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.

Monday, June 06, 2011

Barefoot running, shoes, and born to run

The barefoot running debate: Born to run, shoes & injury: the latest thinking

One of the more interesting, and certainly topical presentations at the recent meeting of the American College of Sports Medicine in Denver was a symposium on barefoot running.  It was led by Irene Davis and Daniel Lieberman, both advocates for barefoot running and top scientists in this field.  Lieberman in particular is something of a 'legend' in the field, and two years ago, he gave the prestigious keynote address at the ACSM meeting.

Since then, the area has moved on, thankfully.  Most recently, Lieberman's group did some fascinating work on the barefoot running concept in runners accustomed to shoes or running barefoot, and that's the focus of this post, along with some thoughts on the concepts underlying barefoot running.  A lot of the time, I'll play devil's advocate, because I believe in Lieberman's findings, and the theory behind barefoot running is sound.  But there are some "loopholes", and I'll end with those.

Not just a fad, and certainly not only for the niche

There are more than a few people who have dismissed barefoot running as a fad.  And many will have labeled it a niche concept, practiced by a very small percentage of runners.  That's only partly true.  If you think that barefoot running has nothing to do with you, think again.  You may not have discarded your shoes, but the truth is the the shoes you are running in have already been influenced by the concepts that drive the barefoot running movement.

That is, the last decade, which has seen more and more evidence come out AGAINST shoes, has also seen a shift in the shoe industry.  Gone are the heavy, bulky motion-control shoes, replaced by shoes that are now marketed to simulate barefoot running.  Nike were apparently the first to do this, though I remember Adidas bringing out "feet you wear" in the 1990s.  But it was the Nike Free that was the first "barefoot shoe", and Irene Davis, in her ACSM presentation, told the story that a famous college coach in the USA was responsible for this because he told a Nike rep who had come to watch his team train that his runners were more comfortable being barefoot. 

The rep rushed back to HQ, reported on the athlete's preference, and it heralded the shift.  Now, almost all the companies are focusing on the 'minimalist concept' of shoes.  There are even new companies (Vibram, Newton).  Of course, there are a few stubborn survivors, but the whole market has shifted, there is no doubt about it.  Why?  Because of the current thinking around running, and the role of footstrike, and our feet, in injury risk during running.  So we're all affected, even if we run in shoes, and here's the theory.

The Born to Run theory - our survival was dependent on being endurance running champs

Barefoot running is fascinating because it has made such a big impact in the mainstream.  And a big part of that is the book "Born to Run", by Christopher MacDougall.  It traces the story of the Tarahumara Indians, and contains chapters dealing with the work on barefoot running - Irene Davis and Lieberman have starring roles.

The fundamental premise behind barefoot running is the theory that we are designed for distance.  Lieberman's first big impact on sports science was his research called "Endurance running and the evolution of homo", where he presented the anatomical evidence that humans are the world's best distance runners.  Let's face it, as sprinters, we're pretty lousy.  Even the super fast Usain Bolt runs half the speed of most big mammals for about a quarter of the duration!  Therefore, if our survival as hunter-gatherers depended on our ability to chase and catch prey, we were pretty much doomed by our lack of speed.  That comes about as a result of many factors (I have a colleague doing research into the muscle and metabolic differences between humans and other animals), but that's for another day.

Compounding this is that we didn't have projectile weapons until fairly recently, so we had to catch our prey up close.   But fortunately, our survival didn't depend on our sprinting ability.  What it did require was endurance ability, and that's where we are the champions.

Thermoregulation, endurance success and persistence hunting

Some of the most important reasons we are so good at endurance relate to thermoregulation.  We are bipedal (which makes us slow), but the linearity is good in terms of helping us lose heat and absorb less from solar radiation.  We are small, we have less hair, and most importantly, we can sweat.  This allows us to exercise at fairly high workrates without reaching what is now known to be a critical body temperature that would force fatigue on us.  This is true, incidentally, of animals, who also stop exercise at a very narrow range of temperature.  And animals can't really lose heat WHILE running - they have to pant, which is not possible during exercise.  So running at the same speed, we have a big thermal advantage.

