ELECTRIC-AI pilot study: The evaluation of patient-reported symptoms in remote monitoring care of ICD patients

Rationale: Remote monitoring (RM) is now considered the gold standard for
follow-up care of patients with an implantable cardioverter defibrillator
(ICD). Despite the potential of RM, there are important limitations of RM in
real-world clinical practice, including a lack of patient-reported outcomes and
symptom status, high rates of non-actionable hospital visits, and a complex
infrastructure that requires a dedicated team. With the ageing population,
rising healthcare costs, and a shortage of healthcare personnel, it is critical
to improve the RM infrastructure. We hypothesize that supplementing the remote
device data with patient-reported symptoms, could improve the efficiency of the
process. Leading to less non-actionable in-hospital visits and earlier
identification of patients at high risk.

Objective: To collect patient-reported symptoms, device data, and the clinical
follow-up action of each remote transmission to develop an automated triage
workflow that can discriminate between actionable and nonactionable
evaluations. In addition, we want to assess the usability of symptom
registration with a mobile application for a diverse group of ICD patients.
Secondary objectives are evaluating the effect of symptom reporting on the
quality of life of patients and investigate trends in symptoms, device data,
and clinical outcomes. Third, we aim to collect a comprehensive remote
monitoring dataset that can be used for the development of an automated triage
algorithm.

Study design: This is a single-center, prospective, and observational study
aimed to collect patient-reported symptoms and device data from study
participants with a Gallant* ICD or Gallant* CRT-D (Abbott). The participating
clinical site is the Amsterdam University Medical Center (AUMC, location AMC
and VUmc). Patients use the Castor Connect smartphone application to report
their symptoms, in addition to their standard-care pathway for one year.
Symptom tracking will occur with a standardized questionnaire developed by the
AUMC based on clinical expertise. A purposive sample of study participants
(n=5) will be invited to participate in semi-structured interviews to gain
insight into how outcomes of triage could in future be presented back to
patients, closing the feedback loop. Preliminary feedback options generated
through co-creation will be evaluated in semi-structured interviews with this
sample of participants.

Participants will not be exposed to risks associated with the use of the
application, as it serves as a data collection tool that is used alongside
regular care. Participants will be asked to use the application at least
weekly, this will take 2 to 15 minutes. The focus on symptom reporting, could
potentially impact the quality of life or anxiety levels of a patient. This
will be evaluated during the study. Participants need to visit the hospital
once, during the baseline visit at the initiation of the study. Halfway through
the follow-up period and at the end of the study the patients will receive a
call from the researcher. If patients are selected (n=5) for the
semi-structured interviews to evaluate the usability of symptom reporting in
the mobile application, they will have to visit the hospital for a meeting of
one hour. Patients will be able to decline participation in the semi-structured
interviews without further consequences for their participation in the main
study.