Comparison of foot strike patterns of barefoot and m inimally shod runners in a recreational road race

Peter Larson

Performance Health Spine and Sport Therapy,91A N.State St.,Concord,NH 03301,USA

Comparison of foot strike patterns of barefoot and m inimally shod runners in a recreational road race

Peter Larson

Performance Health Spine and Sport Therapy,91A N.State St.,Concord,NH 03301,USA

Background:Previous studies of foot strike patterns of distance runners in road races have typically found that the overwhelming majority of shod runners initially contact the ground on the rearfoot.However,none of these studies has attempted to quantify footstrike patterns ofbarefoot or m inimally shod runners.This study classifies foot strike patterns of barefoot and m inimally shod runners in a recreational road race.

Methods:High-speed video footage was obtained of 169 barefootand 42 minimally shod distance runners at the 2011 New York City Barefoot Run.Footstrike patterns were classified for each runner,and frequencies of forefoot,m idfoot,and rearfootstriking were compared between the barefoot and m inimally shod groups.

Results:A total of 59.2%of barefoot runners were forefoot strikers,20.1%were midfoot strikers,and 20.7%were rearfoot strikers.For m inimally shod runners,33.3%were forefootstrikers,19.1%were m idfoot strikers,and 47.6%were rearfoot strikers.Foot strike distributions for barefootand m inimally shod runners were significantly differentboth from one another and from previously reported footstrike distributions of shod road racers.

Conclusion:Footstrike patterns differ between barefoot and m inimally shod runners,w ith forefoot striking being more common,and rearfoot striking less common in the barefoot group.

CopyrightⒸ2014,Shanghai University of Sport.Production and hosting by Elsevier B.V.All rights reserved.

Barefoot;Biomechanics;Foot strike;Gait;M inimal shoes;Running

1.Introduction

Footstrikes during running are typically classified as either (1)rearfoot,in which initial contact is made somewhere on the heel or rear one-third of the foot;(2)m idfoot,in which the heel and the region below the fi fth metatarsal contact simultaneously;or(3)forefoot,in which initial contact is made on the front half of the foot,after which heel contact typically follows shortly thereafter.1Previous research on foot strike patterns in road races indicates that the majority of shod distance runners are rearfoot strikers,w ith percentages ranging from 74.9%of runners in an elite half-marathon race,1to 81% of recreational runners in a 10-km race,2to over 90%of recreational runners in marathon distance events3,4(Table 1).

Available research suggests that multiple factors influence the type of foot strike exhibited by a given runner under a given setof conditions.Forexample,several race studies have found that the percentage of non-heel striking runners increased among faster runners,1,2,4,5suggesting a speed effect.Running surface has also been shown to influence foot strike.Nigg6reports data from an unpublished thesis7show ing that barefoot runners are more likely to forefoot strike on asphalt(76.7%forefoot,23.3%rearfoot),and rearfoot strike on grass(45.7%forefoot,54.3%rearfoot).Gruberetal.8found thatonly 20%of habitually shod runners adopted a m idfootor forefoot strike when running barefoot on a softsurface,versus65%adopting a m idfoot or forefoot strike when running barefoot on a hard surface.

Table 1 Summary of foot strike patterns reported in observational studies of runners in races.To simplify comparisons,data for asymmetrical runners reported by Larson et al.3and Kasmer et al.4are not included here.

Of all potential factors contributing to variation in foot strike type,the role of footwear has perhaps been the subject of mostdebate and research in recentyears.This is in partdue to increased interest in barefoot running,as well as marketing of“barefoot-style”minimally-cushioned shoes by running footwear companies.

Results of studies that have examined the effects of footwear on foot strike include:

·Lieberman et al.9found that habitually unshod Kenyan and American runners typically land on their midfoot or forefoot while running barefoot,whereas habitually shod Kenyan and American runners tend to contact the ground w ith the rearfoot/heel fi rst in both shod and unshod conditions.

