Footwear and Foot Orthotic Effectiveness Research: A New Approach

Footwear and Foot Orthotic Effectiveness Research: A New Approach Mark W. Cornwall, PhD, PT ' Thomas G. McPoil, PhD, PT, ATC Mark W. Cornwall Thomas G. McPoil inematic analysis of the rearfoot has been used extensively in an attempt to document the effectiveness of foot orthotics. The most common method of rearfoot analysis has been measurement of calcaneal inversion and eversion. In these studies, the sub jects are typically filmed from behind as they walk or run either on a treadmill or overground. Markers are typically placed on the posterior leg and calcaneus along a line that bisects each of those segments. The angle between these two lines is calculated and used as a measure of rearfoot complex motion, which includes an adequate approximation of subtalar joint movement. If the subject has shoes on, markers must be placed on the heel counter of the shoe rather than directly on the calcaneus. This type of measurement of calcaneal inversion and eversion has been used in numerous research studies evaluating walking (6,7,13,16,26), running (7,22), shoe design (2,15,16), and foot orthoses effectiveness (6,11,18). The popularity of measuring calcaneal inversion and eversion is probably related to the fact that it is rela- Measurement of calcaneal inversion and eversion during walking is limited when subjects wear shoes. The authors of this study propose the use of transverse tibial rotation as a viable alternative measurement when barefoot assessment is not possible. The purpose of this study, therefore, was to: I) determine the relationship behveen transverse tibial rotation and rearfoot motion during the stance phase of normal walking and 2) demonstrate the usefulness of measuring transverse tibial rotation when evaluating the effect of footwear and insole foot orthotic devices. Part I consisted of eight volunteers (five women, three men) whose rearfoot and transverse tibial motion was videotaped while they walked along a 12-m walkway. The results of this study showed that although absolute values were not comparable, the two motion patterns are related to each other. The correlation between the mean rearfoot and tibial motion patterns of all 16 feet was r =.953. Part 2 investigated the effect of footwear and orthotics on transverse tibial rotation using two case presentations. A video camera was positioned in front of each subject as they walked at a selfselected speed under various footwear or orthotic conditions. The results of the case studies revealed that footwear or foot orthotics decrease maximum tibial internal rotation compared with barefoot walking. In addition, internal tibial rotation velocity and acceleration were decreased by the use of shoes, an accommodative orthosis, and an inflatable medial longitudinal arch support. A rigid orthotic produced a slight increase in transverse tibial rotation and a dramatic increase in iransverse tibial acceleration. It is felt that measurement of transverse tibial rotation may prove useful in evaluating footwear and orthotic effectiveness. Key Words: walking, orthotics, tibial rotation 'Associate Professor, Department of Physical Therapy, Northern Arizona University, P.O. Box 15105, Flagsfa( AZ 8601 1 'Associate Professor, Department of Physical Therapy, Northern Arizona University, FlagstaH, AZ tively easy to use and that it is essentially the same angle measured by clinicians during static weight bearing (12,19) or nonweight bearing evaluations (9). One of the major limitations of these studies has been the difficulty of measuring movement of the foot while it is inside the subject's shoe. If the markers for the calcaneus are placed on the heel counter of the shoe rather than the calcaneus, the movement recorded is more representative of the shoe rather than the calcaneus (1 7,24). Several attempts have been made to compensate for this problem, including cutting a "windown in the posterior aspect of the heel counter (1,4,6,14). Because of the differences found between movement of the heel and of the shoe, it is necessary to make these modifications to the shoe, especially when information relative to injury treatment and/or prevention is desired (23). Although modification of the shoe's heel counter provides similar data compared with barefoot running, it is not a practical solution for clinicians interested in looking at the effect of their orthotic treatment, since a "window" would have to be cut in the heel counter of each patient's shoe. In addition, it is not JOSPT Volume 21 Number 6 June 1995 337

