BA
CHELOR
THESIS
Bachelor Programme in Exercise Biomedicine, 180 credits
Analysis of Changes in Running Technique
Between a Shod and Barefoot Running
Condition.
Matilda Andersson
Exercise Biomedicine, 15 credits
Analysis of Changes in Running
Technique Between a Shod and
Barefoot Running Condition
Matilda Andersson
2016-05-31
Bachelor Thesis 15 credits in Exercise Biomedicine Halmstad University
School of Business, Engineering and Science Thesis supervisor: Hanneke Boon
Abstract
Background: Lately, barefoot running has become popular and there is a debate on the pros
and cons of barefoot running with regards to running injuries. Many factors are causing injuries and one of the factors discussed is the fact that we run in shoes. When we run in shoes the biomechanics of the running technique may and therefore be a possible cause to injury. Aim: The aim of the study was to assess how the foot strike pattern, angle of the knee and ankle joint at time of initial contact, as well as the step length changes between a shod and barefoot running condition in habitually shod runners when running in a pace equivalent to their running pace over ten kilometers. Method: Twenty-seven healthy runners (18 male, 9 female) were included in the study. The study took place at the fitness center of Halmstad University. Subjects ran on a treadmill, in an individual pace equivalent to their running pace over ten km, both in a shod and barefoot running condition. Two-dimensional analysis of the sagittal plane kinematics of the knee joint, ankle joint and foot position to horizontal, foot strike pattern and step length was done. Participants ran for ten minutes with shoes and for five minutes barefoot. Running technique was videotaped using an Iphone 6 camera and landmarks were marked with white tape to ease the analysis. Results: Changes in foot strike pattern was observed. When running barefoot 63% of the subjects adopted a non-heel strike pattern compared to 18.5% when shod (p=0.001). Knee flexion was increased at IC for the barefoot condition, with 164°±6 relative knee angle compared to 167°±6 when shod (p=0.001). Ankle angle at IC did not show a statistical significant difference between conditions (p=0.657). When barefoot the angle was 117°±8 compared to 115°±8 when shod. Foot angle to horizontal showed a flatter foot placement at IC with a less dorsiflexed foot for the barefoot condition (-4°±8) compared to shod (-12°±8), (p=0.001). Step length was decreased for the barefoot condition (0.82m ±0.15) compared to shod (0.85m ±0.13), (p=0.008). Conclusion: Results are consistent with previous findings that barefoot running in some cases change the running technique with a flatter foot placement, an increased knee flexion at IC and a decreased step length. However, caution must be taken when habitually shod runners transition to barefoot running in regards to the biomechanical changes that may occur. To benefit from barefoot running a non-heel strike pattern is required. Further, the running technique may be the more important factor, regardless of wearing shoes or not.
Skillnader i löpteknik mellan traditionell löpning i skor jämfört med
barfotalöpning
Bakgrund: Barfotalöpning har under den senaste tiden blivit mer populär och det debatteras
kring fördelar och nackdelar att springa barfota med hänsyn till löprelaterade skador. Det finns flera faktorer som kan orsaka skador när det kommer till löpning där skor har varit en faktor som diskuterats i samband med de biomekaniska förändringar som sker när vi springer i skor.
Syfte: Syftet med studien var att jämföra de biomekaniska skillnaderna mellan traditionell
löpning i sko jämfört med barfotalöpning hos löpare som är vana vid att springa i skor. Skillnader jämfördes i fotisättning, vinklar i knä, ankel och fotled samt steglängden. Metod: Tjugosju distanslöpare (18 män och 9 kvinnor) deltog i studien. Studien utfördes på Högskolan i Halmstad. Testdeltagarna sprang på ett löpband i en hastighet likvärdig den hastighet de springer tio km på både med och utan skor. Två dimensionell analys från sagittal plan av förändringar i fotisättning, vinklar i knäled, ankelled och fotens position till det horisontella, samt steglängden dokumenterades med videofilmning. Testpersonerna sprang tio minuter med skor och fem minuter utan skor på samma hastighet. Resultat: Förändring i fotisättning visade en signifikant skillnad där 63% av testdeltagarna övergick till en framfotaisättning under barfotalöpningen jämfört med 18,5% med skor (p=0.001). Resultat visade en ökad flexion i knäled vid isätt när testpersonerna sprang barfota. Knävinkeln var 164°±6° relativ knävinkel jämfört med 167°±6° med skor (p=0.00.1). Vinkel i ankeln visade ingen signifikant skillnad mellan barfota (117°±8°) och skor (115°±8°) (p=657). Fotens vinkel till den horisontella linjen visade på en mindre dorsalflexion vid barfota (-4°±8°) jämfört med skor (-12°±8°) (p=0.001). Steglängden var kortare vid barfota (0.82m ±0.15) jämfört med skor (0.85m ±0.13) (p=0.008).
Slutsats: Resultat i studien stämmer överens med tidigare studier gällande de biomekaniska
förändringar som sker vid övergång till barfotalöpning. Barfotalöpning kan ge en framfotaisättning, en ökad flexion i knäled vid isätt, och kortare steglängd. Löpare som övergår till barfotalöpning måste vara försiktiga och en övergångstid kan vara nödvändig för att undvika skador. Tidigare forskning har visat att löparen måste förändra sin löpteknik med kortare steg och en isättning där framfoten landar först, för att få positiva effekter av barfotalöpning.
Table of content
1. Background ... 1
1.2 Biomechanics of barefoot and shod running ... 2
1.2.1 Foot strike pattern in the barefoot versus shod condition ... 2
1.2.2 Forces acting upon the body while running barefoot versus shod ... 3
1.2.3 Changes in step length and step frequency ... 5
1.2.4 Foot strike patterns at different running speeds ... 6
1.2. Aim ... 6 1.3. Research questions ... 7 2. Method ... 7 2.1 Subjects ... 7 2.2 Testing procedures ... 8 2.2.1 Pilot test ... 8 2.2.2 Pre-test ... 8 2.2.3 Testing session ... 9 2.3 Material ... 10 2.4 Data analysis ... 10 2.5 Statistics ... 12 2.6 Limitations ... 12
2.7 Ethical and social consideration ... 12
3. Results ... 13 4. Discussion ... 15 4.1 Discussion of results... 15 4.2 Discussion of methods ... 18 4.3 Conclusion ... 19 References ... 21 Appendix A ... 24 Appendix B ... 26 Appendix C ... 28
1
1. Background
Human ancestors have been running since millions of years ago. Not until around 40.000-50.000 years ago humans started to use footwear. In the 1970’s the running population and popularity increased and the first modern running shoes were invented. However, there is still a large number of running related injuries. Over a year, 19.4 % to 79.3 % of distance runners are likely to suffer from any form of lower extremity injury (Van Gent, Siem, Van Middelkoop, Van Os, Bierma-Zeinstra, & Koes, 2007).
