The Definitive Guide to Pronation: Clinicians Version

The intended audience for this article is for mainly for clinicians, but can also be for other people who are VERY interested in foot and ankle biomechanics and have a relatively good handle on anatomy. It is lengthy and in depth.  A trimmed down, less technical version is available here: The Patient’s version.

Please redistribute the link, but don’t copy it and pawn it off as your own…

The purpose of this article is to delve into the issue of pronation and “over”pronation.  I went into this topic out of my own interest in learning more than I already knew. I am not “pro” or “anti” pronation control nor “pro” or “anti” minimalist. But for decades, the biggest factor in gait evaluation for most runners has been the amount of pronation they have.  Promoted by running shoe stores, running shoe companies, coaches, podiatrists and many other clinicians, “over”pronation has been the labeled as the “cause” of a vast assortment of injuries.

Unfortunately, there is a large amount of conflicting clinical and biomechanical conclusions that leave the end user (patient) muddled and confused as to what they should be doing.  In addition, what is known about foot mechanics now is wildly different than what many clinicians were taught in school.  If these clinicians don’t stay up-to-date (any many don’t) they are leading their patients down the wrong paths.

Due to the length and breadth of this article I have created a sort of “table of contents” with hyperlinks to the heading of the article.  If you have any questions, please email at  Keep in mind; I have no horse in this race.  I’m not trying to promote minimalism and I’m not trying to up-sell you an orthotic footbed in a running store.  I tried to list as many valid research articles on each topic as I could and provide a very brief summary of each research paper.

1. Introduction
2. Pronation and overpronation defined
        2.1    Subtalar Axes of Rotation
        2.2   Subtalar Axes of Rotation and Type of Injury
        2.3    Ankle Joint Transverse Plane Motion
        2.4    Influence of Forefoot Position on Rearfoot Motion
3.  Movement Coupling from the Foot to the Leg
4.  Measurement of Pronation in a Research Setting
5.  Attempts to limit pronation
        5.1   Orthotics limiting Pronation
        5.2   Shoes Limiting Pronation
        5.3   Assigning Shoes Based on Arch Type
6. Does Overpronation Cause Injury?
7.  Do Orthotics Reduce Pain and Injury?
8. Summary and Conclusion

1. Introduction

Ask a runner, an employee at a running store, a clinician, a biomechanist or anyone else, ‘What is pronation?” and you’re going to get wildly different answers.

If you want some fun, take it a step further; ask them “When does pronation become ‘over’-pronation?” or even, “When does ‘over’-pronation become a cause of injury?”  If any of them tell you they know the answer to either of the last questions, you now know they are full of it.

If there is one thing to take away from this article it’s that there are few definitive answers or rules.   Is it because I’m ill informed?  Some might say so, but I’d like to think not!  I think it’s because I’ve read too much and seen way too many patients in my office that I can confidently say: “I don’t know for sure.”  If you meet anyone who says they fully understand pronation, tibiocalcaneal joint coupling and foot mechanics, I’d politely disagree.  This includes salespeople, clinicians and even the top biomechanist researchers in the world.

If you’re looking for a definitive statement on what ‘over’-pronation is, when it becomes a cause of injury, or even what the axes of rotation that pronation is supposed to occur around, you won’t find it here.  Why not?  Do I skip over that part?  No, it is because there are loose definitions, completely different methods of research, debate amongst researchers and an ever changing database of studies.  What I will present is the up to date research on those topics, however be advised:  There are many different answers.

2. Pronation and Overpronation Defined

Before you read any further, you need to have a thorough understanding of the cardinal planes of motion, anatomy and terminology.  This video should help most people through this.  Although many of the statements in the narrative are very outdated and just plain incorrect, it helps people orient themselves to the basic 3D motion.

The textbook definition of pronation of the foot is a combination of movements in all 3 planes – eversion, dorsiflexion and abduction.  This is in the open chain (foot not fixed to the ground).  However, in the closed chain segment of running (foot fixed to the ground in the stance phase), the dorsiflexion of the foot is replaced with the tibia moving forward relative to the foot and the abduction is replaced by some internal rotation of the tibia relative to the foot and the axis of rotation for the frontal plane calcaneal eversion moves inferior.

The textbook definition of ‘over’pronation is a little more difficult. Some people will define it as too much pronation (compared to normal), OR too much velocity of pronation (compared to normal) OR a prolonged period of time spent in pronation (compared to normal). There are inherent flaws and questions that come up with this such as: What is a “normal” amount of pronation, what is a “normal” amount of velocity and what is a “normal” amount of time spent in pronation?

Here is just one example of how the definition of what pronation is can vary: When measuring pronation, do you take into account the position of the foot prior to footstrike (i.e. open chain)? For example, let’s say a person has a 12 degree rearfoot varus in a non weight bearing position but when running, they land with a rearfoot inversion of 3 degrees. Near midstance, the runner at max pronation has 10 degrees of rearfoot eversion. Did he pronate with a rearfoot eversion excursion of 13 degrees (3 degrees of inversion at footstrike plus the max rearfoot eversion near midstance) or did he pronate with 22 degrees of eversion (the 12 degree non-weight bearing rearfoot varus plus the 10 degree inversion at max pronation)?

We have no definition of what “normal” pronation is, so how can we possibly define what ‘over’pronation is? This type of ambiguity is frustrating for clinicians and patients.

2.1 Subtalar Axes of Rotation

Note that I have not used any degrees in describing the axes of rotation.  That’s because there is a good amount of debate in the literature on this topic.  For example, Manter (1941), Isman and Inman (1969) and Van Lengelaan (1983) have all described the subtalar joint axis of rotation to be around 41 degrees from the transverse (horizontal) plane with a wide scope of variation of anywhere from 20 to 68 degrees in the sample population they measured.  However, more recent imaging techniques and intracortical bone markers (metallic beads implanted into the bones of subjects) are available and have reported the axis of rotation to be much different.  In fact, Lundberg (1989) reported the axis to be closer to 30 degrees.

Why is this relevant? (i.e. who cares.  Why are you boring me with this?)

