The €1.5 million grant from EIT Health will fuel a two-year initiative aimed at advancing heart failure monitoring through voice-based detection of acutely decompensated heart failure (ADHF). The randomised controlled trial (RCT) to be conducted in Barcelona and Maastricht will involve around 200 patients. The study aims to evaluate the efficacy of voice analysis in identifying early warning signs and predicting ADHF exacerbations. This pioneering approach has the potential to revolutionise the management of ADHF by providing a non-invasive and cost-effective means of early detection, leading to improved patient outcomes and reduced hospitalisations.
Primary Objective:The voice-based decompensation prediction software detects HF-related decompensation better and at an earlier stage compared to the standard of care, allowing earlier intervention with medical therapy (particular adjustment of diuretic therapy) and resulting in a significant reduction of hospitalisations and worsening of symptoms.Rationale: Heart failure (HF) is a common chronic disease that contains the risk of imminent volume overload, called decompensation. Symptoms usually occur late during the course of decompensation, leaving insufficient time to effectively intervene. Voice-based digital biomarker may detect imminent deterioration significantly earlier without the requirements of implantation of invasive devices such as CardioMEMS.
Objective: The primary objective of the study is that voice-based decompensation prediction software detects HF-related decompensation better and at an earlier stage compared to the standard of care including standard eHealth, allowing earlier intervention with medical therapy (particular adjustment of diuretic therapy) and resulting in better outcome.Secondary objectives are, among others, safety of monitoring using voice-recordings, cost-effectiveness, acceptance and usage experiences of voice-based monitoring, effects on quality of life and the accuracy of detection of volume overload.
.jpeg)
Study design: Randomized double-blinded controlled trial (RCT) at three sites. Patients in both groups will be equipped with a smartphone and the pre-installed application to daily monitor voice-recordings and symptoms. The recording immediately prior to discharge of the initial hospitalization, when patients are recompensated, will serve as baseline (=reference)to personalize the individual voice pattern. During admission while still volume overloaded, patients will perform the first voice recording, which will be continued during the study. All voice samples and the answers to the symptom questions will be transmitted to the NoahLabs server for analysis and will be compared to the reference sample.In both groups, monitoring of symptoms will be equal. If certain thresholds are exceeded(e.g. increase in shortness of breath, peripheral oedema), health care professionals (HCP)will receive notifications about the findings. For patient group A (intervention group), HCPs will additionally receive notifications about deviations of the daily voice recordings from the individual patient’s baseline and a potential risk of HF-related decompensation. If there are no deviations, HCPs will not receive a notification. For group B (control group), HCPs will only receive notifications related to symptoms but not voice-recordings (i.e. standard eHealth monitoring as currently applied in these patients). In both groups, HCPs may adjust therapy in the most appropriate way upon their own discretion. The observation period for patients will be 6 months with fixed evaluation at months 1, 3 and 6. The patients will be seen by a physician knowledgeable of HF that is part of the study team (investigator), but who is blinded to the study group. Additional contacts with healthcare providers will be possible as clinically required.
Study population: 200 patients randomized 1:1 will be recruited from the wards of the participating sites, admitted with acute decompensated HF (ADHF) or de-novo HF(irrespective of left-ventricular ejection fraction), both sexes and aged 18 years or more. Limited exclusion criteria (HF not main problem, intervention planned during study period, inability to use app, not speaking local language) to enable the inclusion of a representative patient population.