Lately, barefoot running has become popular and there is a debate on the pros and cons of barefoot running with regards to running injuries. From an evolutionary perspective running barefoot is not something new, we have been doing it for a long time hence some authors are pointing out it would be a natural way of running (Lieberman, 2012). Researchers are also speculating that barefoot running could increase running performance and decrease running injuries (Lieberman, 2012; Murphy, Curry & Matzkin, 2013; Tam, Wilson, Noakes & Tucker, 2013). However, others think the increased injury rates are due to the increased running population and changes in its composition. In the United states the amount of runners and the weekly running distance has increased tremendously over the past 35 years (Murphy et al., 2013). As an example, the amount of runners participating in the New York marathon 1983 was 14 546, compared to 2011 where there were more than 47 000 runners participating. This increase in total number of runners also suggest that the composition of the group that runs marathons has changed. Slower and not equally trained runners are thought to be more prone to injury due to differences in physical conditioning and strength (Tam et al., 2013).
The topic of running injuries is interesting because of the greatly increased incidence of running injuries. Many factors are causing injuries and researchers have not been able to identify one single cause but have identified factors that could be related to increased injuries.
Some of the risk factors that have been discussed are; repetitive forces on the lower extremity, overuse and previous injury, kinematic variables like foot strike patterns, stride length, step frequency, leg stiffness and joint angles (Saragiotto, Yamato, Hespanhol Junior, Rainbow, Davis, & Lopes, 2014; Shih, Lin, & Shiang, 2013; Murphy et al., 2013). Another factor that has been discussed to be a reason of increased running injuries is the fact that we run in shoes.
When we put our feet into shoes we are changing the mechanics of our running style. The cushion in traditional running shoes makes the runner able to land on the heel with a reduced
2
impact force. Runners tend to take longer steps when running in shoes and to land with a more extended knee. This is in contrast to barefoot running, where the runner tends to take shorter steps and land with a flatter foot placement (De Wit, De Clercq, & Aerts, 2000; Lieberman, 2012; Thompson, Gutmann, Seegmiller, & McGowan, 2014).
The factors causing running injuries are hard to identify due to the wide variability in mechanics between individuals, that there is not a single injury factor, and the differences in study design and methodology done on the topic (Tam et al., 2013). However, a lot of research has been done comparing the differences between shod and barefoot running, and how they could be a cause of injury. These are factors like foot strike pattern, impact forces and joint kinematics that will be discussed further.
1.2 Biomechanics of barefoot and shod running
When studying the differences between shod and barefoot running, the biomechanics of running gait have to be understood. The gait cycle for one leg is divided into four phases: stance phase, early float, mid swing and late float (Lohman, Everett, Balan, Sackiriyas & Swen, 2011). The stance phase is most commonly discussed in the change between shod and barefoot running, and especially the time when the foot makes contact with the ground (initial contact, IC). At this point the foot strike pattern may be observed (Figure 1).
Figure 1. The running gait cycle (Lohman et al., 2011).
1.2.1 Foot strike pattern in the barefoot versus shod condition
A heel strike pattern is when the runner lands on the heel followed by a heel-toe movement during stance phase. Mid-foot strike is when the toe and heel reach contact with ground at the
3
same time and a forefoot strike is when the runner lands on their forefoot before bringing down the heel (Lieberman, Venkadesan, Werbel, Daoud, Andrea, Davis, Ojiambo Mang’Eni & Pitsiladis, 2010). The heel strike pattern is the most commonly used foot strike pattern among distance runners with percentages ranging from 75 % to 90 % found in road races (Larson, 2014). When changing to a barefoot condition some runners have shown to move into a mid-foot or a foremid-foot strike pattern (Figure 2). So the question is; what consequences this change in foot strike pattern have with regards to biomechanics of the running gait and possible injury factors.
Figure 2. Foot strike pattern and Ground Reaction Force of a, heel strike pattern with shoes and b, forefoot strike pattern barefoot (Lieberman et al.,2010).
1.2.2 Forces acting upon the body while running barefoot versus shod
When we run and walk forces act upon our body. Ground reaction force (GRF) is the force that is required to support our body weight. During running the vertical GRF can be as much as two to five times our body weight (Hamill, & Knutzen, 2008). At initial contact with ground, an
a
4
impact peak is created that is most commonly seen when running with a heel strike pattern. When running with a midfoot or forefoot strike pattern the impact peak is lower and in some cases not notable (Hamill, & Knutzen, 2008; Lohman et al., 2011) (Figure 3).
Figure 3. Ground reaction forces in rear foot, mid.foot and forefoot strike pattern during stace phase (Lohman et.al, 2011).
As mentioned earlier, higher vertical GRF (Thompson, et al., 2014) and higher vertical impact peak (Hobara, Sato, Sakaguchi & Nakazawa, 2012) has been associated with an increase of running injuries. When running these forces are repeated with every step and would cause a high cumulative load on the body segments. Since researchers have found that higher impact forces are associated with running injuries, a change to a midfoot or forefoot strike pattern could be of interest for runners (Lieberman, 2012).
Although barefoot running is commonly associated with a forefoot strike (Squadrone, Rodano, Hamill & Preatoni, 2015; Tam et al., 2013), it is not an absolute consequence of barefoot running. When running barefoot some people tend to automatically land with a flatter foot placement. When a forefoot strike is adopted in barefoot running the stress the forces are creating under the heel can be reduced, although the forces under the forefoot and at the metatarsal heads are also greater (Squadrone & Gallozzi, 2009).
