Dialyzing Wisely is a multi-centre, registry embedded, stepped-wedged, interrupted time-series evaluation of the implementation of an evidence-based and best practice acute KRT pathway in the 15 adult general and cardiac ICUs and 3 pediatric general and cardiac ICUs in Alberta that provide acute KRT (Table 1). This implementation plan will follow the principles of the Learning Health System Knowledge to Action Framework (Fig. 2) [16].

Table 1 Alberta ICUs Delivering KRTFig. 2

Outline of learning health system knowledge to action framework. The healthcare system data are used to set local priorities for improvement. Data on priority problems and potential contributors help find the evidence based solutions to improve local health systems (data to knowledge). Using evidence based knowledge, this will inform local quality improvement and implementation science guided efforts (knowledge to performance). Finally, the continuous monitoring of local KPIs and practices will develop practice based knowledge while revealing future opportunities for quality improvement (performance to data). Adapted from Kilbourne et al. [16]

Dialyzing Wisely will utilize a structure-process-outcome framework for quality assessment by implementing a standardized framework to each acute KRT program (Fig. 3) [17, 18]. Each participating ICU will be structured to be led by a team with expertise in KRT (e.g., physician, educator, registered nurse, administrator) who will undergo and further disseminate targeted education strategies prior and during the intervention. Each ICU will receive quarterly performance reports on a minimum suite of essential KPIs tailored to their unit’s specific practice. Through this feedback the team members will work on adjusting and improving KRT practices that do not meet KPI benchmarks. Further, acute KRT prescribers will receive individualized reports outlining their prescription practices and their alignment with best evidence. Follow up with individual ICUs and practitioners will occur to better understand and evaluate prescribing practices. After completion of the intervention period, the impact of the intervention will be measured through select clinical and economic outcome measures (e.g., mortality, lengths of stay, KRT utilization and renal recovery). At this time, using our partners (outlined below) we will transition ownership to local stakeholders.

Fig. 3

Outline of Structure-Process-Outcome framework. The framework of elements which constitute each standardized acute dialysis program is depicted above. While themes remain consistent between programs, each program may use specific elements which work best within their own processes

The SPIRIT checklist is available as an Additional File.

Trial oversight

Dialyzing Wisely will be governed by an Executive Committee. The Executive Committee will be comprised of Leads from 4 participating Strategic Clinical Networks (SCNs) (Critical Care, Medicine (Kidney Health Section), Cardiovascular Health and Stroke, and Maternal, Newborn, Child and Youth). SCNs are province-wide integrated teams that collaboratively identify and solve challenges within their specific area of health. SCN membership includes clinicians, patients, operational leaders, researchers, community and industry partners and other stakeholders [19].

The Executive Committee will be supported by an International Advisory Panel. This panel will be made up of critical care nephrology experts with a specific interest in quality and safety for acute KRT. The International Advisory Panel will provide a high-level review of program performance and will provide unbiased recommendations to ensure ongoing advancement and program success.

The program will also have a Steering Committee with representation from all stakeholder groups (i.e., physicians, nurses, educators, administrators, operational leads, epidemiologists, health economists, informatics specialists and patient-partners). This Steering Committee will review KPI reports and study developments to ensure ongoing appropriate program advancement and oversight.

Finally, key stakeholders have been identified at each individual study site to operationalize and champion the implementation of the acute KRT pathway and enact change. We will also include provincial organizations to facilitate our audit and feedback process and transition ownership of the program to local sites. This will include members listed above as well as the Alberta Medical Association Physician Learning Program (PLP).

All aspects of the governance will be overseen by the program manager, and facilitated by the research assistant. Key reporting and educational messaging will be delivered by Clinical Practice Leads with expertise in knowledge translation strategies (Fig. 4).

Fig. 4

Project governance and roles structure. *Stakeholders will fill the role of Leadership and Educational leads for each Acute Dialysis program, fulfilling the Structure KPI Requirements in Dialyzing Wisely. All project committees will be supported by the Program Manager and Research Assistant. Key reporting and educational messaging will be facilitated by Knowledge Translation Specialists

Population and eligibility

The inclusion criteria will be critically ill patients (i.e., adults and children) receiving acute KRT as part of their routine ICU care. No exclusion criteria will be applied.

Interventions, duration and frequency of follow-up

The acute KRT pathway will consist of two specific interventions:

1.

Monitoring, reporting and audit of acute KRT KPIs.

2.

Provision of individualized prescriber and program reports for acute KRT initiation patterns.

The acute KRT care pathway will be implemented in a stepwise fashion with a pilot followed by randomized stepped wedge roll out at centres across Alberta over the subsequent 21 months. Roll out of ICUs will be performed in clusters and will coincide with individual site activation of a novel, province-wide electronic provincial clinical information system (CIS), Connect Care (EPIC, Verona WI).

