Here’s How To Interpret Heart Rate Variability Data (And How Not To)

How to interpret HRV data, and how to use it when training

Photo of author
Written by
Last Updated:

In Parts I and II of this series, we saw how heart rate variability (HRV) can indicate the stress response and how we can measure it. We have also discussed protocols, best practices for measurement, and technology options.

Now, let’s finally talk about the data.

How do we interpret it? And how do we use it in the context of training?

Keep reading to find out!

A heart rate monitor.

Absolute Vs. Relative Changes In Heart Rate Variability Data

When we measure something, it is normal to compare it to others or ask ourselves, is this a good number? However, when it comes to HRV, we need to understand that absolute values tend to have little meaning.

While it is true that there are associations between very low HRV and poor health outcomes, it is important to remember that these are just associations, and a low HRV does not necessarily lead to poor outcomes.

This is why I believe it is far more useful to look at relative changes in HRV with respect to our own historical data, as opposed to absolute HRV values.

This is not to say that our absolute HRV cannot improve over time. It is possible for HRV to improve, especially if our lifestyle can be improved too (e.g., being a bit more active, eating a healthier diet, or managing large stressors in our lives better, etc.).

However, in the long term, many factors impact our physiology, from seasons (e.g., normally, HRV is lower in the winter months), to aging (as heart rate variability reduces with aging) and other factors.

Thus, in practical terms, I recommend focusing on relative changes so that we can make meaningful adjustments on a day-to-day basis, not necessarily to improve HRV itself but to improve our health and performance.

In other words, we use heart rate variability data to improve our outcomes of interest, and our outcomes of interest should be health and performance, not necessarily HRV itself.

Here, I have used the word improve, instead of increase, when talking about HRV. We’ll see in a minute why, but for now, keep in mind that HRV can improve even without increasing, and this has to do with its stability. 

A heart rate monitor on someone's finger.

Your Normal HRV Range

In the previous section, I mentioned how the most meaningful way to use and interpret heart rate variability data is to analyze relative changes with respect to our historical data. How do we do that? Well, we need a frame of reference.

Unfortunately, HRV is highly individual, so we don’t have a generic frame of reference the way we do for some other variables. Thus, the best way to build a frame of reference is to collect our own data.

Hence, we use our historical data to establish our own frame of reference, or what we often call our normal range or normal values.

This normal range is where we expect the data to be on a given day if we do not have large stressors or poor responses, causing a deviation outside our normal range.

Below is an example of my own data with a few annotated stressors collected using HRV4Training.

You can see how various stressors, such as moving to a new place and then traveling for work to give talks at various conferences, have caused dips in HRV (a low HRV is a sign of higher stress).

The dips in HRV ended up below my normal range, highlighting that this was a meaningful change in the data. 

Here's How To Interpret Heart Rate Variability Data (And How Not To) 1

The normal range is key because there is always variation in physiology: our HRV today will most likely be a bit lower or a bit higher than yesterday, and it is important to understand that most of these day-to-day variations should be of no concern.

When our data is within our normal range, we have a positive response, regardless of having a slightly lower or higher value than a different day.

If we use a tool that does not provide a normal range, it can be hard to understand if a daily value should prompt a change as we have no way to know if the value represents just normal day-to-day variability or if it represents a meaningful deviation from our normal. 

Finally, given what was just discussed, we can associate a suppression in HRV below our normal range with a negative response to stress. It could be training stress or other stressors, as in my screenshot, in which work-related stress had a strong impact on my physiology.

What about a positive response?

A positive response is normally highlighted by a more stable HRV. When things are going well, we don’t necessarily expect HRV to increase, but we simply expect to have fewer suppressions and to see heart rate variability data within our normal range.

Stability is typically a good sign when it comes to HRV.

People analyzing data.

How Do We Use Heart rate variability Data?

So far, we have learned that we should look at relative changes in HRV over time with respect to our own historical data.

We have also learned that we can use our data to build a frame of reference called our normal range, which allows us to determine if a daily value is part of normal day-to-day variability or if it highlights a more meaningful change, for example, a suppression in HRV below our normal range.