All of this combined leads to the theory of persistence hunting, where humans are able to hunt by outrunning the intended prey over hours, rather than seconds.  There are a few good videos showing this - the hunter selects to hunt at the hottest part of the day, and then identifies the biggest animal and runs after it.  Over seconds, it's a no contest, but the hunter just continues to run, driving the animal out of the shade and into the sun.  This continues for hours, but eventually, physiology wins the day and the animal reaches that critical temperature.  If you have ever seen what happens at this point, you'll know the scene - the animal staggers as if drunk, losing motor control and eventually, it just stops completely, lying down for the hunter to make the kill.

What's the relevance of all of this, you may be wondering?  Well, this leads to the theory of barefoot running, because Lieberman and others argue that humans are designed for distance - it was essential and the "normal" state.  By extension, wearing shoes is "abnormal", and so we must return to what we were intended for - long distance running WITHOUT shoes.  And that's where the evidence comes in!

Shoes and injury

One of the main lines of evidence is actually a lack of evidence.  That is, there is not a single study that has shown that shoes reduce the risk of injury.  In fact, there is now evidence that the risk of injury is unchanged, maybe even higher in the expensive shoes.  Injury rates haven't come down even a little bit since the 1970s, the period which has seen the explosion in the running shoe industry - back in the 70s, the shoes resemble today's minimalist shoes, but about 70% of runners were getting injured, the same rate as today.

There are studies showing that runners who run in the more expensive shoes are more likely to get injured, even when you correct for distance and previous injury history.  There is also evidence that prescribing shoes according to the shape of the foot (high arch gets a neutral shoe, flat foot gets a motion-control shoe) does nothing to the risk of injury in a sample of 2,000 in the military.

So all in all, it doesn't look great for shoes.  There are a couple of confounders in this line of thinking, which I'll address at the end, but the lack of evidence for shoes doesn't hurt the idea that maybe the foot is best, and we should be discarding the shoes and trusting our feet.

The evidence for barefoot running

But what the theory needs is evidence.  And some of that has come from Lieberman, and was published in Nature last year.  He looked at barefoot and shod running with the important realization that familiarity (which may determine "skill") would have an important effect on how their mechanics (and hence impact forces) might be altered. He therefore tested two groups, one which was habitually shod, and another that was accustomed to running barefoot.

Below are some graphs showing the key concepts, and this is really the best evidence for what Lieberman and Davis are saying, so it’s worth explaining carefully (these graphs came up in all three presentations, it is really the only research so far, but it's very compelling).

So the first graph (above) shows the force measured during a single ground contact period in a person who is barefoot and heel-striking. I’ve highlighted the key point on the graph – the impact transient, a spike in the force up to around 2.5 times body weight within the first 50ms of ground contact. That impact transient represents a sudden increase of force, and the rate at which it is applied (which is basically the slope of the line from contact to that peak) as well as the size of that force are associated with injury, in particular stress injuries of the tibia (Irene Davis spoke of this at length in her presentation).

Shoes - spreading the load

The concept of shoes is that they spread that impact out, absorbing some of the force and reducing both the size of the impact transient and also the rate of loading. That’s what the graph below shows. The top panel is the barefoot situation where you land on the heel, as shown above, and the bottom panel shows what shoes do.   I’ll show more data shortly.

 Forefoot running - the body's cushion, removing the transient

The barefoot condition in someone who is forefoot striking is shown next.  Here, the impact transient has disappeared – the forefoot landing allows the impact peak to be absorbed by the calf, ankle and Achilles tendon and the result is a smoother profile, and significantly reduced loading rates and forces.

 The problem - not all barefoot running is equal, and the consequences are extreme

So that’s great news, right? It says that barefoot running is the way to go? In theory. The problem is that making that transition from wearing shoes to running barefoot involves significant risk, and Lieberman's research shows why.  And if you’re not careful, then you actually end up INCREASING the loading rate and the impact force, because you run “badly” without the protective elastic shoe dissipating that force as was shown in that middle graph above.

