Lets go back to the perennial question on foot muscle strength and traditionally shod running vs barefoot/minimalist running with the publication of this new research:
The effect of minimal shoes on arch structure and intrinsic foot muscle strength
Elizabeth E. Miller, Katherine K. Whitcome, Daniel E. Lieberman, Heather L. Norton, Rachael E. Dyer
Journal of Sport and Health Science; Available online 2 May 2014
This prospective study explored the effects of endurance running (ER) in minimal versus standard running shoes on the foot’s superficial layer intrinsic muscles and the function of the longitudinal arch. Our hypothesis was that running in minimal shoes would cause hypertrophy in these muscles and lead to higher, stronger, stiffer arches.
The hypothesis was tested using a sample of 33 healthy runners randomized into two groups, a control group shod in traditional running footwear and an experimental group shod in minimal support footwear, whose feet were scanned in an MRI before and after a 12-week training regime. Running kinematics as well as arch stiffness and height were also assessed before and after the treatment period.
Analysis of anatomical cross-sectional areas and muscle volumes indicate that the flexor digitorum brevis muscle became larger in both groups by 11% and 21%, respectively, but only the minimally shod runners had significant areal and volumetric increases of the abductor digiti minimi of 18% and 22%, respectively, and significantly increased longitudinal arch stiffness (60%).
These results suggest that endurance running in minimal support footwear with 4 mm offset or less makes greater use of the spring-like function of the longitudinal arch, thus leading to greater demands on the intrinsic muscles that support the arch, thereby strengthening the foot.
This study took a group of traditionally shod runners and allocated them to one of two groups: one group kept doing what they were doing and the other group were transitioned to minimalist shoes. At baseline and follow-up they measured arch height parameters, foot strike pattern and used MRI to measure the volume and cross-sectional size of selected foot muscles (as a proxy for strength).
The authors claim to have found:
- more of the minimalist group where not heel striking at follow-up
- increases in muscle volume were found at follow up in the flexor digitorum brevis in both control and minimalist groups; and in the abductor hallucis in the minimalist group
- increase in muscle cross sectional area in the abductor digiti minimi in the minimalist group
- no changes in arch height at follow-up (consistent with what I argued here)
Of the six muscle strength measurements (3 muscles measured by both volume and cross-sectional diameter); 1/6 showed an increase in the traditionally shod group; the other 5/6 did not change. In the minimalist group, 3/6 increased. That is hardly a compelling result and is not necessarily consistent with the conclusion reached by the authors in the abstract above and in the paper (though they did use the word ‘suggest’ in the abstract). Given that one muscle allegedly got stronger in the traditionally shod group, that is certainly more evidence that running shoes do not weaken muscles. This does not support the premise that these muscles are weak in running shoes or were weak to start with (which is what they imply), it just shows that they might get stronger in the minimalist condition.
Can the results they got be believed?
There are many red flags in this one and the study falls well short of the accepted gold standard and consensus for the reporting of trials (ie what the CONSORT statement and every textbook on randomized trials says you should do). This information is needed for the transparent reporting of trials so that a proper appraisal of the methods and results can be done. I have reviewed several studies recently that have fallen short on this: here, here, here and here.
What are the issues:
- I see no evidence of an a priori sample size calculation. They may have done one, but did not report the justification. That does not mean I have a problem with the sample size, its just they did not report how it was arrived at. However, the cynic in me can not help but point out that the fan boys will have to reject the results of the above study due to the sample size – they rejected the recent Vibram and bone stress injury study based on its sample size of 36 being too small (it wasn’t) – its just if you going to reject that study for that reason, then you have to reject this one for the same reason (I don’t).
- there is no indication if the study was single or double blinded. Did the participants know what the study was looking for (single blinding) and did those doing the measurements know which group the participant was in (double blinding). This should have been reported.
- the handling of drop-outs in the study was not what is considered gold standard. Intention to treat analysis (recommended by the CONSORT consensus) was not done or at least a rationale or justification as to why it was not done was not given in the paper.
- probably the most significant issue is that the analysis presented in the tables and how the results were reported in the abstract and full paper were based on within groups comparisons. That is NOT how you analyse a randomized trial. The whole point of randomized trials is to compare the outcome between the two groups, which they did not do. They just compared the outcome within each group.
These issues should have been addressed in the pre-publication editorial and peer review process. In this case, it appears to have failed. Until these issues are addressed the results can not be trusted. There are so many positives about this study, such as the way they transitioned the minimalist group; the choice of outcome measures; and the length of follow-up. It is a shame that a potentially good trial was ruined by not adhering to the consensus by which randomized trials should be reported and analysed.
Equally of concern is the biased introduction to the paper and those biases in the discussion which represent bad science. Many references cited in the introduction do not actually say what the authors are claiming they say. Those familiar with the references used in the paper will easily see the use of the wishful thinking fallacy. On other occasions authors have cherry picked references to support what they want to claim, ignoring many other references that show the opposite or do not support their claims. This really is bad reporting of science.
The whole basis of what they are discussing is based on the assumption that increasing the strength of the intrinsic muscles of the foot is a good thing and that minimalist running does this. I have no doubt that increasing muscle strength is a good thing and we know it goes a long way to preventing injury. But, is the premise that changing the running technique to increase that muscle strength a good thing? The whole thrust of the paper is that it is (ie they started this with a preconceived bias and that is not a good way to do science). It could be argued that if a particular running technique increases muscle strength, then it does so by increasing the activity of the muscles. Surely that is a bad thing? Surely an increase in muscle activity is a sign of an inefficient gait? I would have thought that an efficient gait would have less muscle activity. The authors were clearly biased that this was not the case. It may or may not be the case, but good reporting of research would be without those biases and address all the issues.
As always: I go where the evidence takes me until convinced otherwise, and this was potentially a good trial if the issues I raised re the reporting were done according to the CONSORT statement and if there were no preconceived biases and cherry picking of citations going on. This is a good example of bad science reporting.
POSTSCRIPT: I have had a few comments on Twitter and Facebook; basically:
I want to compare outcomes between doing x and y = good approach to doing science
I want to prove that x is better than y = bad approach to doing science.
Miller, E., Whitcome, K., Lieberman, D., Norton, H., & Dyer, R. (2014). The effect of minimal shoes on arch structure and intrinsic foot muscle strength Journal of Sport and Health Science DOI: 10.1016/j.jshs.2014.03.011