single vs multiple sets and muscle hypertrophy: meta-analysis

[TPT]

just published for more chatter.


Single vs. Multiple Sets of Resistance Exercise for Muscle Hypertrophy: A Meta-Analysis

Krieger, James W
The Journal of Strength & Conditioning Research. 24(4):1150-1159, April 2010.
doi: 10.1519/JSC.0b013e3181d4d436

Abstract:

Krieger, JW. Single vs. multiple sets of resistance exercise for muscle hypertrophy: a meta-analysis. J Strength Cond Res 24(4): 1150-1159, 2010-Previous meta-analyses have compared the effects of single to multiple sets on strength, but analyses on muscle hypertrophy are lacking. The purpose of this study was to use multilevel meta-regression to compare the effects of single and multiple sets per exercise on muscle hypertrophy. The analysis comprised 55 effect sizes (ESs), nested within 19 treatment groups and 8 studies. Multiple sets were associated with a larger ES than a single set (difference = 0.10 +/- 0.04; confidence interval [CI]: 0.02, 0.19; p = 0.016). In a dose-response model, there was a trend for 2-3 sets per exercise to be associated with a greater ES than 1 set (difference = 0.09 +/- 0.05; CI: -0.02, 0.20; p = 0.09), and a trend for 4-6 sets per exercise to be associated with a greater ES than 1 set (difference = 0.20 +/- 0.11; CI: -0.04, 0.43; p = 0.096). Both of these trends were significant when considering permutation test p values (p < 0.01). There was no significant difference between 2-3 sets per exercise and 4-6 sets per exercise (difference = 0.10 +/- 0.10; CI: -0.09, 0.30; p = 0.29). There was a tendency for increasing ESs for an increasing number of sets (0.24 for 1 set, 0.34 for 2-3 sets, and 0.44 for 4-6 sets). Sensitivity analysis revealed no highly influential studies that affected the magnitude of the observed differences, but one study did slightly influence the level of significance and CI width. No evidence of publication bias was observed. In conclusion, multiple sets are associated with 40% greater hypertrophy-related ESs than 1 set, in both trained and untrained subjects.
(C) 2010 National Strength and Conditioning Association

Go to Full Text of this Article

[TPT]

this was a valuable study because most of the previous studies and meta-analyses were on strength. this was a study on muscle hypetrophy. something that should be discriminated as strength and hypertrophy are not one and the same. thus, should not be interpretated as such.

multiple sets per exercise were associated with significantly greater modifications in muscle size during resistance programs. specifically, multiple sets had 40% greater hypertrophy-related effect sizes than a single set. there was a trend toward dose effects. i.e., increasing muscle size with increasing number of sets.

thus, the study suggests that we should be performing a minimum of 2-3 sets per exercise. possibly, 4-6 sets would be more effective in achieving maximal hypertrophy.

[Abraxas]

You know,im curious to know how going to failure fits in all of this.
Especially since going to failure efficiently, takes time to develop.

I would like to see some studies where they use athletes
who are trained to efficiently go to failure,let them do one all out set.

And see if those people get a just as significant growth response as the subjects who did multiple sets in this study,and see if the ''failure'' group did more sets,if that would have any significant benefit over doing one set as far as hyperthrophy goes.

[Abraxas]

I believe,high intensity,low volume,high frequency coupled with a periodisation scheme were each workout the muscles are subjected to maximum yet substantially different loads,as far as weight and rep execution goes,is they way to go.

Just a theory..

[ob205]

Mike Mentzer would be pissed! I would like to see this study done on identical twins, where one trains 1 set to failure and the other multiple. Ideally, they would be say 21, with no previous weight training experience and similar athletic backgrounds. Assuming all things being equal (diet, rest, extra ciricular activities), our greatest variable would be intensity of EFFORT. If you give 2 people 80% of the 1rm and say go to failure, you are going to get some very different results, based on their motivation and ability to generate emotional intensity. This is where I find the study, although useful, not completely accurate.

I trained for several years Heavy Duty style and made great gains, while at the same time being told my the Ivory Tower types in the Exercise Phys dept that it is not possible to stimulate hypertrophy with only 1 set.

Looking around any commercial gym I see most people doing a great number of sets, but hardly any of them look productive. 1 all out set from a person trained in the high intensity techniques and motivated easily exceeds 4-6 sets of an individual half assing the work.

Good Post, love the topic and the debate!

[Costco77]

How can a study really judge intensity and going past mental and physical failure?

[Abraxas]

How can a study really judge intensity and going past mental and physical failure?

I agree,there should be some way though..

[JamesKrieger]

You know,im curious to know how going to failure fits in all of this.
Especially since going to failure efficiently, takes time to develop.

I am the author of this paper.

I used a random effects model to analyze the data. When combining the results of different studies, you can't account for all possible sources of variation between studies. A random effects model essentially takes this into consideration. What that means is that it's harder to see significant differences with a random effects model, because you are automatically building in more uncertainty.