If the axis of rotation for the subtalar joint is truly around 41 degrees that means that if the subtalar joint pronates in a closed chain position, the tibia will internally rotate close to a 1:1 ratio with the subtalar joint.  This refers to the way too often cited “miter joint” analogy which if true, has huge implications (but it’s not true).

It implies that if you pronate excessively, you will also internally rotate the tibia excessively in a 1:1 ratio with the subtalar transverse plane motion and thus end up with knee and hip rotational torques and possibly injury.  This is referred to as “movement coupling”.  Recent research isn’t supporting the idea of a 1:1 miter joint coupling and thus, this terminology is out of date.  This will be discussed further in the “Movement Coupling” section.

MANTER JT, 1941. Movements of the subtalar and transverse tarsal joints. Anat Rec 80: 397.
ISMAN RE, INMAN VT 1969. Anthropometric studies of the human foot and ankle. Bull Prosthet Res 11: 97.
VAN LANGELAAN EJ 1983.  A kinematic analysis of the tarsal joints: an x-ray photogrammetric study. Acta Orthop Scand 54: 5.
LUNDBERG A 1989.  Kinematics of the ankle and foot. In-vivo roentgen stereophotogrammetry. Acta Orthop Scand Suppl 233:1-24.

2.2 Subtalar Axes of Rotation and Type on Injury

Closed chain subtalar joint pronation is triplanar (front to back, side to side and rotational).  The movement in the frontal plane mainly occurs via inversion/eversion of the calcaneus (see here for video of what eversion/inversion is), while movement in the transverse plane occurs mainly through a combination of adduction in the talonavicular joint and rotation in the tibiotalar joint.  In addition, as noted above, variation in the axis of rotation from the transverse plane (degrees “from” horizontal) can vary from 20 to 68 degrees in different individuals in different studies.  Since subtalar pronation/supination is a triplanar motion this has implications on injuries.  For example, Root (1977) theorized that an individual with a high axis transverse plane subtalar pronation would have increased adduction/abduction in the talus and thus, increased internal/external rotation of the tibia and be more vulnerable to lower leg/knee injury.

Conversely, an individual with an axis of rotation lower to the transverse plane (if you have trouble picturing this, think lower “in” the saggital plane) would have more motion in the frontal plan (eversion/inversion of the calcaneus) and thus, be more vulnerable to foot injuries.  This injury site correlation to subtalar axes was confirmed by Tomaro (1996), however the study was done non-weight bearing and with static measurements.  Later, this theory had a setback when Williams (2001) discovered that those individuals with pes planus (i.e. “flat feet”) had a larger ratio of rearfoot frontal plane eversion to internal tibial rotation (as expected), but they also had a higher rate of knee injuries – not foot injuries.  In other words, this was opposite of what was thought to be found by Tomaro and hypothesized by Root.   Again, Williams study was confirmed by Nawoczenski (1998) who found that those runners who had high arches and lower rearfoot eversion to internal tibial rotation ratios suffered from more foot injuries.   More on this later…

Since researchers are coming up with different ways that the foot moves and different types of injuries for feet that do move the same way, it is difficult to assume what type of injury a person will have, based on their foot type, or what their subtalar axis of rotation to the transverse plane is.  There are a vast number of intrinsic and extrinsic factors to consider.

ROOT ML et al., 1977. Clinical Biomechanics: Normal and Abnormal Function of the Foot, Vol II, Clinical Biomechanics Corp, Los Angeles
TOMARO et al., 1996. Subtalar joint motion and the relationship to lower extremity overuse injuries.  J Am Podiatr Med Assoc. 86(9):427-32
WILLIAMS, D.S. et al., 2001. Lower extremity kinematic and kinetic differences in runners with high and low arches. J. Appl. Biomech. 17, 153–163.
NAWOCZENSKI DA et al., 1998. The effect of foot structure on the three-dimensional kinematic coupling behavior of the leg and rear foot. Phys Ther. 78:404-416. 1998

2.3 Ankle Joint Transverse Plane Motion

To complicate matters further, it has been revealed that there is quite a bit of transverse plane motion (rotation) in the ankle joint itself.  Previously we had always believed that the ankle joint (talocrural, not subtalar) was uniaxial with only plantar and dorsiflexion occurring and there was no rotation in the transverse plane.  Again, more recent research shows that there is plenty of transverse plane motion in the ankle joint.  For frontal and transverse planes of motion respectively, Nester (2007) found 15.3 and 10 degrees, Arndt (2007) found 12.2 and 8.7 degrees for slow running and Lundgren (2008) found 8.1 and 7.9 degrees for walking.  This is in stark contrast to what I was taught in kinesiology undergrad and in chiropractic school. (back in the day)

Again, you may be wondering, “Why do I care if there is transverse plane motion in the ankle joint?”  Well, it goes back to the previous section regarding how much pronation is actually transferred to the rest of the lower extremity via internal tibial and femoral rotation, which I will discuss in the “movement coupling” section.

NESTER et al., 2007. In vitro study of foot kinematics using a dynamic walking cadaver model. J Biomech, 40(9):1927-37.
ARNDT A et al., 2007.  Intrinsic foot kinematics measured in-vivo during the stance phase of slow running. J Biomech, 40(12):2672-2678.
LUNDGREN P, et al., 2008.  Invasive in-vivo measurement of rear-, mid- and forefoot motion during walking. Gait Posture, 28(1):93-100.

2.4 Influence of Forefoot Position on Rearfoot Motion

Here is one of biomechanics research major shortcomings:   Forefoot position significantly influences rearfoot mechanics, yet the vast majority of research only looks at rearfoot mechanics.

In an observational study involving 385 subjects, Gross (2007) found that the non-weight bearing position of the forefoot was associated with hip pain.  The greater the forefoot varus alignment, the greater the likelihood of hip pain and total hip replacement.  Rearfoot position was not associated with any hip pain.  Keep in mind that this was a static, non-weight bearing study (you can’t measure forefoot varus in weight bearing!)  However, it was a large study and cannot be ignored.  Did the forefoot varus precede the hip pain?  We cannot know from this study.