Henceforth, Cheung and Rainbow (2014) studied the acute effect on foot strike pattern and loading rates when habitual shod runners first attempted a barefoot running condition. Twenty out of 30 of the subjects showed a transition from heel strike to a non-heel strike pattern. The average vertical loading rate and instantaneous vertical loading rate were reduced when running barefoot for the non-heel strikers. However, the remaining ten subjects showed a mixed landing pattern and the average vertical loading rate and instantaneous vertical loading rate
5
were then higher (Cheung et al., 2014). This is consistent with Shih et al., (2013) who studied the average and maximal loading rate over different foot strike pattern. Results showed that a forefoot strike generated a lower average and maximal loading rate in both a barefoot and shod condition. However, the loading rates were greater for those who used heel strike when running barefoot compared to those who used heel strike when shod (Shih et al., 2013). Moreover, barefoot running with a forefoot strike increases the peak pressure under the forefoot, which might lead to an overuse or injury of the metatarsals (Murphy et al., 2013).
1.2.3 Changes in step length and step frequency
Other factors that have been shown to reduce the vertical impact peak in running are increased step frequency (Hobara et al., 2012) and decreased stride length (Thompson et al., 2014). Interestingly, these two factors have also been shown to change upon a transition from heel to midfoot strike (De Wit et al., 2000; Squadrone et al., 2009) and thus possibly in the transition from shod to barefoot running. With a shorter stride length, the runner will land with the foot located more underneath the body and closer to the center of mass. A shorter stride length can further result in a decrease in vertical GRFs (Thompson, et.al, 2014). When running barefoot, researchers have found a decrease in step length. This could be a result of the flatter foot placement at initial contact, which also affects the angle of the knee before touchdown, causing the heel to be placed closer to the center of the body (De Wit et al., 2000). However, a shorter step length and an increased step frequency may increase the accumulated load, due to more gait cycles, which could be another cause of injury (Hall, Barton, Jones, Morrissey, 2013).
Changes in the stride length and the foot strike pattern could influence the kinematics of the lower extremity during running. Previous research has found kinematic differences between shod and barefoot running where the largest difference seems to be the ankle angle at initial ground contact, as a result of a change in foot strike pattern (De Wit et al.,2000; Shih et al., 2013; Valenzuela, Lynn, Mikelson, Noffal & Judelson, 2015). A forefoot strike has been shown to correlate with greater plantar flexion at initial ground contact. A greater plantar flexion in the ankle has also been shown to change the kinematics in the knee and hip to be less extended (De Wit et al., 2000; Shih et al., 2013; Valenzuela et al., 2015).
However, Williams, Green and Wurzinger (2012) did not find any changes in the knee or hip joints in the barefoot or forefoot strike running compared to the shod condition with a rear foot strike pattern. As there were no changes in the knee joint kinematics, this study did not show a
6
decrease of stride length between barefoot and shod running, which could be a result of the participants not being used to run barefoot or with a forefoot strike pattern. Following this, it can be concluded that the kinematics in other joints can in some cases, if the runner is adapting a barefoot style, (i.e. mid-foot or forefoot strike pattern, shorter step length), be affected by the foot strike pattern (Williams et al., 2012).
1.2.4 Foot strike patterns at different running speeds
What has been found is that the foot strike pattern also depends on the running speed. A higher speed could influence the landing pattern in habitually rear foot strikers by moving more toward the forefoot during landing (Breine, Malcolm, Frederick & De clercq, 2014). As a result of increased speed the peak pressure and maximal force acting upon the body are increasing regardless of foot strike pattern (Cooper, Leissring, & Kernozek, 2015). Cooper et al. (2015) found that the subjects who changed their foot strike pattern to a non-rear foot strike showed a decrease in total forces. Although, when speed increased and foot strike pattern moved towards the forefoot, forces were greater towards the forefoot.
Several studies have compared shod and barefoot running biomechanics to find any differences in possible risk factors of running injuries. Although a lot of work has been done comparing the change in foot strike pattern, changes in joint angles and stride length over fast paces, researchers have in most cases used a set pace when analyzing these factors. When using a set pace for all runners this pace may not be on the same exertion level for all runners. Only a few studies have used a relative pace to the participant. Therefore, this study will observe the kinematic differences between a shod and barefoot condition during running at an individual pace that is equivalent to the pace they would run in over a ten km distance.
1.2. Aim
The aim of this study was to assess how the foot strike pattern, sagittal angles of the knee and ankle joints at initial contact time, as well as the step length, changes between a shod and barefoot running condition in habitually shod runners when running in a pace equivalent to their running pace over ten kilometers.
7
1.3. Research questions
How does foot strike pattern, step length and step frequency change when moving from a shod condition to a barefoot condition in habitually shod runners?
How do the sagittal angles of the knee, ankle and foot in the contact phase differ between a shod compared to barefoot condition in habitually shod runners?
2. Method
The study was a cross-sectional, descriptive study and data was collected in two conditions with the same participants during a single test.
The study took place at the fitness center of Halmstad University (Idrottscentrum). All participants attended two sessions. During the first session, participants were familiarized with treadmill running and the environment. Their normal foot strike pattern and running pace was recorded. During the second meeting, participants performed the main test, which included treadmill running in a shod and a barefoot condition where both conditions were video recorded.
2.1 Subjects
Twenty-nine (18 male and 11 female) healthy runners volunteered to participate in the study. Out of the 29, only 27 participants completed both test. Therefore 18 male and 9 female (age: 38 ± 12 years; height: 175 ± 11 cm; weight: 68.8 ± 12 kg; BMI= 22.3 ± 1.9 kg/m2) were included in the study. Participants were recruited from local running clubs around Halmstad and other local experienced runners from the university’s gym. This was done through a Facebook group (Facebook, Inc, California, 2004) created by the test leader. Local runners were invited to the group to get more information about the test (Appendix A). Inclusion criteria to participate in the study were the following: Participants should normally run in traditional running shoes, have a training pace between 4.30 and 6 minutes per km over a ten kilometer distance and a training volume of at least ten kilometers per week. All participants had to be free from running overuse injuries during the past six months and other chronic and orthopedic conditions that may affect running biomechanics, such as arthritis, osteoporosis, coronary disease, cancer, anterior cruciate ligament injury and acute musculoskeletal injuries. Participants had to be comfortable running for a minimum of 30 minutes on the treadmill.
8
2.2 Testing procedures
2.2.1 Pilot test
To ensure that the planned method would work a pilot study was performed one week in advance of the first testing session. The pilot took place in the fitness center at Halmstad University to ensure that the equipment was working with the environment of the center. Participants were marked with white tape on anatomical landmark as shown in figure 4. Participants warmed up for five minutes, and then performed the first part of the running test in a shod condition running for ten minutes. This was followed by a rest period of three minutes, during which they removed their shoes. After, they performed the second part of the running test in the same speed as when shod but in a barefoot condition for ten minutes. Video recordings were taken from a side angle 1.5 meter away from the treadmill with a height of 70 centimeters from the floor using an IPhone 6 mobile camera 720p and 240 frames per second (Apple Inc. California).