clearly known what effect cutting of the heel counter ha. upon the shoe's stability or the ramification of such modified footwear on foot orthoses effectiveness. It has been known for some time that the movements of the rearfoot and tibia are closely related. As the rearfoot pronates in closed kinetic chain activities, the tibia internally rotates. The tibia externally rotates in conjunction with rearfoot supination (5). Because of this known anatomical relationship between transverse tibial rotation and rearfoot motion, the authors propose measurement of transverse tibial rotation as a tool for indirectly measuring rearfoot motion when barefoot measurements are not possible. This problem is punctuated by foot orthotic effectiveness research because some type of footwear must The tibia externally rotates in conjunction with rearf'oot supination. be worn. The purpose, therefore, of this study was two-fold: I) to demonstrate the relationship between transverse tibial rotation and rearfoot motion during the stance phase of walking and 2) to illustrate the utility of transverse tibial rotation measurement in evaluating the effectiveness of foot orthoses. PART 1. RELATIONSHIP BETWEEN TIBIA1 ROTATION AND REARFOOT MOTION METHODS Instrumentation Rearfoot inversion and eversion motion was recorded using a Super VHS camcorder (Model AG-450, Panasonic Corp., Cypress, CA) posi- Camera #1 I Camera #2 lh Lh \ Timing Lights 7 FIGURE 1. Schematic diagram showing the experimental data collection set-up. tioned 5 m behind the subject, at a height of 28 cm from the floor and perpendicular to the plane of the motion (Figure 1). The video recording speed was 60 fields per second with a shutter speed of 1/500 sec. The camera was calibrated using a known linear distance in the field of view to correct for any lens aberrations that might affect the kinematic measurements. Transverse tibial rotation was measured using a tibial pointer device which was secured to the subject's tibia at the level of the tibial tubercle with velcro straps (Figure 2). The tibial pointer device consisted of a 25 X 25 X 25 mm aluminum block with two 110-mm aluminum rods positioned at a 90" angle to each other. Fifteen-millimeter diameter spherical reflective markers were attached to the ends of each aluminum rod. The movement of the tibial pointer device was recorded using a video camera, operating at a speed of 60 fields per second and a shutter speed of 1/500 FIGURE 2. Drawing of the Tibia1 Pointer Device used to measure transverse tibial rotation during walking. sec (Model BD400, Panasonic Corp., Cypress, CA). It was equipped with a 12.5-75 mm zoom, fl.2 lens, and attached to a video recorder (Model AG1960, Panasonic Corp., Cypress, CA). The camera was positioned 5 m in front of the subject, at a height of 53 cm from the floor and perpendicular to the plane of motion (Figure 1). The camera was also calibrated using a known linear distance in the field of view. The resulting videotaped images from both cameras were digitized using the Peak Performance two-dimensional automated analysis software (Peak Performance Technologies, Englewood, CA). Gait velocity was measured using two infrared photo cell timing devices (Tandy Corp., Fort Worth, TX), positioned 6 m apart and connected to a digital timer (Model 54030, LaFayette Instrument Company, Lafayette, IN). Subjects Eight volunteers, (five women, three men) with a mean age of 26.3 years (SD = 3.9, range = 23-35), served as subjects for this study. The mean weight and height for the eight subjects was 65.8 kg (SD = 9.8, range = 52-82.5) and 165.4 cm (SD = 6.4, range = 155-1 72.5), respectively. Criteria for participation in this study included no history of congenital deformities of the lower extremities, severe orthopaedic or neurological injuries to the lower extremities, or traumatic injury to either lower extremity at least 6 months prior to participating in the study. The procedures used in this study were Volume 21 Number 6 June 1995 JOSPT

