulieu M, Alam A, Kim SJ, Sood MM, Manns B. Effects of a Knowledge-Translation Intervention on Early Dialysis Initiation: A Cluster Randomized Trial. J Am Soc Nephrol. 2021 Jul;32(7):1791-1800. doi: 10.1681/ASN.2020091254. Epub 2021 Apr 15.

Manns BJ, Garg AX, Sood MM, Ferguson T, Kim SJ, Naimark D, Nesrallah GE, Soroka SD, Beaulieu M, Dixon SN, Alam A, Allu S, Tangri N. Multifaceted Intervention to Increase the Use of Home Dialysis: A Cluster Randomized Controlled Trial. Clin J Am Soc Nephrol. 2022 Apr;17(4):535-545. doi: 10.2215/CJN.13191021. Epub 2022 Mar 21.