JMIR Publications

ABSTRACT

Background:

Chronic heart failure accounts for approximately 1-2% of healthcare expenditures in most developed countries. These costs are primarily driven by hospitalizations and comorbidities,. Telemonitoring has been proposed to reduce the number of hospitalizations and decrease the costs of treatments for patients with heart failure. However, the effects of telemonitoring on healthcare utilization remains unclear.

Objective:

The aim of this systematic review is to study the effect of telemonitoring programs on healthcare utilization and costs of patients with chronic heart failure. We assessed the effect of telemonitoring on hospitalizations, emergency-department visits, length of stay, hospital days, non-emergency department visits and healthcare costs.

Methods:

We searched Pubmed, Embase and Web of Science for Randomized Controlled Trials (RCT) and non-randomized studies on non-invasive telemonitoring and healthcare utilization. We included studies published between January 2010 and August 2020. For each study, we extracted the reported data on the effect of telemonitoring on healthcare utilization. We used p-values smaller than .05 and confidence intervals not including 1.00 to determine if the effect was statistically significant.

Results:

We included 16 RCTs and 13 non-randomized studies in this review. Inclusion criteria, population characteristics and outcome measures differed between the included studies. The majority of the studies showed no effect of telemonitoring on healthcare utilization. Number of hospitalizations were statistically significant reduced in 9 out of 24 (38%) studies whereas emergency department visits were reduced in 1 out of 8 studies. An increase in non-emergency department visits (6 out of 9 studies) were reported. Healthcare costs showed ambiguous results with three studies finding an increase of healthcare costs, three studies finding a reduction and four studies finding no significant differences. Healthcare cost reductions were realised through a reduction of hospitalizations whereas increases were caused by the high costs of the telemonitoring program or increased healthcare utilization.

Conclusions:

A majority of telemonitoring programs do not show clear effects on healthcare utilization measures, except for an increase in non-emergency outpatient department visits. This may be an unwarranted side-effect rather than a prerequisite for effective telemonitoring. The consequences of telemonitoring on non-emergency outpatient visits should receive more attention from regulators, payers and providers. This review further demonstrates the high clinical and methodological heterogeneity of telemonitoring programs. This should be taken into account in future meta-analyses aiming to identify effective components of telemonitoring programs.