·Hatala etal.10reported thathabitually barefoot individuals from the Daasanach tribe of Kenya,a group without a strong running tradition,overwhelm ingly tended to land on the rearfoot at a slow jogging pace(2.01—3.00 m/s; 83% rearfoot strikes).Percentage of m idfoot strikes increased w ith running speed among the Daasanach(60% of all foot strikes were classified as m idfoot at 6.01—7.00 m/s).

·Squadrone and Gallozzi11found that strike index was sim ilar among habitually barefoot runners when running unshod or in m inimally cushioned shoes(MCS;Vibram Fivefingers,Vibram USA,Concord,MA,USA)as compared to when they ran in conventional cushioned shoes.

·Ham ill et al.12found that habitually shod subjects typically sw itched to a m idfoot strike when running barefoot, but landed initially on the heel in all shod conditions (including a m inimally cushioned shoe).

·Tenbroek et al.13found that habitually shod runners exhibited a flatter foot strike when running barefoot or in m inimally cushioned footwear compared to moderate or thickly cushioned shoes,but all contacted fi rst on the heel.

·Bonacci et al.14found that habitually shod runners exhibited a less dorsiflexed ankle at contact when running barefoot compared to when they ran in three shod conditions(conventional shoe,racing flat,and moderately cushioned shoe).

·Lieberman15found that Tarahumara runners from Mexico who habitually wear m inimally cushioned huarache sandals tend to m idfoot or forefoot strike when they run, whereas conventionally shod Tarahumara typically land on the rearfoot.

·Pontzer et al.16found that the presence of footwear (m inimally cushioned sandalsvs.barefoot)had no effect on strike type among Hadza hunter-gatherers.

·Several additional studies have demonstrated that even if habitually shod runners continue to rearfoot strike when barefoot,they tend to land with reduced dorsiflexion of the ankle at contact than when shod.17—19

A limitation of existing studies of foot strike in barefoot and minimally shod runners is thatmosthave been conducted on small sample sizes of subjects in a laboratory setting or along a shortoutdoor runway.None have examined footstrike patterns of barefoot/m inimally shod runners in a race setting on a hard,asphalt surface.

The goals of this study are thus(1)to determ ine the frequency of forefoot,m idfoot,and rearfoot striking in a comparatively large sample of barefoot and m inimally shod runners in a recreational road race;(2)to compare foot strike distributions between barefoot and m inimally shod runners; and(3)to compare foot strike distributions observed here to those reported in previous studies of recreational distance runners.The nullhypotheses tested are:(1)footstrike patterns do notdiffer between barefootand minimally shod runners in a recreational road race;(2)footstrike patterns exam ined heredo not differ from those reported previously in the literature for conventionally shod runners in road races.

2.M ethods

2.1.Procedure

Runners were videotaped at the New York City Barefoot Run on 25 September,2011.This event involved loops around Governor’s Island in NYC.Runners were videotaped about 350 m from the starting line as they passed by on a flat,asphalt road surface.Only data from the fi rst loop are reported here since many runners only ran one loop around the island(one loop on the course was approximately 3.25 km long).

Video recording was carried outw ith a Casio Exilim EX-F1 digital camera(Casio America,Inc.,Dover,NJ,USA)at a fi lm ing rate of 300 Hz.The camera was mounted on a tripod near ground level,and was oriented perpendicular to the passing runners so that they could be videotaped in the sagittal plane.The camera was obscured next to a patch of vegetation so that runners would be unlikely to notice it as they approached.The race course was approximately 10 m in w idth at the fi lm ing location.Thus,distance of runners from the camera was variable,but most runners were sufficiently close to allow clear visualization of the location of initial foot contact.Because there was no formal timing for this event,it was not possible to identify individual runners by their bib numbers or finish times.