Monitoring and reporting of acute KRT KPIs

The monitoring, reporting and audit of acute KRT KPIs will be done by means of automatically generated reports delivered directly to KRT stakeholders at each ICU at quarterly intervals (Table 2). These unit-level aggregate summaries of KPIs will be benchmarked to other ICUs in Alberta and will be delivered in an electronic fashion. KPIs captured in the reports for CKRT will include: 1) filter life, 2) downtime, 3) delivered dose, 4) ultrafiltration realized and 5) number of access alarms; and for IKRT: 1) treatment completion, 2) delivered dose, 3) solute clearance, 4) ultrafiltration realized and 5) catheter malfunction.

Table 2 Intervention definitions and parameters - Key Performance Indicators (KPIs)Prescriber reporting

An ICU-specific prescriber and program report will be provided to each prescriber and ICU on a quarterly basis. The report will include: 1) the number of acute KRT initiations, 2) acute KRT initiations based on conventional indications (i.e., hyperkalemia, acidosis, metabolic status, oxygenation status and cumulative fluid status) termed ‘appropriate initiations’, 3) general guidelines of KRT initiation criteria as per STARRT-AKI standard initiation arm protocol. (Table 2) [5] These will be benchmarked against prescriber patterns from providers in the same ICU, as well as similar ICUs across Alberta.

Stakeholder education

Prior to implementation of the reports, each ICU will receive education strategies specifically tailored to their site. The education strategies will be informed by local acute KRT leaders, champions, and stakeholders and will serve to identify barriers and facilitators to the program (Table 3). Initial education strategies will likely contain similar themes across all sites and will be managed by our study team, however after receiving feedback through the KPI reports, each site will be encouraged to facilitate and conduct their own audit and educational activities to address any unit specific shortcomings identified in their acute KRT KPI performance.

Table 3 Components of the multi-faceted intervention and knowledge implementation strategy

Select outlying prescribers will be contacted by the study team to further evaluate either well performing or poorly performing prescriber patterns as they relate to most recent evidence.

Primary outcomes

Measurement of change across acute KRT KPIs that will include the following:

∘ CKRT: 1) filter life, 2) downtime, 3) delivered dose, 4) ultrafiltration realized and 5) number of access alarms

∘ IKRT: 1) treatment completion, 2) delivered dose, 3) solute clearance, 4) ultrafiltration realized and 5) catheter malfunction

Number of appropriate acute KRT initiations

Number of patients entering maintenance KRT programs

Acute KRT and healthcare systems costs

Secondary outcomes

Length of KRT

ICU and hospital lengths of stay

ICU and 90-day mortality

Rates of KRT dependence at 90 days

Health-related quality of life measurement (i.e., EQ-5D-5L and PedsQL) and patient-related outcome measures (PROMs) (i.e., ESAS-r and IPOS-renal)

Data management

Effects of prescriber and acute KRT program patterns in the initiation of acute KRT will be determined by monitoring resource use associated with initiation of KRT at both a program level and healthcare system level. This will be done by determining first the units of each resource and then by assigning costs to each unit. At the program level we will capture the number of acute KRT initiations as well as total patient-days of acute KRT per specific modality (i.e., intermittent or continuous). These acute KRT initiations will also be adjusted for severity of illness to enable translation across units with varying case-mix, acuity and workload. Acute KRT disposable costs will include KRT filters, catheters, replacement and anticoagulation solutions.

New intake of critically ill survivors with severe AKI into ESKD dialysis programs will be reviewed on a quarterly and yearly basis to determine any changes relevant to use of this resource to the Alberta healthcare system. We will ensure that any changes in intakes reflect acute KRT initiations based on adherence to best-evidence practices.

Data will be collected on patient characteristics: (i.e., demographics, type of admission [medical, surgical, trauma]), clinical status (i.e., comorbid diseases including chronic kidney disease, primary diagnosis), acuity (i.e., APACHE II, SOFA, CFS), ICU treatment (i.e., duration of renal replacement therapy, mechanical ventilation, vasoactive therapy), ICU and hospital lengths of stay, and outcomes (i.e., renal recovery and mortality); and KRT-associated resource data: (i.e., filter use, prescription/dose, machine alarms/down time, anticoagulation, re-hospitalizations, progression of renal disease). Data variables to be captured are summarized in Table 4.

Data sources will include multiple Alberta Health Services administrative databases, the Nephrology Information System (NIS) and the Patient Based Renal Information System (PARIS) (Supplementary Table 1) [20, 21].

Co-enrollment

Co-enrollment into other clinical research studies will be evaluated on a case-by-case basis.

Statistical analysesHealth outcome measures

The patients’ characteristics and the target clinic outcomes in the baseline and intervention periods will be summarized. Mean (SD) and/or median (IQR) will be used for continuous variable; frequency will be used for categorical variables. To compare pre- and post-intervention difference, p-values will be provided by t-test (for normally distributed variables), non-parametric Wilcoxon tests (for non-normally distributed variables) or Chi-square test (for categorical variables). Interrupted time series (ITS) analyses will use autoregressive integrated moving average (ARIMA) models to determine changes in the KPI performance following the implementation of the acute KRT pathway.