What I'm saying here, is that my statistical model automatically accounts for small variations among studies in regards to training to failure.

Also, all the studies included in this analysis involved subjects training to failure, so the point is moot.

And see if those people get a just as significant growth response as the subjects who did multiple sets in this study,and see if the ''failure'' group did more sets,if that would have any significant benefit over doing one set as far as hyperthrophy goes.

This was a meta-analysis of studies comparing single sets to multiple sets, so there are no people in this study, per se. The experimental units in this study are the studies and treatment groups.

I would like to see some studies where they use athletes
who are trained to efficiently go to failure,let them do one all out set.

What do you mean, "efficiently go to failure?"

What I've never understood is why failure is considered some sort of "magical" or "holy grail" concept. It only means that the muscles, as a whole, can no longer produce enough force to move the resistance. That's it. However, from the concept of stimulating protein synthesis (which is the entire goal if you're trying to get bigger), it doesn't mean much.

[JamesKrieger]

Assuming all things being equal (diet, rest, extra ciricular activities), our greatest variable would be intensity of EFFORT.

Intensity of effort is a subjective quantity that cannot be directly measured. While there are self-reported ratings of intensity of effort (like Rate of Perceived Exertion or RPE), you are essentially taking the subject's word for it.

Second, intensity of effort is not the greatest variable. This is easily illustrated by looking at someone doing a 100-RM set. Doing 100 reps to failure certainly will require a great intensity of effort, but it will do little to change muscle size.

If you give 2 people 80% of the 1rm and say go to failure, you are going to get some very different results, based on their motivation and ability to generate emotional intensity.

You are going to get different results based on a variety of factors, including genetics and diet. This is where measures of variation (standard deviations and standard errors) and statistics come into play. Basically, statistics tell you whether the magnitude of a difference between groups (1 set and multiple sets, in this case) is greater than the variability observed in those groups.

This is where I find the study, although useful, not completely accurate.

The statistics automatically take into consideration the variability, so yes, the study is accurate in that sense.

I trained for several years Heavy Duty style and made great gains, while at the same time being told my the Ivory Tower types in the Exercise Phys dept that it is not possible to stimulate hypertrophy with only 1 set.

My paper showed that you do stimulate hypertrophy with 1 set. It's just that the effects of multiple sets were greater.

[TPT]

I am the author of this paper.

I used a random effects model to analyze the data. When combining the results of different studies, you can't account for all possible sources of variation between studies. A random effects model essentially takes this into consideration. What that means is that it's harder to see significant differences with a random effects model, because you are automatically building in more uncertainty.

What I'm saying here, is that my statistical model automatically accounts for small variations among studies in regards to training to failure.

Also, all the studies included in this analysis involved subjects training to failure, so the point is moot.

'repetitions to failiure' is a relevant variable and should be controlled as such.

especially, as it fits into the study of single vs multiple sets. or the arguments of high intensity/heavy duty vs volume training. i.e., single sets with repetitions to failure might be argued as more effective for muscle strength or hypertrophy than multiple sets without repetitions to failure.

im suspicious of whether many of the cited studies actually controlled repetitions to failure as they stated since i didnt see any reliability measures and some studies appeared to discriminate 'failure' more than others. e.g., rhea et al. (2002) described a much more explicit training program with undulating periodization with no supervision. while others including ronnestead et al. (2007) used a different program with initial supervision. i understand this was part of a meta-analysis and will not be so critical of the inherrent limitations of meta-regressions. however, speaking of implicit confounds should still be appreciated.

This was a meta-analysis of studies comparing single sets to multiple sets, so there are no people in this study, per se. The experimental units in this study are the studies and treatment groups.



What do you mean, "efficiently go to failure?"

What I've never understood is why failure is considered some sort of "magical" or "holy grail" concept. It only means that the muscles, as a whole, can no longer produce enough force to move the resistance. That's it. However, from the concept of stimulating protein synthesis (which is the entire goal if you're trying to get bigger), it doesn't mean much.

nothing magical, but relevant as it may have differential effects on muscular endurance, strength, hypertrophy, and the central nervous system.

[Will Brink]

multiple sets are associated with 40% greater hypertrophy-related ESs than 1 set, in both trained and untrained subjects.

That was established a long time ago. If they have funding, too bad they didn't look to answer a new question.

[JamesKrieger]

especially, as it fits into the study of single vs multiple sets. or the arguments of high intensity/heavy duty vs volume training. i.e., single sets with repetitions to failure might be argued as more effective for muscle strength or hypertrophy than multiple sets without repetitions to failure.

