Paper Review: A New Interval Design To Improve Time At VO2max?
The Study
High-intensity decreasing interval training (HIDIT) increases time above 90% VO2peak
Quick Synopsis
This study looks at a new ‘progressively decreasing’ interval design for developing VO2max, which begins with longer efforts, and gradually reduces both the interval length and recovery interval as the session goes on.
The researchers looked at whether this new design improves time spent training close to VO2max, and/or the amount of work that can be done before reaching fatigue, when compared to sessions with exclusively short or long intervals.
The ‘progressively decreasing’ design is based on two important concepts:
First is the concept of so-called ‘VO2-on’ kinetics. When you abruptly increase your workload at the start of an interval, it takes some time for your aerobic system to ‘boot up’, and for oxygen consumption to rise to meet demand. This is particularly notable during the first interval in a session (see here for more discussion on VO2 kinetics).
Second is the concept of ‘critical power’ and W’ (pronounced ‘W-prime’). Critical power is similar to threshold power or FTP. It’s the power that you’ll tend towards when riding at a high intensity as you begin to fatigue. Above the critical power, fatigue hits much more rapidly than below the critical power. W’ is the total amount of work that can be done above the critical power before you reach exhaustion. You can think of this like a battery. In an interval session W’ is partially depleted with each interval, and then ‘reconstituted’ or ‘re-charged’ during a recovery interval. You can read more about critical power and W’ here.
The theory behind the ‘progressively decreasing’ design was that:
Having a long first interval would be useful in speeding up so-called ‘VO2-on’ kinetics, allowing participants to elevate oxygen consumption towards VO2max faster compared with short intervals like 30-sec hard, 20-sec easy.
Much like a battery, W’ recharges faster when it’s lower, and more slowly when it’s higher. Thus, early on in an interval session (when W’ hasn’t been depleted as much), it makes sense to implement a longer work interval (to deplete W’ more) and a longer recovery interval (to allow for the slower recharging rate). Later in an interval session, when W’ is more significantly depleted, shorter recoveries are needed in order to substantially ‘re-charge’ W’, and shorter work intervals are also prudent, because W’ is already low.
Methods
The study involved 12 ‘amateur’ cyclists who were training at least 3x per week. Average VO2peak was 57ml/kg/min (note - VO2peak is the maximum oxygen consumption reached by the participants in a ramp test at the point of exhaustion, but may not be their true VO2max).
At the start of the study, all participants underwent various testing sessions (on separate days) to determine their ‘VO2peak’, critical power and W’.
The participants then performed three interval sessions in a randomised order, and on separate days to allow adequate recovery:
Long intervals: 3-mins ‘hard’, and 2-mins ‘easy’ until the point of exhaustion.
Short intervals: 30-sec ‘hard’ and 20-sec ‘easy’ until the point of exhaustion.
Progressively decreasing intervals: interval length starting at 3-mins, and then decreasing to 2-min, 1-min, 45-sec, and finally 30-sec. Once 30-sec intervals were reached, these were repeated until the point of exhaustion. Like the long and short intervals, a 3:2 work:recovery interval was used, so for example, after the 3-min hard interval, there was a 2-min easy recovery period, after the 2-min interval, there was a 1-min 20-sec recovery, and so on.
In all sessions, the 'hard' intervals were performed at an intensity predicted by the critical power model to reach exhaustion after 5-mins, and this was individualised to each participant. On average, the intervals were performed at 117% critical power (which would be roughly 120% FTP).
The 'easy' intervals were performed at the reciprocal intensity about the critical power. So for example, if the 'hard' intervals were performed 17% above the critical power, the recovery was performed 17% below the critical power (i.e. 83% critical power, which would probably be around 87% FTP or thereabouts).
We’ve included an extract from the study which shows the format of the intervals:
During each interval session, oxygen consumption was measured, as well as perceived effort, heart rate and lactate.
Key Findings
The key finding of this study was that time spent above 90% VO2peak (which is widely considered to be a good marker for an effective VO2max development session), was improved with the progressively decreasing interval design.
The average time spent above 90% VO2peak for the different designs was:
Progressively decreasing intervals: 312 seconds (5 minutes 12 seconds)
Short intervals: 183 seconds (3 minutes, 3 seconds)
Long intervals: 179 seconds (2 minutes, 59 seconds)
Interestingly, despite this difference in time spent close to VO2max, there was no difference in time to exhaustion and total work performed, nor lactate levels and perceived effort at the end of the session.