2.2.Participants

Foot strike was classified for a total of 241 runners.This sample included 169 barefoot runners,and 42 runners who were wearing Vibram FivefingersⓇ(VFF)running shoes (remaining runners wore a mixture of sandals,conventionally cushioned running shoes,and other brands of m inimally cushioned shoe sample sizes for each were insufficient to allow statistical comparisons).VFF is a m inimally cushioned line of footwear characterized by individual pockets for the toes.It was not possible to identify specific shoe models w ithin the brand,butallare minimally cushioned(0—4 mm of m idsole cushioning for models available at the time of the race).Pacing for most runners could be subjectively described as an easy jog.

The sample of runners exam ined in this study was a m ixture of habitual barefoot runners,those who had recently begun running barefootor in minimal running shoes,and those who may have been running barefoot for the fi rst time in this event.Many had traveled from a distance to attend the race, which was part of a weekend-long event that specifically catered to barefoot enthusiasts.

2.3.Data collection

Video for each runner was analyzed frame-by-frame in Apple Quicktime Pro(Apple Inc.,Cupertino,CA,USA).Foot strike categorization followed the methods described by Hasegawa et al.1and Larson et al.3A rearfoot strike was defined as one in which fi rst contact of the foot w ith the ground was made on the heel or rear one-third of the sole (Fig.1A).A midfoot strike was defined as one in which the heel and the region below the fi fth metatarsal contacted the ground simultaneously(Fig.1B).A forefootstrike was defined as one in which initial contactof the footw ith the ground was on the fronthalf of the sole,w ith no heel contactat footstrike (Fig.1C).Foreach runner,footstrike was classified for the left foot(which was closest to the camera).In most cases,the fi rst foot strike observed for a given runner once they entered the frame of the video was analyzed.Deviations from this procedure occurred only if the initial left footstrike was obscured (e.g.,by another runner).A subsetanalysis that Iconducted on a sim ilar sample of video footage in a previous study found that the method of footstrike classification utilized here had a repeatability of 98%.3

2.4.Statistical analyses

Footstrike frequency distributions were compared between barefoot and m inimally shod runners using chi-square(X2) analysis.Foot strike distributions obtained here were also compared to those of traditionally shod recreational runners reported by Larson etal.3and Kasmer etal.4using chi-square analysis(asymmetrical runners in those studies were excluded from the analysis since this study did not attempt to quantifyfrequency of asymmetry).Comparisons are only made to those two studies because they exam ined a sim ilar recreational sample of runners.A Bonferroni correction was applied to correct for multiple independent comparisons.Thus,frequency distributions compared in chi-square analyses were considered significantly different atp＜ 0.01.All statistical analyses were carried out using NCSS 8(NCSS,LLC, Kaysville,UT,USA).

Fig.1.Examples of foot strike patterns captured from high-speed video of the 2011 New York City Barefoot Run.(A)Rearfoot strike;(B)midfoot strike;(C) forefoot strike.

3.Results

A totalof 59.2%of barefoot runners were forefootstrikers, 20.1%were m idfootstrikers,and 20.7%were rearfootstrikers. For the m inimally shod runners,33.3%were forefoot strikers, 19.1%were m idfootstrikers,and 47.6%were rearfootstrikers (Table 1).

Results of chi-square analyses indicate that observed foot strike frequency distributions differ significantly between barefoot and m inimally shod runners(X2=13.5,df=2,p＜0.01).The foot strike frequency distribution for barefoot runners in this study differs significantly from those recorded for traditionally shod road racers in Larson et al.3(X2=571.63,df=2,p＜ 0.0001)and Kasmer et al.4(X2=751.86,df=2,p＜ 0.0001).The foot strike frequency distribution for m inimally shod runners in this study differs significantly from those recorded for traditionally shod road racers in Larson etal.3(X2=149.2,df=2,p＜0.0001) and Kasmer et al.4(X2=265.88,df=2,p＜0.0001).

