Foot morphology in barefoot vs shod runners and its effects on biomechanics

A few weeks ago there was this publication that looked at morphological differences between habitual barefoot and shod runners and it was only of mild interest due to the ethnic differences between the groups. Were the morphological differences due to the barefoot vs shod or where that due to Indian (barefoot) vs Chinese (shod)? I was going to review the study, but then the same group of authors published another study that looked at how the morphological differences between barefoot vs shod runners affected a number of biomechanical parameters. That one appears much more interesting to review:

A comparative biomechanical analysis of habitually unshod and shod runners based on a foot morphological difference
Qichang Mei, Justin Fernandez, Weijie Fu, Neng Feng, Yaodong Gu
Human Movement Science; Volume 42, August 2015, Pages 38–53

Running is one of the most accessible physical activities and running with and without footwear has attracted extensive attention in the past several years. In this study 18 habitually male unshod runners and 20 habitually male shod runners (all with dominant right feet) participated in a running test. A Vicon motion analysis system was used to capture the kinematics of each participant’s lower limb. The in-shoe plantar pressure measurement system was employed to measure the pressure and force exerted on the pressure sensors of the insole. The function of a separate hallux in unshod runners is analyzed through the comparison of plantar pressure parameters. Owing to the different strike patterns in shod and unshod runners, peak dorsiflexion and plantarflexion angle were significantly different. Habitually shod runners exhibited a decreased foot strike angle (FSA) under unshod conditions; and the vertical average loading rate (VALR) of shod runners under unshod conditions was larger than that under shod conditions. This suggests that the foot strike pattern is more important than the shod or unshod running style and runners need to acquire the technique. It can be concluded that for habitually unshod runners the separate hallux takes part of the foot loading and reduces loading to the forefoot under shod conditions. The remaining toes of rearfoot strike (RFS) runners function similarly under unshod conditions. These morphological features of shod and unshod runners should be considered in footwear design to improve sport performance and reduce injury.

This paper was very confusing to read and I am concerned at the number of parameters that they compared the two groups on relative to the sample size (ie the probability of finding a difference in some of the parameters based on chance is pretty high); and it is also possible that the differences in the biomechanical parameters were due to ethnicity differences and not necessarily the barefoot vs shod; but here goes anyway:

To help me get around the confusion reading the paper and not getting into details nor a critique of what they did, here are my notes (and comments) summarizing and simplifying what they found:

  • Stride length: both barefoot and shod runners had a shorter stride length when barefoot compared to shod  (not unexpected)
  • Stride time (time of right foot strike to next strike): both barefoot and shod runners had a shorter stride time when barefoot compared to shod
  • Contact time: both barefoot and shod runners had a shorter contact time when barefoot compared to shod
  • Foot strike angle (FSA): Shod group in shoes had a FSA of 16.1 degrees and barefoot it was 4.8 degrees (not unexpected);  barefoot group  without shoes the FSA was negative 8.2 degrees and with shoes it was negative 2.9 degrees. The negative indicates a more plantarflexed position   (not unexpected).
  • Stance period (% of mean contact time taken in the mean stride time): Shod group in shoes it was 37.5% and barefoot it was 36.8%; Barefoot group on shoes it was 29.1% and barefoot group when barefoot it was 28%.
  • The unshod group had greater eversion than the shod group
  • greater hip flexion angle at touchdown in the shod group when running barefoot
  • (there were some more kinematic differences in different conditions, but they getting difficult to understand and interpret graphs – the full paper is here for those inclined to dig into it)
  • Medial forefoot pressures: Shod group when barefoot had greater peak pressures and pressure time integrals compared to the barefoot group when both shod and barefoot running
  • Lateral forefoot pressures: Highest peak pressures and pressure time integral were found in the barefoot group when running barefoot (I may want to interpret that as a negative thing in the context of Bojssen-Mollors theory of high gear vs low gear propulsion)
  • Hallux pressures:  Lowest peak pressure and pressure time integral was found in the barefoot group when running barefoot; when shod, the barefoot group had higher pressure
  • Other toes: Highest peak pressure was seen in the shod group running barefoot

As I said above, I do have concerns over the shear numbers of parameters what they measured (and I have not mentioned them all) relative to the sample size and they did not do any sort of adjustment of the analysis methods because of that, so the above findings need to be interpreted in that context. To avoid confusion and point out what the authors considered most important, here is the authors summary of their results:

In summary, both habitually unshod and shod runners show significant differences while running under shod and barefoot conditions…The coronal and internal-external ankle angle was also significantly different between shod and unshod runners, particularly in the stance period. Habitually shod runners also have a greater vertical loading rate while attempting running barefoot. Under shod conditions, the separate hallux of habitually unshod runners and the other toes of habitually shod runners show greater peak pressure and pressure time integral with reduced value compared to the medial forefoot and lateral forefoot while comparing with running with shoes

What does all that mean? How should we interpret all this?
I not sure. The authors have provided a lot of data to wade through, perhaps too much data that the wrong interpretations could be made. A lot of the findings are what would be expected and have been reported before, so no issues there.

What is new is the interpretation that authors put on the plantar pressure data:

The morphology-related function (separate hallux of habitually unshod runner) is distinct under unshod and shod conditions. The separate hallux of habitually unshod runners shares part of the foot loading and reduces the loading focused on the forefoot region during stance period under shod conditions. The other toes of habitually shod runners show a similar function…

I guess how you want to interpret this as a good or bad thing depends on how much weight you want to give to the results, in the context of the ethnicity differences and the high number of parameters measured. The authors consider the pressure pattern seen in the barefoot group has being “better” than the shod group, but in the context of Bojsen-Mollors concepts it could be considered a “worse” thing (I discussed his concepts here).

Whichever way the results are interpreted, I fail to see how they managed to reach this conclusion:

Knowledge of this can improve footwear design and aid prevention of metatarsal stress fractures and plantar fasciitis related injuries.

As always: I go where the evidence takes me until convinced otherwise, and its hard to get excited about this study, let alone improve footwear design because of it.

Mei, Q., Fernandez, J., Fu, W., Feng, N., & Gu, Y. (2015). A comparative biomechanical analysis of habitually unshod and shod runners based on a foot morphological difference Human Movement Science, 42, 38-53 DOI: 10.1016/j.humov.2015.04.007

Last updated by .

2 Responses to Foot morphology in barefoot vs shod runners and its effects on biomechanics

  1. Jordan July 20, 2015 at 4:22 pm #

    Very interesting take on this information. I agree that they seem to have measured too many things given the sample size! But I do hope a study like this will lead to more like it in the future – perhaps with more people involved. Thanks for sharing!

  2. Vincent August 5, 2015 at 2:07 am #


    This website is a fantastic resource. I red a lout of your posts on barefoot vs shod running and was wondering..

    1) If it is clear that running form is different between barefoot and shod running…

    2) this means that the form depends (in part) on the way the foot strikes the ground (pressure, location, sensitivity etc. I dont have your vocabulary or expertise)… then…

    3) Since running shoes models are getting “modified” every year, get worn off, of are replaced by a new “model”…

    4) (conclusion) would it not be better to run barefoot just because it will stay the same over time?

    Maybe it is “naive” or “straight dumb”. If you have a second I would love for you to “set me straight”! I do run right now in pegasus but im am wondering if that would not be a + for barefoot running. Also english is not my first language so forgive any typos and clunky style.

    Thank you

Leave a Reply