It is hypothesized that the forefoot varus will result in internal tibial and femoral rotation due to the coupling effect.  This internal femoral rotation will hypothetically place increased load on the external hip rotators and thus, predispose the hip to pain and increased stress.  Whatever the mechanism, this study found a significant association.

Gross KD, et al. Varus foot alignment and hip conditions in older adults. Arthritis &Rheumatism 2007; 6: 2933-2938

3. Movement Coupling from the Foot to the Leg

Historically, pronation of the foot has been thought to cause internal rotation of the tibia.  This is called movement coupling.  With the foot and tibia, the classic ratio has been taught to be 1:1 and the analogy of a “miter hinge” has been the standard.

Given the information noted in the previous paragraphs and chapters in this post with respect to the axes of rotation for the subtalar and ankle joint, it becomes apparent that the 1:1 miter joint analogy is far from accurate.

One important factor that seems to be lost in gait analysis is the total body movement.  Consider the gait cycle where the right foot is in midstance.  The foot is pronated which should internally rotate the femur.  Also consider the left leg in its swing phase.  The left leg is imparting a rotation on the pelvis toward the right which will create an external rotation of the right femur.  You also have other muscles of the hip that are imparting an external rotation and abduction moment on the femur.  So, you have an internal rotation moment on the tibia and an external rotation moment on the femur.  The two cancel each other out…somewhere along the kinetic chain.

Getting back to the 1:1 miter hinge analogy, recent research is not supporting this 1:1 ratio.  For example, Powers (2002) found that the ratio of rearfoot eversion to tibial rotation during walking is closer to 2.5:1, and McClay (1996) found it to be 1.5:1 while Nawoczenski (1998) found the ratio to be 1.8:1 for running.  Ratios of up to 2.5:1 of rearfoot motion to tibial rotation are far from the 1:1 ratio most of us were taught in school.

More boldly, two new studies have taken it a step further:  Kernozek (1993) found that there was little to no correlation between the dynamic Q angle  and the amount of rearfoot pronation in the subjects they studied.  This study has since been supported by Reischl (1999) found the magnitude and timing of peak pronation did not correlate to that of tibial and femoral rotation and that “The lack of a relationship between peak foot pronation and the rotation of the tibia and femur is contrary to the clinical hypothesis that increased pronation results in greater lower extremity rotation”.

Newer studies not only show that the subtalar axis of motion from the transverse plane may not average 41 degrees, and also that the axis of rotation changes as the foot moves through the gait cycle.  Dr. Kevin Kirby, DPM (2001) notes, “the most accurate research to date has demonstrated that the subtalar joint axis no longer can be accurately described as a solitary axis of rotation but is best described as a multitude of axes of rotation that all have different spatial locations.”

In addition to the constantly morphing findings mentioned above, it has been found by Nigg (1999) that movement coupling is less pronounced by those with pes planus (flat feet) than those with pes cavus  (those with high arches).   This may be due to ligament laxity.  Despite these findings, conventional running shoe/orthotic footbed prescription is as follows:  If an individual has high arches (pes cavus), that means that they lack the natural cushioning associated with pronation and so they should not be given any anti-pronation devices.  On the other hand, is someone has pes planus (flat feet), that means that they should be prescribed devices to push up the medial arch and prevent pronation since they lack the normal arch in their anatomy.  However, this is on the assumption that excessive pronation will cause excessive internal tibial rotation via joint coupling.  The excess internal rotation will cause knee injuries.  However, as Nigg discovered, the movement coupling between the foot and the tibia is lessened in those with pes planus.  In other words, you attempt to reduce the internal rotation of the tibia in people with flat feet because you assume they pronate excessively, however your attempt is muted simply due to the fact that movement coupling is already reduced in those very individuals you are trying to “fix”.

To add even more confusion to the issue, another study by Nigg (1993) found that the amount of rearfoot eversion wasn’t different in high arched vs. low arched individuals, however there was greater internal tibial rotation in the high arched individuals.  This means that there was a lower rearfoot eversion to internal tibial rotation ratio in higher arched individuals.  Agreement to this study came by Nawoczenski (1998) who found the exact same thing: no difference in calcaneal eversion between those with high vs. low arches, but a higher amount of internal tibial rotation in those with higher arches, which, again would confirm a lower rearfoot eversion to internal tibial rotation ratio.  So, it’s settled, right?  Wait!  Not so fast.  Williams (2001) also found a lower rearfoot eversion to tibial int. rotation ratio, but opposite to the previous studies, the Williams (2001) study said it was because there was more eversion of the calcaneus and less int. tibial rotation.

One study that seems to correlate somewhat with these previous ones the study by Pohl (2007) who found that foot/tibial coupling is significantly less in walking than it is in running.

So we come to a cross-roads:  If the research can’t agree on how, why or how much the foot movement influences the rest of the lower extremity, maybe it’s because the foot doesn’t influence the lower extremity as much as we’ve been lead to believe.  I’m not discounting coupled movements here.  What I’m asking is the crucial question that is in the back of most every clinician’s and researcher’s mind: “Does the foot change the kinematics of the leg, or is there more to it….does the leg drive the kinematics of the foot?”

One significant, but rarely discussed paper investigated this very question: Bellchamber et al.,(2000) collected data from subjects walking and running by using force plates, angular velocities of the tibia and intersegmental joint moments at the ankle.  They calculated the “power flow” and found that in walking, there was a large power flow from the tibia to the foot.  In other words, the leg was driving the foot.  During running it was a bit different, but control was still mostly from the leg to the foot.  For the first 15% of stance, there was little power flow in either direction.  From 20-60% of stance, there were periods of negative power flow (distal to proximal, or foot controlling the leg), but mostly positive power flow (proximal to distal, or leg controlling the foot motion).  Then the remainder of stance (from 60-100%) was all positive power flow.  I’m not sure why this paper is not discussed more often, but it appears that for the most part, lower extremity control when walking or running is from proximal to distal.

