Smartwatch data could offer new way to assess cardiovascular health

The answer to your heart health may be on your wrist, a new study suggests. Researchers have developed a new way to assess cardiovascular health based on information routinely collected by smartwatches, according to a study being presented at the American College of Cardiology's Annual Scientific Session (ACC.25).

According to the findings, dividing the average daily heart rate by the number of steps taken per day provides a more reliable indicator of a person's cardiovascular fitness compared with either heart rate or step count alone.

The metric we developed looks at how the heart responds to exercise, rather than exercise itself. It's a more meaningful metric because it gets at the core issue of capturing the heart's capacity to adjust under stress as physical activity fluctuates throughout the day. Our metric is a first attempt at capturing that with a wearable device."

Zhanlin Chen, medical student at Northwestern University Feinberg School of Medicine in Chicago and study's lead author

Heart disease is the leading cause of death in the United States. While several screening tests can provide early warning of heart disease risk, many people do not undergo recommended screenings. Researchers said that taking advantage of the information collected by smartwatches could offer a new strategy to identify people at higher risk and encourage them to speak with a clinician about their heart health.

For the study, researchers analyzed data from nearly 7,000 U.S. adults who provided data from their Fitbit and electronic health records to the All of Us research program, a nationwide prospective study supported by the National Institutes of Health. Together, the data reflected 5.8 million person-days and 51 billion total steps taken.

Calculating the relationship between participants' average daily heart rate per step (DHRPS) and a variety of cardiovascular outcomes, data showed that people with elevated DHRPS (in the top 25^th percentile) were about twice as likely to have Type 2 diabetes, 1.7 times as likely to have heart failure, 1.6 times as likely to have high blood pressure and 1.4 times as likely to have coronary atherosclerosis (a buildup of plaque in the heart's arteries), compared with people who had lower DHRPS. No relationship was found between DHRPS and the risk of a stroke or heart attack.

The results also showed that DHRPS was more strongly associated with cardiovascular disease diagnoses than either daily heart rate or step count alone. Additionally, in a smaller subset of 21 participants, DHRPS was more strongly linked to maximum metabolic equivalents (METs) achieved on a treadmill stress test than either daily heart rate or step count alone.

Based on the findings, the researchers suggest that DHRPS could be used as an early indicator of who might benefit from more screening tests or cardiovascular conditioning to improve their heart's functioning. According to Chen, the metric is simple enough that an individual could calculate it on their own based on the data collected by a smartwatch, or it could potentially be built into smartwatch applications.

However, he added that the study offers only an initial validation of the DHRPS approach, and the cross-sectional study design did not allow researchers to determine when the Fitbit measurements were taken relative to when cardiovascular disease outcomes were diagnosed. Moving forward, the researchers hope to conduct more prospective studies with a higher temporal resolution, tracking DHRPS at the scale of minutes rather than aggregated data across days.

With further refinement and validation, Chen said that DHRPS or a similar metric could ultimately be incorporated into the standard heart disease risk assessment that clinicians use.

"Wearables are welcomed by the consumer and worn throughout the day, so they actually have minute-to-minute information about the heart function," Chen said. "That is a lot of information that can tell us about a lot of things, and there's a need to further study how this detailed information correlates with patient outcomes."