Welcome to the Science of Sport, where we bring you the second, third, and fourth level of analysis you will not find anywhere else.

Be it doping in sport, hot topics like Caster Semenya or Oscar Pistorius, or the dehydration myth, we try to translate the science behind sports and sports performance.

Consider a donation if you like what you see here!

Did you know?
We published The Runner's Body in May 2009. With an average 4.4/5 stars on Amazon.com, it has been receiving positive reviews from runners and non-runners alike.

Available for the Kindle and also in the traditional paper back. It will make a great gift for the runners you know, and helps support our work here on The Science of Sport.

Tuesday, November 29, 2011

The Barefoot running round-table discussion from UKSEM: Thoughts from "inside"

Thoughts from the Barefoot running round-table discussion at UKSEM: An inside view

Many of you will probably know by now that at the recent UKSEM conference in London, I chaired a session called "Natural Running – advantages and disadvantages. A Round Table Discussion".

The protagonists in the debate were:
  • Prof Daniel Howell, an anatomy professor from Liberty (USA), known as the "barefoot professor"
  • Simon Barthold, who formerly worked as a podiatrist but who now works in biomechanics and is Asics global research consultant
  • Prof Benno Nigg, one of the world's leading biomechanists
  • Dr Mathias Marquard, a clinician and running coach (who would go on to become the voice of reason in many of the more hostile aspects of the debate, as I'll describe!)
  • Prof Daniel Lieberman, evolutionary biologist from Harvard, who as you may know, recently published the Nature studies looking at how habitually shod and barefoot runners differ, and who wrote a key paper on how humans are adapted (skeletally and physiologically) to run long distances
The debate concept

That's a pretty high-profile "cast", including some of the world leaders in their fields.  Then there was me, chairing a debate which everyone knew could easily become an argument!  To begin with, academics don't enjoy this method of getting theories out.  I know this because three of the five on the panel said as much before and after, and I suspect it's mostly because scientists like to work according to a linear 'template' that says you first introduce the question, then you describe the gaps in the literature, then you systematically plug those gaps using your experiments, then you present data and move towards understanding.

A debate, however, is not linear, but circular, or more like a "vortex", in that different threads are whirling around together, and I think it can be an uncomfortable way to discuss data.  The risk is always that every statement made by one person contradicts another's views, and they want to respond to it, so we would basically get sucked down and never move forward.  

I can appreciate this, but I think it's also an excellent way to accelerate understanding for the audience (but then you'll have to tell me this if you were there), because it super-condenses a big topic into a discussion and for that reason, I think it works rather well.

Controlling it, however, was a nerve-jangling prospect.  Before the debate, Asics (who sponsored the UKSEM conference, which included naming rights to this debate, which is commendable) had worked with PR teams to try to manage it, because they were understandably concerned about excessive hostilities (lively debate is good, outright hostility is not!) and also about one or two of the members dominating the discussion.  "Everyone must get a say".  So I had to ensure that neither happened, and with this being such a polarized topic, and knowing that there were pro- and against- academics on stage, I confess to being quite anxious about it.  My approach to this of course is to joke and try to entertain (why be dull when you can liven it up?), but that didn't stop me from almost forgetting the names of the first two speakers as I introduced them!  

Buy or sell?  The first provocative question

Nevertheless, I got past that first little hurdle, and then the debate kicked off with a simple question, based on the media portrayal of the barefoot debate.  The question was: "If a runner picks up a magazine or newspaper, they are seeing the following statement: 'Shoes are evil.  They do not help, they may even cause injury.  Barefoot running is natural, and will help prevent injury, and therefore everyone should be encouraged to run barefoot'.  Do you buy or sell this concept?"

That's a very provocative question, and as mentioned, I hate how this issue has been polarized.  In fact, if there's anything you take out of this website, it's that when people polarize a debate into one of two extremes, they're both wrong.  In science, there's always middle-ground, and a significant "but", whether it's related to barefoot vs shod running, training vs talent, dehydration vs overhydration, doping control, carbs vs fat in diet.  But an extreme question was necessary to get the ball rolling.