In a paper by DeLeo et al., (2004), the authors state, “Based on the majority of these studies, it appears that variations in [the joint coupling ratio] ratio can be attributed to tibial internal rotation excursion to a greater extent than rearfoot eversion excursion.”  In other words, joint coupling ratio is mostly due to what happens in the tibia, not what happens at the foot.

If that’s the case (leg drives the foot), we can take things even one step further.  Souza et al., (2010) found that hip internal rotation timing and rearfoot pronation were strongly linked and hip external rotation and rearfoot supination were also strongly linked.  This fits in line with what many clinicians and several researchers have been suggesting lately – the hip and core kinematics seem to be the driving force and possible error source for more injuries that foot kinematics are.

That is not to say that footwear or orthotics don’t change things.  Please see the “Attempts to Limit Pronation” section for more on this

POWERS CM. et al., 2002. Comparison of foot pronation and lower extremity rotation in persons with and without patellofemoral pain. Foot Ankle Int. 23:634-640.
MCCLAY I et al., 1996.  The subtalar angle: a proposed measure of rearfoot structure. Foot Ankle Int. 17:499-502.
NAWOCZENSKI DA et al., 1998. The effect of foot structure on the three-dimensional kinematic coupling behavior of the leg and rear foot. Phys Ther. 78:404-416.
REISCHL SF et al., 1999. J. Relationship between foot pronation and rotation of the tibia and femur during walking. Foot Ankle Int. 20:513-520.
KERNOZEK TW et al., 1993 Quadriceps angle and rearfoot motion: relationships in walking. Arch Phys Med Rehabil. 74:407-410.
KIRBY KA. 2001: Subtalar joint axis location and rotational equilibrium theory of foot function. JAPMA 91: 465.
Nigg, B.M., Nurse, M.A., Stefanyshyn, D.J., 1999. Shoe inserts and orthotics for sport and physical activities. Medicine and Science in Sports and Exercise 31 (Suppl. 7), S421–428.
Nigg BM et al., 1993.  Effects of arch height of the foot on angular motion of the lower extremities in running. J Biomech 26: 909-916.
WILLIAMS, D.S. et al., 2001 Lower extremity kinematic and kinetic differences in runners with high and low arches. J. Appl. Biomech. 17, 153–163.
POHL et al., 2007. Forefoot, rearfoot and shank coupling:effect of variations in speed and mode of gait. Gait Posture 25 (2), 295–302.
BELLCHAMBER TL, van den Bogert AJ, 2000. Contributions of proximal and distal moments to axial tibial rotation during walking and running. Journal of Biomechanics 33, 1397-1403.
DELEO et al., 2004. Lower extremity joint coupling during running: a current update. Clinical Biomechanics (Bristol, Avon), 19, 983–991.

4. Measurement of Pronation in a Research Setting

There are numerous ways that pronation is measured in research studies.  To me, this is one of the largest sources of discrepancy in the results.  Some studies look at static positions others look at dynamic movement, some look at weight bearing others look at non-weight bearing, some look at dynamic motion based on markers placed on the shoe others place markers on the skin while others use metallic beads implanted in the bones, some look at live people, others look at cadavers.

Terms like “arch index”, “longitudinal arch angle”, “valgus index”, “arch height ratio”, “navicular drop”, “foot posture index” and others are used to describe how much an individual pronates, but only seem to complicate the issue.  Some have good intra-examiner reliability, some don’t.  Some have good inter-examiner reliability, some don’t.

It is not the purpose of this paper to dissect each of these techniques.  My point is that there is no consistency in the manner in which research is performed.  When there is no consistency in methods, you cannot expect consistency in results.  Therefore, we get some studies saying there is a correlation between pronation and injury while other studies say there is poor correlation (see pronation and injury section)

In a study by Reinschmidt (1997), intracortical bone markers (metallic beads or pins placed within the bones) as well as external shoe markers were used in subjects running on a treadmill.  The average calcaneal eversion with the shoe markers was found to be 16 degrees, while the measurements with the intracortical markers were only 8 degrees.  The authors concluded, “The rotations derived from external shoe and shank markers typically overestimate the skeletal tibiocalcaneal kinematics.”

This overestimation with externally mounted markers has been confirmed in a study by Stacoff (2000) using intracortical bone markers, subjects ran barefoot, with shoes and with 3 shoe soles and 2 different types of orthotics.  In the end, differences in tibial and calcaneal movement between barefoot and running in shoes were small and disorganized.  The authors concluded, “The result of this in vivo study contrasts with previous investigations using skin and shoe mounted markers and suggests that these discrepancies may be the result of the overestimation with externally mounted markers.”

Along with external shoe markers, markers that are placed on the skin have been evaluated in comparison to more reliable methods.   List (2012), compared ankle motion evaluated by skin markers vs. videoflouroscopy in 11 subjects walking uphill, level and downhill.  They concluded, “The differences between skin marker assessed rearfootshank and the fluoroscopic assessed isolated TAA motion were neither consistent between subjects, nor motion planes, nor conditions.”

One other study I will mention here is that of Benoit (2004) which compared skin mounted vs. intracortical pins to measure knee kinematics of running and cutting.  They concluded that their study “indicates that skin mounted reflective markers display significant limitations in predicting 3D kinematics of the knee joint.” And that “Although skin-marker derived kinematics could provide repeatable results this was not representative of the motion of the underlying bones.”

Taking these factors into consideration, it is difficult to place equal value on previous research which has used external markers on shoes or even skin markers, as compared to intracortical measurements and fluoroscopic analysis which appear to be markedly different and reliable.  So do we throw out all previous research?  Obviously not, but I think their value is less than studies that incorporate better analysis techniques.  I will usually value a study done on live subjects rather than cadaver feet and more value on studies with intracortical markers than skin or external shoe markers.  (check out this video for cadaver measurements and here for intracortical live subects)

REINSCHMIDT et al., 1977.  Tibiocalcaneal motion during running: measured with external and bone markers. Clin. Biomech. 12:8 –16.
STACOFF A et al., 2000. Tibiocalcaneal kinematics of barefoot versus shod running. J Biomech. 33(11):1387-1395.
LIST et al., How well can skin marker analysis detect the kinematics of a total ankle arthroplasty? – a comparison to videofluoroscopy. Journal of Foot and Ankle Research 2012, 5(Suppl 1):O35

5. Attempts to Limit Pronation

Based on the notion that “over”pronation (whatever that is defined as) is bad, clinicians have attempted to find ways of limiting the pronation (whether it’s the pronation angle or the pronation velocity).  This is mainly in the form of posted orthotics, orthotic footbeds, or shoes which limit pronation via dual density midsoles, medial flaring of the midsole, arch supports and other means.  So let’s take a look at the effectiveness of these methods…

5.1 Orthotics Limiting Pronation

If you have read through this article up to here, congratulations.  You are a much better person now, but you will also understand what I am about to say about the question “Do orthotics limit pronation (angle or velocity)?”  Here’s my Clintonesque answer:  Maybe.  It depends what you are referring to.