KPI and interrupted time series analysis

Interrupted time series (ITS) analyses will be done using autoregressive integrated moving average (ARIMA) models to account for temporal trends and to determine whether there were changes in the process and clinical outcomes at the intervention period (compared with the baseline period) associated with implementation of the evidence-based acute KRT pathway. Each KPI will be assessed separately, as well as in aggregate with other KPIs. Autocorrelation, partial autocorrelation, and inverse autocorrelation functions will be assessed for model parameter appropriateness and seasonality. Stationarity will be assessed using the autocorrelation function and the augmented Dickey–Fuller test. The presence of ‘white noise’ was assessed by examining the autocorrelations at various lags, using the Ljung–Box χ2 statistic.

Healthcare system costs analysis

The primary health economic evaluation will be a within-study analysis of the cost-effectiveness of the Dialyzing Wisely program. Subject to available resources we will consider additional model-based analyses of Dialyzing Wisely over a longer time horizon.

The within-study analysis will be conducted on resource use and outcomes occurring during the study period. It will include total quarterly acute KRT-associated costs for each specific ICU following the implementation of KPI reporting. The cost analysis will include 1) utilization costs of CKRT filters, CKRT fluids, KRT anticoagulation (if any),and KRT catheters. Costs will be calculated in part using acute KRT process measures captured by our acute KRT KPIs.

We will also conduct an analysis of healthcare systems costs including those associated with total ICU and hospital stay and ongoing new ESKD costs (i.e., long-term, maintenance dialysis costs, total healthcare costs. Healthcare system costs will be reported as costs of acute KRT in ICU as a proportion of total ICU costs per quarter. Modeling analysis will capture costs to the health service, social care providers and patients so as to provide cost estimates from a societal perspective. We will determine models for averted and delayed acute KRT and averted new end-stage kidney disease requiring chronic dialysis based on observed changes in practice based on prescriber reports and adherence to best-evidence. Results will be reported as the incremental net benefit and incremental cost-effectiveness ratios. Uncertainty will be captured in the analyses through probabilistic sensitivity analysis and reported using cost- effectiveness acceptability curves, showing the likelihood the intervention will be cost-effective over a range of values of willingness-to-pay thresholds for specific outcomes.

Health-related quality of life and patient-reported outcome measures

Cost-effectiveness will be analyzed by estimating incremental cost and effectiveness based on, patient life-years gained and quality-adjusted life years [QALY]) gained. These will also be modeled based on adherence to best evidence-based practices and anticipated outcomes. QALYs will be calculated based on health-related quality of life as measured by the EQ-5D-5L and the PedsQL in children.

PROMs analysis will be conducted by quality-of-life assessment using the ESAS-r and/or the IPOS-renal scales [22]. This data will be collected at first chronic dialysis session. Quantitative analysis will be done to assess changes in scores using a linear mixed effects model with the baseline score as a fixed effect co-variate, and dialysis unit, with each cluster as a random intercept.

All statistical analyses will be done using SAS Enterprise Guide 7.1 (Cary, NC), TreeAge Pro (TreeAge Software Inc., Williamstown MA) and Excel (Microsoft, Redmond VA).

Subgroup analysis

Pre-specified subgroup analysis will include ICU patients to 1) adult vs. children, 2) female vs. male, 3) academic vs. community/regional ICUs, 4) cardiovascular ICUs vs. Cardiac ICU vs. medical/surgical ICUs, 5) high KRT volume vs. low KRT volume centers (i.e., as per quartiles).

We will perform the above analyses for health economic evaluations, patient and process of case measures to include our pre-specified primary and secondary outcomes for each subgroup. Each analysis will be accompanied by a test for interaction between treatment and subgroup to ascertain whether effects differ significantly between subgroups.

Ethics approval and consent to participate

This evaluation was reviewed by the University of Alberta Health Research Ethics Board (HREB) and a waiver of consent was granted based on the premise this project represents health services implementation and evaluation compatible with a quality assurance and improvement initiative.

Knowledge translation and dissemination

Dialyzing Wisely is largely a program focused on knowledge translation and implementation of an acute KRT pathway based on evidence-based best practices.

Throughout the project period Clinical Practice Leads will review sub optimally performing KPIs across acute KRT programs and provide targeted education to individual ICU teams to best understand opportunities to drive improved performance. The Clinical Practice Leads will also facilitate the implementation of evidence-based best practices into clinical practice through individualized prescriber reports regarding acute KRT initiations and bedside provider education.

We will publish two peer-reviewed manuscripts for the Dialyzing Wisely program, one outlining the protocol, and the second disseminating study results. Additional manuscripts may be developed to present subgroup findings as well as patient-centered outcomes including PROMs. In addition, the results of the Dialyzing Wisely program will be presented at local, provincial, and national critical care and nephrology meetings.