So from my vantage point, this is what I saw from the five responses:

Only Daniel Howell outright bought the concept.  He explained that he has been LIVING barefoot for 6 years, spending 95% of his time without shoes.  He is an advocate not only for barefoot running, but for barefoot living.  His main argument, which I'll get to shortly, is that barefoot running is the "natural state"

All the other speakers were relatively non-committal.  Prof Benno Nigg was most neutral, saying that every year, he asks his students this question in a final exam: "Does barefoot running prevent injuries?", and the only answer he accepts for a good grade is "I don't know because we don't know".  

It became clear right away that Prof Nigg was not about opinions.  At all.  He is perhaps the world's leading biomechanist, and has had in excess of 300 publications on the subject, plus dozens of books, and is really all about the evidence.  Which is a bit of a problem in a round-table discussion, but his absolutely neutral answer did two things a) it highlighted that this is a debate that really does lack evidence, and b) that he was going to be the "go to guy" for scientific fact, not opinion!  If I wanted to kill the debate, ask Prof Nigg for his opinion!  

Professor Daniel Lieberman said the same thing - we don't have the evidence yet, but there is enough theory there, as well as the 'birth' of a line of evidence that may begin to steer us towards it.  At this point, they mostly agreed with one another.

"Buy, but keep the receipt for a refund"

Let me digress and state my view on that question, since I didn't get to state it at UKSEM!  I believe that EVERYONE can benefit from some barefoot running.  That is, I think that barefoot running is, at worst, a good training modality that may have benefit for running performance, even when wearing shoes.  We know from research and simple experience that there are significant differences in muscle activation and loading patterns when running barefoot, and these are all potentially favourable, even if barefoot running is used only as a training method.  In fact, I'd go so far as to encourage all runners to try barefoot running, even if it is only during a warm-up or cool-down, or once a week for a short time.

For some people, I do believe that barefoot running may be the answer to their injury problems.  I think there is enough there to suggest that some individuals who struggle in shoes will fare much better without them.  However, here's the catch - we don't fully know who they are, and more importantly, why they benefit.  We can surmise that it has to do with the change in loading on different joints (as shown by Lieberman and countless others), the proprioception, the strengthening of joints, and so forth.  But we simply don't know.

By extension then, there may well be people who simply cannot adapt to barefoot running.  In fact, I'm certain this will be the case.  They break down and get new injuries, usually of the ankle, calf, Achilles tendon or foot.  And these individuals may never take fully to barefoot running.  I still think that the fact that they do pick up these injuries indicates the 'stress' and if the body adapts positively to stress, then they too can benefit from barefoot training, if not fully immersing themselves in it.  However, for them, it must be recognized that shoes may be the only thing enabling them to run (regardless of whether it is "natural" or not). 

And the one thing I would implore the "barefoot evangelists" to recognize is that just because it works for them, does NOT mean it will work for everyone, and so don't make the same mistake we often  accuse shoe companies of making when they gave everyone motion-control and stability devices.

The key thing, I believe, is that barefoot running allows us to study shod running better.  It invites the realization that perhaps it is running form/technique that is crucial, and by comparing and contrasting the two, we might understand why people run the way they do, and where the risks may originate.

So in short, my answer to that question is "Buy barefoot running as a concept, try it out as a training modality, but keep the receipt so that you can return it if you don't find the "fit" right for you.  At worst, you'll discover a new muscle activation pattern, a new and effective training method, and potentially, changes to running form that will help you run better, in shoes"

Back to the debate...

The tale of two Daniels, and confusing a hypothesis with evidence

The first real point of disagreement in the debate came with a theoretical discussion of "natural running".  That's a vague, all-encompassing term, and we could have debated it for an hour, all by itself.

But let's just go with it at a superficial level, for now!

Prof Daniel Howell, the barefoot professor, was asked to elaborate on the evidence for barefoot running.  Remember, the panel had all agreed that evidence was lacking, so the next question I put is "what evidence do you need, and what do you have?"

Howell's response was that "barefoot running is natural".  We are not born with shoes, our ancestors did not run in shoes, and it is therefore natural for us to run barefoot too. To live barefoot, in fact.  What is not always as clear is that somewhere along this logic, "natural" becomes a synonym for "better".  Howell at one point challenged Simon Barthold, asking him to justify why he said that people need shoes (I agree with Barthold on this one, by the way.  At least for some people). 