If that seems equivocal, it’s not because I’m ill informed, it’s because the research is equivocal.  Below, I have listed a few studies to give you a taste of what I’m talking about.  Some show orthotics change the motion of the rearfoot and tibia and some studies that show the opposite.  After a few of these, I realized that for every research article showing one thing, it will be countered by a study showing the opposite.  I think this goes back to what I had mentioned earlier in this paper:  There are so many different methods that researchers use, nothing is consistent, so how can you expect results to be consistent.  Overground running, treadmill running, skin markers, shoe markers, intracortical bone markers, precise placement of markers, elite runners, novice runners, etc.

When it comes to paying hundreds of dollars for custom orthotics, one would think that they are paying for a precise, specific foot orthotic that will benefit their anatomy.  However, how accurate are the orthotics?  There are so many different theories on the “best” way to make custom orthotics, how do you know who is right?

One recent study by Chuter (2003) took 10 experienced and 10 inexperienced clinicians and had them independently make a cast of a single individual’s right foot in what they deemed as a neutral position.  The variability was huge: “The range of the forefoot-to-rearfoot relationship across all groups was from 10.0 degrees everted to 6.5 degrees inverted, indicating that there is a wide range in neutral-position casting of the foot”  So, the casts had a range of 16.5 degrees for rearfoot eversion/inversion.  If such large variations in measurements exist, how can we say custom orthotics are specific or precise for doing anything?

Even if these clinicians were consistently specific at making orthotics in a static position, various studies have shown that the static position that the cast is made in doesn’t reflect the architecture of the foot in its dynamic state.  In other words, the foot arch changes when walking, running or cutting (Hamill 1998, Dicharry 2009, Benke 2012) so why are orthotics being made for $500 when they are not designed to do what they should and as we’ll see below, less effective at controlling pronation than taping is?

So, here are a few studies showing the positives and negatives of orthotics altering foot and leg kinematics: (don’t skim past each study, these are interesting)

Studies Showing Orthotics make a Change (but it’s variable and unsystematic):

1) Nawoczenski et al., (1995) looked at 20 runners with various lower extremity pains.  The runners running biomechanics were recorded at the beginning of the study and then each runner was casted for a posted orthotic in a non-weight bearing position.  Each runner progressively increased the use of the orthotics for a 3-4 week period and then was videotaped on a treadmill using markers placed on the skin.  They found a significant reduction in internal tibial rotation in both pes planus and pes cavus foot types in the first 50% of stance, however, no change was found in the frontal plane motion (eversion) of the rearfoot.

2) Rodrigues et al., (2012) looked at 16 asymptomatic runners and 17 runners with knee pain and studied their kinematics with and without medially posted insoles.   They used skin surface markers to measure with.  Contrary to the Nawoczenski et al., (1995) article listed above, Rodrigues et al., (2012) found that the medially posted insoles reduced the rearfoot eversion excursion, peak rearfoot eversion and also the rearfoot eversion velocity, yet this study found small and insignificant changes in tibial rotation.  Yes, that’s exactly opposite to the Nawoczenski study.

3) Liu et al., (2012): This is one of the only studies using intracortical markers to evaluate kinematic changes associated with orthotics and unfortunately, it only involved 5 subjects (I’d imagine people aren’t lining up to have pins stuck in their bones).  The authors found that, “Changes in calcaneus-tibia motion were comparable with those described in the literature (1°-3°)… However, the nature and scale of changes were highly variable between subjects.”  They found that sometimes, the changes were in the subtalar joint and sometimes in the ankle joint.

CHUTER V et al., 2003 Variability of neutral-position casting of the foot. JAPMA, 93(1):1-5.
Hamill et al.,
Dicharry et al., 2009.  Differences in static and dynamic measures in evaluation of talonavicular mobility in gait .J Orthop Sports Phys Ther. Aug;39(8):628-34
Bencke et al., 2012 Measuring medial longitudinal arch deformation during gait. A reliability study. Volume 35, Issue 3, Pages 400–404
NAWOCZENSKI et al., 1995.  The effect of foot orthotics on three-dimensional kinematics of the leg and rearfoot during running. J Ortho Sp Phys Ther, 21:317-327.
RODRIGUES et al., 2012.  Medially posted insoles consistently influence foot pronation in runners with and without anterior knee pain.  Gait and Posture Nov 5, 2012
LIU et al., 2012.  Effect of an antipronation foot orthosis on ankle and subtalar kinematics. Med Sci Sports Exercise. Dec;44(12):2384-91

Studies Showing Orthotics Don’t Make a Change:

1) Stacoff (2000) is another study using intracortical markers and medially posted orthotics, but again the subject pool was only five subjects.  This study was close to what Nawoczenski (1995) found in that there were insignificant effects for rearfoot eversion.  With respect to tibial rotation, there were significant, but very small changes.  They concluded, “This in vivo study showed that medially placed foot orthoses did not change tibiocalcaneal movement patterns substantially during the stance phase of running.

