CKD Specific Telemonitoring Platform to Minimize Adverse Outcomes in High Risk CKD Patients

Background: Chronic Kidney Disease (CKD) is an increasingly prevalent condition that represents a major national public health burden. The health transition from advanced CKD to the start of dialysis is a period of heightened vulnerability for many patients. Our data has demonstrated that during the 30-day period before the start of dialysis, there is a 10-fold increase in Emergency Department (ED) visits and hospitalizations related to kidney failure and its complications. Importantly, this spike occurs despite patients having frequent contact with their nephrology healthcare teams. For stable CKD patients, virtual monitoring could facilitate the communication of accurate and reliable data between patients and providers helping to avoid unnecessary ED visits and facilitate more optimal dialysis starts.

Design& Method: The VIEWER study is a national, pragmatic, multicenter randomized controlled trial across 5 Canadian sites from two provinces including:

1. Seven Oaks General Hospital (SOGH), Winnipeg, Manitoba.

2. St. Boniface hospital (SBH), Winnipeg, Manitoba.

3. Health Sciences center (HSC), Winnipeg, Manitoba.

4. Health Sciences Center, London, Ontario.

5. Scarborough Health Network (SHN), Scarborough, Ontario.

VIEWER is a CKD-specific virtual care platform that integrates data from a wireless Blood Pressure (BP) cuff, weight scale, transcutaneous O2 sat monitor, and wearable motion tracker. The patient-facing component of VIEWER (patient portal), is a custom application based on a mobile tablet that guides patients through a daily self-assessment routine using the connected devices. BP measurements, weight and oxygenation (as surrogates of volume status), and step counts (as a surrogate of functional status), provide semi-continuous longitudinal data on patient physical status. Additionally, patients are prompted to fill out a weekly symptom survey (Edmonton Symptom Assessment Score- revised; ESAS-r), a validated instrument of kidney failure symptoms. Patient data and ESAS scores are automatically uploaded to fully PHIA/HIPPA compliant servers where they are made available to the patients' care team through a secure, web-based provider portal. Provider notifications (flags) are generated for out-of-range values (i.e. BP, weights for volume management) and a secure messaging component allows for direct patient-provider communication. Participants will be trained to use the VIEWER platform by a member of the research team. On the provider side, two members of the care team at each site will be trained as "superusers" of the provider portal. These users will check and respond as needed daily (Monday to Friday) for flags and patient messages, and will respond or communicate as needed to the health care team as per local clinic standard operating procedures. Trends in measurements will be reviewed at multidisciplinary team rounds, which typically occur weekly. During routine clinic visits, trends in all objective and subjective measures will be available to review to inform clinical decision making. Consistent with our pragmatic design, we will not be proscriptive in how monitoring data are interpreted; rather, we ask that care teams consider this enriched data stream and incorporate that information according to their own protocols and clinical judgement.

Patients randomized to the control group will see their multidisciplinary CKD care teams as per usual care.

Follow-up visits and Focus Groups (or individual interview): Patient Reported Experience Measures (PREM) and Kidney Disease Quality of Life-Short Form (KDQOL-SF) will be measured at baseline, 3, 6, 9 and 12 months in all paticipanats. Usability, acceptability, and patient and provider perspective on the intervention will be assessed at the end of the trial using the System Usability Scale (SUS). In addition, two virtual focus groups will be conducted at the end of the study with a subset of participants and healthcare providers who used the VIEWER platform.

Statistical Methods:The primary outcome (time to first hospitalization or ED visit, censoring at dialysis or death) will be assessed using univariate Cox proportional hazards models and a Kaplan-Meier analysis with a log-rank test, as will the secondary outcomes of all-cause mortality, ED visits, hospitalization, and acute inpatient dialysis initiation. Change in overall PREM, KDQOL-SF and SUS scores will be assessed with two-sided t-test or Wilcoxon Rank Sum test as distributionally appropriate, and interpreted relative to their minimal important differences. All outcomes will be assessed at a two-sided alpha= 0.05. Subgroup analyses will be prespecified and limited to 1) eGFR greater or less than 10ml/min, 2) sex and gender (described below) 3) diabetes.