TOOLBOX: We have talked about using heart rate variability to guide your training program. Today, let’s talk about the use of HRV during your training with one of the newest tools in sports science now: detrended fluctuation analysis alpha 1 (DFA-alpha1)!
Resting heart rate
In the last couple of Pez Toolbox articles, I have discussed the concept of Heart Rate Variability during rest, and how it can be applied to guide your training program – identifying when you are primed and ready to go, as well as days that your body is not optimally rested/recovered for High Intensity.
Although cycling is comprised of a very wide range of efforts, from high-intensity Zwift races to long & easy weekend café rides, most of our training can be boiled down into three separate training ‘zones’: low, moderate, and high intensity. The moderate and high intensity zones are separated by the anaerobic threshold, which can be measured in a lab from blood lactate concentrations or respiratory data. However, it is more commonly approximated through Functional Threshold Power (FTP).
Perhaps slightly lesser-known, but equally important is the aerobic threshold – the boundary between low and moderate intensity zones. Dr. Stephen Seiler and other researchers have done considerable research on the distribution of training in elite athletes and their performances and has shown that performing 70-80% of training volume in a low intensity zone improves performance outcome of endurance athletes. Therefore, correct identification of the upper limit of the low intensity zone is important to endurance athletes.
Just like the anaerobic threshold, the aerobic threshold can be estimated using blood lactate or respiratory data in a research laboratory. However, who has access to a research lab during a global pandemic? Furthermore, the transition from easy to moderate intensity does not occur at fixed percentages of max HR, VO2max, or maximum aerobic power for all athletes, so applying a universal estimation might not be appropriate to all athletes.
What is DFA-alpha1? How is it measured?
Recently, researchers have gained greater insight into how our body responds to exercise through non-linear dynamics of Heart Rate Variability data. In particular, the non-linear detrended fluctuation analysis (DFA) has recently become a viable, non-invasive measure of the aerobic threshold. These are calculated by looking at long term patterns in the beat-to-beat HR data, where the R-R intervals of a given time interval are partially dependent on the previous segment. More specifically, DFA’s short-term scaling exponent alpha1 can be used for exercise because only a relatively short window (~2 min) is needed, and it can be calculated during non-steady state exercise.
Without getting too deep into the analysis, a DFA-alpha1 coefficient value of ~1.0 at rest indicates normalcy, while deviation from that value indicates either increased randomness which occurs during moderate/high intensity training (DFA-alpha1 < 0.75), or increased correlation, which occurs during low intensity exercise (DFA-alpha1 > 0.75).
In simple terms, DFA-alpha1 of 0.75 represents the aerobic threshold. Values above 0.75 indicate low intensity exercise, while values below 0.75 indicate moderate to high-intensity exercise.
You don’t need a research lab
What do you need?
There are a few hardware requirements to incorporate this into your training. However, these are relatively inexpensive, especially compared to the cost of heading into a research lab. Like measuring HRV during rest, you will need to use a chest strap HRM – unfortunately optical HRM’s cannot be used. Secondly, you will need software to analyze the raw HRV data from your HR strap. Currently, the only app able to display DFA-alpha1 in real time is the HRVLogger (available on both iOS & Android). The general recommendations are to use a high-quality HRM (such as Polar H9/H10 or Garmin Dual HRM) strap, connected via Bluetooth (avoid ANT+, if possible).
Practical Application of DFA-a1
Adding in DFA-alpha1 to your low intensity rides is relatively easy (especially indoors on a trainer). Determining your aerobic threshold using either power (Watts) or HR (bpm) lets you set the upper limit for your low intensity/recovery rides. This sort of test will be different than an FTP or ramp test since it will not be a maximal effort.
To do this, you will want to complete a few 4-6 min intervals from approximately 60-80% of FTP and monitor/record the intensity (in Watts or bpm) where DFA-alpha1 reaches 0.75. For example, for an FTP of 270 W, you would consider a ramp of 6 min each from 60-80% of FTP, with each ramp step increasing about 5% in intensity. This would then include a warmup (self-paced), followed by 6 min at each 162W, 176W, 190 W, 203 W, & 216W.
I recently performed a very similar ramp test with DFA-a1 and returned a value of 0.75 at approximately 200 W. With that information, I know to keep my power below 200 W on my easy rides.
When performing a low-intensity session indoors, I use the HRVLogger to monitor my DFA-alpha1. To begin with, I open the HRVLogger app and pair my Polar H10 (detailed instructions can be found in the HRV FAQ. Once that is done, I start my cycling training app and start to free ride – I generally like to avoid doing structured workouts when monitoring DFA-alpha1. After starting my ride and starting a recording in the HRVLogger, I use the Real-Time view on the app and closely monitor the live DFAa1 data, which is updated every 2 min. I can then free ride at my own, self-selected intensity and ensure that my easy rides are truly low intensity by keeping DFA-a1 above 0.75.
We have seen how HRV can now be used during an activity to monitor the demands of that exercise on our body. During low intensity exercise, DFA-alpha1 will remain above 0.75. As exercise intensity increases, DFA-alpha1 will drop below 0.75. This offers a relatively inexpensive and non-invasive method to monitor low intensity exercise.
As always, I hope you learned something – ride safe and stay healthy!
Interested? Learn More!
Both Bruce Rogers and Marco Altini are very responsive on Twitter – posting interesting content and responding to questions from curious individuals like me!
de Godoy, M. F. (2016). Nonlinear analysis of heart rate variability: a comprehensive review. J. Cardiol. Ther. 3, 528–533. doi: 10.17554/j.issn.2309-6861.2016.03.101-4
Gronwald T, Rogers B, Hoos O. (2020). Fractal Correlation Properties of Heart Rate Variability: A New Biomarker for Intensity Distribution in Endurance Exercise and Training Prescription? Front Physiol. 11:550572. doi: 10.3389/fphys.2020.550572. PMID: 33071812; PMCID: PMC7531235.