2) Ferber (2011) had 20 participants walk in either semi-custom moulded, non-moulded or no orthotics.  Skin markers were used and the researchers found no differences between rearfoot eversion, tibial rotation or medial longitudinal arch angle.  There was, however a reduction in the change in distance between the calcaneal tubercle and the metatarsal heads, which lead the authors to conclude that there should be a concurrent reduction in the strain on the plantar fascia.  My opinion regarding this is that it seems over-simplistic to assume that a reduction in the change of the distance between the calcaneal tubercle and the metatarsal heads would equate to a reduction in the plantar fascia strain, since most plantar fascia strictures insert distal to the MTP joints.  I have been in personal communication with Dr. Ferber regarding this and while he seems to agree with my opinion, he states that the reduction in plantar fascia strain has been confirmed in an unpublished study in Sweden by using strain gauges in the plantar fascia.  I have no idea on how this unpublished study was conducted.  For the purposes of this article however, the point remains that through the use of semi-custom orthotics, there was no change in rearfoot or tibial kinematics.

3) Brown et al., had 24 subjects walking on a treadmill in either 3 conditions: over-the-counter orthotics, custom casted orthotics or no orthotics.  Skin markers were used to collect the measurements.  In the end, there was no difference in maximum pronation, calcaneal eversion, and total pronation between the three conditions. The pronation velocity and time to maximum pronation were deemed “unreliable” in this study since the authors stated that those two factors were heavily influence by subjects who had previously worn orthotics and there was a “learned behavior.”  The point was however, “Based on the results of this study, padded arch supports nor biomechanical orthoses can be preferentially recommended for their ability to control maximum pronation, calcaneal eversion, and total pronation during walking.

So in the end, the effects of orthotics are highly variable and inconsistent.  This is also the conclusion of Nigg et al., when they tested 15 different runners with 5 different types of orthotics – either neutral, full lateral, full medial, half lateral, and half medial postings.  Using 3D kinematic camera systems and force plate data, the authors concluded, “The effects of the chosen insert interventions on the individual movement characteristics were small and not systematic” and, “The results of this study also showed that subject specific reactions to the tested inserts were often not as expected.  Additionally, reactions were not consistent between the subjects.”

STACOFF et al., 2000. Effects of foot orthoses on skeletal motion during running. Clin Biomech, Jan;15(1):54-64.
FERBER R, BENSON B. Changes in multi-segment foot biomechanics with a heat-mouldable semi-custom foot orthotic device.  J Foot Ankle Res. 2011 Jun 21; 4(1):18.
BROWN et al., 1995 The effect of two types of foot orthoses on rearfoot mechanics. J Orthop Sports Phys Ther. 1995 May;21(5):258-67
Nigg BM et al., Effect of shoe inserts on kinematics, center of pressure and leg joint moments during running. Med Sci Sports Exerc 2003;35(2):314–‐319.

5.2  Shoes Limiting Pronation

To organize this topic into a question of “do they” vs “don’t they” would be a monumental undertaking, given the different ways of dealing with the idea of reducing pronation (higher medial arch, dual density midsole, medial flaring of the midsole etc).  In addition, you have the well documented problem that simply having a softer midsole in a shoe causes both increased pronation excursion as well as faster pronation velocity (Hamill et al., 1992; Wit et al., 1995; Kersting & Bruggemann, 2006).  So, what I’m saying is that as soon as you introduce a shoe with cushioning (which every shoe has to some degree) you are also increasing some amount of pronation.  This makes the act of “controlling” this pronation even more difficult (no, I’m not a barefoot runner).

In order for the shoe to limit pronation, we have to assume that the foot will couple it’s movement with the shoe (move with the shoe).  This was investigated by Stacoff et al., (2000) who found that shoe/calcaneus coupling is not strong in the first half of stance during running, but the coupling significantly increases in the second half.  This would mean that since most pronation occurs in the first half of stance yet shoes/calcaneus coupling is less in the first half, the anti-pronation effects of the shoe are lessened.

So, rather than get into a review of studies, I will refer to a 2011 study by Cheung et al., that looked at all available research and narrowed their review down to 29 different studies which looked at 3 different conditions aimed at lessening pronation: orthotics, motion control shoes and athletic taping (not the colorful stretchy tape).  The authors concluded that overall, all 3 measures were able to reduce pronation to some degree compared to no intervention.  In order of effectiveness of reducing rearfoot eversion, the worst was orthotics (mean change of 2.24 degrees), then motion control shoes (mean change of 2.52 degrees) and the most effective was taping (mean change of 2.64 degrees).

With respect to the anti-pronation measures of the shoes, dual density midsole was the most effective while heel counters or wedge modifications were not significant.

HAMILL, J et al., (1992). Timing of lower extremity joint actions during treadmill running. Medicine and Science in Sports and Exercise, 24(7), 807-813.
Wit B.D., Lenoir M. (1995). The effect of varying midsole hardness on impact forces and foot motion during foot contact in running. Journal of Applied Biomechanics, 11(4), 395–406.
Kersting, U. G., & Bruggemann, G. P. (2006). Midsole material-related force control during heel-toe running. Research in Sports Medicine, 14(1), 1-17.
STACOFF et al., Movement coupling at the ankle during the stance phase of running. Human Performance Laboratory, The University of Calgary, Canada Foot Ankle Int 21:232-9. 2000
Cheung RTH et al., Efficacies of different external controls for excessive foot pronation: A meta-analysis. British Journal of Sports Medicine 2011; 45: 743-751.

5.3  Assigning Shoes Based on Arch-Type

One point that I feel needs to be made here is the idea that a type of shoe should be prescribed based on the height of one’s arch.  The so called “wet-footprint test” has been the basis for shoe prescription for the past 3 decades and despite the evidence showing its inappropriateness, many shoe retailers and clinicians continue to look at the shape of a client’s/patient’s arch and use this shape as a basis for prescribing a “motion control”, “stability” or “neutral” shoe.   Walk into most  running stores and you will see the shoes arranged on the wall in categories based on these 3 shoe types.  There are two well designed studies that have conclusively disproven this notion:

Study#1: Ryan et al., (2011) categorized 81 female runners into 3 groups: Neutral, pronated or highly pronated.  Rather than prescribing a shoe based on standard practice of a neutral shoe for a neutral foot, “stability” for a pronated foot and “motion control” for a highly pronated foot, the authors of the study randomly assigned shoes to all the runners, despite their foot type.   What they found was that (over a 13 week half-marathon training program) if the runners happened to be given the correct type of shoe for their foot, they were more likely to be injured than if they were given the wrong shoe based on their foot type.  For example, “In neutral feet, the neutral shoe reported greater values of pain while running than the stability shoe; in pronated feet, the stability shoe reported greater values of pain while running than the neutral shoe.”  The authors concluded, “The findings of this study suggest that our current approach of prescribing in-shoe pronation control systems on the basis of foot type is overly simplistic and potentially injurious.”

