Julia Kurps, our Real World Data Consultant specialized in OHDSI and RADAR-base, contributed to a publication published in Nature Medicine, as part of our contribution to the European Union Innovative Medicines Initiative BigData@Heart program (grant no.116074). The full publication can be read here.
The study explored the potential added value of consumer-grade wearable technology as support in conventional clinical research and patient management within the RATE-AF trial (ClinicalTrials.gov identifier: NCT02391337). This trial aimed to compare heart rates in elderly, multimorbid patients with permanent atrial fibrillation and heart failure who were treated with either a drug commonly used to treat heart failure or beta-blockers. Over 20 weeks, 53 participants (average age 75.6 years, 40% women) provided extensive heart rate and physical activity data via a wrist-worn wearable device linked to a smartphone.
For this project, the open-source RADAR-base platform, which was developed under the European Union Innovative Medicines Initiative RADAR-CNS program (grant no. 115902), operated on a virtual machine hosted by Amazon Web Services in the Europe (London) region, managed by The Hyve. Participants registered their devices through the Fitbit developer application, enabling automated and secure data collection through the RADAR-base platform. Participants also had access to view their individual heart rate and step counts.
The study concluded that the drug commonly used to treat heart failure and beta-blockers have similar impacts on heart rate in patients with atrial fibrillation. Besides, the study clearly demonstrated the strength of continuous remote data collection by providing an average of two to three million data points per patient collected throughout a 20-week period. A neural network model based on the wearable sensor data collected via the RADAR-base platform showed similar performance for predicting future health status as conventional measures used in clinical trials. This suggests the potential for reducing the need for frequent in-person clinical assessments.