How Fatigue Affects Running Form: Implications and Applications

  There are many changes in running form that are seen during running induced fatigue.  It is logical to assume that these adverse changes in mechanics may increase injury risk due to muscles decreasing ability to absorb energy (Mair et al., 1996, Dutto et al, 2002).  Whether that hypothesis is true or not is debatable.   Some deterioration in form may be an intelligent strategy designed to decrease ground reaction forces and loading rates, while other changes may simply be fatigue and thus, take away your ability to properly stabilize.  It appears to be a double edged sword.  If you have the interest/attention to read this article in it’s entirety, you will see what I mean.

  This blog post will first look at the changes in running form that have been revealed by research.  Following each change in running form will be implications and applications of that knowledge that can potentially make you a better runner with less injuries.

  There are many nutritional and physiological aspects of fatigue that are intriguing, however they are not topics that I would consider to be anything I am particularly knowledgeable about, despite some formal educational background in those topics.  So, I will focus strictly on the biomechanical changes that occur to most runners when fatigued.

1) Pronation:

  If you have followed my blog or twitter feed, you will already understand the very tenuous link between “over pronation” and injury.  It is not as straightforward as it would seem (see a lengthy explanation here), so read this section with that in mind…

  I have found 4 studies that found rearfoot eversion increases with running induced fatigue (Dierks et al., 2010, Dierks et al, 2011, Cheung et al., 2007 and Koblbauer 2013) (rear foot eversion is technically not pronation, but for the purpose of this blog, it will construed as pronation)

  The Koblbauer article found an increase of 1.6 degrees of pronation in the non-dominant foot and 1 degree increase in the dominant foot, whereas Cheung found a 6.5 degree increase in rearfoot eversion.  However, in both studies, they placed the markers on the shoe, rather than the skin.  It has been shown in multiple studies that marker placement on the shoe increases apparent rearfoot eversion (Reinschmidt 1997, Stacoff 2000).  Regardless, since the marker placement was the same in the fatigued state and the non-fatigued state and there were net changes in both studies, we can infer that there was at least some increase in pronation/rearfoot eversion.  By comparison, Dierks 2010 found a net 1.5 degree increase in rearfoot eversion during a fatigued state and they cut holes in the shoe in order to use skin placement for the markers.  I am unclear as to why the Koblbauer and Dierks study showed similar results, but Cheung was much different.  I have been unable to attain full text of the Cheung article.

Implications:  Increasing pronation by 1.5 degrees may be statistically significant, but whether 1.5 degrees is relevant for injury is debatable.  For the vast majority of people, I doubt 1.5 degrees of pronation will be the impetus for injury.  On the other hand, an increase of 6.5 degrees would likely be more prone to cause injury, however as I stated earlier, the relationship of pronation to injuries is tenuous and we don’t know the threshold of when pronation becomes “overpronation” or when it even matters.  An increase of 1.5 degrees may be “statistically significant” but whether or not it is “clinically significant” is a different issue.

Applications:  If you are suffering from tibialis posterior tendiopathy or shin splints, then pronation may be more relevant to you.  If you are concerned about limiting pronation, then you may want to use a shoe with more stability/pronation control on longer runs when you will be more fatigued and possibly pronating more than usual.  On the other hand, I wrote a blog post about the “Unintended Consequences of Limiting Pronation“, so you may want to read that before you start switching your shoes up.

2)  Forward Trunk Lean

  There is really only one relevant research paper on this topic, but another worth mentioning.  The one only worth mentioning is a study “way back” in 1980 by Elliot.  8 collegiate runners kinematics were examined and it was found that following 2900 meters, they tended to “carry the trunk further forward during the running cycle”

  In the other, more significant paper (to which I will reference often in this blog) was by Koblbauer 2013 and found that the trunk flexion angle increased from 9 degrees in the pre-fatigued state to 13 degrees when running induced fatigue set in.  A near 50% increase in trunk flexion angle in the fatigued state.


   The “one size fits all” approaches such as Chi, Pose and Evolution all promote a “forward lean”. In my experience, runners tend to implement this as a forward lean from the waist. Proponents will argue that runners are instructed to lean from the ankles, but my experience tells me it doesn’t usually work that way (and yes, I’ve taken both Chi and Pose clinics out of curiosity). Even if it did, a forward shift of the center of gravity would place even greater strain on the muscles in the posterior chain (particularly the low back extensors and gluteals) if the runner is leaning from the waist. As we see in these two studies, fatigue increased trunk inclination during running. If runners are already arbitrarily told to lean forward in a “one size fits all approach”, they are essentially being told to exaggerate the way they will get when they are fatigued.  As always, running form advice should be done on a case-by-case basis, rather than a “one size fits all” approach


   It would be very easy for me to say that runners should start performing trunk stability exercises and increasing trunk extension endurance.  However, the very interesting thing about the Koblbauer study is that “On the contrary, a positive relation between kinematic changes and core endurance was found”.  In other words, the runners “who displayed better core endurance exhibited larger trunk kinematic changes when fatigued”.  Yes you read that correctly, those who had better core extension endurance (as evaluated by this test) ended up losing their posture more than others.  The authors have theoretical explanations:

  1. That the trunk lean is compensatory for fatigue elsewhere in the legs (much more on this later)
  2. Their measure of core endurance was done statically, whereas running is dynamic, so perhaps static testing is a poor measuring tool.  They suggest that strength, rather than endurance should have been measured (again, much more on this later)
  3. That when you combine #1 and #2, a strength program for the lower extremities would be a better way to combat the form deterioration typically seen in running induced fatigue (umm…you guessed it – more on this later)

3) Decreased Stride Length and/or Stride Frequency (stride rate)

   Our running speed is determined by the length of each stride multiplied by the frequency/rate of our strides (cadence).  In other words, Velocity=Stride length X stride rate.

   For the most part, studies show both stride frequency and stride length decrease as we fatigue, resulting in slower velocity.  However, some of these are variable. For example, Hanley 2006, Verbitsky 1998, Girard 2013, Landers 2011 and Hobara 2009 all reported decreases in stride frequency in fatigued states, while Nummela 2006 showed no change and Elliot 1980 showed an increase in stride frequency when fatigued.

   With respect to stride length, the same variability is seen; Elliot 1980Nummela 2006, Girard 2013, Landers 2011 and Hobara 2009 all reported decreases in stride lengths in fatigued states. However, Hanley 2006, Williams 1991 and Siler 1991 showed increases in stride length.  However all those studies that showed an increase in stride length during fatigue were performed on treadmills.  Hmmm…..

   When the numbers are pooled together in each individual study, we see trends toward decreases in stride length and stride frequency, however individuality exists with runners.  For example, Dutto 2002 found 10 of the 15 runners had statistically significant changes in stride frequency and they ranged from –3.7 to 4.4%.  Yes, that means that as fatigue set in, some runners increased stride frequency and some decreased


   Different people employ different strategies with stride frequency and stride length as fatigue sets in.  There are many theories as to why this is.  Some authors feel that when a decrease stride length and frequency is seen, it is due to neuromuscular fatigue.  Others have stated that increasing the stride length during fatigue is a strategy designed to reduce ground reaction forces via reduced leg stiffness.  This is contrary to others such as Verbitsky 1998 who found that by decreasing stride rate, the amplitude of the initial shock wave was increased.  Theories are out there, but the fact of the matter is, we just don’t know.


   There is a trend toward decreasing stride length and a stronger trend toward decreasing stride frequency, but individual variations exist.  Because of these trends and because of the proportionally larger decreases in stride frequency vs. stride length, it has been suggested that runners focus on maintaining their stride frequency as fatigue sets in.  For example, in the Hobara 2009 paper, they found a 17% decrease in stride frequency and a 10% decrease in stride length.  The authors concluded, “When runners feel fatigue, they might benefit from paying more attention to the maintenance of stride frequency, rather than focusing on maintaining stride length”.

   Much of this deterioration in stride frequency may be due to a deterioration in the “spring-mass” characteristics of running.  This model is further explained in the next section on increased ground contact time.  By improving neuromuscular recruitment through periodized strength training, we can reduce the loss of stride length and other factors.  This was shown by Esteve-Lanao 2008 who witnessed stride length reductions during running induced fatigue in those runners who did not do a periodized running-specific strength program and saw no reduction in stride length in those runners who did do a periodized running-specific strength program twice per week for 8 weeks.

4) Contact time:

   Ground contact time (GCT) is the amount of time your foot spends on the ground during a gait cycle   There are many studies (and yes, this is consistent) that GCT increases as fatigue sets in – Hanley 2006, Girard 2013, Hobara 2009, Elliot 1980, Nummela 2006 and others.


  GCT is a contentious topic, in my opinion.  I made a different blog post about it earlier.  On one hand we have good evidence that more elite runners tend to have shorter GCT than slower runners.  That has convinced many coaches to implore their runners to try and shorten their GCT.   However, reducing GCT comes at an increased metabolic cost (which may be why GCT lengthens as we fatigue).  This is great if you’re a sprinter, but a long distance runner?

   In addition, reducing GCT means that you have less time to generate the same amount of force (i.e. higher rate of force production) which could potentially increase risk of injury.  Shorter GCT has been linked to increased risk of injury in at least one study – Bredeweg 2013.


   The spring-mass model of running is an important concept here, but I’m sure attention spans are waning at this point of the article, so I’ll keep it brief.  Essentially force production during running can be seen as a series of muscle and springs.  We get a lot of force production from the “springiness” of our tendons.   The Achilles tendon recovers approximately 35% of the mechanical energy that the body generates with each step while the arches of the foot can get about 17%, as cited by Perle 2012.  The better the muscles “stiffen” at impact, the more energy storage and release by the tendons.  If the muscles have a poor ability to stiffen, there is less “springiness”.  Less stiffness in the legs (and the body in general) leads to longer GCT.  I made a short video about this:


   My personal viewpoint on this, (which I elaborate on much more below…) is that consciously trying to decrease GCT is a bad idea.  However, by training the neuromuscular system to better recruit muscles and training explosive strength (via heavy weight lifting, plyometric and sprint training), your entire neuromuscular system will be better tuned to naturally maintain proper GCT as you fatigue.

