Study claims heart rate monitor Fitbit wearables have 20bpm deviation

Research would show that the heart rate measurement of the Fitbit Surge and Charge HR deviates by an average of twenty heartbeats per minute. For the study, the results of the fitness trackers were compared with an electrocardiogram. Fitbit contradicts the claims.

The researchers provided 43 participants with the two Fitbit fitness trackers and performed an EKG at the same time. Heart rate was then measured at rest and while performing various training conditions such as running, cycling and climbing stairs. According to the researchers, the heart rate measurement is unreliable, especially during medium to high intensity exercises. The difference with the fitness trackers and the professional ECG equipment is about 20 bpm on average.

Deviations in the reported heart rate can be potentially dangerous, because users, for example, train with a heart rate that is too high, without realizing it. The research doesn’t pinpoint the cause of the anomalies, but suggests it could be related to Fitbit’s algorithms.

Fitbit uses optical sensors for heart rate measurement for its PurePuls technology and combines the measurements with algorithms. The company claims that its technique is suitable for determining the intensity of a workout on the basis of the heart rate.

The study was conducted by researchers at California State Polytechnic University, on behalf of law firm Lieff Cabraser. That office is pursuing a class action lawsuit against Fitbit for allegedly inaccurate heart rate readings.

The results are consistent with a previous independent study, which was conducted on a smaller scale. In that test, a Fitbit Charge HR tracker reported a heart rate of 68, while a pulse oximeter recorded a heart rate of 91.

Update 16.30: Fitbit contradicted the research in a response, published by Gizmodo. According to the maker of the wearables, the research was not carried out correctly and was biased, because it was commissioned by the law firm. Fitbit states that the study was conducted with a ‘consumer grade’ ECG and not with a medical device.