Session rating of perceived exertion as an efficient tool for individualized resistance training progression

Original article written by Rodrigo Gomes, Manoel Lixandrão, Carlos Ugrinowitsch, Alexandre Moreira and colleagues.

Background

Progression during resistance training is essential to help drive adaptation. However, prescription and progression during resistance training is often strangely implemented. For example, training is traditionally prescribed with a set (e.g., 3 sets) and repetition (e.g., 8 repetitions) structure that requires a predetermined weight to be used (e.g., 80% of 1 repetition maximum (1RM)). Furthermore, progression or regression in loads and volumes tend to occur at the end of a training cycle once a new round of maximal testing occurs. The issue with this is that that this does not account for changes in strength that occur during a training cycle, nor does it consider the fatigue that athletes experience throughout a training cycle. For this reason, practitioners occasionally use autoregulatory methods of prescription that help account for these ‘ups and downs’ throughout a mesocycle.

The use of rating of perceived exertion (RPE) is a common method of monitoring, and occasionally regulating, training. The benefit of using RPE is that it is in arbitrary units and provides a global number that can indicate the intensity of exercise across different forms of training (e.g., field and gym-based training). This may be particularly valuable for resistance training as we tend to be rather poor in quantifying volumes and intensities in this form of exercise. It should be noted that RPE can be used following every set to guide loads and volumes (refer to great work by Helms et al.^[1,2] for further information regarding this type of use) or after every session as a monitoring tool. Furthermore, due to its relatively simple application, it can be used by practitioners to guide whether individuals are finding sessions easy/hard and whether progression/regression is required ^[3].

What the authors did

In this study by Gomes et al.^[4], rating of perceived exertion was used to guide progression of resistance training. Specifically, they compared the effects of rating of perceived exertion and traditional fixed progression (i.e., sessions were prescribed at the start of the mesocycle and they were progressively completed) on 1RM strength and changes in muscle cross sectional area (CSA) across a six week period.

What the authors found

This manuscript suggested, albeit with a bit of sub-group analysis, that it is plausible to use this method of prescription for progressing training. Additionally, changes in strength and muscle CSA are similar after six weeks of training. They also found that training participants with repetition maximum prescription and high numbers of repetitions per sets (e.g., 12-15RM as they did in ‘protocol A’) is extremely hard. Six out of 10 participants from the RPE group did not manage to progress beyond the first training protocol (i.e., they did not manage to even do protocol ‘B’ or ‘C’) which suggests additional piloting of the study would have been prudent. Furthermore, due to the need to do sub-group analysis (which seemed a little strange as it wasn’t suggested in the aims…), they found that the participants that did not progress from the 12-15RM training protocol may have had a small increase in muscle size but the lack of statistical power and the potential noise of this measure (coupled with the potential small changes) seemed to really limit the conclusions.

Limitation

I’ll admit, I love looking at novel methods of autoregulatory prescription. Therefore, seeing this manuscript in this month’s issue of Journal of Strength and Conditioning Research, I was looking forward to having a read. However, after reading, I was a bit underwhelmed. It felt like the authors had a great idea but perhaps hadn’t fine-tuned some of the finer points of the methodology. Furthermore, it felt like they tried to make the best of a bad situation by trying to do a sub-group analysis (which probably wasn’t originally planned) and then suffered from the lack of statistical power.

One of the biggest problems with this study was the way that they prescribed and then planned to progress the load. There were three different protocols (Protocol A, B, and C) and each had a different number of repetitions and sets (Refer to Table 1 from the manuscript). For the RPE group, they then progressed training by using the below method:

“All subjects started with the same exercise session (i.e., protocol A; 2 sets of 12–15RM). For the next training session, exercise workload was adjusted individually according to its respective RPE response to the last training session. If the RPE values were ≤5 (i.e., moderate), the workload was progressed for the next training session (e.g., protocol A→B; protocol B→C). If the values of RPE were ≥6 and ≤8 (i.e., hard), the workload was kept the same for the next training session. If a subject rated the session ≥9 (even if in protocol A) for 2 consecutive sessions, we reduced one exercise set for the next raining session, and so on up to the minimum exercise volume (i.e., one set per exercise).”

But this causes a serious issue with progression (and something that the authors soon found out!). If you have one group progressing or regressing based off how hard a session was and you’re asking them to do repetition MAXIMUM training, then you’re going to have a lot of participants stating that the session was maximal! Furthermore, asking participants to complete 12-15 repetitions to failure (even if it is just for two sets) with 120 seconds break would have been brutal. Therefore, six out of the 10 participants in the RPE group did not progress past the first protocol (i.e., they spent six weeks doing the same session and rating it maximally every time!). Such a serious flaw in the methods likely caused the researchers to scratch their heads and ask how they could make the most out of the data that they had. This then likely lead to the sub-group analysis which used only the six participants who did not progress (which was a bit ironic that the participants that were focussed on in the end were participants that did not progress their training) and this likely caused substantially underpowered analysis.

Moving on from the methodological issues, one final limitation would have to be the lack of outcome measures. Unfortunately, only squat 1RM (in a Smith machine) and CSA were the only adaptive outcomes. It would have been nice to see a higher velocity-based outcome (e.g., CMJ) or performance (e.g., sprint) outcome included. Additionally, it would have been good to see the mean RPE values for each group across the training period and looked at additional considerations such as fatigue/recovery and enjoyment of the different methods of training (although my suspicion is that after six weeks, the participants that did not progress would have been unhappy).

What this means for coaches

Honestly, the practical limitations from this study should be taken with a pinch of salt for this one. Indeed, it does demonstrate that it is possible to progress or regress training based off RPE in this unique way (although I’m a little less optimistic if it was done with RM zones). Additionally, it demonstrates that coaches can see adaptations in strength and CSA with different training prescriptive methods (although some participants who progressed in the RPE group did not see changes in CSA which was attributed to their poor understanding of the rating anchors…). However, that’s about where the direct practical takeaways end and the considerations around the appropriateness of autoregulatory practices begin. If you are going to programme autoregulatory methods, it is essential to consider whether the progression system works and makes sense. When fatigue is high, asking participants to complete high-volume training is inappropriate. Alternatively, constantly requiring athletes to train with RM zones causes large amounts of fatigue (i.e., repetition maximum is still a maximum irrespective when you fail).

Reviewer’s comments

Overall, this one has been a real head scratcher for me. At first, I want to commend the authors for going to the effort of completing a training study. They are exceptionally difficult to do and are extremely valuable to the industry. But, when I read a little deeper into the methods, I begin to wonder why would a team of researchers go to such a large effort to train individuals and only have two main outcome measures? Furthermore, how did such an important methodological consideration sneak through the conceptualisation phase when planning the project? Alas, I don’t know if I’ll ever get the answers to this one. In the meantime, I challenge anyone to do two sets of 12-15RM with 120 seconds rest in the back squat and tell me that it was less than 9 out of 10 on the RPE scale.

Recommended resources

Book – Periodization: Theory and methodology of training – Tudor Bompa & Carlo Buzzichelli