So below are the two key graphs. First, a graph of impact force in three conditions. Start in the middle – the yellow bar is the impact force running in shoes, while heel-striking, which is what most people do.  In fact, Lieberman found that 100% of people who were accustomed to shoes would land on the heel when they had to run in shoes.  

Now move right – that green bar is the force for people who run barefoot, but who are “good at it”, and have become accustomed to it.  Lieberman found that in this group, the regular barefoot runners, about 90% land either on the forefoot or midfoot.  As explained, their impact forces are significantly lower. 

But those who are not accustomed to running barefoot are the key group, shown in blue.  83% of runners who usually run in shoes would land on their heel even when running barefoot, at least at first.  And shown by the blue bar, their impact force is very slightly higher.

But that’s not the major difference. That comes when you look at the impact transient and the initial rate of loading, shown below.

Now you see that if you run barefoot and are NOT ACCUSTOMED to it, you land on the heel, as shown by the blue bar (again, 83% of his normally shod runners did this).  And the result is an impact loading rate that is SEVEN times greater than running in shoes with the same landing.  If the theory is that this impact transient, and the rate of loading, is related to injury risk, then what this is showing is that people who are accustomed to shoes who then run barefoot (perhaps following unqualified advice in a magazine) are exposing themselves to higher risk of injury as a result of massively increased loading rates.

So that’s the first problem – the purported reduction in injury risk (which is probably real) is dependent on the technique.  If you get it, and land on the forefoot, the impact transient and loading rate are lower (see the green bar above).  But if you get that wrong, the consequences may be severe (further research is required on this).

Learning - skill will improve and mechanics will be adjusted.  But not without risk...

Now, we’re pretty amazing machines, and I can assure you that most people will be able to learn how to run barefoot effectively. That is, with a little practice, they’ll begin to make the adjustments that move them from that high risk group who heel-strike to a more cushioned forefoot strike.

You can experience this for yourself if you take off your shoes and run – within 50m, you’re already starting to make those adjustments, because running barefoot on your heels, well, it hurts.  So we "feel" the ground, more than we want to at first, and over time, we adjust our mechanics and shift to a forefoot landing.  That's why a comfortable majority - 90% - of people who are accustomed to running barefoot will land on the forefoot.

But that brings us to injury risk number two – the adjustments you make to help cushion the landing have consequences. They transfer the impact forces and loading to the posterior muscles, those muscles at the back of the leg. You land with your toes more pointed away from you, and so your calves are contracted, and catching body weight every landing.  You may reduce the impact forces, but those forces that are left are being applied to very specific muscles.  Muscles that may have never been asked to do that kind of work.

The eccentric load on the Achilles tendon and calf muscles is enormous. Your feet and ankles are working much harder, doing what is probably their job, what they are designed for, but they haven’t done it for many years, perhaps a lifetime! The end result is that the strain on the muscle and tendon systems is just enormous, and people break down very quickly.

This is the same outcome as you get from introducing things like Pose and Chi running either too quickly or incorrectly.  Your mindset is "land on the forefoot" and that's a pretty tough skill, and it's a pretty challenging demand to cope with for muscles that have for years not been asked to do it!  My own personal experience reflects this - I was regularly running upwards of an hour in shoes, no pain.  I went running in Vibrams, and did 30 minutes consisting of 1 min jog, 4 minutes walk.  And for three days, I could barely walk, my calves, hamstrings and glutes were so tight - it was typical DOMS from eccentric loading that muscle was unaccustomed to.

I know a few who have experienced the same thing.  And if they overcome the initial stiffness, there is a big risk waiting a week or two down the line, because the constant load on unfamiliar muscles is a huge factor that sees a lot of people get calf or Achilles injuries when trying to run barefoot.