Howell believes that we don't, because it's natural to be barefoot, and that this must be better.  I'm paraphrasing of course (I'm sure I'm open to criticism about context here, but that's basically his position, as anyone who heard it will say, I'm sure).

There are fundamental problems with this idea.  First, he makes a big error of confusing the hypothesis with the evidence.  All he has at this early stage is a theory that can lead to a hypothesis.  Prof Lieberman (the other Daniel in the discussion) has a better understanding of this.  I had a long lunch with Lieberman the day before, and we discussed the entire debate, and this came up.  My point is that we didn't have anti-biotics until recently either, and the result was that many people died as a result of "natural" causes, and the invention of these medicines was clearly a positive step.  To equate "natural" with "better" is a very basic mistake to make.

Second, the problem that I think Howell has is that he has not recognized that being barefoot as a runner exists in a larger context, and that context includes about 100 things that make us different from our ancestors.  For example, we sit at desks for 8 hours a day, we sleep on comfortable mattresses, we drive, and we "hunt" our food in supermarkets and not in bushlands, we play in shoes (when we're not playing on computer games), and we grow up in them and then at 30, we are faced with a possible change (as a result of this debate).  Not one of those things happened before, but every one of them COULD be a contributing factor to injury risk.  In other words, weakness of supporting muscles and tendons as a result of years of disuse and TV-watching might mean that being "natural" is a more risky option that being in shoes.  There is a real possibility, as stated earlier, that some people need shoes in order to run.  The notion that being barefoot works for everyone today because it may have worked for everyone a long time ago is a leap of faith.

Lieberman recognizes this, and it means that he can appreciate that the anthropological finding about what we had on our feet many years ago is not proof of what we should wear today, it's only a starting point for a hypothesis that can be tested.

The skill aspect of running

The consequences of making the over-simplification of "natural = better" are significant.  For example, I presented on barefoot running last week, and suggested that barefoot running is a skill that has to be learned.  If, like Howell, you believe that natural barefoot running is better, then you don't need to recognize the skill aspect of running.  In fact, we know this because he called this skill idea "bull" in a Twitter post recently.  The problem is this: The scientific evidence produced by Lieberman shows very clearly that people who have run in shoes for many years do NOT run barefoot the same way as people who have been barefoot for a long period.  Thus, there is some learning, some adaptation that takes place, and whether we can all achieve this adaptation remains to be seen.

That is, take the shoes off and you get a pretty dire picture - these individuals continue to heel-strike, at least for a short time, which predisposes them to very high ground reaction forces and a huge vertical loading rate, both of which are surmised to be linked to injury risk.  Also, the muscles and tendons are unconditioned for barefoot running, and are then suddenly loaded differently, which further increases the injury risk.

Anecdotally, and from my own coaching (and Lieberman's observations which he shared with me over lunch), new barefoot runners make some fundamental errors because they don't adopt what seems to be the optimal barefoot running gait right away.

If they are running "naturally", however, and we buy into the theory that it's how it was intended by nature, then I fear that we're missing a huge piece of the puzzle, because it is quite clear that not all barefoot running is equal either.  And so when people "fail" when barefoot and are surprised, it's probably (this is opinion at this stage - evidence will come) because of faults in the gait, the most obvious of which seems to be over-striding and deliberately forcing a forefoot landing by plantar-flexing at the ankle (pointing the toe down).

This is a recipe for disaster, since it loads the ankle joint on a contracted muscle, and probably led to so many Pose runners breaking down when we monitored a group who'd just learned this technique.  I suspect the same risk exists for barefoot running, but it happens "naturally" and if you adopt the historical hypothesis as "proof", then you are blind to this possibility.  On the whole, I think that Howell does a disservice to his own advocacy by being blind to the evidence.

But very importantly, if making the transition to barefoot running should be viewed as a skill that has to be learned, then why not view all running as a skill?  This is an interesting question and kind of leads into where this debate will go in the future, I think, but more on this later.

The cushioning debate

The biggest point of difference came around a discussion on cushioning and impact forces.  Lieberman had the day before presented his Nature study findings, where the impact transient was absent when running barefoot with a forefoot landing, and explained this using an effective mass model.  Basically, what he is saying is that when you run in this way, and land forefoot, a lower effective mass decelerates on ground contact, than when you land on the heel.  To illustrate this, he used the analogy of a pen falling vertically to the ground compared to a pen falling at an angle of 45 degrees.  A greater effective mass "stops" when the pen lands vertically.

