Periodic Reporting for period 3 - RETENTION (heaRt failurE paTient managEment and iNTerventIOns usiNg continuous patient monitoring outside hospitals and real world data)
Berichtszeitraum: 2024-05-01 bis 2025-10-31
RETENTION is a HORIZON 2020-SC1-BHC project, funded under the topic SC1-DTH-12-2020: "Use of Real-World Data to advance research on the management of complex chronic conditions". It has started on 01/05/2021 and will end on 31/10/2026, after a project extension granted in early 2025.
In RETENTION the health condition that we are dealing with is Heart Failure (HF). HF is a chronic disease, caused by the inability of the heart muscle to pump adequate amount of blood into the circulation. Practically, the heart cannot respond adequately to its workload. This in turn leads to secondary problems, like metabolic disorders and, more importantly, to oxygen deficiency, which becomes more intense as the disease progresses and whenever the need for oxygen increases -like e.g. during physical activity. HF is associated with high mortality rates and frequent hospitalisations, imposing also a serious burden on health systems that is expected to rise in the coming years as the population ages.
The solution we are testing is technology platform consisting of two parts. In one part, the system is used to easily acquire and transmit a set of important clinical and environmental parameters from the patients’ home and present them to the physician on a regular basis or upon request. In this context, it is a telemonitoring system, that increases patient-physician interaction and is expected to improve the monitoring efficiency. On the other part, the system applies Machine learning technologies to analyse the data collected, aiming to identify patterns and correlations that would hopefully enable the prediction of acute episodes or side effects, and allow for personalized interventions.
With regard to Real World Data (RWD), there is increasing evidence in the literature that this type of data, can play a role in clinical decision making. It is foreseen that this type of data can generate added value to the predictive models, when analysed in correlation with the clinical measurements of the patients, and lead to more safe and accurate predictions and more effective and personalised interventions to be used in the daily clinical practice.
Overall, the goal of the project is to improve the management of the disease, to support clinical decision making, to help the prevention of acute episodes and allow physicians to timely perform effective interventions before symptoms escalate. In this way, it is expected that it will have a positive impact on the primary and secondary outcomes that have been defined in the project, namely in reducing mortality and hospitalisation rates, improving quality of Life as well as safety and well-being of the patients.
The RETENTION platform will be tested and validated in a Randomised Clinical Trial, conducted in 6 hospitals in Europe, with a power of 450 patients divided in three subgroups: patients with heart failure (HF), patients with Left Ventricular Assist Devices (LVAD) and heart transplant patients (HTx), to be monitored under the RETENTION study protocol for an average period of 18 months.
In RETENTION the health condition that we are dealing with is Heart Failure (HF). HF is a chronic disease, caused by the inability of the heart muscle to pump adequate amount of blood into the circulation. Practically, the heart cannot respond adequately to its workload. This in turn leads to secondary problems, like metabolic disorders and, more importantly, to oxygen deficiency, which becomes more intense as the disease progresses and whenever the need for oxygen increases -like e.g. during physical activity. HF is associated with high mortality rates and frequent hospitalisations, imposing also a serious burden on health systems that is expected to rise in the coming years as the population ages.
The solution we are testing is technology platform consisting of two parts. In one part, the system is used to easily acquire and transmit a set of important clinical and environmental parameters from the patients’ home and present them to the physician on a regular basis or upon request. In this context, it is a telemonitoring system, that increases patient-physician interaction and is expected to improve the monitoring efficiency. On the other part, the system applies Machine learning technologies to analyse the data collected, aiming to identify patterns and correlations that would hopefully enable the prediction of acute episodes or side effects, and allow for personalized interventions.
With regard to Real World Data (RWD), there is increasing evidence in the literature that this type of data, can play a role in clinical decision making. It is foreseen that this type of data can generate added value to the predictive models, when analysed in correlation with the clinical measurements of the patients, and lead to more safe and accurate predictions and more effective and personalised interventions to be used in the daily clinical practice.
Overall, the goal of the project is to improve the management of the disease, to support clinical decision making, to help the prevention of acute episodes and allow physicians to timely perform effective interventions before symptoms escalate. In this way, it is expected that it will have a positive impact on the primary and secondary outcomes that have been defined in the project, namely in reducing mortality and hospitalisation rates, improving quality of Life as well as safety and well-being of the patients.
The RETENTION platform will be tested and validated in a Randomised Clinical Trial, conducted in 6 hospitals in Europe, with a power of 450 patients divided in three subgroups: patients with heart failure (HF), patients with Left Ventricular Assist Devices (LVAD) and heart transplant patients (HTx), to be monitored under the RETENTION study protocol for an average period of 18 months.