Study #2: Knapik et al. (2010) had 408 men and 314 women military recruits assigned shoes based on their foot type, while they had a control group of 432 men, 257 women assigned stability shoes regardless of their foot type.  Over a 12 week training program, there were no differences in injuries between the two groups.

RYAN et al.,  The effect of three different levels of footwear stability on pain outcomes in women runners: a randomised controlled trial. Br J Sports Med, 2011, 45, 715–721.
Knapik et al., Injury reduction effectiveness of assigning running shoes based on plantar shape in marine corps basic training. Am J Sports Med. 2010;38(9):1759-1767.

6. Does “Over”Pronation Cause Injuries?

If you perform a search on the web for “pronation” and “running injury”, you will find quite a few websites that list studies linking pronation to certain types of injuries.  I would exercise caution for many of those, since a great deal of them are based on static measurements.  As I have discussed, static measurements don’t correlate well to dynamic positioning during running or walking.  Conflicting evidence is present in these studies.

I’d rather not go through and dissect each study, so thankfully, Van Gent (2010) already did it for us by performing a systematic review of the literature and concluded, “there was limited evidence that a higher heel valgus [rearfoot eversion] was protective against knee and foot injuries, while a lower heel valgus and a higher right arch index were protective factors only for knee injuries. There was limited evidence that static biomechanical lower limb alignment was not related to lower limb injuries.”  Yes, you read that correctly – a higher heel valgus was “protective” against foot and knee injuries.  That seems like a departure from what I’ve read, but as it the theme with this article…everything is conflicting.

So, moving on to actual kinematic studies…

  • Rabbito (2011) found that in the 24 runners they examined, those with tibialis posterior strains were associated with greater and prolonged pronation during running.  Cause or effect?  We don’t know, but an association at least.
  • Ryan (2009) found that in the 48 runners they examined, those who already had Achilles pain were more likely to have greater rearfoot eversion.  Cause or effect?  We don’t know, but an association at least.
  • Messier (1988) looked at 19 non-injured vs. 13 injured runners who suffered from plantar fasciitis, ITB syndrome or shin splints.  They found that there was a “non-significant” trend for a higher rate of injury in those runners who had greater maximum pronation, total rearfoot movement and maximum velocity of pronation
  • Viitasalo (1983) looked at 13 uninjured runners and 35 runners who had medial tibial stress syndrome and found that those who were injured with MTSS had significantly greater rearfoot eversion excursion.  Again…cause or effect?  We don’t know.
  • Messier (1991) found that in 20 non-injured runners vs 16 runners with patellofemoral pain syndrome (PFPS), there were no differences in rearfoot kinematics when running.

So, these kinematic studies seem to imply that greater pronation increases risk of injury.  You may be wondering why I keep writing “Cause or effect?  We don’t know”  That is because we don’t know if the pain was causing the runner to somehow alter their gait.  For example, the study by Viitasalo found that injured runners with MTSS have greater pronation, but a different study by Hreljac (2000) did not find this.  His was a much differently designed study where he took a group of runners who had never sustained an injury throughout their running careers and a group of runners who had sustained multiple lower extremity injuries below the knee in their running careers.  The key point to his study is than the “injured” runners were not injured at the time of the study and had to have been running injury free for the 3 months prior to the study.  What the Hreljac (2000) found is that the group of runners who had NEVER sustained an injury actually had a non-significant trend to have greater pronation velocity and increased supination at touchdown compared to the previously injured runners.  This is contradictory to the Viitasalo study.

What I’m getting at is that these are studies that are either retrospective (looking at runners with previous injuries) or those with current injuries.  The best design would be a “prospective” study – a study that looks at the biomechanical parameters of runners and then tracks them for a while to see who gets injured.

So, looking at prospective, kinematic studies…

  • Willems et al., (2006) took 3D kinematic running data of 400 undergraduate students.  They were followed for 3 years and played sports including a wide variety of sports including badminton, judo, soccer, swimming, track and field and others.  There were many students who had “sports injuries” to the lower legs, and many students who were injury free.  Of those who ended up injured, they tended to have a more central heel strike (rather than a lateral heel strike) when running, greater rearfoot eversion and abduction excursion and rolled off their feet more laterally at toe off than the uninjured group.  However, they did not see increased tibial rotation in the injured group, leading them to surmise that the abduction of the rearfoot was “absorbed by musculoskeletal structures in the lower leg itself.”  Given what I wrote earlier in this paper regarding rearfoot/tibial coupling, that point is debatable.
  • Thijs et al., (2007) performed a similar prospective study on 84 military recruits.  In this case however, they didn’t look at 3D kinematics but instead, looked at plantar pressure measurements.  If a subject has a more medially placed plantar pressure during stance, one can infer that it means they were more pronated.  During the 6 week follow up, there were 36 subjects who developed anterior knee pain.  In those who developed anterior knee pain, there was actually less pronation than the group that was not injured.