4.Discussion

Available published data from road race studies conducted to date indicate thatapproximately 75%—95%of runners land on their rearfoot when initially contacting the ground1—4(Table 1).It is reasonable to presume that the vast majority of the runners in these studies were habitually shod and wore some type of cushioned running shoe during the race,though exactshoe properties mightdiffer among running populations (e.g.,racing flats for elite half-marathoners,conventionally cushioned running shoes for recreational marathoners).In support of this presumption,only two of the 936 runners exam ined by Larson etal.3were wearing minimally cushioned running shoes(VFF for both;no runners were barefoot).

In contrast to the above studies,Lieberman et al.9observed that initial contact on the midfoot or forefoot is typical for habitually barefoot Kenyan adolescents on a dirt road(88%of foot strikes)and habitually barefoot American adults in the laboratory(75%of foot strikes).Incidence of rearfoot striking in this same population of habitually barefootAmerican adults increased from 25%to 50%when shod, and habitually shod Kenyans and Americans tended to rearfoot strike regardless of whether they were wearing shoes.9These results suggest that footwear may influence foot strike patterns.

Footstrike distributions for barefoot runners observed here were significantly different from those observed previously for shod road racers.Larson et al.3and Kasmer et al.4observed that less than 10%of runners in their samples were symmetrical forefoot or m idfoot strikers.In this study,79.3%of barefoot runners were forefoot or m idfoot strikers.This is fairly close to the percentages observed forhabitually barefoot American adults and Kenyan adolescents running w ithout shoes.9It is also similar to the pattern observed foradultmale Hadza hunter-gatherers running in sandals or barefoot.16However,it differs markedly from habitually barefoot Kenyans of the Daasanach tribe,10Hadza juveniles,and adult Hadza women.16

It is possible that speed,surface properties,and running experience are confounding variables when it comes to comparing foot strike patterns among studies.The habitually barefoot Kenyans in the studies mentioned above were all running on dirt or sand surfaces that are presumably softer than a laboratory runway or an asphalt road.Gruber et al.8found that m idfoot and forefoot striking was more common in barefoot runners on a hard surfacevs.a softer surface. Furthermore,the adolescents studied by Lieberman et al.9were experienced runners and were running at a fast pace (5.5 m/s).When the Daasanach,who are not considered frequent runners,ran at this speed or faster,frequency of m idfootand forefootstriking increased to the pointwhere they were more common combined than rearfoot striking.10In the Hadza tribe,adult women and children typically rearfoot strike,whereas adultmen typically m idfootstrike.16This latter finding suggests that running experience may also influence running form and foot strike type since adult Hadza men tend to run more often while hunting game as compared to Hadza women who primarily gather plant foods.Taken together,results from these studies suggest that determ ination of foot strike type is multifactorial,w ith m idfootand forefootstriking being most likely when experienced runners run barefoot on harder surfaces and at faster paces.

Foot strike distribution for m inimally shod runners was significantly different from both barefoot runners observed here and from shod runners observed in previous road race studies.A total of 52.4%of minimally shod runners were forefoot or midfoot strikers.Thus,frequency of forefoot and m idfoot striking in m inimally shod runners on an asphalt road is lower than in barefoot runners,but higher than in traditionally shod runners.It seems that at least in terms of foot strike,running in a minimally cushioned shoe may encourage kinematic patterns that are different than running in a traditionally cushioned shoe,but may not always encourage kinematic patterns sim ilar to that typically observed in barefoot running.The response may be very subject-specific.

Studies have observed significantly higher vertical impact force peaks and loading rates in rearfoot striking barefoot runners.9,18Given this,it is somewhat surprising that runners wearing VFF,a shoe thatprovides m inimal impactprotection to the foot,would continue to land on the rearfooton a hard surface like an asphalt road.There are a few possible explanations for this.First,it is possible that runners attending this“barefoot”race who were wearing minimal shoes were less experienced w ith barefoot running and thus wore shoes for protection(i.e.,they were not com fortable running fullybarefoot).It has been demonstrated that foot strike patterns in m inimal shoes can change w ith experience,and inexperienced m inimally shod runners may exhibit different gait mechanics than those who have had greater acclimation time.9,20