2.2.2 Pre-test
Participants were given a written explanation of the test and had to sign an informed consent (Appendix B) before the test begun. Information about anthropometric data, running experience, running training volume and running pace over a ten km distance was collected. Participants were then informed about the procedures and had the opportunity to ask questions if anything was unclear.
Participants were informed to bring all different running shoes they had for the test leaders to choose which shoes to wear during the test session. The shoe closest to a standard running shoes was chosen. The Minimalist Index (MI) by Esculier, Dubois, Roy and Dionne (2014), was used as a guideline to categorize the shoes as standard running shoes. Standard running shoes were defined as follows; cushioned sole, moderate to high resistance to longitudinal flexibility, moderate to high resistance to torsional flexibility (Esculier, 2014).
Participants started with a five minutes warm up on the treadmill in their own selected pace to get used to the treadmill. After five minutes the pace on the treadmill was increased to a pace equivalent to their ten km pace, which was decided by their training pace over ten km and race time over ten km. Video recordings were taken from a side angle to analyze their preferred foot strike pattern when running in shoes.
9
2.2.3 Testing session
All material was setup and the participants were informed about the procedures in the test. Participants were marked with white markers on anatomical landmarks including the greater trochanter, mid-thigh (20 centimeters inferior of the greater trochanter in line with the femoral bone), lateral femoral condyle, lateral malleolus of the ankle, calcaneus of the foot and the fifth metatarsal, as is shown in figure 4 (Fredericks, Swank, Teisberg, Hampton, Ridpath, & Hanna, 2015; Lieberman, Castillo, Otarola-Castillo, Sang, Sigei, Ojiambo & Pitsiladis, 2015).
Figure 4. Markers were placed on anatomical landmarks to ease the analysis (Fredericks et.al, 2015).
Participants warmed up in the shoes selected from the pre-test for five minutes on a self-selected pace. Then the pace was increased to the set pace from pre-test, with zero incline, running for ten minutes in shoes. This was followed by a rest period of three minutes were participants were informed to take their shoes off and new markers were placed on the socks at the same landmarks as with shoes. During the second condition participants had two minutes to get used to the barefoot condition but gradually the pace was increased to the same pace as when shod. When they reached the same pace participants ran for five minutes without shoes. Due to a burning sensation under subjects’ feet, the time running barefoot was reduced to five minutes.
Participants did not receive any form of verbal coaching on how to run. Only if they tended to run too close to the front of the treadmill this was adjusted to make sure that they were in the right position in relation to the camera. Every minute the participants were asked about their perceived exertion using a Borg scale to make sure the pace was relative to an exertion within
10
“fairly light” to “somewhat hard” on the scale (Borg, 1982). Data was collected through videotaping over 15 seconds during minute 9-10 in the shod condition and during minute 4-5 in the barefoot condition. This to make sure the runners had time to adapt to the treadmill and barefoot running.
2.3 Material
The test environment for this study was the university’s gym. Participants ran on a treadmill (Free Motion Fitness, FMTL8255P-EN.1) at a pace, defined as a pace the participant run a ten km distance in, (from now called ’ten km pace’). An IPhone 6 camera (Apple Inc. California) and Dartfish Classroom Analysis Software 5.5 (Dartfish Company, Fribourg, Switzerland) was used to analyze foot strike pattern, angles of knee, ankle and foot at IC and step length. The camera recorded in a slow motion setting, taking 240 frames per second with 720p resolution and was placed 1.5 meters away from the treadmill and 0.7 meters above the floor. The participants ran in their personal running shoes with tight fitted running clothes in a dark colored fabric. A Borg scale of perceived exertion was used to control that the pace for each individual was within “fairly light” to “somewhat hard” on the scale (Borg, 1982).
2.4 Data analysis
Analysis of foot strike pattern, step length, angles of the ankle joint, knee joint and foot angle to horizontal was done using Dartfish Classroom Analysis Software 5.5 (Dartfish Company, Fribourg, Switzerland). Only the right lower extremity was analyzed, as this was the closest one to the camera. Joint angles and foot strike pattern were analyzed at initial contact. Initial contact (IC) was defined as the first frame when the foot made any visible contact with the ground. Participants’ foot strike pattern was determined as follows: heel strikers, when the heel or posterior one third of the foot first made contact with ground, or non-heel strikers, when the heel and the forefoot simultaneous made contact with ground or when the anterior one third of the foot first made contact with ground (Cheung et al., 2014). This definition was used for both the shod and barefoot condition. Step length was measured from fifth metatarsal at IC to fifth metatarsal at toe off (Fredericks et al., 2015). A visual scale was determined for each participant using the measured distance between the lateral femoral condyle and the mark on the mid-thigh. This measure was then used as a reference mark for distance in Dartfish to get the step length (Fredericks et al., 2015; Lieberman et al., 2010).
11
The knee joint flexion angle at IC was analyzed using a relative angle between the line from the malleolus to the lateral femoral condyle and the line from the lateral femoral condyle to the greater trochanter of the femur (Lieberman et al., 2010).
The ankle joint angle at IC was analyzed by drawing one line from the fifth metatarsal to the lateral malleoli and from the lateral malleoli to the condyle of the femur, creating an angle between the two segments (Lieberman et al., 2010). The angle of the foot to horizontal at IC was measured from the lateral malleoli to the fifth metatarsal, creating an angle with the horizontal of the ground. This angle was then corrected by the same angle in mid-stance, where the foot was flat to the ground. A positive angle indicated a plantar flexion of the foot and a negative angle a dorsiflexion of the foot. Measurement of the angles are shown in figure 5 and 6.
Figure 5. Analysis of knee, ankle and foot angles at initial contact.
12
2.5 Statistics
Data was analyzed using the statistical software program IBM SPSS statistics 20.0. Data was tested for normal distribution using the Shapiro Wilks test. As data for the step length was not normally distributed, change in step length and angles of knee, ankle and foot was analyzed using the non-parametric Wilcoxon signed-rank test to determine whether there was a difference between the shod and barefoot condition. Data for foot strike pattern was analyzed using the non-parametric Chi-square test for nominal and ordinal data. Statistical significance was set at P ≤ 0.05 and all data are presented as median with min – max and mean ± standard deviation (SD).