In the first reporting period, the project's objectives focused on efficient management, coordination, and technical progress. Technically, user requirements were collected, and the RETENTION platform's architecture was developed iteratively with cross collaboration among clinical and technical partners. A data model was defined using widely used ontologies to develop data repositories and transmission mechanisms. Medical devices and sensors were tested and selected, Patient Edge devices were integrated and the mobile application was created. The CSB Dashboard was integrated with the Security Component, ML training workflows, and the DSS, using a micro-services architecture and continuous integration. The platform was tested and fine-tuned in a virtual Pilot of Pilots (vPoP) process.
In the second reporting period, significant technical advancements included the completion of the integrated RETENTION platform, with the mobile app supporting multiple health devices. The CSB Dashboard saw major improvements, and new DSS functionalities were added. The Security Component and data repositories were updated, and the ML Tool was redeployed. Extensive testing ensured system reliability, with dockerised containers enabling distinct platform instances for each hospital. Preparation for the RCT commencement progressed well with device acquisition and patient recruitment. The first patients were enrolled at the end of this period.
In the third reporting period, the project mainly focused on the RCT conduct. From a technical perspective, a distinct instance of the platform was deployed for each hospital on the same virtual machine using dockerised containers and a reverse proxy. From the moment the clinical trial started, on 24 April 2024, a problem reporting mechanism was established to report any technical issue encountered. The scope of this mechanism extended beyond initial problem reporting, to also cover for platform improvements and additional features requested by the clinicians. From a clinical perspective, each clinical team focused on the Trial conduct in their own hospital, involving patient recruitment, enrolment and training, setting-up the set of home devices, assuring ethics compliance, implementing the procedures for the proper implementation of the clinical protocol and patients’ follow-up. At the end of this period, the patient recruitment phase concluded with a total of 390 patients enrolled, comprising 219 HF, 64 LVAD and 107 HTx. Although slightly below the original total target of 450 patients by a percentage of 13.3%, the final sample size is supported by the revised statistical power analysis and remains adequate to assess the study’s primary objectives. Project assessment regarding the Ethical use of AI has been introduced in this reporting period, to ensure adherence to the EU framework for ethical development and use of AI. A phased implementation of the European Commission’s Assessment List for Trustworthy AI (ALTAI) was designed and executed.
Alongside technical and clinical work, the horizontal activities of project management, dissemination, communication, standardisation, sustainability and ethics monitoring are being performed in line with the Description of the Action (DoA) and the evolving project environment throughout all project phases.
In the second reporting period, significant technical advancements included the completion of the integrated RETENTION platform, with the mobile app supporting multiple health devices. The CSB Dashboard saw major improvements, and new DSS functionalities were added. The Security Component and data repositories were updated, and the ML Tool was redeployed. Extensive testing ensured system reliability, with dockerised containers enabling distinct platform instances for each hospital. Preparation for the RCT commencement progressed well with device acquisition and patient recruitment. The first patients were enrolled at the end of this period.
In the third reporting period, the project mainly focused on the RCT conduct. From a technical perspective, a distinct instance of the platform was deployed for each hospital on the same virtual machine using dockerised containers and a reverse proxy. From the moment the clinical trial started, on 24 April 2024, a problem reporting mechanism was established to report any technical issue encountered. The scope of this mechanism extended beyond initial problem reporting, to also cover for platform improvements and additional features requested by the clinicians. From a clinical perspective, each clinical team focused on the Trial conduct in their own hospital, involving patient recruitment, enrolment and training, setting-up the set of home devices, assuring ethics compliance, implementing the procedures for the proper implementation of the clinical protocol and patients’ follow-up. At the end of this period, the patient recruitment phase concluded with a total of 390 patients enrolled, comprising 219 HF, 64 LVAD and 107 HTx. Although slightly below the original total target of 450 patients by a percentage of 13.3%, the final sample size is supported by the revised statistical power analysis and remains adequate to assess the study’s primary objectives. Project assessment regarding the Ethical use of AI has been introduced in this reporting period, to ensure adherence to the EU framework for ethical development and use of AI. A phased implementation of the European Commission’s Assessment List for Trustworthy AI (ALTAI) was designed and executed.
Alongside technical and clinical work, the horizontal activities of project management, dissemination, communication, standardisation, sustainability and ethics monitoring are being performed in line with the Description of the Action (DoA) and the evolving project environment throughout all project phases.
Up to now, no measurable impact could be calculated, as the RCT is ongoing and no interim assessment of the RETENTION care model can be made, due to the standards of the blinded RCT conduct. Still, the effective use of the platform within the trial centres is expected to yield advanced data analytics and learning capabilities for large scale analysis of the comprehensive datasets collected, as well as innovative security control techniques. These developments are expected to contribute in the research field of data analytics beyond the state of the art. Regarding the potential impacts, the involvement of the stakeholders potentially impacted by the project results is ongoing and proceeds along the project execution timeline, aiming to assess the results into medium-term outcomes and long term impacts relevant to patient clinical outcomes, reduction of healthcare pressure, healthcare costs' reduction and cost-efficiency in patient management against standard of care.