Another possible explanation for the difference in foot strike distribution between barefoot and m inimally shod runners is that the m inimal amount of cushioning present in VFF is enough to moderate impact force such that the body does not respond by making a gait change.Though a recent metaanalysis found that increased vertical impact loading rate is a risk factor for tibial stress fractures,21Nigg22has questioned whether impact force peaks or loading rates are a significant contributor to running injury.It may be the case that the loading rates experienced by some rearfootstriking m inimally shod runners are w ithin a normal range of tolerance for the human body.

A third possibility is that vertical impact force is not the only stimulus for foot strike change,and that some other factor besides a need to reduce impact force contributes to the higher frequency of m idfoot and forefoot striking in barefoot runners.For example,Robbins and Gouw23proposed that gait modifi cations in barefoot runners may in part be associated w ith horizontal loads applied to the plantar surface.In the barefoot condition,gait adaptations may be required to reduce shear forces between the foot and ground surface in order to protect the plantar skin of the foot.It seems reasonable to assume that the presence of even a m inimally cushioned shoe sole would both reduce plantar sensation and provide protection from ground shear,and thus stimulus for gait change may not be as strong as when running fully barefoot.

Differences in foot strike patterns observed here between barefoot and m inimally shod runners may have implications w ith regard to running injury.Failure to allow a gradual adaptation to running in m inimally cushioned shoes to accommodate gait and tissue adaptation could be potentially injurious.Giuliani et al.24reported a case study of two runners who developed 2nd metarsal stress fractures after transitioning to m inimally cushioned shoes.Ridge et al.25found that approximately 50%of runners who they studied developed marrow edema in at least one foot or ankle bone in concert w ith a 10-week adaptation to running in m inimally cushioned shoes(VFF).Two of their subjects developed stress fractures(2nd metatarsal for one,calcaneus for the other). This suggests increased remodeling of foot bones associated w ith a change to m inimal footwear,which could progress to bone damage in the form of a stress fracture.Unfortunately, Ridge et al.25did not report data on running form before or after the transition,and it is uncertain at which point in the gait cycle forces become potentially injurious for individual bones during a transition to minimal shoes(e.g.,impact, m idstance,toe-off,etc.).It is worth noting that Ryan et al.26gradually progressed runners into VFF over 12 weeks and found no elevated risk of injury compared to a conventionally cushioned running shoe(increased calf/shin pain was the only significant difference in the minimal shoe).Unfortunately kinematic data were not reported in that study so the role of form could not be addressed.

This study has a few lim itations that should be discussed. Because this race was partofa weekend-long barefoot running“festival”,many of those attending had participated in form clinics and barefoot running sem inars on the day prior to the race.Thus,it is possible that some runners were consciously running according to how they had been taught the previous day.However,since both barefootand m inimally shod runners had the opportunity to attend the same form sem inars,the comparisons between barefootand minimally shod runners in this race should nothave been affected.

It is possible that the overall frequency of m idfoot and forefoot striking was inflated by subjects forcing their form to meet their perception of how they should run when barefootor in minimal footwear.It is for this reason that Ichose to fi lm in a discrete way 350 m from the starting line.The intentof this protocolwas to allow runners time to settle into the run and to m inim ize the likelihood that they would notice that they were passing a camera.Despite this concern,itshould be noted that frequency of forefoot and m idfoot striking observed here are not inconsistent w ith results of other studies that have observed barefoot runners on hard surfaces.8,9,27