In this latter case, implementing some changes might be a good idea. But what changes are we talking about here? 

We can think of stress and recovery from two angles.

First, we can try to reduce the stressors, i.e., the sources of stress likely causing issues. This can be easy if the stressor is training, as we can simply manipulate training and make some adjustments to our training plan (we’ll see in a minute exactly which adjustments are probably more important).

However, in the case of non-training-related stressors, it might be hard to adjust them at times. For example, if we are traveling for work, have a stressful week at the office, or some other issue at home, we might not be able to reduce our stressors.

In these cases, it can be helpful to look at the problem from a different angle: if we cannot reduce the stressors, maybe we can try to prioritize recovery and restful activities.

This could be as simple as trying to eat better despite being on the move, trying to get some extra sleep, or perhaps taking a walk in nature.

Depending on our lifestyle, habits, and interests, we can always try to think about stress and recovery from these two points of view: either we try to reduce the stressors or give ourselves an extra opportunity to recover from them.

A runner looking at their watch.

Adjusting Training

Let’s get to training stress now. As endurance athletes, this is typically the one aspect we can control quite well. How should we modulate our training when we have a negative response (for example, HRV suppressed below our normal range)? 

Keep in mind that HRV is a generic marker of stress. This means that all stressors impact HRV, not only training, as we can see from my work-related stress example.

Our capacity to handle stress is limited, as there is only so much we can take. This is why heart rate variability can be useful in adjusting training regardless of what caused the suppression in HRV.

When we are in a highly stressed state, we are likely unable to respond positively to a certain training stimulus, and we might be better off with a different course of action.

In particular, the adjustment you make will depend on a number of things, for example, your training history and athletic level. However, we can derive some general points from published scientific literature on heart rate variability data-guided training.

For example, we know that low-intensity training (e.g., training done below the first ventilatory threshold, the intensity that allows us to chat without catching our breath) typically does not impact HRV.

A person taking their heart rate.

This means that our HRV tends to renormalize very quickly after a low-intensity session.

On the other hand, training performed at higher intensities, for example, a tempo run at moderate intensity or high-intensity intervals, will cause a long-lasting disruption in HRV.

The fitter we are, the shorter the suppression, but we still have a disruption in autonomic activity after training of intensities higher than the first ventilatory threshold.

Research from Stephen Seiler has also shown that if we keep the intensity low, training for more time (e.g., 2 hours instead of 1 hour) does not seem to impact our autonomous nervous system negatively, as HRV remains stable. 

These studies tell us that adjusting intensity more than volume seems preferable when we have a suppression in HRV.

In the past decade, more studies have tested this theory, splitting runners or cyclists into two groups, one following a certain training plan and another one following a similar plan but adjusting the plan based on HRV.

In most occasions, the HRV-guided group ended up performing fewer higher-intensity sessions during the study, as, at times, when HRV was suppressed below an individual’s normal range, intensity would be reduced.

However, the HRV-guided group consistently performed better or showed better physiological adaptations. This should be no surprise as we intuitively understand that not only the type of stressor but also the timing of the stressor matters.

Important Training takeaways

HRV is a stress marker that we can use to capture how we respond to training and lifestyle stressors.

Once we have collected some data, we can build our normal range and understand if, on a given day, our score is showing a positive response (e.g., an HRV within our normal range) or a negative one (e.g., a suppression below our normal range).

On occasions in which our HRV is below our normal range, it might be a good idea to make small adjustments to our training plan, for example, reducing intensity, but not necessarily the volume (or duration) of our session.

A heart.

 

Photo of author
Marco holds a PhD cum laude in applied machine learning, a M.Sc. cum laude in computer science engineering, and a M.Sc. cum laude in human movement sciences and high-performance coaching. He has published more than 50 papers and patents at the intersection between physiology, health, technology and human performance. He is the co-founder of HRV4Training, advisor at Oura, guest lecturer at VU Amsterdam, and editor of the Wearables department of IEEE Pervasive Computing. He loves running.

Leave a Comment

This site uses Akismet to reduce spam. Learn how your comment data is processed.