FIGURE 3. Drawing of the position of the refleaive markers and the rearfoot angle calculated from the two-dimensional video image. approved by the Institutional Review Board at Northern Arizona University, and all subjects signed an informed consent prior to participation. Procedures Following the initial screening, each subject was asked to lie on their stomach so that the calcaneus and lower one-third of each leg could be bisected using a felt-tip pen (12). Once the bisection lines were completed, the subject was instructed to practice walking at a self-selected pace along a 12-m walkway for 15 minutes. The subject was instructed not to look at the ground while walking. Once the subject's between-trial gait velocity was consistent (variation of no more than 5%), the tibial pointer device was positioned over the tibial tubercle, and four light reflective markers were positioned with two-way tape to the end points of the previously drawn bisection lines on the leg and calcaneus (Figure 3). All bisection lines and marker attachments were performed by the same individual. For filming, subjects initially stood relaxed with their feet in their normal angle and base of sup port. This position was filmed for approximately 30 seconds by both cameras and was used as a determination of the subject's resting standing JOSPT Volume 21 Number 6 June 1995 position. The subject then walked along the walkway while the reflective markers were again simultaneously videotaped by both cameras. A total of three consecutive walking trials were collected on each extremity for all subjects. A single stride was recorded for each walking trial. During data collection, gait velocity was continuously monitored. Those trials in which the gait velocity varied more than 5% were repeated. Data Analysis A software program was written to project lines containing the posterior calcaneal and the leg markers. The rearfoot angle was then calculated as the angle formed by the intersection of these two lines (Figure 3). A second software program was written to calculate the position of the tibia in the transverse plane. The algorithm was similar to that used by Sutherland and Hagy (25) and involved comparing the perceived length of the pointers with their known length. Figure 4 illustrates the specific algorithm used in this study. Because of the lapid external rotation of the tibia during the later portion of the stance phase, the procedures outlined by Cornwall and McPoil were used to minimize the errors seen with two-dimensional analysis of the rearfoot (3). These procedures primarily involve calculating rearfoot motion for only the initial 60% of the stance phase of walking. The calculated rearfoot motion from each of the three trials were expressed in terms of the percentage of each trial's stance phase duration and then averaged across trials (3). This standardization procedure allowed for compensation of small miations in walking speed between trials and between subjects. Finally, each motion pattern was expressed relative to the person's resting standing position. The following variables were determined from the resulting motion patterns: transverse tibial rotation angle at heel strike, rearfoot angle at heel strike, maximum rearfoot eversion angle, maximum internal tibial rotation angle, time-to-maximum rearfoot eversion angle, and time-tomaximum internal tibial rotation angle. Between-trial reliability for each of the above variables was assessed using a type 1,2 intraclass correlation coefficient (ICC) (21). The consistency of the rearfoot and transverse tibial rotation motion patterns were assessed using the standard error of measurement (SEM) (20). A measure of the degree of relationship between transverse tibial rotation and rearfoot motion was assessed using a Pearson correlation coefficient. RESULTS The ICC values for each of the dependent variables measured in this part of the study are presented in Table 1. As can be seen, the ICC values ranged from 332 for time-te maximum internal tibial rotation to.965 for time-to-maximum rearfoot angle. The average SEM values for rearfoot motion ranged fiom.14 to.57" with a mean of.26". The average SEM values for transverse tibial rotation ranged from.10 to.69" with a mean of -43". The group mean movement patterns for transverse tibial rotation and rearfoot motion were observed to be similar (Figure 5). Pearson correlation coefficients between the two motion patterns ranged from r =.526 to r =.960. The correlation between mean rearfoot and tibial motion patterns of all 16 feet was r =.953. PART 2. THE EFFECT OF FOOT ORTHOSES ON TRANSVERSE TlBlAL ROTATION METHODS Instrumentation The same equipment used to measure transverse tibial rotation in Part 1 was also used in this portion of the study.