2.6 Limitations
At first, the criteria to participate was females between 20-30 years but due to challenges in finding subjects the criteria had widen to both genders and age between 20-50 ending up with the mean age 38 ± 12 years.Three participants dropped out of the study after testing session one and therefore their results are not included in the study.
This study used a 2D data analysis technique, which only presents the angles visible in the sagittal plane. Any rotation occurring in the joint angles cannot be estimated, as this motion was not visible from the sagittal plane in a 2D analysis.
2.7 Ethical and social consideration
All participants were informed about the procedures of the tests both in written and verbal form. They were given time to read through the procedures and ask any upcoming questions. Before participation participants signed an informed consent to acknowledge that they had read and understood the procedures of the test as well as been informed about the risks of the tests and their rights of participation in the study. They were also informed that participation in the study is voluntary and that they could at any time drop out of the study with no explanation. The informed consent was approved from supervisor prior testing. All personal information and data was handled with discretion according to Personuppgiftslagen (PuL), (1998). The results were presented as a whole group and the participants were informed about this.
This study could help the running population to get more knowledge about how barefoot running may change the biomechanics of a persons’ running style. Conclusions that barefoot
13
runners will forefoot strike are well oversimplified and cannot be transformed to all cases of barefoot running and shod running and there is still a large variance between individuals in running biomechanics (Tam et al., 2013; Shih et al., 2013; Cheung et al., 2014). A lot of research has been published on barefoot running, where the changes in biomechanics from a shod condition have been widely spread. A lot of training and evaluation could be needed if someone that is a habitually shod runner have interest in changing to the barefoot running.
3. Results
Participant characteristics
Participant characteristics, training experience, volume and pace are described in table 1. Participants’ mean (± SD) for age, height, weight, were 38 ± 12 years, 175 ± 11 cm, 68.8 ± 12 kg respectively. Mean BMI were 22.3 ± 1.9 kg/m2, mean training volume and running experience were 37.6 ± 28.3 km/week and 17.5 ± 13.5 years respectively and mean pace during test were 12 (± 0.85) km/h.
Table 1. Anthropometric data and running experience among participants (N=27).
N=27 mean ± (SD)
Age 38 ± 12
Height (cm) 175 ± 11
Weight (kg) 69 ± 12
BMI (kg/m2) 22 ± 2
Running mileage (km/week) 38 ± 29
Running experience (years) 18 ± 14
Pace during test (km/h) 12 ± 0.85 SD = standard deviation
Foot strike pattern
Participants’ foot strike pattern was determined as follows: heel strikers, when the heel or posterior one third of the foot first made contact with ground, or non-heel strikers, when the heel and the forefoot simultaneous made contact with ground or when the anterior one third of the foot first made contact with ground (Cheung et al., 2014). Results for foot strike pattern are presented in table 2. Change in foot strike pattern showed a statistical significant difference between the two conditions where 22 participants heel striked when shod and ten participants heel striked when barefoot (p≤0.001).
14
Table 2 Foot strike pattern for shod and barefoot conditions. There was a statistical difference between conditions (p≤0,001), with more participants using heel strike compared to non-heel strike in the shod condition.
Foot strike pattern N= 27
Heel strike Non-heel strike
Shod 22 5
Barefoot 10 17
Out of the 27 participants, when running shod, 22 subjects (81.5%) showed a heel strike pattern and 5, (18.5%) of the participants a non-heel strike pattern. When changing to the barefoot condition 17 participants (63%) transitioned to a non-heel strike pattern whereas 10 (37%) still heel striked (figure 7).
Figure 7. Foot strike pattern over the two conditions. Showing the percentage of heel strikers and non-heel strikers.
Kinematic variables
Results for knee, ankle and foot angles as well as step length are presented in table 3. Results showed a statistical significance of increased knee flexion at initial contact for the barefoot condition (p≤0.001). Median flexion angle of the knee at initial contact when barefoot was 164° (min 153° - max 177°) compared to 167° (min 156° - max 178°) when shod. No statistical significance was found regarding the ankle joint angle between the shod and barefoot condition. Median angle of the ankle joint at initial contact when shod was 114° (min 102° – max 129°) compared to 118° (min 100° – max 135°) when barefoot (p= 0.657). However, results did show
81,5% 37% 18,5% 63% 0,0% 10,0% 20,0% 30,0% 40,0% 50,0% 60,0% 70,0% 80,0% 90,0% Shod Barefoot % O F S UBJE CT S
Foot strike pattern
15
a significant difference between the two conditions in the foot placement to horizontal (p≤0.001). In the shod condition the median angle of the foot was -14° (min -24° – max 2°) and for the barefoot condition -2° (min -19° – max 8°). A positive angle indicated to a more plantar flexed foot and a negative angle to a more dorsiflexed foot.
Step length was statistically significant different with a decrease in step length for the barefoot condition compared to the shod condition (p=0.008) with a median of 0.76 (min 0.62 – max 1.18) meters for barefoot and 0.86 (min 0.65 – max 1.19) meters for shod.
Table 3. Results of median (±SD) for joint angles at initial contact as well as step length between the shod and barefoot conditions.
Condition Median (min - max) Mean (± SD) P-value
Knee flexion shod at IC (°) 167 (156 - 178) 167 ± 6 0.001
Knee flexion barefoot at IC (°) 164 (153 - 177) 164 ± 6 0.001
Ankle angle shod at IC (°) 114 (102 - 129) 115 ± 8 0.657
Ankle angle barefoot at IC (°) 118 (100 - 135) 117 ± 8 0.657
Foot to horizontal shod at IC (°) -14 ((-24) – 2) -12 ± 8 0.001
Foot to horizontal barefoot at IC (°) -2 ((-19) – 8) -4 ± 8 0.001
Step length shod (m) 0.86 (0.65 - 1.19) 0.85 ± 0.13 0.008
Step length barefoot (m) 0.76 (0.62 - 1.18) 0.82 ± 0.15 0.008
SD= standard deviation IC= initial contact
4. Discussion
4.1 Discussion of results
The purpose of the study was to assess how the kinematics changes at initial contact when habitually shod runners run barefoot. In regards of foot strike pattern, angle of the knee joint, angle of the ankle joint and foot angle to horizontal, as well as the step length. The running pace was set as an individual pace similar to the pace they would keep when running a ten km distance. The main finding in this study was the change in foot strike pattern. Out of the 27
16
participants, 22 showed a heel strike pattern when running with shoes and five participants a non-heel strike pattern. When changing to the barefoot condition twelve participants (44.5%) transitioned to a non-heel strike pattern, ten (37.0%) kept a heel strike pattern and the five (19.5%) who showed a non-heel strike pattern when shod retained the same foot strike pattern. In total, 63% of all runners showed a non-heel strike pattern when running barefoot. This is consistent with some of the previous studies where a change in foot strike pattern towards a forefoot or midfoot strike pattern was observed when changing to a barefoot condition. Cheung et al., (2014) reported that two-thirds of their participants transitioned to a non-heel strike pattern when running barefoot on their first attempt. Similarly, Cooper et al., (2015) found that 60 % of their subjects adopted a non-heel strike pattern when running barefoot.