This is sound logic, of course.  But it led to an argument, because I think Simon Berthold misunderstood the point of the analogy.  He had printed off Lieberman's website explaining barefoot running and adamantly criticised Lieberman's explanation.  I think it's fairly clear what the analogy was meant to illustrate, and I think there is no doubt that landing on the heel does involve a significantly higher impact transient (just look at the difference in magnitude - it's 700% higher for heel-striking than forefoot landing).

There are some very theoretical questions about the use of this model, and Benno Nigg commented on this, but overall, this was an argument that didn't help the debate, because it obscured the point about impact forces.  I think an analogy was mistaken for a literal explanation and Lieberman's website became the focus of argument when we might have been discussing the mechanics a little better.  I eventually had to dismiss this discussion and move on, because nothing good was going to come out of it, because Berthold had pursued it down a blind alley to a point where Lieberman couldn't defend the analogy anymore, and Lieberman was getting flustered as a result.  End of discussion.

There was some disagreement over cushioning as well.  Lieberman's "model" is that part of the benefit of being barefoot is that it reduces the loading rate and effectively removes the impact transient.  For this to be beneficial, as opposed to having purely academic value, it has to be shown that these forces on landing are linked to injury. There is some evidence of this from Irene Davis' work, and Lieberman mentioned in the debate that the higher impact forces and loading rates have been linked to injuries like shin-splints and potentially knee problems.  

Benno Nigg was of the opinion that it wasn't the impact forces, but rather the forces in mid-stance that were more important.  His work suggests that the active forces may be more important, and these are very similar for shod vs barefoot running.  One of his big lines of evidence, of course, is to show that the degree of cushioning in the shoe (or running surface) actually doesn't change the impact forces.  His explanation for this was perhaps a little rushed, but has to do with the idea that muscle can be "tuned" by activation levels to make it optimal for a given surface.  The end result is that whether you run on hard or soft surfaces, the impact is relatively "benign".  This became a fairly high-brow biomechanical discussion, which definitely doesn't work in a round-table debate, and so wasn't explored as well as it perhaps needed to be.

Voice of reason: What do shoes really need?

The voice of reason in this debate, as I mentioned, was Mathias Marquard.  A highly acclaimed German author of running/coaching books, and a clinician, he adopted a very neutral and sensible view in the debate.  His experiences as a runner and a coach had brought him full circle, from going fully barefoot 15 years ago, to now recognizing the value of barefoot running, but not prescribing it.  He seems to have found the practical balance, and complemented the scientific discussion very well.  He made this point very eloquently on many occasions, and as a result, when I felt the debate was getting off track, he was the "go to guy" to bring it back with pragmatic viewpoint.  He was very valuable, mostly because of his pragmatism (and humour!)

It was Marquard who brought up a really interesting question when he said that we need to ask very seriously what shoes actually need to have for running?  Do they need massive cushioning?  Do they need stability devices?  Do they need motion control gadgets and built-up medial arch supports?  Do they need rigidity?  The answer to all these questions, in his opinion, was "No", and that was one of the most important points to come out of the discussion.  It was a point that the whole panel agreed on. There is a perception of needing all these aspects, but no evidence for them, and a real possibility that we're better off without them.

On the cushioning, Nigg and Lieberman both agreed, for their different reasons, and I think on the side of massive motion-control, it's become increasingly clear that we don't need all the devices that used to be common.  The shoe industry has already picked up on this, incidentally, and the number of heavy, bulky shoes available has, at least in my estimation, come down enormously compared to a decade ago.

The practical approach

The final point of debate was the practical approach to transitioning barefoot.  It was a thread throughout the debate, and right upfront, Simon Barthold asked me the question "If I were to design an experiment to test barefoot running, where a group of runners will do 45 minutes of barefoot running, would my University's Ethics Committee approve that research?".

The answer of course, is no, unless they didn't know any better, because we know that 45 minutes of barefoot running in a population of shod runners is guaranteed to cause injury!   This was put forward to Barthold, presumably to illustrate the risks of barefoot running, which is quite true.  However, it doesn't say anything about whether barefoot running is good or bad - that's a separate question.  For example, if I wrote a proposal saying that I would be putting a group of overweight heart-attack victims on exercise programmes consisting of 30 min a day, that study would also be rejected, but we know that exercise is excellent and even prescribed for this group!