It shows that a) running in shoes requires next to zero eccentric work in those muscles (relative to shoes), and b) when you start barefoot running, you may have to go way back to basics, beyond beginner level, in order to undo what 20 to 50 years of shoes may have caused. 

Making the switch - the practical problem

And herein lies the key - the shift to barefoot running is theoretically sound.  Lieberman and Davis have a compelling case.  But practically, it may be too big an adjustment for some to make.  People have different abilities when it comes to learning a skill, and skill is what it will take to shift out of the shoes.  It also takes different mechanics, possibly muscle strength, and so out of 100 people, it is quite conceivable to me that you'll get a range of responses.

Some will respond brilliantly, and will be able to run big distance barefoot and have no problem.  Some will really battle, and may have to return to beginner level, but they'll learn it.  Eventually.  Whether they have the mindset or willpower and discipline to do it is another question.  Irene Davis presented her recommendation, which started with taking someone up to 30 minutes of barefoot walking, followed by progressive introduction of jogging.  She started at 1 min jog, 9 min walk for 30 minutes.  That eventually worked its way to 9 min jog, 1 min walk.  Which is all good and well, but if you're running 40 to 80 miles per week, the idea of going that far back...not likely.

Then out of our 100 people, I do think there are some who just will not succeed barefoot.  Perhaps they have a structural problem somewhere, perhaps 20 to 40 years of shoes have caused changes that simply cannot be reversed.  Perhaps they have muscle weaknesses elsewhere, and barefoot running is not sufficient to overcome those.  One of the problems I have in this debate is that those who are advocating barefoot running are basically treating it as medicine (if you have a condition, take a drug.  If you have an injury, take barefoot running...).  The problem is that this is not done with any idea of dosage, "contra-indications" or exclusion criteria, and some might just not be able to do it, which makes 'one-size fits all' advice unhelpful.

The point is that there should never to a single approach to an injury problem.  I think there's no doubt that someone who is chronically injured may have their best chance in trying barefoot running.  But this cannot come at the expense of a holistic view (and I must emphasize, Davis is not guilty of this), and so I'd put barefoot running forward as part of a solution, something to try, and if you are one who succeeds, then go with it.  But if not, then look elsewhere, and don't worry, it's not as simple as some are suggesting.

High performing runners - barefoot may just not be practical

The other group that I have to mention is high-performing runners.  Not necessarily only the elite, but even those who train for a fast marathon, or Ironman, or even high mileage 10km runners.  They are doing 120 to 200 km per week, and a lot of it is fast running.  Much faster than persistence hunting would have required - obviously, there is no data, but it's unlikely that humans have ever tried to cover 200km in a week, a lot of it at 7 min/mile or faster, even getting down under 5 minutes/mile for long periods.

If you are doing this kind of distance, then there is a real question over barefoot running.  Some will argue that the reduction in loading rates and impact transients makes it more likely that you can succeed barefoot, especially at high mileage.  But there's a confounder here too - muscle fatigue.  The third presentation in the symposium showed some really interesting evidence that the loading on the joints and bones was HIGHER as muscles fatigued.  This stands to reason, of course - muscle absorbs much of the impact force, and so tired muscle loses that ability, exposing the joints.

So those who are training for performance may struggle because of a muscle fatigue issue - the muscle is working differently, and harder in certain muscles, when barefoot, and that may be limiting.

The bottom line - individualized approach, but sound theory

On the whole, barefoot running, or at least minimalist shoes, is a sound concept. Lieberman’s theories regarding our ability to run are solid, and I do believe that the days of bulky, motion-control shoes are numbered. I think that barefoot running will be very difficult to implement, if not impossible, for some people and probably doesn’t work in the extremist view that some people are offering for it.

I believe that it may be PART of the solution for SOME of the cases of injury. For SOME, it may be ALL of the solution, the solve-all. For others, it will be completely ineffective, and for other still, it will be the cause of their problems.


So if I had to give some recommendations, speaking now both as a scientist and a coach:
  • If you are are injured, or struggle with chronic injury problems, then give barefoot running a try.  It may be especially helpful if you have knee problems, or any anterior injuries (anterior shin pain, for example), because going barefoot will switch the load. 

