Impact of Remote Monitoring on Clinical Outcomes in Defibrillator Patients During the COVID-19 Pandemic: An Interrupted Time Series Analysis

Background Remote monitoring (RM) has become the standard of care in many Cardiac Implantable Electronic Device (CIED) clinics across North America and Europe. Some clinics are even adopting an alert-based, RM-only strategy for select patients. However, it remains unclear whether RM, compared to usual non-remote care, impacts the total number of CIED clinic visits, emergency department visits, hospitalizations, and all-cause mortality. Because such outcomes and their relationship to RM may be magnified in the presence of a global COVID-19 pandemic, we aimed to perform an interrupted time series analysis to observe trends in these outcomes in response to the declaration of the COVID-19 pandemic in patients with implantable cardioverter-defibrillators (ICD).

Methods In this retrospective study, we utilized existing electronic provincial databases maintained by Alberta Health Services (AHS) and Alberta Health (AH) to determine CIED visits, emergency room visits, cardiovascular (CV) hospitalizations, and all-cause mortality. We performed Interrupted Time Series (ITS) analysis to compare outcome trends in ICD-patients with and without RM in Alberta, both during the COVID-19 pandemic and the pre-pandemic period. We defined the time-period of the pandemic as March 17, 2020, to July 17, 2021. Pre-pandemic was defined as March 17, 2018, to March 16, 2020. We compared best model fits using the Akaike Information Criterion (AIC), selecting the model with the lowest AIC for each outcome. The best-fitting models were plotted. Outcomes between RM and non-RM groups were compared using regression models, with differences reported using 95% confidence intervals.

Results The CIED population consisted of 6,183 ICD patients from March 17, 2018, to July 17, 2021. Of these, 2,989 (48.3%) had access to RM. Our study found that access to virtual consultations sharply increased at the onset of the pandemic in both cohorts, though this trend was significantly higher in the RM group. Conversely, a sharp decline in in-person visits was observed for RM patients. Compared to those without RM, patients with RM showed no significant differences in all-cause mortality, hospitalizations, or emergency room visits, and these trends were not impacted by the COVID-19 pandemic.

Conclusion In ICD patients with and without RM, the number of virtual consultations increased while in-person visits decreased during the pandemic. However, no significant changes in the trends of cardiovascular hospitalizations, emergency room visits, or all-cause mortality were observed in either group during this period. This suggests that RM did not significantly impact key health outcomes for ICD-patients during the pandemic in Alberta.