Heart rate variability explained: what the number measures and what moves it

A heart beating at 60 beats a minute does not beat once every 1,000 milliseconds. One gap might be 980 ms, the next 1,040, the next 1,010. Heart rate variability (HRV) is the size of that variation. It sounds like noise. It is actually a signal of how the autonomic nervous system, the part that runs heart rate without conscious control, is balancing its two branches.

What HRV reflects

The vagus nerve, part of the parasympathetic branch, slows the heart and can change the gap between beats within a single breath. The sympathetic branch speeds it up and acts more slowly. When the vagal brake is strong, as it tends to be in rested, fit people during sleep, the beat-to-beat intervals vary more. When the body is under strain from hard training, illness, alcohol, poor sleep or psychological stress, vagal activity usually drops and HRV falls with it.

That is why wearables measure it at night or first thing in the morning. Movement, posture, talking and digestion all change heart rate during the day. Sleep is the closest thing to a standard measuring condition most people get.

RMSSD and SDNN: the two numbers you will see

There are dozens of HRV metrics. Two show up on consumer devices and in most papers.

The practical point: an RMSSD number and an SDNN number are not interchangeable. A 45 ms reading on one device and a 60 ms reading on another may describe the same night. Shaffer and Ginsberg's 2017 review notes that RMSSD is more influenced by the parasympathetic system than SDNN and is largely unaffected by slow trends in a long recording, which is one reason it suits overnight measurement. The conventional minimum recording is five minutes, although researchers have tested windows as short as 10 to 60 seconds.

Published values

The most cited reference set for short-term HRV is a 2010 systematic review by Nunan and colleagues. They pooled 44 studies with 21,438 healthy adults, using recordings of about five minutes taken at rest.

Those are daytime laboratory values from ECG, not overnight readings from a ring. Nocturnal RMSSD on a wearable is often higher, because vagal activity peaks during sleep. Age also matters a great deal, which is covered on the what is a good HRV page. Nunan's team also flagged very large differences between individuals, which is why any single population number is a weak yardstick for one person.

What moves HRV

The research on what shifts HRV is consistent on direction even when the size of the effect varies.

• Age. HRV falls through adulthood. In data from 8,203,261 Fitbit users, high-frequency power (a vagal measure) fell by about 82 percent in men and 81 percent in women between the ages of 20 and 60.
• Physical activity. The same Fitbit analysis found that more daily activity was associated with better HRV across age and sex in a dose-dependent way. It was cross-sectional, so it shows association, not cause.
• Alcohol. In 4,098 Finnish employees wearing beat-to-beat monitors, an HRV-derived recovery measure in the first three hours of sleep fell by 9.3, 24.0 and 39.2 percentage points after low, moderate and high alcohol intake compared with the same people's nights without alcohol.
• Training load. Hard blocks of training lower HRV for a day or more; adaptation tends to bring it back up. Case studies in elite triathletes found a falling 7-day average of RMSSD in the run-up to non-functional overreaching.
• Illness. Infections commonly lower HRV and raise resting heart rate and respiratory rate, sometimes before symptoms. See respiratory rate.
• Body weight. A 2019 meta-analysis of diet and exercise studies found weight loss tended to shift autonomic balance towards parasympathetic activity, although the evidence was rated low quality.
• Some medicines. GLP-1 receptor agonists raise resting heart rate, and one crossover trial found liraglutide lowered HRV. See wearable data on a GLP-1.

How wearables measure it

Most rings and wrist devices use photoplethysmography (PPG): a green or infrared light reads pulses of blood under the skin and estimates the time between beats. An ECG reads the electrical signal directly and is the reference. PPG works well when you are still and the signal is clean, which is another reason devices report overnight values. Each company also chooses its own window. Oura says it takes five-minute samples through the night and reports their average. WHOOP says it records HRV only during sleep and calculates it as RMSSD. Garmin's HRV status uses readings taken while you sleep and compares a seven-day average with your baseline. The wearable accuracy page covers how close each gets to ECG.

What HRV is not

Lower HRV is associated with higher cardiovascular risk in population studies, but a single low night is not a diagnosis, and a high number is not a clean bill of health. Some heart rhythm problems produce irregular intervals that inflate HRV. If a device reports unusually high and erratic HRV with palpitations, that is a reason to see a clinician, not a reason to celebrate.

Reading your own number

The useful comparison is you against your own baseline, from the same device, measured the same way. Look at the 7-day average, not last night. The reading wearable trends page explains baselines and noise.

Logging it

Peptide AI pulls overnight HRV from Oura, WHOOP and Garmin directly, and from Apple Health or Health Connect, through wearable integrations. It sits on the same timeline as the dose log, the training log and lab results, so a dip in HRV can be read against what changed that week.