    The benefit (in a backward kind of way) of starting injured is that you're pretty much compelled to go back to beginner level and build up, so your chances of staying beneath that "injury risk line" are better!

  • If you are not injured, but fancy trying it, then by all means, go for it, but be very careful.  I would suggest the best way to approach it is to think of barefoot running as a training modality.  Just like you'd go to Pilates to improve core strength, or spend time in the gym on upper body or leg, think of barefoot running as a session. 

    There is some evidence for this, incidentally.  Pieter Bruggemann (of Oscar Pistorius testing fame) actually did a study on the Nike Frees, and found that just using them in the warm-up improved lower leg strength, balance and agility within five months.  So this points out one way to do it - do warm-ups either barefoot or in minimalist shoes.  Or do five minutes at a time, building very, very slowly, so that you don't affect your other training but gradually develop the strength.  If you find that you enjoy it and don't seem likely to be injured, then push on.

  • If you are a high-mileage runner, then think carefully about tinkering with barefoot running, or about changing your technique to land on the forefoot.  For one thing, if you are a competitive runner, or even aspiring runner (going for PBs, that is), then you're the person most likely to overdo it!  It's part of what makes you competitive!  So again, I'd advise that you consider incorporating barefoot running into the programme as a training aid, because it will help your feet, calves and ankles.  But as with any training aid, phase it in very slowly.

  • If you are constantly battling calf, hamstring, foot or ankle problems, then consider barefoot as treatment, but take the most conservative guideline you can think of and halve it - do 50% less than what that says.  The rationale is that someone with a hamstring injury can't just avoid strengthening the muscle - it's part of the rehab.  So for chronic calf and ankle/achilles problems, running barefoot may be exactly what you need.  But you are the kind of runner who has to start with five minutes of walking, not running, and hold back massively.  If you can succeed, then hopefully this will help your return to running in shoes, and maybe, eventually, minimalist shoes.
Then, no matter which category you fall into, golden rule - try not to control your landing.  You shouldn't be thinking about how your foot hits the ground.  Just "listen" and "feel" the contact, and your body will gradually adapt it for you.  The mistake that is made is to think "forefoot" and this leads to excessive loading, as explained above.  You will occasionally land on the heel, but the natural response is to cushion the landing.  The danger is trying to catch the landing, so just relax and focus on not reaching, not pointing, not catching.

    And I’d end my opinion on barefoot shoes by just cautioning its advocates to avoid making the same mistake you are accusing the shoe industry of having made for many years. The shoe industry, it is said (with good reason) advocated for many years that shoes were the “answer” – one size fits all (pardon the pun), and that simply putting someone in the right shoe would prevent or cure injury. Now, the barefoot movement is in danger of making the same error – learn from the past and recognize that individuals need individual solutions. So don’t put everyone in a barefoot box.


    P.S.  A footnote...(pardon the pun)

    The devil's advocate view

    Just to end, I want to point out three counter-arguments to three arguments made by Dan Lieberman in his talk.  Mostly, this is to give you some ammunition when you get overwhelmed by a barefoot advocate who is arguing it as the sole solution, the be-all, end-all of running.  It's likely far more complex, and here are some responses...

    1.  Shoes are abnormal, barefoot running is normal

    Lieberman said this often often in his talk.  He says that when we see a guy running barefoot, maybe we should think of him as normal and ourselves as abnormal.  And it may well be true.  But the problem is that "normal" doesn't imply good, and "abnormal" is not necessarily bad.  Think back 100 years, life expectancy was about 50.  Go back another 50 years, and you expected to live to your 40s!  People died young all the time, particularly mothers during childbirth.  Today, it's up near 80 in a lot of countries.  You might say that in 1911, you were abnormal if you lived to 80.  Today you'd be normal.  Is "normal" wrong, or just different?  It tells nothing of what is good or bad.  Sometimes "normal" changes as part of progress.