In summary of whether or not “over”pronation is associated with injury, there is a trend that shows more pronation is associated with increased risk of injury; however, this is inconsistent and conflicting.  The majority of these studies show an association, but not necessarily a cause.  We need more prospective biomechanical studies, rather than studies that measure static alignment or retrospective studies or studies looking at runners who are currently injured

VAN GENT et al., Incidence and determinants of lower extremity running injuries in long distance runners: A systematic review. BrJ Sports Med 2007;41(8):469-480.
Rabbito et al., Biomechanical and clinical factors related to stage I posterior tibial tendon dysfunction. J Orthop Sports Phys Ther. 2011 Oct;41(10):776-84.
RYAN et al., Kinematic analysis of runners with achilles mid-portion tendinopathy.
Foot Ankle Int. 2009 Dec;30(12):1190-5.
MESSIER et al., 1988. Etiologic factors associated with selected running injuries. SP, Pittala KA Med Sci Sports Exerc. 1988 Oct; 20(5):501-5.
VIITASALO et al., 1983. Some biomechanical aspects of the foot and ankle in athletes with and without shin splints. Am J Sports Med. 1983 May-Jun; 11(3):125-30.
HRELJAC A et al., 2000. Evaluation of lower extremity overuse injury potential in runners.
Med Sci Sports Exerc. 2000 Sep; 32(9):1635-41
Messier et al., 1991. Etiologic factors associated with patellofemoral pain in runners.
Med Sci Sports Exerc. 1991 Sep; 23(9):1008-15
Willems et al., A prospective study of gait related risk factors for exercise-related lower leg pain. Gait Posture. 2006 Jan;23(1):91-8
Thijs et al., A prospective study on gait-related intrinsic risk factors for patellofemoral pain. Clin J Sport Med. 2007 Nov; 17(6):437-45

7. Do Orthotics Reduce Pain and Injury?

Here is what this paper has uncovered so far:

a) There is inconsistent and contradictory information about the axes of rotation that pronation occurs around
b) There is inconsistent and contradictory information about how much movement coupling exists between the foot and the knee/hip and whether abnormal foot mechanics can lead to lower extremity injury
c) There is inconsistent and contradictory information about how much of it orthotics and motion control shoes can control pronation, how they do it and what effects they have on the rest of the lower extremity
d) It appears that foot control is dictated by the rest of the leg, not the other way around.  In other words, the leg drives the foot mechanics
e) The fabrication of custom made orthotics is an inexact science and studies show the same patient’s orthotics can vary as much as 16.5 degrees in the frontal plane when made between 20 different clinicians.

Given that information, the answer to the question of “Do orthotics reduce pain or injury?” is obvious…right?  Orthotics can’t possibly help foot and lower extremity pain…right?  Wrong.

For some reason, they can actually help for some people with some types of pain and injury.

I am not in the business of performing research reviews on the hundreds of studies that have been performed, so in this case I will defer to two different systemic reviews of the literature that have already been performed.

Review #1: Collins et al., (2007) whittled 3192 studies on orthotics down to 22 based on weeding out those with poor design or methods.  Here were some of their conclusions:

a) There is a surprising lack of difference in effectiveness between off the shelf orthotics and the much more expensive custom orthotics
b) There appears to be some evidence for the use of orthotics for prevention of first time injuries
c) Due to lack of, as well as conflicting evidence, it is difficult to recommend or refute the use of orthotics for individuals who already have an injury

Review #2: Hume et al., (2008) performed a similar review but were a little stricter in studies they deemed as appropriate to include.  In the end, they included only 17 studies: 11 controlled studies and 6 uncontrolled studies.  Of the controlled studies, they found (according to NHS):

a) One study of prefabricated semi-rigid orthotics showed a moderately beneficial short-term effect compared with customized soft sham orthotics for the treatment of foot pain and foot function for plantar fasciitis.
b) One study showed a small harmful effect of prefabricated semi-rigid orthotics compared with prefabricated soft orthotics for treating plantar heel pain.
c) A moderate benefit of orthotics (semi-rigid versus anti-inflammatories and prefabricated soft (silicone) versus stretch) was found in two studies for plantar fasciitis.
d) One study found small to moderate harm for prefabricated soft versus anti-inflammatories for plantar fasciitis.
e) Three studies reported benefits of orthotics in reducing the prevalence of injury.
f) None of the studies found a beneficial effect of orthotics for preventing chronic lower limb injury.

Collins N, Bisset L, McPoil T, Vicenzino B: Foot orthoses in lower limb overuse conditions: a systematic review and meta-analysis. Foot Ankle Int 2007, 28(3):396-412.
Hume P, Hopkins W, Rome K, Maulder P, Coyle G, Nigg B: Effectiveness of foot orthoses for treatment and prevention of lower limb injuries : a review. Sports Med 2008, 38(9):759-779.

8.  Summary and Conclusion

Due to the inconsistencies and conflicting studies, I think many people will be flustered and more confused than they were before they read these 17 pages.  I can understand why.  However, there is a lot to be learned by the inconsistencies of the research.  Below, please find some take home points:

  1. Since the axes of rotation for pronation are so highly variable between individuals, it is difficult to make broad statements about who will be at risk from pronation abnormalities and who would be helped by trying to fix them
  2. The influence of pronation on injury has been broadly and exceedingly exaggerated.  There is too much emphasis placed upon pronation and injury so it needs to stop.
  3. Control of the foot appears to be top down.  In other words, the leg controls the foot, not the other way around.  This has the potential to explain why there is such large variability in results for these studies.  There is a shocking scarcity of research that examines foot mechanics and how it relates to anything above the tibia.  If the hip mechanics are driving the foot mechanics, no wonder why orthotics show an inconsistent ability to control the foot and tibia.
  4. When choosing a shoe, go with what has worked for you in the past.  Don’t be lured into a shoe based on what a salesperson tells you you need because they can “see” you ‘over’pronate.  Go with what feels right and comfortable for you and if it doesn’t work, try something different.  Trial and error will work it out.
  5. If you are still getting injured, get a full body gait analysis and look into the credentials of the person looking at your form. Do they know anatomy and mechanics, or did they read “chi running”. No, this statement isn’t based on research reviewed in this paper, but since the research suggests that the mechanics are “top-down”, you need to look into the stability and mobility of the hips, pelvis and core. You may have some hip or pelvic mechanics that need changing.
  6. If you have tried a variety of shoes and feel you want to try orthotics, it appears off the shelf orthotics are just as beneficial (and 1/5 the price) as custom made, casted orthotics.

I hope you have enjoyed this paper and use it for reference.  I hope to be coming back and adding to it as more research is released.  Please feel free to forward it to your friends.

© 2013, Dr. Kevin Maggs