Clinical Scorecard: Remote care may reduce heart failure events

At a Glance

Key Highlights

• In a 24-week, multicenter, open-label phase 3 randomized trial of 506 patients, the primary composite endpoint occurred in 17% with mobile health management and 41% with usual care.
• The primary endpoint was time to first cardiovascular death or worsening heart failure; the reported relative reduction in likelihood of a first event was 65%, with a number needed to treat of 3.8.
• Worsening heart failure occurred in 14% of the mobile health group and 39% of the usual-care group; cardiovascular mortality occurred in 4% and 8%, respectively.
• Endpoint proportions favored mobile health management across prespecified subgroups, including patients aged 75 years or older, those classified as frail, and those with reduced or preserved ejection fraction.
• Mean adherence was 77% for biometric transmission and 67% for daily symptom questionnaires; 20% of intervention patients ended assigned follow-up prematurely.

Guideline-Based Recommendations

Diagnosis

• The trial enrolled patients hospitalized for heart failure decompensation who were preparing for discharge or had been discharged within the preceding 30 days.
• Worsening heart failure events included acute decompensation requiring intravenous decongestive therapy in an outpatient or emergency setting, unplanned hospitalization, or decompensation complicating another hospital admission.

Management

• The intervention combined daily smartphone-based transmission of blood pressure, heart rate, and body weight with symptom questionnaires, team review, automated alerts, and video consultations.
• Usual care followed each center's established heart failure protocols, including nurse-led education, symptom monitoring, and face-to-face follow-up; guideline-directed treatment adjustments were encouraged in both groups.
• The authors stated that studies to establish the full implementation requirements for mobile health technology in heart failure care teams are warranted.

Monitoring & Follow-up

• The mobile health platform generated 11,012 alerts during follow-up.
• The trial did not systematically evaluate alarm response times, staffing requirements, training needs, or workflow effects of alerts and unplanned contacts.

Risks

• The trial was open-label and did not systematically collect adverse-event data; it was not designed to systematically assess intervention-related safety.
• Fourteen intervention patients were unwilling to continue using the technology and 14 were unable to use it.
• Differences in standard care across centers could have introduced variability, although sensitivity analyses accounting for center-level clustering were consistent with the primary findings.

Patient & Prescribing Data

The trial randomized 506 adults across 9 hospitals and 1 primary care service in Spain: 255 to mobile health management and 251 to usual care. Patients had been hospitalized for heart failure decompensation and were preparing for discharge or had been discharged within 30 days.

Over 24 weeks, the primary composite endpoint occurred in 43/255 patients (17%) assigned to mobile health management and 102/251 (41%) assigned to usual care. The intervention group had more unplanned contact with care teams. Mean adherence was 77% for biometric transmission and 67% for symptom questionnaires; 51 intervention patients (20%) ended assigned follow-up prematurely. All randomized patients were included in the intention-to-treat analysis.

Clinical Best Practices

• The intervention used daily monitoring of blood pressure, heart rate, body weight, and symptoms, with health care team review and automated alerts for abnormal or missing data.
• Both trial groups received established heart failure management, and guideline-directed treatment adjustments were encouraged.
• The authors noted that operational requirements for broader implementation remain to be determined, including staffing, training, alarm response, and workflow effects.

Related Resources & Content

• The Lancet Digital Health trial report (source identified in article; URL not provided)