Frontal View Superior View FIGURE 4. Diagram illustrating the measurement of transverse tibial rotation during walking with the Tibial Pointer Device, where: d = the perceived distance behveen the reflective markers as measured from the video image; h = the known distance between the reflective markers as measured directly from the tibial pointer device; a = the angle defined by the relationship between the known and perceived distance between the reflective markers [a = Cos-' (dm]; and fi = the angle of transverse tibial rotation (fi = 45 - a). Negative values denote internal rotation and positive values denote external rotation. vaname Tibial rotation angle at heel strike Rearfoot angle at heel strike Maximum internal tibial rotation angle Maximum rearfoot wenion angle Time-to-maximum rearfoot angle Timeto-maximum internal tibial rotation angle TABLE 1. The between-trial KC values ior the six dependent variables calculated from the calcaneal inversion/ eversion and transverse tibial rotation movement patterns. ILL Subjects A series of two single-subject experimental designs were chosen to accomplish this study's second purpose. In Case l, a 32-year-old female volunteer sewed as the subject. Her height was 160 cm and she weighed 49.5 kg. She had no history of foot or ankle problems, but was classified as having a pronatory foot type for which foot orthotics might be indicated. In Case 2, a 26-year-old female volunteer served as the subject. Her height was 177.8 cm and she weighed 65.6 kg. She had no current foot or ankle problems but had a bunionectomy several years prior to this study. She was also clasified as having a pronatory foot type. Procedure The transverse tibial rotation for each case was measured using an almost identical procedure employed in Part 1 of this study. Each subject was filmed from the front while she walked at a self-selected speed under each of four experimental conditions. The experimental conditions for Case 1 were: I) barefoot, 2) shoes only, 3) shoes with accommodative orthoses, and 4) shoes with a selfadjusting inflatable medial longitudinal arch support (Reebok International, Stoughton, MA). Ten trials were recorded in each conditions. The experimental conditions for Case 2 were: I) barefoot, 2) shoes only, 3) shoes with rigid orthoses, and 4) shoes with accommodative orthoses. Five trials were recorded in each condition. A fewer number of trials were collected on Case 2 because a review of the results from Case 1 showed that a greater number of trials did not significantly improve the analysis. For each case, the sub ject's own shoes were used for conditions 2 through 4. The accommodative orthotics used by each subject were constructed in our clinic. The orthotics consisted of a molded layer of Plastazote I1 (Alimed, Inc., Ded- Volume 21 Number 6 June 1995 JOSPT

J.. -m- Tibia I ham, MA) with Nickelplast arch reinforcement (UCO International Ltd, Prospect Heights, IL) and covered with PPT (UCO International Ltd, Prospect Heights, IL). Finally, the rigid orthotic used by the subject in Case 2 was her own and had been previously prescribed for her by her physician. 0 10 20 30 40 50 60 Stance Phase Duration (%) FIGURE 5. Mean readoof motion and transverse tibial rotation during the stance phase of walking for all 16 feet. The dotted lines indicate one standard deviation about the mean. Variable BF SO ACH) IMLAS - - - - X SEM X SEM X SEM X SEM MlTR (degrees) -11.4 1.4-1.9 1.6-0.9 0.3 1.8 0.6 tmv (O/W) -125.3 11.4-82.6 14.8-107.9 6.6-55.3 4.3 MlTA (O/sec2) 5639.3 485.6 3541.8 317.7 5045.4 216.2 3003.4 221.9 - BF = Baretoot. SO = Shoe only. ACFO = Accommodative orthotic. IMMS = Inflatable medial longitudinal arch support. SEM = Standard error of measurement. TABLE 2. Mean values for maximum transverse tibial displacement (MTTR), velocity ( MW, and acceleration (MTTA) for the subject in Case 1. Negative values for MTTR denote internal rotation while positive values denote external rotation. - BF = Baretoot. SO = Shoe only. ACFO = Accommodative orthotic. RO = Rigid orthotic. SEM = Standard error of measurement. TABLE 3. Mean values for maximum transverse tibial displacement (MTTR), velocity ( MW, and acceleration (MTTAI for the subject in Case 2. Negative values for M77R denote internal rotation while positive values denote external rotation. Data Analysis The same software program used to calculate transverse tibial rotation in Part 1 was also used in this part of the study. Transverse tibial rotation was calculated for the entire stance phase for these two cases rather than just the initial 60% as was done in Part 1. This was done because there is no distortion in the measurement of tibial rotation during the later portion of the stance phase as happens when measuring calcaneal inversion and eversion. Statistical Analysis The following variables were calculated from the mean transverse tibial rotation motion patterns of each subject: maximum internal tibial rotation during the entire stance phase, maximum transverse tibial rotation velocity during the first 30% of the stance phase, and maximum transverse tibial rotation acceleration during the first 30% of the stance phase. The first 30% of the stance phase was used for determining maximum velocity and acceleration values because it was felt that this would be the portion of the stance phase best representative of foot pronation and, therefore, most important to evaluate an orthotic's effectiveness. RESULTS Case 1 The results indicate that maximum transverse tibial rotation was decreased in all of the experimental conditions compared with barefoot JOSPT Volume 21 Number 6 June 1995