    The key is that in this case, normal is meant to convey that our feet should be doing what they're designed for.  And that may be true, but it's still short on evidence.  We can't discard things that are "abnormal" just because they're unnatural...

    2.  Our ancestors were great runners, and being barefoot is the reason

    This is key to the debate, because it is being suggested that hunter-gatherers, minus the state of the art shoes, were great runners.  And therefore, the lack of shoes is the reason.

    So two things.  First, I'm not necessarily convinced that all of them were great runners.  Let's take a group of 100 people today.  There may be 50 who can run at a reasonable level - the finish marathons, half marathons, in the top half of the field.  They get injured maybe once every two years, but if they're good about training (strength work especially), they keep going.  30 of our runners might really struggle.  The do 5 hour marathons, get injured often, but can get through on isolated occasions.  10 might be absolutely incapable of running - they get injured in the warm-up and just cannot finish 5km, let alone marathons.  And 10 are the best runners.  One of them is Ryan Hall, an elite.  Nine are good, competitive runners.  They push the limits and get injured, but recover well, and they are the "runners".

    Now apply this to the Kalahari, or the Tarahumara.  Is it not possible that in a village of 100, there are 10 "runners" who are the best, 50 others who are good enough, and that they provide the running for the community.  Maybe there are 40 of those 100 who just suck at running!  They become artisans, doing the building or hut construction, maybe they become fishermen!  But they can't run.  I simply do not know the answer to this one - I don't know, for example, whether 70% of Kalahari hunters get injured every year.  Maybe 2 in 3 Tarahumara get injured, take a few weeks off, and then return to running.  Not knowing this means I'm less convinced by the argument that being barefoot is better.

    Then the next point - why is it that we've identified being barefoot as the key difference between us and our ancestors?  We look at them, and we say they were great runners who rarely got injured (a point I'd question, as explained above).  Then we make the next deduction, and we say "they didn't wear shoes, that must be why".

    Well, I can think of plenty of reasons why they may have been injured.  For one, they didn't sit at a desk 9 hours a day.  They didn't have Playstation and TV and computers.  I mean, I've been sitting down for 90 minutes typing this - my posture and core muscles have been asleep.  But that wasn't true for Kalahari - they climbed trees, slept on uneven surfaces, walked, were physically active.  All of this may lead to improved strength, especially in the stabilizers - the core and glutes.

    That may well be as much a reason for their apparent ability without injury, and barefoot running may only be part of it.

    3.  Injury rates haven't changed, so the shoes clearly don't work.

    The argument here is that injury rates were 70% in the 70s and 70% in the 2000s.  So, despite shoe technology, which is hugely different, we still get injured just as much.  And sure, this is true, and when viewed together with the other evidence about shoes, I think it's a good case.

    But just remember that the people who are running in the 2000s are not necessarily the same as those running in the 70s.  When I think of running in the 70s, I think of Amby Burfoot - small, wiry, probably ultra-economical (sorry Amby!  Thanks for reading!).

    When I think of running in 2011, I think of Oprah Winfrey... sorry Oprah.  Simply, the people who run marathons today are not the same kind of runner.  They are heavier, slower, and probably do less training but then still run the marathon distance.  When viewed this way, it's perhaps not surprising that 70% are getting injured.  If anything, it suggests that shoes may be helping, because many of the runners of 2011 are walking/jogging/running injury risks!

    So if you're going to attack the shoes, then I'd rather do it on the basis of the evidence that they don't change mechanics/pronation, that they don't alter risk in controlled studies, not because of this historical comparison!

    That's all folks.  I know it's a lengthy piece, thanks for reading.  It really is just an overview (this is the summary!).  For more on this, check out the featured series on running technique and shoes here!  I discuss a bit more of that evidence on shoes, which I just touched on in this post.


    PPS:  Thank you for all the comments in the discussion below.  I will do my best to respond, but bear with me if it takes some time!


    Anonymous said...

    Disqus seems to be hiding...