Automated Text Messaging After HF Hospitalization

Quick Takes

  • A heart failure (HF) telemonitoring strategy delivered through text messages did not improve NT-proBNP levels or clinical outcomes when delivered as part of a randomized clinical trial in Brazil.
  • A text message telemonitoring strategy delivering HF education and enabling physiologic monitoring was acceptable to patients and improved HF self-care.

Study Questions:

What is the efficacy of text message-based telemonitoring on clinical outcomes when delivered after a heart failure (HF) hospitalization?

Methods:

The MESSAGE-HF study was a prospective, parallel-group multicenter randomized clinical trial carried out across Brazil. The study enrolled patients with HF with reduced ejection fraction at the time of hospital discharge or within 30 days of admission for decompensated HF. Participants randomized to the intervention arm received text messages 4 times daily during the first 30 days of the study. Messages would be either educational or ask for feedback on HF signs/symptoms, which could then trigger a phone call by patients’ care teams or automatic diuretic adjustments. Message frequency decreased after the first month of the trial. The primary outcome was change in N-terminal pro–B-type natriuretic peptide (NT-proBNP) from baseline to 180 days. A hierarchical win-ratio analysis including adjudicated clinical events (cardiovascular death, first HF hospitalization) and variation in NT-proBNP levels was also performed. An additional secondary outcome included change in HF self-care.

Results:

Between July 2019 and July 2022, 699 patients were enrolled and randomized (65.8% male, mean [standard deviation] age 61.2 [14.5] years, 84.3% New York Heart Association class II or III HF symptoms). Guideline-directed medical therapy was optimized in the majority of participants. There was no significant change in the ratio of the change in geometrical means of NT-proBNP from baseline to 180 days (ratio of change telemonitoring to standard care, 0.92; 95% confidence interval [CI], 0.77-1.11; p = 0.39), and there were no significant effects in any of the prespecified subgroups. The win ratio for the hierarchical analysis was 1.04 (95% CI, 0.86-1.26). At 30 days, self-care scores were significantly improved in the telemonitoring group (mean difference, -2.2; 95% CI, -3.7 to -0.7; p < 0.001). This benefit was sustained at 180 days. Red flags were common in the telemonitoring group during the first 30 days of the trial, triggering automatic diuretic adjustments for 43 participants and 249 telephone contacts to 110 participants.

Conclusions:

An intensive, text-message based telemonitoring strategy delivered to HF patients after an HF hospitalization did not improve NT-proBNP levels or clinical outcomes, although it did improve HF self-care.

Perspective:

The care of HF patients has become increasingly complex in light of newer pharmacotherapies, devices, and lifestyle-based interventions. There has been growing concern that telemedicine will be unable to scale to meet the needs of HF patients given its resource-intensive nature and the growing prevalence of HF in the United States. There has thus been interest in using mobile health technologies to enable HF telemonitoring at scale. In meta-analyses, mobile health technologies deployed for HF monitoring and treatment engagement have improved HF outcomes, including symptoms, patient self-management, HF hospitalizations, and mortality. While the current trial failed to meet its clinical endpoints, it did lead to improvements in HF self-care.

Strengths of the MESSAGE-HF trial include the extended duration of follow-up relative to prior mobile health trials (6 months), deployment at a population level, and the collection of important process measures necessary to understand the impact of the intervention on clinical workflow (e.g., number of downstream telephone calls). Important limitations of the trial are worth noting, however. First, a large portion of the intervention was focused on physiologic monitoring to detect early decompensation. A growing body of literature, however, supports moving beyond physiologic monitoring and incorporating behavior change strategies into mobile health interventions. Second, participants in the trial may not be generalizable to the larger population of HF patients. Of note, almost all participants were on optimal guideline-directed medical therapy, which is rarely observed in contemporary observational studies and clinical trials in HF.

Clinical Topics: Heart Failure and Cardiomyopathies, Acute Heart Failure, Heart Failure and Cardiac Biomarkers

Keywords: Heart Failure, Text Messaging


< Back to Listings