# Validity of Heart Rate Measurement Using Wearable Devices During Cardiopulmonary Exercise Testing in Patients With Cardiovascular Disease: Prospective Pilot Validation Study

## TL;DR

This study tested how accurately the Fitbit Inspire 3 measures heart rate during exercise in patients with heart disease and found it to be mostly reliable, though less so during intense activity.

## Contribution

The study provides new evidence on the accuracy of Fitbit Inspire 3 for heart rate monitoring in cardiovascular disease patients during exercise.

## Key findings

- Fitbit Inspire 3 showed strong correlation (r=0.90) with electrocardiography for heart rate measurements.
- Mean absolute percentage error was 5.40%, indicating acceptable accuracy overall.
- Underestimation of heart rate was more common during high-intensity exercise and in patients with heart failure.

## Abstract

Wearable devices offer a promising solution for remotely monitoring heart rate (HR) during home-based cardiac rehabilitation. However, evidence regarding their accuracy across varying exercise intensities and patient profiles remains limited, particularly in populations with cardiovascular disease (CVD) such as those with heart failure (HF).

The objective of this study was to evaluate the accuracy of HR measurements obtained using the Fitbit Inspire 3 during cardiopulmonary exercise testing (CPX) in patients with CVD, including those with HF.

In this single-center, prospective pilot study, we enrolled 30 patients with CVD undergoing CPX. HR was simultaneously recorded using electrocardiography and the Fitbit Inspire 3 at 1-minute intervals across various CPX phases: rest, exercise below and above the anaerobic threshold (AT), and recovery. The correlation between the two methods was assessed using the Pearson correlation coefficient. Measurement error was quantified by mean absolute error and mean absolute percentage error (MAPE), with a MAPE of ≤10% defined as the threshold for acceptable agreement.

All data points were 630 points per minute. The Fitbit Inspire 3 device demonstrated a strong overall correlation with electrocardiography-derived HR (r=0.90; IQR 0.88‐0.91) and an acceptable MAPE of 5.40% (SD 8.33%). The total error was 14.9% (94/630), with overestimation and underestimation of 37 (5.8%) points and 57 (9%) points, respectively. The rate of HR underestimation reached 19 (16%) points during exercise above the AT, compared to 1 (3%) point at rest. When stratified by HF stage (B vs C), underestimation was more pronounced in patients with HF (14/275, 5% points vs 40/355, 11.2% points).

The Fitbit Inspire 3 provides acceptable validity for HR monitoring during CPX in patients with CVD. However, clinicians should interpret HR data with caution during high-intensity exercise, especially in patients with HF.