As above results show, it is clear that not all barefoot runners will forefoot strike, and furthermore, there is a large variance between individuals in running biomechanics (Tam et al., 2013; Shih et al., 2013; Cheung et al., 2014). In the current study 37 % of the subjects retained a heel strike pattern when running barefoot. When it comes to running injuries and loading rates, Shih et al., (2013) found a significant reduction in loading rate when running with a forefoot strike pattern, regardless of wearing shoes or not. Therefore, the change of foot strike pattern might be more important than changing the shoe condition (Shih et al., 2013).
The change in joint angles showed a statistical significant difference in the knee joint with a slightly more flexed knee at initial contact in the barefoot conditions. Although this was statistically significant, the difference observed was only three degrees between the conditions. It can be discussed whether this small change has any major effect on the running biomechanics. A similar change has been observed by other researchers where a more flexed foot is a preparation for the body to land with a flatter foot placement (De Wit et al., 2000; Lieberman et al., 2015; Squadrone et al., 2015). With a greater knee flexion at contact time and a flatter foot placement other studies have found that the runner absorbs the forces better compared to a more extended leg (Williams et al., 2012; Shih et al., 3013). Given this, a small change in knee flexion might be beneficial in preventing injuries related to high forces.
As mentioned earlier, previous studies have found a significantly greater plantar flexion angle in the ankle joint at IC when running barefoot (De Wit et al., 2000; Shih et al., 2013; Valenzuela et al., 2015). Since there was a significant difference in foot strike pattern in current study, one would assume a kinematic difference in the ankle joint as well. However, results did not show
17
a statistical significant difference between the two conditions in the ankle joint at IC. This could be due to a methodological error as the markers placed on the participant’s anatomical landmarks had to be replaced when changing from the shod to the barefoot condition. Another reason could be that when running in cushioned shoes with a forefoot strike pattern, runners have been shown to land with a more plantar flexed foot than when running barefoot. Further, barefoot running has shown to lead to a flatter foot placement or mid-foot strike (Shih et al., 2013; Williams et al.,2012). Interestingly, the foot angle to horizontal did show a statistical significant difference between conditions. When running in shoes, the foot angle indicated a more dorsiflexed foot with a median angle of -14° (min -24° – max 2°). In the barefoot condition this angle was -2° (min -19° – max 8°). Since a degree of zero indicate a midfoot-strike pattern, these findings are consistent with previous findings that barefoot running has shown to generate a mid-foot strike pattern (Shih et al., 2013; Williams et al.,2012).
Step length was decreased when running barefoot as consistent with previous findings (De Wit et al., 2000; Squadrone et al., 2009). The main reason for this could be a result from the change in foot strike pattern where 63% out of the 27 subjects showed a non-heel strike pattern when running barefoot. As De Wit et al. (2000) explained the change in step length could be due to a change in foot strike pattern. When a forefoot or midfoot strike pattern is adopted the foot will fall closer to the center of mass. However, not only the participants who changed their foot strike pattern showed a decreased step length. Therefore, the step length may also have decreased as a result of the change in knee flexion where a more flexed knee will make the foot fall closer to the center of mass at IC (De Wit et al., 2000).
Results of the current study are consistent with previous findings. Runners adopted, in 67% of cases, a different running technique with a change in foot strike pattern to a non-heel strike pattern, a greater flexion at the knee joint at initial contact and a shorter step length. However, these findings were not the same for all runners. Therefore, barefoot running might not have positive kinematic effects for all runners and some would need practice to benefit from barefoot running (Murphy et al., 2013; Tam et al., 2013). Previous research also shows that there is not one single factor that causes a running injury, and running barefoot would not prevent running injuries for all runners. Furthermore, the characteristics of a person's running technique seem more important than footwear (Lieberman, 2012; Tam et al., 2013; Shih et al., 2013; Thompson et al., 2014). When changing one component of the running biomechanics, other components seem to follow. Therefore, a runner who changes the foot strike pattern, will possibly also
18
change the sagittal angles of the joints at initial contact and the step length. Because the body has to adopt to a new running technique, fatigue might occur earlier. Future research should focus on how these components are changing when the runner fatigues.
4.2 Discussion of methods
There are some limitations to this study, which may have affected the results. This study used a 2D analysis of the kinematic differences in shod and barefoot running. Thus, only angles in a sagittal plane was observed, and potential movements in the frontal and transverse planes in the knee, ankle and foot were not possible to measure (Maykut et al.,2015; Squadrone et al., 2009). The running analysis was done on a treadmill with recreational runners whom were used to run outdoor. Although the pre-test was performed to get the participants familiarized with treadmill running, their running style may have been affected of the treadmill, adopting a different running pattern compared to outdoor running where the speed is self-selected. However, since both conditions were observed on a treadmill, using the same running speed and under the same conditions, current method is a reliable way to control for biomechanical changes between the shod and barefoot condition (Riley, Dicharry, Franz, Della Croce, Wilder & Kerrigan, 2008).
The treadmill that was used had sidebars that covered the hip of the subjects’, therefore the study had to focus on the lower part of the lower limbs of the body such as the knee, ankle and foot as well as step length and foot strike pattern. This made it difficult to draw any conclusion on how the lower limb kinematic could influence changes in the hip and torso.
Since the test was done in a gym environment at different times of the day the surrounding environment of people and noise could not be controlled. The setup of material had to be reset at each time of the testing sessions, which may have influenced the placement of the camera in relation to the treadmill and the participant’s position to the camera. However, all spatial measurements were performed within the same video recording. The environment in the gym also affected the position of the camera making it difficult to place the camera further away from the treadmill. The camera was set to the same height from the floor for all participants where the camera would be focused on different heights of the subjects’ i.e. comparing a short runner to a tall runner. These are factors that could affect the variations of the results with a greater standard deviation in angles of the knee joint, ankle joint and foot to horizontal.