But for all the studies included, training protocols were identical other than the number of sets. Any studies that compared 1-set-to-failure to multiple-sets-not-to-failure were excluded.

im suspicious of whether many of the cited studies actually controlled repetitions to failure as they stated since i didnt see any reliability measures and some studies appeared to discriminate 'failure' more than others. e.g., rhea et al. (2002) described a much more explicit training program with undulating periodization with no supervision. while others including ronnestead et al. (2007) used a different program with initial supervision. i understand this was part of a meta-analysis and will not be so critical of the inherrent limitations of meta-regressions. however, speaking of implicit confounds should still be appreciated.

As I mentioned in my previous post, it is impossible to include covariates in the model for every single potential source of variance between studies. This is why a random effects model is used; the statistical model automatically accounts for unaccounted sources of variance between studies.

Also, the Rhea paper did not specify whether there was supervision or not. Thus, it cannot be stated that the subjects were not supervised.

nothing magical, but relevant as it may have differential effects on muscular endurance, strength, hypertrophy, and the central nervous system.

In regards to hypertrophy, why?

If I I do 9 repetitions of a 10 RM weight (hence stopping 1 rep short of failure), why would that 1 repetition really make that much of a difference in regards to stimulation of protein synthesis? I think people tend to "micromanage" this aspect of training too much, when it probably makes little to no difference in the grand scheme of things.

The only thing that happens at "failure" is the fact the muscle can no longer produce enough force to move the weight. There's nothing that significant about it. I've always found the statements by HITers a bit incredulous, because they'll say that if someone does multiple sets to failure, they aren't reaching "true" failure. Or if a study happens to find multiple sets to produce superior results, they'll make an excuse like "the single set group wasn't really reaching full mental and physical failure." Then they offer up some vague definition of what "true" failure really means. To me, that's just an excuse to ignore data that doesn't fit with their beliefs.

It's similar to people that believe in ESP. There have been dozens of studies that fail to support the existence of ESP. The ESP believers will then say, "Well, the researchers hostile thoughts are interfering with ESP transmission which is why they're getting negative results." Basically, they've constructed a non-falsifiable hypothesis. They've already made up their minds, and no amount of conflicting data will overturn that.

Essentially, the hardcore HITers have developed a non-falsifiable hypothesis. For every study that shows multiple sets to produce better results, they'll always say, "Well, the single set group isn't REALLY going to failure."

[JamesKrieger]

That was established a long time ago. If they have funding, too bad they didn't look to answer a new question.

In what study was this established?

[Abraxas]

But for all the studies included, training protocols were identical other than the number of sets. Any studies that compared 1-set-to-failure to multiple-sets-not-to-failure were excluded.




As I mentioned in my previous post, it is impossible to include covariates in the model for every single potential source of variance between studies. This is why a random effects model is used; the statistical model automatically accounts for unaccounted sources of variance between studies.

Also, the Rhea paper did not specify whether there was supervision or not. Thus, it cannot be stated that the subjects were not supervised.





In regards to hypertrophy, why?

If I I do 9 repetitions of a 10 RM weight (hence stopping 1 rep short of failure), why would that 1 repetition really make that much of a difference in regards to stimulation of protein synthesis? I think people tend to "micromanage" this aspect of training too much, when it probably makes little to no difference in the grand scheme of things.

The only thing that happens at "failure" is the fact the muscle can no longer produce enough force to move the weight. There's nothing that significant about it. I've always found the statements by HITers a bit incredulous, because they'll say that if someone does multiple sets to failure, they aren't reaching "true" failure. Or if a study happens to find multiple sets to produce superior results, they'll make an excuse like "the single set group wasn't really reaching full mental and physical failure." Then they offer up some vague definition of what "true" failure really means. To me, that's just an excuse to ignore data that doesn't fit with their beliefs.

It's similar to people that believe in ESP. There have been dozens of studies that fail to support the existence of ESP. The ESP believers will then say, "Well, the researchers hostile thoughts are interfering with ESP transmission which is why they're getting negative results." Basically, they've constructed a non-falsifiable hypothesis. They've already made up their minds, and no amount of conflicting data will overturn that.

Essentially, the hardcore HITers have developed a non-falsifiable hypothesis. For every study that shows multiple sets to produce better results, they'll always say, "Well, the single set group isn't REALLY going to failure."

How do you quantify intensity if not going to failure? (you could say minus one rep to failure, fine by me)

Im not a HIT proponent,im just trying to make all the pieces fit together.

What im postulating is,whether you go to failure or not,there will always be variables in the amount of force someone can produce relative to their own strength, in a set. (or in multiple ones for that matter)

Since most people (since most people do bodybuilder-type volume routines) do not ever subject themselves to training themselves to produce a maximum effort in a single set,it would be natural for them not to be good at it,and would need more sets to illicit a maximal hypertrophic response because of that fact alone.

So im using the term failure here as a way of confirming a high intensity effort.

[TPT]

That was established a long time ago. If they have funding, too bad they didn't look to answer a new question.

not sure what youre speaking to.