Limitations
One slight concern with the results of this study is that they may have been skewed by two ‘outliers’ or strong-responders, as illustrated in the figure below, taken directly from the paper, with our annotations overlaid.
The straight lines show the time spent above 90% VO2max for each of the 12 individuals and the bars show the average time spent above 90% VO2max for the group as a whole.
We can see that there were two people who seemed to respond really well to the progressively decreasing design. However, for all other participants, the effect was more modest, and in fact there were 4/12 participants where the progressively decreasing interval design wasn’t the most favourable.
Another limitation is that the intensity used in the ‘recovery’ intervals was really quite high (roughly 83% critical power, which would translate to nearly 90% FTP). This was chosen in order to keep oxygen consumption high while still allowing some recovery and ‘re-charging’ of W’.
So while at first glance, this study appears to compare the progressively decreasing design to two fairly classic VO2max sessions, the abnormally high intensity used for the recoveries limits the comparability with these well-used interval designs.
Indeed, as the authors of the study acknowledge, other research, has shown short ‘microburst’ type intervals to be more favourable than long intervals in increasing time to exhaustion and time above 90% VO2max, which is in disagreement with the findings of the present study. The present study also used a 3:2 work:recovery ratio rather than the 2:1 ratio that’s normally used for short microburst intervals, which may also account for some of the differences seen.
Finally, it’s worth noting that this study only looks at time spent training close to VO2max, and doesn’t investigate which interval design actually results in the best improvements in VO2max. It’s often assumed that spending time near VO2max is a good way to develop this key fitness attribute. However, we also know that other types of training - such as long, low-intensity riding - are also great ways to build VO2max. So it doesn’t necessarily follow that more time at VO2max = better improvements in VO2max.
Our Take-Aways
1. Evidence is promising for progressively-decreasing efforts
The progressively decreasing interval design is one we’ve seen suggested on a few occasions previously. In particular, the concept of shortening your recovery interval to only as much time as is necessary to recover has been suggested when using a Moxy (or other near infrared spectroscopy) device, which can measure oxygen supply at the muscles. It’s good to see some scientific research backing up this training approach. More recently, we’ve seen the results of this study largely replicated among runners, giving us some extra confidence in this training approach.
Overall, it’s definitely an interesting interval design, based on good physiological theory, and is something we’ll be experimenting with over coming weeks. We’re also considering adding the session to our Complete Workout Library once we’ve done some of our own testing (so watch this space!).
2. How to structure session in practice
It’s interesting to note that the amount of time the participants were able to sustain the intervals before they reached exhaustion was surprisingly low for all protocols (roughly between 10-13 minutes). In practice, you’d probably look to include perhaps 2x blocks of intervals, with a longer period of recovery between, in order to allow more total training time spent close to VO2max overall.
A session we’ll be testing and finessing will look something like:
2x blocks of:
3M hard 2M easy
2M hard 1M20S easy
1M hard 40S easy
30S hard, 20S easy
Where the hard intervals are done at roughly 110-125% FTP or 5-min max power and the easy intervals are done around 80% FTP.
5-mins gentle recovery between blocks.
We’re proposing to skip the 45-second interval that preceded the 30-second interval in the present study. This is in order to simplify the design, and also slightly shorten each block of efforts so that people hopefully aren’t going completely to the point of exhaustion at the end of each block.
3. Individual Variability
As mentioned in relation to the study limitations, there was considerable variability in terms of how different athletes responded to the three types of session. This highlights the importance of including some variety in your training (although you should avoid including too much variety as this makes it hard to gauge progress and refine your pacing strategy!).
In addition, it’s worth noting that there was also wide variation in the percentage of critical power at which the efforts were paced. The hard intervals were pitched at each participant’s estimated 5-min maximal power. The standard deviation in workloads was 6%, which means that the majority of participants had a workload ranging between 111% to 123% of critical power. This is quite a broad range in and of itself. In addition, a minority (roughly 4-5/12 participants) will have had powers that were above or below this range.
It’s always really important to individualise your interval sessions to suit your own physiology and abilities. We have an article on methods you can use to do this here. A good strategy for this particular session would be to use data from a 5-min max effort to pace your work intervals.
Get Fast, Faster:
The Complete Workout Library
75+ proven, structured cycling workouts for road, MTB, CX/gravel and other major disciplines.
Drawn from science and honed through experience. Designed to improve key components of cycling fitness, including endurance, threshold power, VO2max etc.