19
However, even though the camera could have been in a different angle between subjects and their test sessions, the position of the camera was not changed within the test session of one subject. Therefore, the results should be reliable.
Before the second test participants were marked with white tape markers. Markers were placed by the same test leader at the anatomical landmarks to minimize sources of error. Markers placed on the fifth metatarsal and calcaneus had to be replaced when changing from a shod condition to a barefoot condition which need to take in consideration of the results. Participants were asked to wear tight fitted clothing. However, some of the markers placed on the clothes could have moved whilst running. This could have influenced the results of the angles in the knee and ankle with a greater variation of the results. Similarly, the markers used to analyze the visual scale in Dartfish could have moved when the participants’ ran on the treadmill. This could have had an impact on the measured step length in Dartfish with a greater variation of the results.
All participants wore their own running shoes, which were assessed at the first testing session. If a participant had more than one pair of running shoes the ones most similar to a traditional running shoe was used during the test. All participants were used to run in cushioned shoes and no one were used to barefoot running. The Minimalist Index (MI) by Esculier, Dubois, Roy and Dionne (2014), was used as a guideline to categorize the shoes as standard running shoes to get the shoes as standardized as possible.
The pace was decided by the time subjects run a 10 km race in and the pace they would keep over 10 km on training. This to be able to find a pace that was slower than earlier studies used. Some of the participants were used to run a longer distance than ten km, making their pace relative fast. This could have changed their foot strike pattern and step length. However, the majority of subjects ended up with a pace around 12 km/h (11.97 ± 0.85).
4.3 Conclusion
The results of this study are consistent with previous findings that barefoot running can in some cases change the running technique by a flatter foot placement, an increased knee flexion at IC and a decreased step length. These are factors that could minimize the risk of running injuries, by decreasing the impact force. However, caution must be taken when
20
habitually shod runners are planning to move to barefoot running. Further, the running technique may be more the more important factor, regardless of wearing shoes or not. Future research should focus on how fatigue influences running technique when habitually shod runners have an interest in transitioning to barefoot running.
21
References
Breine, B., Malcolm, P., Frederick, E. C., & De Clercq, D. (2014). Relationship between running speed and initial foot contact patterns.Medicine & Science in Sports & Exercise,46(8), 1595-1603.
Borg, G.A. (1982) Psychophysical bases of perceived exertion. Medicine and Science
in Sports and Exercise 14, 377-381.
Cheung, R. T. H., & Rainbow, M. J. (2014). Landing pattern and vertical loading rates during first attempt of barefoot running in habitual shod runners. Human Movement Science,
34, 120.
De Wit, B., De Clercq, D., & Aerts, P. (2000). Biomechanical analysis of the stance phase during barefoot and shod running. Journal of Biomechanics, 33(3), 269-278.
Fredericks, W., Swank, S., Teisberg, M., Hampton, B., Ridpath, L., & Hanna, J. B. (2015). Lower extremity biomechanical relationships with different speeds in traditional, minimalist, and barefoot footwear. Journal of Sports Science & Medicine, 14(2), 276-283.
Hall, J. P. L., Barton, C., Jones, P. R., & Morrissey, D. (2013). The biomechanical differences between barefoot and shod distance running: A systematic review and preliminary meta-analysis. Sports Medicine, 43(12), 1335-1353.
Hamill, J., & Knutzen, K. (2008). Biomechanical basis of human movement (3rd ed.). Philadelphia: Wolters Kluwer Health/Lippincott Williams and Wilkins.
Hobara, H., Sato, T., Sakaguchi, M., & Nakazawa, K. (2012). Step frequency and lower extremity loading during running. International Journal of Sports Medicine, 33(4), 310.
Larson, P. (2014). Comparison of foot strike patterns of barefoot and minimally shod runners in a recreational road race.Journal of Sport and Health Science,3(2), 137-142.
22
Lieberman, D. E. (2012). What we can learn about running from barefoot running: An evolutionary medical perspective. Exercise and Sport Sciences Reviews, 40(2), 63-72.
Lieberman, D., Castillo, E., Otarola-Castillo, E., Sang, M., Sigei, T., Ojiambo, R.. . Pitsiladis, Y. (2015). Variation in foot strike patterns among habitually barefoot and shod runners in Kenya.Plos One,10(7)
Lohman, 3., Everett B, Balan Sackiriyas, K. S., & Swen, R. W. (2011). A comparison of the spatiotemporal parameters, kinematics, and biomechanics between shod, unshod, and minimally supported running as compared to walking.Physical Therapy in Sport: Official Journal of the Association of Chartered Physiotherapists in Sports Medicine,12(4), 151.
Lieberman, D. E., Venkadesam, M., Werbel, W. A., Daoud, A.I., D’Andrea, S., Davis, I. S., Ojiambo Mang’Eni, R., & Pitsiladis, Y. (2010). Foot strike patterns and collision forces in habitually barefoot versus shod runners. Nature, (463), 531-535.
Maykut, J. N., Taylor-Haas, J. A., Paterno, M. V., DiCesare, C. A., & Ford, K. R. (2015). Concurrent validity and reliability of 2d kinematic analysis of frontal plane motion during running. International Journal of Sports Physical Therapy, 10(2), 136-146
Personuppgiftslagen. (1998). Retrieved 2016-01-14, from
http://www.datainspektionen.se/lagar-och-regler/personuppgiftslagen/
Riley, P. O., Dicharry, J., Franz, J., Della Croce, U., Wilder, R. P., & Kerrigan, D. C. (2008). A kinematics and kinetic comparison of overground and treadmill running. Medicine
and Science in Sports and Exercise, 40(6), 1093-1100.
Saragiotto, B. T., Yamato, T. P., Hespanhol Junior, L. C., Rainbow, M. J., Davis, I. S., & Lopes, A. D. (2014). What are the main risk factors for running-related injuries? Sports
Medicine, 44(8), 1153-1163.
Shih, Y., Lin, K., & Shiang, T. (2013). Is the foot striking pattern more important than barefoot or shod conditions in running? Gait & Posture, 38(3), 490-494.
