HRV measurements: context & time of the day

Blog post by Marco Altini.

This post is the first of a series on best practices for short heart rate variability (HRV) measurements typically used to monitor physiological stress by apps like HRV4Training. In future posts I will cover a few aspects: context & measurement time of the day, measurement duration, measurement type (lying, standing or orthostatic), paced breathing, and measurement frequency (how often?). Let's start with the measurement's time of the day.

Context  & time of the day

HRV is affected by many factors beyond the ones we are trying to measure. HRV4Training users are interested in the relation between training intensity and recovery, which can be captured as a reduction in parasympathetic activity after intense trainings. This relation between training intensity and recovery can therefore be determined by looking at drops in HRV features representative of parasympathetic activity, such as rMSSD or HF.

However, many other factors also impact HRV. For example, physical activity performed before (or during) the measurement tends to increase heart rate (HR), since the heart needs to pump more oxygen to the body's muscles, and together with increased HR, comes reduced HRV. Another example is mental stress, when under stress our body triggers the fight or flight response, which results again in higher HR and reduced HRV (a good discussion on this can be found on the mybasis blog, here). 

Let's look at some data. The first obvious situation we can look at, is physical exercise. Here we can see a plot of about 30 minutes including a few minutes before training, a run and a cool down phase. Clearly, while running HRV (regardless of what feature we look at) is reduced, as we can see from the almost constant RR intervals in the central phase of the recording:

However, consistent reductions in HRV can be seen even when physical exercise is almost completely absent. For example in the next figure I plotted data recorded while giving a presentation at a conference (simply standing), showing a similar pattern to the one we had before. Again, we have an initial phase (before going on stage), a central phase with reduced HRV (during the talk) and a "recovery" phase:

And finally another condition in which we have no physical activity, but where we have increased HRV / parasympathetic activity, meditation:

All of these measurements were taken using a Polar H7, the HRV Logger app, and on the same person. In the next figure we can see how the three different conditions compare across all time and frequency domain HRV features:

As expected, rMSSD and HF are very low during training or public speaking. The conditions just analyzed are rather "extreme" and therefore serve the purpose well, when it comes to understanding the importance of context, for both mental and physical stressors and their impact on HRV.

However, it should be noted that even much more common conditions such as eating (and digestion) or drinking, as well as mental stress (even at "lower levels" compared to public speaking) can cause changes in HRV that can make interpretation of our measurements more complex, if we are interested in monitoring recovery.

See for example this simple test I've done using Camera HRV, measuring HR and HRV before and after drinking coffee. As expected, HR increases while HRV (rMSSD in this case) decreases:

Even when we limit external stressors and physical activity, changes in HR and HRV naturally occur during the day due to the circadian rhythm, our "natural clock". 

So how do we get to measurements that we can interpret, even though HRV can be influenced by so many different factors?

The morning measurement

We need to isolate context, providing stable and repeatable conditions where the influence of all external factors is kept to the minimum. In the medical world, where many of these measurements originated, the influence of external factors is limited by performing standard protocols, typically in the morning.

This is the case in many situations, for example measuring your weight before breakfast, measuring blood pressure in standard conditions (sitting, arm position, etc), and similarly, measuring HRV to determine the impact of trainings on recovery and physiological stress.

In many studies, the protocol for HRV measurement consisted in morning measurements while lying down, sometimes after an initial "adaptation phase" that is afterwards discarded. 

For these reasons the best moment to take your HRV measurement should be right after waking up, possibly while still in bed, and before reading your email or start thinking about work or other aspects of your life that might cause additional stress or anxiety. 

Alternatively, a routine that involves a few more steps, e.g. go to the bathroom, empty the bladder, stand or lie down again and take the measurement, also works, as long as it is as consistent as possible during different days. Another option are night long measurements, which are however still rather unpractical, but follow a similar principle, i.e. isolate context and measure under the same, stable, repeatable conditions on a daily basis.

A few more tips on the ideal morning routine and measurement procedure can be found in two podcasts recently published on thequantifiedbody.net by Andrew Flatt.