It is also important to point out that this study only classified the initial contact pointof the footw ith the ground into three broad categories.It was not possible to examine the forces associated w ith ground contact or accurately assess kinematic variability w ithin the discrete categories.W ide variation in initial contact position has been recognized for a long time,28and such variation may influence patterns of force application.Forexample,Logan etal.29reported a high degree of variability in force measurements among rearfoot-striking runners in a comparison of gait mechanics between cushioned running shoes,racing flats,and distance spikes;they suggested individual differences in initial contact location as a possible explanation for this variation.Altman and Davis27found that visually assessed m idfoot strikers were often classified as forefoot or heel strikers by the strike index method. Recent research also suggests that runners who contact fi rston the heel exhibit variation in the location of maximal vertical impact loading,w ith as many as 25%—33%of runners who contacton the heelexperiencing maximal vertical loading rate when the center of pressure is under the m idfoot.30

Despite the potential for variation in force measurements w ithin visually assessed foot strike categories,a recent laboratory study found that foot strike angle at contact correlates well w ith kinetic measures of foot strike such as the strike index.27It is difficult to obtain accurate measurements of joint angles from video recorded in the field,but the rearfootstrikes observed here for barefoot and m inimally shod runners appeared subjectively to be m ilder(less inclination of the sole relative to the ground at contact)than those observed in the shod runners of Larson et al.3This is supported by studies show ing that foot positioning at contact in runners wearing minimal footwear is more similar to the barefoot condition than to the conventional shoe condition even if they continue to contact fi rston the heel.11,13It is thus possible that runningform varies between footwear conditions in subtle ways that were not measured here,and future studies that attempt to undertake finer scale measurement of kinematic variables in the field are needed.

5.Conclusion

The results of this study provide insight into the role of footwear in determ ining footstrike pattern.They indicate that the majority of barefoot runners tend to contact the ground on the m idfootor forefootwhen running on an asphalt road.This contrasts w ith the typical rearfoot striking pattern observed in conventionally shod runners on hard surfaces.Results also show that a m inimally cushioned running shoe may not perfectly simulate barefoot running,w ith frequency of m idfoot and forefoot striking being approximately equal to rearfoot striking.

1.Hasegawa H,Yamauchi T,Kraemer WJ.Footstrike patterns of runners at the 15-km point during an elite-level half marathon.J Strength Cond Res2007;21:888—93.

2.Kerr BA,Beauchamp L,Fisher V,Neil R.Footstrike patterns in distance running.In:Kerr BA,editor.Biomechanical aspects of sport shoes and playing surfaces:proceedings of the international symposium on biomechanical aspects of sport shoes and playing surfaces.Calgary:University Press;1983.p.135—42.

3.Larson PM,Higgins E,Kam inski J,Decker T,Preble J,Lyons D,et al. Footstrike patterns of recreationaland sub-elite runners in a long-distance road race.J Sports Sci2011;29:1665—73.

4.Kasmer ME,Liu X,Roberts KG,Valadao JM.Foot-strike pattern and performance in a marathon.IntJSportsPhysiolPerform2013;8:286—92.

5.Hayes P,Caplan N.Foot strike patterns and ground contact times during high-calibre m iddle-distance races.J Sports Sci2012;30:1275—83.

6.Nigg BM.Biomechanics ofsports shoes.Calgary:Topline Printing;2010. p.300.

7.Herzog W.The influence of velocity and playing surfaces on the load on the human body in running.Zurich,Switzerland:ETH Zurich;1978 [Dissertation].

8.Gruber AH,Silvernail JF,Brueggemann P,Rohr E,Ham ill J.Footfall patterns during barefoot running on harder and softer surfaces.Footwear Sci2013;5:39—44.

9.Lieberman DE,Venkadesan M,Werbel WA,Daoud AI,D’Andrea S, Davis IS,et al.Foot strike patterns and collision forces in habitually barefoot versus shod runners.Nature2010;463:531—5.

10.Hatala KG,DingwallHL,Wunderlich RE,Richmond BG.Variation in foot strike patterns during running among habitually barefoot populations.PLoS One2013;8:e52548.http://dx.doi.org/10.1371/journal.pone.0052548.

11.Squadrone R,GallozziC.Biomechanicaland physiologicalcomparison of barefoot and two shod conditions in experienced barefoot runners.J Sports Med Phys Fitness2009;49:6—13.

