Heart rate variability: what your wearable measures

About the authorFiled by The BioHacking Awards editorial panel. Evaluations run against the published methodology. Corrections: contact page.

Filed March 17, 2023 · Article library · Methodology

Heart rate variability is the small change in time from one heartbeat to the next. Your heart does not tick like a clock. At 60 beats a minute, one gap might be 0.9 seconds and the next 1.15 seconds. A wearable records those gaps, turns them into one number, and shows it to you each day.

This guide explains what that number is, how a watch, band or ring gets it, and why two devices can show different values for the same person.

What the number actually is

The gap between two beats is called an RR interval. It is the time between two R peaks, the sharp spikes you see on an ECG trace. It is measured in milliseconds. Heart rate variability, or HRV, is a summary of how much those gaps differ.

There is more than one way to do the math. In 1996, a Task Force of the European Society of Cardiology and the North American Society of Pacing and Electrophysiology wrote the main standards. It named two common measures. SDNN is the standard deviation of all the normal beat gaps. RMSSD is the square root of the average of the squared differences between neighbouring gaps. SDNN reflects the whole recording. RMSSD reflects the quick, beat to beat changes. The Task Force said the two cannot replace each other, and that the method should match the purpose.

Sources: Task Force of the European Society of Cardiology and the North American Society of Pacing and Electrophysiology, "Heart rate variability: standards of measurement, physiological interpretation, and clinical use," European Heart Journal 17, 354-381 (1996), also published in Circulation 93, 1043-1065 (1996).

How a wearable gets its reading

A hospital ECG uses sticky electrodes on the skin. Most wearables do not. They use a small green or red light and a sensor that reads the pulse in the blood vessels, a method called photoplethysmography, or PPG. Each pulse wave stands in for one heartbeat. The device marks the time between waves and treats them like RR intervals.

This is an estimate. A pulse at the wrist or finger is not the same as the electrical signal in the heart. Movement, a loose fit, and cold fingers can all add noise. Because of that, devices usually measure while you are still, most often during sleep.

Sources: Kinnunen and Koskimaki, "The HRV of the Ring: Comparison of nocturnal HR and HRV between the OURA ring and ECG," Oura Health white paper (2018).

What two popular products say they do

WHOOP says it calculates HRV with RMSSD. Its guide, dated August 11, 2021, says the value is a dynamic average taken during sleep, weighted toward your last slow wave sleep stage of the night. The company says this gives a steadier baseline to follow.

Oura's ring sits on the finger and reads inter-beat intervals from the PPG signal. Its 2018 white paper says the ring records RMSSD for every five minutes of the night. A beat is used only if eight normal beats in a row were found around it. In a test of 10 healthy adults, the ring's nightly average RMSSD matched an ECG with a correlation of 0.984. The authors reported that about 90 to 95 percent of the five-minute blocks across a night had enough clean data.

That is a small test and the makers ran it. It shows the method can track an ECG under lab conditions. It does not promise the same match for every person or every night.

Sources: WHOOP, "Everything You Need to Know About Heart Rate Variability (HRV)," August 11, 2021 (whoop.com, Wayback Machine capture of October 28, 2021); Kinnunen and Koskimaki, Oura Health white paper (2018).

Why readings differ between devices

Several things can make two numbers disagree, even if both are honest.

Different math. One product may report RMSSD and another SDNN. Those are different quantities, so the values do not line up.
Different windows. The Task Force noted that it is not appropriate to compare SDNN from recordings of different lengths. A five-minute block and a full night are not the same test.
Different timing. A value from the deepest sleep of the night is not the same as one from a morning check or a daytime spot reading.
Different sensors and filters. Each maker decides which beats to keep and which to throw out as noise.

Oura's authors also point out that HRV is affected by age, hormones, lifestyle, heart rate and the time of day. They advise people to build their own baseline instead of leaning on a single reading.

Sources: Task Force (1996); Kinnunen and Koskimaki, Oura Health white paper (2018).

How to read your own number

WHOOP's guide shows that the middle half of its members aged 20 to 25 have averages of roughly 55 to 105 ms, while those aged 60 to 65 sit around 25 to 45 ms. The same guide stresses that these are wide ranges and that a friend's higher number does not mean much for you. It suggests following your own trend over weeks.

That is a sound habit for any device. Keep one device. Wear it the same way each night. Look at the direction over time, not a single morning. And treat the number as a data point from a consumer gadget. It is not a diagnosis, and this article makes no health claims about what any value means for you. If a reading worries you, a clinician with proper equipment is the right person to ask.

Sources: WHOOP, "Everything You Need to Know About Heart Rate Variability (HRV)," August 11, 2021.

Sources

Task Force of the European Society of Cardiology and the North American Society of Pacing and Electrophysiology. Heart rate variability: standards of measurement, physiological interpretation, and clinical use. European Heart Journal 17:354-381 and Circulation 93:1043-1065, 1996.
WHOOP. Everything You Need to Know About Heart Rate Variability (HRV). August 11, 2021.
Kinnunen H, Koskimaki H. The HRV of the Ring: Comparison of nocturnal HR and HRV between the OURA ring and ECG. Oura Health, 2018.