5) Compensation Strategies for Injuries

   While researching this article, I found this part to be pretty cool:

   When looking at running mechanics, there are a few altered mechanics that correlate to injuries.  For example, patellofemoral pain syndrome (PFP) is correlated in both retrospective and prospective studies to increased knee valgus (the knee travelling toward midline).  However, in a recent study, Dierks 2010 found that the PFP runners actually compensated very well at the beginning of the run, but as fatigue set in, they became unable to hold their compensation strategies.  The authors state, “the runners in this PFP subgroup seem to have been compensating by limiting motion in their joints in an attempt to stabilize the leg to prevent further dynamic malalignment and reduce pain. However, the peaks of these motions, including knee valgus, increased at the end of the run compared with the beginning, which corresponded with increased pain at the end of the run.”


   This is fascinating to me as someone who is very interested in biomechanics and injuries.  There are many refuting studies out there on linking faulty biomechanics and injury.  Perhaps researchers need to take a step back when doing their research and look at runners in a fatigued state in order to get a better appreciation of the way runners get injured.  Maybe we could all get a better understanding of linking injuries to faulty mechanics.


   If you have an injury history and you are getting a gait analysis, try and get yourself filmed when in a fatigued state (unless your injury prevents that).


   So with respect to the application of this knowledge of longer GCT, loss of posture, increased pronation, decreased stride length and rate, I have tried to take a global look at all of it together.  My conclusion is similar to Nummela 2006 who had had an interesting viewpoint.  In their study, they concluded that neuromuscular force production plays a large role in fatigue.  They contend that physiologists traditionally point to limitations of endurance capacity being mediated by either the idea of the central governor (brain limiting endurance capacity in order to maintain homeostasis) or the limits in the body’s ability to transport and utilize oxygen (Vo2 max).  However, Vo2 max cannot fully explain all the differences seen in endurance capacity.  They point to various studies that show improvements in neuromuscular recruitment via strength training lead to a concurrent increase in endurance running capacity.  They concluded that their literature review, as well as their own research, supports the idea “that the neural control and the ability of the neuromuscular system to produce force and power provide additional information to the energetic model of distance running performance”  They also state that, “neuromuscular capacity to produce power is more decisive than oxygen utilization in distance running race during the final lap.”

   The Koblbauer 2013 article had similar conclusions (remember, this is the recent study that found runners bend forward at the trunk as fatigue set it, however, it was worse in those who had better core endurance).  They stated that measuring the static endurance may have been a poor test, given the dynamic nature of running.  They point to the Leetun 2004 paper – a prospective study that found hip strength was much more applicable to runners than trunk endurance.  The authors of that study concluded, “in a more athletic population, this study suggests that isometric hip strength measures, particularly in external rotation, are more accurate predictors of back and lower extremity injury than trunk endurance measures.”

   If we try and group everything together, we can look at the loss of posture, increased GCT, increased pronation and decreased stride length and rate as caused by one main source: loss of limb/body stiffness.  With regard to this, there are two unanswered questions in my opinion: 

   1) Is the loss of limb/body stiffness a result of neuromuscular fatigue, inhibition via the central governor, or limitations in oxygen uptake and utilization…or a combination of all three? 

   2) Is the loss of stiffness consciously adjusted, or is it affected by fatigue.  In other words, is it beneficial, or detrimental?  Perhaps the increased loading rate that goes along with increased limb stiffness contributes to injury.  What I am saying is, if you are able to maintain shorter GCT via limb stiffness for a longer period, you may be more vulnerable to injury.  So maybe we make conscious decisions to lessen impact rate by reducing limb stiffness.  There is evidence to support this:

  1. We know pronation is a way we absorb shock, therefore, increasing pronation when we are fatigued may be a strategy to lessen loading rates
  2. We know that increasing trunk flexion during jumping is a strategy to improve shock attenuation (Saha 2008, Weinhandl 2011)
  3. One study on 87 female runners found that as runners fatigued, ground reaction forces (GRF) decreased – as we would expect, given increased trunk flexion, increased pronation, longer GCT etc.  However, there was a link between runners with previous injuries and higher GRF when they were fatigued.  In other words, the trend was for all fatigued runners to have decreased GRF, however, those with previous injuries didn’t decrease their GRF’s as much as those who did not have previous injuries

   In its entirety, the take home message that I have come away with (please let me know if you have come to different conclusions) is twofold:

  1. We see biomechanical changes in running induced fatigue that seem to indicate a strategy to reduce impact.  This is likely because of muscle fatigue.  In other words, muscles are supposed to absorb shock, but as they fatigue, we utilize other strategies such as increased trunk flexion, increased pronation, increased GCT, decreased stride length etc.
  2. That strength training, plyometrics and explosive speed training can assist in your nervous system’s ability to better recruit muscles and thereby be more efficient in warding off the biomechanical changes typically seen in running-induced fatigue.