Page 18 Heart Failure Remote Monitoring Evidence From the Retrospective Evaluation of a Real World Remote Monitoring Program

Although mortality rates increased in the CCCP group after program completion, the overall effect was
still beneficial compared to controls over the 1-year follow-up. This finding is similar to results from eight

meta-analyses published between 2007 and 2013 evaluating the effect of remote monitoring on
mortality [11-17]. These studies reported that compared to usual care, remote monitoring reduced
mortality with overall effects ranging from 17% to 51%. The variations in these effects could be
explained by the difference in type (structured telephone support vs telemonitoring), speed of feedback

(rapid vs non-rapid), invasiveness (invasive vs non-invasive) of remote monitoring, duration of follow-
up, study designs, and severity of disease. While a majority of the studies included in these meta-
analyses evaluate the effect of remote monitoring only within the monitoring period, this evaluation
further monitored participants 8 months beyond the regular 4-month program duration to evaluate the

residual effects of the telemonitoring and educational intervention. While the evidence of the association
of remote monitoring and reductions in mortality has been consistent over time in meta-analyses, the
same effect has not been demonstrated for hospitalizations. The majority of the meta-analyses
referenced above reported reductions in hospitalizations except Clarke [14], who did not find a

significant reduction in all-cause hospitalization. Likewise, some recent prospective trials have also
demonstrated varied effects, ranging from no effect [18-20] to reduction in hospitalizations rates by
37% [21]. Among many other reasons that may account for the incongruity between mortality and
hospitalization effects, it is generally easier to assess mortality, which is a hard endpoint that is difficult

to miss, unlike hospitalizations, which may be unreported, misclassified, or missed in controls.
Traditionally, remote monitoring is seen as a short-to-medium term adjunct to regular care to empower
patients for self-management following hospitalization. Long-term use is not usually feasible due to cost.
However, based on our findings, we speculate that increasing the duration of the program to enable

patients to develop self-competency may improve outcomes. Although this may not be cost-effective for
all participants, risk stratification to identify patients who will benefit from prolonged monitoring may be
needed. Alternatively, because patients have built disease self-competency on the program and
accompanying monitoring devices, a graduated removal of program components (i.e., keeping patients

pg. 18