Wearables for Cardio Monitoring May Yield Inaccurate Data Based on Skin Tone

Daniel Koerber, MD

Data collected by wearable technology may be less accurate for detection of both heart rate and rhythm in individuals with darker skin tone, according to results of a systematic literature review presented at the American College of Cardiology’s (ACC) 71st Annual Scientific Session, held April 2-4, 2022.

The findings, say study authors from the University of Alberta in Edmonton, Canada could suggest the potential for racial bias in consumer wearable devices and limit the deployment of the technologies for medical research.

“People need to be aware that there are some limitations for people with darker skin tones when using these devices, and the results should be taken with a grain of salt,” said study co-lead author Daniel Koerber, MD, resident physician at the University of Alberta in Edmonton, Canada, in an ACC statement. “Algorithms are often developed in homogeneous white populations, which may lead to results that are not as generalizable as we would like.”

The authors explain that heart rate measurement and detection of arrhythmia may be altered by the effect of the skin’s melanin content on absorption of light—the mechanism by which wearable devices record these data.

“Ongoing research and development of these devices should emphasize the inclusion of populations of all skin tones so that the developed algorithms can best accommodate for variations in innate skin light absorption,” said Koerber.

The specific focus of the current study was to summarize the accuracy of cardiac data collected specifically by wrist-worn devices among individuals of varying skin tones.

For the literature review, Koerber and colleagues searched for studies across Embase, MEDLINE, CINAHL, and Cochrane from database inception to July 28, 2021, and identified 622 records that stratified heart rate and rhythm data for consumer wearable technology according to participant race and/or skin tone (expressed using the Fitzpatrick score [range 1-6]). After screening there were 10 studies meeting the inclusion criteria with a total of 469 participants.

Six of the 10 studies (n=293) reported the frequency-weighted Fitzpatrick score, according to the study abstract. Investigators calculated the mean participant score to be 3.5, with a range of 1-6.

The Canadian team reports that 40% (4/10) of the studies reviewed found “a significant reduction in accuracy” of heart rate measurement with the wearable devices among darker-skinned individuals when data were compared with that recorded from lighter skinned participants. The reduction in accuracy was also seen in comparisons of the wearable data with “gold standard” measurements obtained with ECG or chest strap monitoring in the same darker-skinned participants.

In one study Koerber et al found that the wearable devices evaluated recorded significantly fewer data points for individuals with darker skin tones but that no discrepancy was recorded in heart rate accuracy, according to the abstract.

Of the 10 studies they evaluated, the researchers say only 1 assessed ECG changes and it found the accuracy of R-R interval measurement vs ECG data in dark-skinned individuals was significantly reduced (r = 0.98, p<.05).

The authors say the findings are concerning for a number of reasons, a key one being the increasing interest in using consumer wearable devices not only for medical research but for early detection of cardiovascular issues.

“There are a lot of claims that these devices can detect heart rhythm issues like tachycardia, bradycardia and even atrial fibrillation,” said Koerber in the ACC statement. “We want to be able to inform health care providers about whether these are reliable sources for collecting data in all patients, regardless of skin tone.”

Further, Koerber cites recent studies that have found reduced performance in individuals with darker skin tones with other devices, such as pulse oximeters.

“It is important to explore alternative options to make sure we can create a more equitable solution in health care and not just in the consumer industry,” Koerber said. For example, there is some evidence to suggest that certain wavelengths of light, particularly green light, are more accurate in people across all skin tones, he said.

Study limitations cited in the ACC statement include the relatively small number of published studies with relevant data and importantly, the variability in populations, devices, and outcomes assess across the final studies included.

Koerber will present the study, “The Effect of Skin Tone on Accuracy of Heart Rate Measurement in Wearable Devices: A Systematic Review,” on Sunday, April 3, at 11:45 a.m. ET / 15:45 UTC in Poster Hall, Hall C.