12.Hamill J,Russell EM,Gruber AH,M iller R.Impact characteristics in shod and barefoot running.Footwear Sci2011;3:33—40.

13.Tenbroek TM,Rodrigues P,Frederick EC,Hamill J.Effects of unknown footwear m idsole thickness on running kinematics w ithin the initial six minutes of running.Footwear Sci2013;5:27—37.

14.Bonacci J,Saunders PU,Hicks A,Rantalainen T,Vincenzino BT, Spratford W.Running in a minimalistand lightweightshoe is not the same as running barefoot:a biomechanical study.Br JSports Med2013;47:387—92.

15.Lieberman DE.Strike type variation among Tarahumara Indians in m inimal sandals versus conventional running shoes.J Sport Health Sci2014;3:86—94.

16.Pontzer H,Suchman K,Raichlen DA,Wood BM,Mabulla AZP, Marlowe FW.Foot strike patterns and hind limb joint angles during running in Hadza hunter-gatherers.J Sport Health Sci2014;3:95—101. 17.M cNair PJ,Marshall RN.Kinematic and kinetic parameters associated w ith running in different shoes.Br J Sports Med1994;28:256—60.

18.De Wit B,De Clercq D,Aerts P.Biomechanical analysis of the stance phase during barefootand shod running.J Biomech2000;33:269—78.

19.Bishop M,Fiolkowski P,Conrad B,Brunt D,Horodyski M.Athletic footwear,leg stiffness,and running kinematics.JAthl Train2006;41:387—92.

20.Warne JP,Warrington GD.Four-week habituation to simulated barefoot running improves running economy when compared w ith shod running.Scand J Med Sci Sports;2012.http://dx.doi.org/10.1111/sms.12032. [Epub ahead of print].

21.Zadpoor AA,Nikooyan AA.The relationship between lower—extremity stress fractures and the ground reaction force:a systematic review.Clin Biomech2011;26:23—8.

22.Nigg BM.Impact forces in running.Curr Opin Orthop1997;8:43—7.

23.Robbins SE,Gouw GJ.Athletic footwear and chronic overloading:a brief review.Sports Med1990;9:76—85.

24.Giuliani J,Masini B,A litz C,Owens BD.Barefoot-simulating footwear associated w ith metatarsal stress injury in 2 runners.Orthopedics2011;34:e320—3.

25.Ridge ST,Johnson AW,M itchell UH,Hunter I,Robinson E,Rich BS, et al.Foot bone marrow edema after a 10-wk transition to m inimalist running shoes.Med Sci Sports Exerc2013;45:1363—8.

26.Ryan M,Elashi M,Newsham-West R,Taunton J.Exam ining injury risk and pain perception in runners using m inimalist footwear.Br J Sports Med;2013.http://dx.doi.org/10.1136/bjsports-2012-092061.[Epub ahead of print].

27.A ltman AR,Davis IS.A kinematic method for footstrike pattern detection in barefoot and shod runners.Gait Posture2012;35:298—300.

28.Cavanagh PR,Lafortune MA.Ground reaction forces in distance running.J Biomech1980;13:397—406.

29.Logan S,Hunter I,Hopkins JT,Feland JB,Parcell AC.Ground reaction force differences between running shoes,racing flats,and distance spikes in runners.J Sports Sci Med2010;9:147—53.

30.Breine B,Malcolm P,Frederick EC,DeClercq D.Initial foot contact patterns during steady state shod running.Footwear Sci2013;5:S81—2.

Received 25 September 2013;revised 4 February 2014;accepted 10 March 2014

E-mail address:peter.larson@performancehealthnh.com

Peer review under responsibility of Shanghai University of Sport.

2095-2546/$-see front matter CopyrightⒸ2014,Shanghai University of Sport.Production and hosting by Elsevier B.V.A ll rights reserved. http://dx.doi.org/10.1016/j.jshs.2014.03.003