[Oldmanmatt]

http://jap.physiology.org/content/jap/113/1/71.full.pdf

Having just read through this (only once, so far), I note the authors conclude that there is parity in strength gains (as well as hypertrophy) when applying a ~30% 1RM program or a ~80% 1RM; where those programs are multi set and "to failure" in nature.

However, when I read the results (as evidenced in fig 3, for instance), there seems to be significantly greater strength gains from the 80% 1RM program, than the 30%?
I note an improvement in torque, though that seems a smaller increase than the increase in 1RM induced by the 80% program.

Otherwise, if I'm missing something (and given my whole 20 minutes of intense study (not) this is entirely possible), does this not contradict, substantially, Tidow et al?


All posts either sarcastic, tongue-in-cheek or mildly mocking-in-a-friendly-way unless otherwise stated. Looking at you, here, Dense. 

[BicepsMou]

Couple of comments w.r.t study design and outcomes:

There are 3 test protocols involved in the study:
1)   1 set to failure at 80% of 1RM
2)   3 sets to failure at 80% of 1RM
3)   3 sets to failure at 30% of 1RM

So the study tries to look at 2 effects on strength (and hypertrophy):
a)   Effect of number of sets – 1 vs 3 sets
b)   Effect of intensity level in % of 1RM – 30% vs 80% of 1RM

Results (w.r.t max strength gains – I’m not interested in hypertrophy):
•   The study does confirm a difference in isotonic (as opposed to isometric) strength gains between the 80% and the 30% protocols:
<<Isotonic maximal strength gains were not different between 80%-1 and 80%-3, but were greater than 30%-3 (P   0.04)>>
•   As the training protocol was not an isometric one but a concentric/eccentric protocol (knee extensions), it is to be expected from a training specificity standpoint (=neural movement adaption) to see more adaptions in dynamic strength as opposed to static strength, so no surprise here
•   There was no (significant) difference btw. the 3 sets @80% and the single set @80%

Comments on study design:

The 18 young men (with their 36 test legs) selected for the study where assigned to 2 protocols each (1 leg per protocol).

1 - So we have n=12 individuals (à 1 leg) for each protocol; a fairly low number, which explains why some of the differences in outcome that may be observed do not pass the (artificial) statistical criteria of significance at a set P-level. In other words: there could be common sense evidence of a difference in outcome, even if the statistical criteria are not sufficient to back this up from a purist statistical standpoint.

2 – The test subjects are untrained (no former lifting activity over the last year); as we know that with untrained subjects, nearly ANY protocol results in good strength gains, so it is to be expected that also the 30%-protocol yields strength gains. But as we also know, the intensity threshold needed to yield similar strength adaptions with ongoing training status and history continuously increases, so this result cannot be applied to trained individuals. In other words: if you already have a strength training history as a climber (as most of us here may have), I would not believe this to be true for the training community on UKB.

3 – The application of 2 different test protocols for both legs of one test subject is also interesting: There is a known transfer from 1-sided training towards the other, untrained limb. This effect is often deliberately used when preventing strength losses during injury periods. E.g. training the healthy fingers 1-sided prevents (to a certain degree) strength loss in the injured (and thus immobilized) hand / fingers. So we could put up the hypothesis that there may be a strength transfer from the e.g. leg trained @80% to the leg trained @ 30% that biases the protocol comparisons (just realise  that they acknowledge this in the discussion part of the paper)

Ups... quite long post. Got carried away 

[Nibile]

Interesting comments.
Not factoring nutrition in, is a big Bias in the study in my opinion.

[roddersm]

•   There was no (significant) difference btw. the 3 sets @80% and the single set @80%

Sorry just skimmed your post but are you saying 1 set is as good as multiple when training strength? If so that would be handy

[finbarrr]

2 – The test subjects are untrained (no former lifting activity over the last year); as we know that with untrained subjects, nearly ANY protocol results in good strength gains, so it is to be expected that also the 30%-protocol yields strength gains. But as we also know, the intensity threshold needed to yield similar strength adaptions with ongoing training status and history continuously increases, so this result cannot be applied to trained individuals. In other words: if you already have a strength training history as a climber (as most of us here may have), I would not believe this to be true for the training community on UKB.

that^
when you start climbing , anything will make you better and stronger. after that every plateau may need a different solution

[duncan]

•   There was no (significant) difference btw. the 3 sets @80% and the single set @80%

Sorry just skimmed your post but are you saying 1 set is as good as multiple when training strength? If so that would be handy

That's what we would all like to hear but BicepsMou nails it. The findings are not applicable to ukb (a trained population). The findings are likely to be a "Type II" error: no statistically significant difference seen between the groups because there are not enough participants. Crucially, there is no sample-size calculation - the number of participants needed to give a good chance of detecting a difference if there really is a difference.

Lore, it would be hard to control participant's nutrition over the whole period of the study but they all had the same post-exercise snack and fluid. This is another reason for needing more than 18 participants, the known and unknown biases become less influential on findings as the effect of the odd individual who (for example) has a weird diet are diluted by the bigger group.

[BicepsMou]

•   There was no (significant) difference btw. the 3 sets @80% and the single set @80%

Sorry just skimmed your post but are you saying 1 set is as good as multiple when training strength? If so that would be handy

No – sorry, I’m not saying that :-)

The study concludes this within the specific circumstances of the design – but with the stated limitations (subjects’ training status, number of subjects and thus effects on significance testing), I don’t thinks this is applicable to the situation of typically UKB ambitioned training climber…

So no shortcuts. We’ll still have to put in some honest work… 

Edit: overlap with duncan's reply, who was quicker...

[Dexter]

May have missed this while skimming but were the leg pairings controlled or randomised? If, as suggested, what the left leg does may affect the strength/hypertrophy of the right, then how these parings were made would be a potential variable. I'm assuming that as the sample numbers are low this was randomised, and that infering any additional information would further reduce samples sizes per variable and thereby may any meaningful statistics even more difficult.