Overload injuries in barefoot/minimal footwear running

My interpretation of the preponderance of the research on general injury rates between barefoot/minimalist vs traditionally shod runners and forefoot/midfoot vs rearfoot striking is that there is no difference in the rates. That does not mean that I will not change my mind or evolve my interpretation as more research is made available. For that reason, I welcome new research that adds to the mix and helps our understanding. The research below, I got just over a week ago and have not deliberately delayed writing about it as I have not been well and have a couple of sick kids taking over my life. Before writing about the research, I did a search to see if anyone in the crankosphere blogosphere had spun or misinterpreted this research to suit their purposes. (Un)fortunately no one had. That evil cynical voice coming from on my left shoulder was telling me that I should wait before writing about it to see what interpretations gets put on this research, then I could lambaste the loons (like this one) … but I thought the better of it. Here is the abstract of the study in question:

Overload injuries in barefoot/minimal footwear running: evidence from crowd sourcing
Martin Daumer​, Christine Kleinmond, Christoph Stolle, Christian Lederer, Marc Hilgers, Markus Walther
PeerJ PrePrints 2:e250v1 2014
Background: The incidence of injuries in minimal footwear or barefoot runners compared to runners using conventional running shoes is still being discussed.
Methods: We focus here on methods and results of a online survey in a community of minimal footwear/barefoot (MF/B) runners. The aim of the survey was to investigate the viability of the method and to record first insights on running behavior, distance performance and injuries.
Results: In total 226 runners answered the questionnaire, 15 subjects had to be removed due to invalid data. A total of 211 (94%) subjects (152 male, ages 15-71 years [mean=40]) were included in the analysis. The risk to suffer a running related injury was significantly increased during the time period of changing from shod running to MF/B running (see Table 2). The injury rate per km was markedly lower – about one half – in MF/B than in shod running, but threefold higher during the transition period.
Discussion/Conclusion: Future research into the right “dosage” of barefoot/minimal footwear running in the transition period is warranted. We speculate that special adaptations – which may take years and thousands of kilometers to become effective – of the neuromuscular control play a major role, very similar to the thousands of hours a person needs to play and practice playing the piano before becoming a musician.

Essentially what they did was to recruit runners via a barefoot running website to fill out an online survey. The criteria was if they used to run regularly in traditional running shoes and after a period of transition are now running in minimalist shoes or barefoot. They found:

  • mean km/week prior to transition was 27.2 (+ 23.0); during the transition it was 19.5 (+ 18.0); and after the transition it was 35.3 (+ 26.6).
  • they calculated the number of injuries per 10 000km to be 12.1 (+ 55.7) before transitioning; during the transition it was 35.8 (+ 101.5); and after the transition it was 5.4 (+ 22.0). This lead the authors to claims that the injury rate was much higher during the transition and lower after the transition, and based on the data, that is what it appears to show.

My observations:

  • firstly the journal that this was published in is very low level and has no impact factor (academics will appreciate the significance of this).
  • its was a self selected online survey that was largely only promoted to one barefoot running community. No one familiar with research methodology would accept the results from such a survey. The potential for bias is way too great (it was the same issue that was the epic fail in the interpretation of a similar previous online survey that led to me call out a loon!). This is not just my “biased” interpretation of this study; every textbook on research methodology will point out the same thing; every epidemiologist, who do these kinds of studies for a living, will point out the same thing.
  • recall bias is also a significant factor in these sorts of study designs.
  • I not comfortable with the way they converted the data to a “homogeneous” data set (eg if there were “comments like ‘too many’ or ‘many’ were counted as three injuries“) – this raises questions over the calculation of the standard deviations (look at the size of them in the results above).
  • They obviously did not include runners to tried to transition and failed to compare them to, nor did they include those who did not transition to compare the number of injuries. I assume that they were not included due to the source that was used to recruit the participants (ie biased). In other words,  there were no comparison or control groups!
  • Bizarrely, despite acknowledging the recall bias issue (and not the other potential biases), the authors still manage to conclude “we do consider the finding about a reduction of injury risk during the barefoot phase, as compared to the shod phase, as relatively strong.” I don’t think their data even points to a weak relationship (though the relationship may actually be strong, but you can not conclude that from the data in this study due to the huge potential biases in the data collection and lack of a control or comparison group).

All this study really is, is a collection of n=1 anecdotes from a biased sample. The study tells me of those who chose to respond, if their recall is correct, that they are getting less injuries now running minimalist/barefoot then previously running shod. The plural of anecdotes is not data.

As always, I go where the evidence takes me until convinced otherwise, and the evidence still tells me that there are no general differences in injury rates between barefoot/minimalist vs traditionally shod runners and forefoot/midfoot vs rearfoot striking.

POSTSCRIPT: here is a similar commentary on these sorts of poor study designs and how those with agendas cherry pick the poor studies: http://theness.com/neurologicablog/index.php/health-of-vaccinated-vs-unvaccinated

Last updated by .

One Response to Overload injuries in barefoot/minimal footwear running

  1. Marc-Antoine Doré March 4, 2014 at 6:31 pm #

    Craig, I really like your interpretation of this study. I think there’s even more you can add like what’s their definition of transition and minimalist shoes… I don’t know if that was specified in the full article.

    Now even though I totally agree with what you are pointing here, my opinion is: Is this study proving anything we didn’t know? Even without all the bias, we know that during transition your injury rate will increase. And that’s not specific to running and barefoot vs shod…It’s the same for gym training and for everything we do in our life. So yes it was biased but it also didn’t prove anything new to me! (if I come to NZ, my transition from left seated driver to right seated will probably increase my injury/death risk also!)

    Thanks for your good work!

Leave a Reply