So the point is that it's not bad just because it's risky.  It's that it's risky.  Simple as that.  There is risk and reward, and the practical implication of this is "How do I make the transition?"

This is where, once again, I believe it's vital to recognize the skill aspect, or at least the learning process, and to understand that we don't all learn the same way (and nor should we).  Daniel Howell was of the impression that going barefoot first is the best approach.  Others, like Barthold, would advocate that you run in minimalist shoes first, lightweight trainers perhaps, then racing flats, to manage the transition.  There is really no right or wrong answer here.  I think it can work either way, as long as one is very cautious.  

You could, for example, build up to say 40 minutes over 3 months, but basically viewing yourself as a beginner runner, starting out with something as basic as 1 min run, 1 min walk for 10 minutes.  And then systematically increase as you adapt.  Or, like Lieberman did, you can do your normal run, but within sight of home, just take off your shoes and finish the last few minutes barefoot.  Do this every second run, each time from slightly further out, and you'll be up to a full run in about the same time.

Change management and running form

I think the key is that while there is no prescribed way, there is a concept, and the concept is that you have to manage the change as though you were doing a training regime for the very first time.  It's almost impossible to tell a guy who is running 70km a week to go back down to 10km for a few weeks.  He won't do it - he might try, but he'll still err on the high side, and then I think many runners will become injured as a result.  So again, it takes recognition that barefoot running is not the solution simply because it's natural, but rather that it has to be learned and adapted to, and then not to simply run barefoot because it's natural and assume that it'll work itself out.

For example, I think it's important to condition the calf muscles before even running.  I also think you have to be aware of over-striding and avoid the temptation to actively force the landing onto the forefoot.  Let gravity handle the landing.  In fact, I think the worst thing to do is to cognitively tinker with running technique, particularly how the foot strikes the ground.  I think incremental change will work for most people, whereas wholesale changes that work at a cognitive level equal disaster for most (which is the problem I have with Pose).

There are many other points about running form, and this is probably where this debate will go in future.  Nobody knows what "perfect running form" is just yet, and the problem is that it may be individualized based on a set of say 50 different inputs.  So what is perfect for me is unlikely to work for you, and this is the reason that some runners are injury free and others are not, I suspect.  A runner with glut. medius weakness for example, might succeed with one form, but will fail using "perfect" or better running form, and so on.  Injuries are multi-factorial (flexiblity, imbalances, strength etc) and so running form to prevent them will certainly be multi-factorial too.

However, I do think it is wise to at least consider HOW you run.  As mentioned, barefoot running is not by itself the answer.  It's a means to discover the answer, perhaps, and for some people, it may go on to become the solution.  But for most, it's a good way to accelerate the discovery of better running, to strengthen and condition differently, and then to benefit from that later on.

Conclusion - evidence to fill the space between what is known and needs to be known

To wrap up the debate, I said something along the lines of that at that moment, there is a great debate going on, but with many gaps.  There is a space between what we know and what we hypothesize, and that gap will be filled by future research.  Some of that is on the go already - my lunch with Lieberman was heavily focused on research that he is now doing, and the research that I will soon be doing to get to the bottom of the 'skill' aspect of barefoot running (and thus running as a whole) and also on the long-term injury prospects of barefoot running.  That research is coming!

In the meantime, this kind of debate is very valuable, if anyone was there and has some feedback or comments, I'd welcome them.  I'm sure my perspective from the round-table will differ from yours in the audience.  So as always, thoughts welcome!

UKSEM wrap

The next thing to do is to discuss UKSEM Day 2, which is the day that featured some of the highlights of the conference.  Prof Yorck Olaf Schumacher presented on the biological passport, Daniel Coyle presented on better ways to practice and learn, and so I need to summarize those.  And of course, there was David Millar's excellent talk on his doping.

But that will come in due course!


For those not yet saturated by the barefoot topic (which I'll leave alone for now!), check out the following articles from this site:

This is actually quite a cool concept, the Facebook Q & A, so look out for more of those in the future!