23
Squadrone, R., & Gallozzi, C. (2009). Biomechanical and physiological comparison of barefoot and two shod conditions in experienced barefoot runners. The Journal of Sports
Medicine and Physical Fitness, 49(1), 6.
Squadrone, R., Rodano, R., Hamill, J., & Preatoni, E. (2015). Acute effect of different minimalist shoes on foot strike pattern and kinematics in rearfoot strikers during running.
Journal of Sports Sciences, 33(11), 1196-1204.
Tam, N., Astephen Wilson, J. L., Noakes, T. D., & Tucker, R. (2014). Barefoot running: An evaluation of current hypothesis, future research and clinical applications. British Journal
of Sports Medicine, 48(5), 349.
Thompson, M. A., Gutmann, A., Seegmiller, J., & McGowan, C. P. (2014). The effect of stride length on the dynamics of barefoot and shod running. Journal of Biomechanics,
47(11), 2745-2750.
Valenzuela, K., Lynn, S., Mikelson, L., Noffal, G., & Judelson, D. (2015). Effect of acute alterations in foot strike patterns during running on sagittal plane lower limb kinematics and kinetics. Journal of Sports Science and Medicine, 14(1), 225-232.
Van Gent, R. N., Siem, D., van Middelkoop, M., van Os, A. G., Bierma-Zeinstra, S. M. A., & Koes, B. W. (2007). Incidence and determinants of lower extremity running injuries in long distance runners: A systematic review. British Journal of Sports Medicine,41(8), 469-480.
Williams, 3rd, D S Blaise, Green, D. H., & Wurzinger, B. (2012). Changes in lower extremity movement and power absorption during forefoot striking and barefoot
24
Appendix A
Information till deltagare
Hej!
Vi, Matilda Andersson och Linnea Nyberg studerar på Högskolan i Halmstad på biomedicin programmet- inriktning fysisk träning. Vi vänder oss till er då vi nu i vår ska skriva
examensarbeten som kommer att behandla biomekanik inom löpning och undrar om ni är intresserade av att delta i våra studier.
Studierna kommer att utföras i laboratoriet på Högskolan i Halmstad samt på närliggande anläggning, Idrottscentrum. Du som testperson kommer att behöva avsätta tid för att delta under två tillfällen under veckorna 10– 13. Testtillfälle ett kommer att ta ca 30 minuter och testtillfälle två ca 60 minuter. Studien kommer att behandla biomekanik där syftet är att undersöka kroppens rörelse vid löpning i två olika förhållande på löpband, med och utan skor.
Kriterier för att delta
Vi söker vana långdistanslöpare mellan 20-40 år samt mellan 50-60 år med en träningsvolym på minst 10 km i veckan. Du bör hålla ett normaltempo på 4.30 - 6 min/km på en 10
kilometers sträcka. För att delta ska du varit skadefri senaste 6 månaderna och vara bekväm med att springa i ditt 10 kilometer tempo under minst 30 minuter. Det krävs även att du är fri från sjukdom eller andra åkommor som kan påverka resultatet.
Inför testtillfällen
Testtillfälle ett
Under testtillfälle ett kommer du att få bekanta dig med miljön och de material vi kommer att använda. Du kommer också att få springa en kortare tid på löpband för analys av ditt löpsteg.
Skulle det vara så att du har olika slags löparskor vill vi gärna att du tar med alla för att bestämma vilka skor som ska användas under andra testtillfället.
Testtillfälle två
25
- Du ska ha svarta eller mörka åtsittande kläder, skulle du inte ha en åtsittande överdel kan du behöva ta av den inför testet.
- Inget intag av tobak, alkohol eller koffein 24 timmar innan testet. - Inget intag av fast föda 2 timmar innan testet.
- Ingen form av tung eller intensiv träning 24 timmar innan testet.
Du som deltagare får avbryta testet när som utan att ange orsak och insamlad data kan då tas bort om så önskas.
Sekretess
Deltagarens personuppgifter kommer att behandlas enligt Personuppgiftslagen (PuL). Inga personliga uppgifter kommer att publiceras utan resultatet kommer att publiceras i
gruppform. Du kommer att ha möjlighet till att ta del av ditt personliga videomaterial och resultat efter studien om så önskas.
Huvudman för studien är Högskolan i Halmstad.
Detta är ett studentarbete och deltagande sker på egen risk. Har du några frågor angående tester eller studien är du välkommen att kontakta oss.
Med Vänliga Hälsningar, Matilda Andersson
andersson89.ma@gmail.com
Linnea Nyberg
26
Appendix B
Samtycke till deltagande i forskningsstudie.
Nedan ger du ditt samtycke att delta i studien som behandlar biomekanik inom löpning. Vänligen läs igenom informationen noga och ge ditt medgivande genom att signera längst ner på sidan.
- Jag har tagit del av informationen kring studien och förstår vad den innebär.
- Jag har fått ställa de frågor jag önskar och vet vem jag ska kontakta om fler frågor dyker upp.
- Jag deltar frivilligt i studien och kan när som avbryta deltagandet utan att ange orsak.
- Jag tillåter att resultat från studien publiceras i vetenskaplig tidskrift, under förutsättning att sedvanlig sekretess upprätthålls.
Jag intygar att jag läst informationen och förstått vad deltagande i studien innebär.
Ort och datum:_______________________________________________________
Namn:______________________________________________________________
Underskrift:_________________________________________________________
Ansvariga för studien är:
Matilda Andersson
27 Högskolan i Halmstad 0725885900 andersson89.ma@gmail.com Hanneke Boon hanneke.boon@hh.se Handledare Högskolan i Halmstad
28
Appendix C
Information om testdeltagare
Namn:____________________________________________________________ Ålder:______Vikt:_______Längd:_______________________________________ Benlängd:_________Trochanter major - femorala condylen:____________________
Erfarenhet av löpning (antal år):________________________________________
Träningsvolym i veckan (antal km löpning):________________________________
Träningstempo på 10 km:____________________________________________
Tid på 10 km i nuläget:_________________När sprang du på denna tid? ________________
Tempo under tester (km/h):____________________________________________
Skomodell du löptränar i:______________________________________________
PO Box 823, SE-301 18 Halmstad Phone: +35 46 16 71 00
E-mail: registrator@hh.se www.hh.se
Mitt namn är Matilda och är 27 år gammal och avslutar mina tre år på Högskolan i Halmstad med en uppsats om löpning.
Nu ska jag ut i världen och guida människor i löpning och yoga! Namaste
