A new study from UHN, featured on the cover of Nature Medicine, shows that wearable technologies such as Apple Watch can help monitor heart failure by providing early signs of deteriorating health where medical attention may be needed.
Heart failure—when the heart does not pump blood adequately to meet the body’s needs—is a global health crisis that leads to hospitalizations, increased use of health care services, and reduced life expectancy. Despite recent medical advances, patients with heart failure still face a high risk of poor outcomes. This highlights the need to improve risk assessment and better guide timely interventions.
“For patients with heart failure, periods of stability are often interspersed with flare-ups of symptoms such as shortness of breath or fatigue. These episodes may require medical attention to prevent hospitalization and improve quality of life,” says Dr. Heather Ross, Clinician Investigator at UHN and co-senior author of the study. “However, risk assessments for heart failure patients often rely on scheduled clinical visits or evaluation tools that take measurements at only one point in time. They don’t account for the changing, episodic nature of heart failure.”
Therefore, doctors and clinicians need to find better ways of monitoring heart failure symptoms and predicting when medical intervention may be needed. Wearable technology, such as Apple Watch, can continuously track important health and fitness measures, such as heart rate and blood oxygen levels, making it an excellent candidate for monitoring. However, until recently, it has been unclear how these measurements might be used to understand day-to-day changes in people living with heart failure.
To investigate this, the research team, led by researchers at UHN’s Peter Munk Cardiac Centre and the Ted Rogers Centre for Heart Research, initiated the Ted Rogers Understanding Exacerbations of Heart Failure study (TRUE-HF) in collaboration with Apple in the fall of 2020. The study examined whether biometric data from Apple Watch could be used to predict peak oxygen uptake (pVO2) in patients with heart failure in their daily lives for a three-month period. PVO2 is the highest amount of oxygen the body uses during intense exercise, serving as a key indicator of cardiorespiratory fitness. It is typically measured in a clinical setting using Cardiopulmonary Exercise Testing (CPET), where patients are evaluated while exercising to maximal capacity.
“We created an AI model, called TRUE-HF, trained on data from 154 patients and then validated on 63 patients, to estimate individuals' daily peak oxygen uptake using measurements from Apple Watch,” says Dr. Chris McIntosh, Senior Scientist at UHN and co-senior author of the study. “We found that when participants went about their daily routines while wearing an Apple Watch, our smartwatch-based pVO2 estimates strongly correlated with lab-derived ones from CPET.”
What’s more, they found that each 10% drop in the TRUE-HF–estimated fitness measure (pVO2) was linked to a more than threefold higher risk of an unplanned medical event. These events typically occurred approximately a week after the drop first appeared. A modified version of the model also predicted unplanned use of medical services around 21 days after the first drop in predicted pVO2. These results were further validated in a public cross-platform FitBit dataset from the National Institutes of Health (NIH) All of Us Research Program.
These results suggest that daily smartwatch measurements can provide early warning signs of worsening health and help predict when unplanned medical care may be needed for people living with heart failure. Identifying real-time changes in health through wearable technology, without requiring additional tests or added effort from patients, could enable faster, better care.
Yuan Gao is a doctoral candidate at UHN and co-first author of the study.
Dr. Yas Moayedi is a Clinician Investigator at UHN and co-first author of the study.
Dr. Christopher McIntosh, Senior Scientist at UHN and Dr. Heather Ross, Clinician Investigator at UHN, are co-senior authors of the study. Dr. McIntosh is an Assistant Professor in the Department of Medical Biophysics at the University of Toronto (U of T). Dr. Heather Ross is a Professor at the Institute of Medical Sciences at U of T.
Other study authors from UHN include Farid Foroutan, Bhavish Verma, Ben Kim, Enza De Luca, Margaret Brum, Darshan H. Bhrambhatt, Joe Duhamel, and Anne Simard.
This work was supported by the Ted Rogers Centre for Heart Research, the Natural Sciences and Engineering Research Council of Canada (NSERC), the Canadian Institutes of Health Research (CIHR), the University of Toronto, and UHN Foundation.
Dr. Ross is the Loretta Rogers Chair in Heart Function.
Apple Incorporated provided 200 iPhones and Apple Watch devices for the study, provided feedback on the manuscript, and collaborated with all authors to build the study-specific mobile application. All authors are investigating patenting the TRUE-HF model described in the manuscript.