walking (Figure 6, Table 2). Similar results were seen for maximum velocity and maximum acceleration of the tibia (Figure 6, Table 2). As can be seen, the inflatable medial longitudinal arch support demonstrated the greatest reduction in maximum rotation angle, velocity, and acceleration. It should be noted, however, that the subject could not use the inflatable medial longitudinal arch support for more than two continuous hours without discomfort in her medial longitudinal arch. Case 2 in transverse tibial measurement compared with rearfoot measurement is most likely the result of skin movement under the tibial pointer. The authors, therefore, feel that the above information is sufficient to conclude that either measurement technique is reliable. Information concerning the amount of measurement error is important in order to judge whether mean changes are significant. It is particularly important in case reports or singlesubject studies such as the present investigation. The final step in the development of a new measurement technique is the The results of this case study showed that shoes alone or an accommodative orthotic combined with shoes reduces maximum transverse tibial rotation compared with barefoot walking. The reduction, however, is less than 2" (Figure 7 and Table 3). In addition, the use of shoes alone or either type of orthotic combined with shoes reduces maximum tibial rotation velocity. Finally, either shoes alone or the accommodative orthotic with shoes reduces maximum acceleration. The rigid orthotic in conjunction with shoes increased maximum tibial rotation and tibial acceleration. This increase was even greater than that found during the barefoot condition. DISCUSSION The first issue to resolve when evaluating any new measurement technique is whether or not the measurement has adequate between-trial reliability. Without such criteria, it has little or no application. The ICCs obtained in the present study for specific variables derived from rearfoot and transverse tibial rotation motion patterns indicate excellent reliability (8). In addition, the average standard error of the measurement (SEM) for each technique was less than.43". This indicates that there was less than a half of a degree of error in measurement. The slightly greater error -4 Barefoot Shoes Soft Orthosis Inflatable Arch Experimental Condition Barefoot Shoes Soft Orthpsis Inflatable Arch Experimental Condltlon FIGURE 6. Kinematic analysis of Case 1 while walking during each of the experimental conditions. A) Mean maximum transverse tibial rotation angle measured during the stance phase. B) Mean maximum transverse tibial rotation velocity obtained during the first 30% of the stance phase. C) Mean maximum transverse tibial rotation acceleration obtained during the first 30% of the stance phase. Error bars indicate standard errors. Volume 21 Number 6 June 1995 JOSPT

establishment of its validity. The movement patterns obtained in the present study for transverse tibial r e tation are similar to those reported in the literature (10,25). Further studies, however, are needed and are currency being conducted in our laboratory. The results of part 1 demonstrate that there is a good correlation between two-dimensional rearfoot and two-dimensional transverse tibial rotation motion as measured in this study. The authors feel confident that the relatively high correlations obtained in this study indicate that the two measurements are recording essentially the same thing. This being the case, twe dimensional measurement of transverse tibial rotation can be used as an effective alternative to the traditional method of two-dimensional rearfoot motion. The use of such a tibial pointer for measuring transverse tibial rotation has several advantages over other foot analysis methods, such as navicular height or calcaneal inversion and eversion. These advantages include: I) the use of a two-dimensional analysis system which is less expensive; 2) the markers are not obscured by other body parts or shoes during walking; and 3) distortion of the angle is not affected by toeing-in or toeing-out of the subject. This means that tibial rotation can be recorded throughout the stance phase rather than just until heel-off, as recommended by Cornwall and McPoil for calcaneal inversion and eversion (3). The results of part 2 from this study demonstrate that the tibial pointer device can be an effective tool to assess the effectiveness of footwear as well as foot orthoses. The findings of the two case studies indicate that maximum transverse tibial rotation is reduced compared with barefoot walking when shoes, an arch support, or an orthotic device is used. The amount of reduction, however, may be related to body mass. The subject in Case 1 was 49.3 kg and demonstrated a dramatic reduction. The subject in Case 2 was of shoes can effectively reduce tibial 65.6 kg and showed a much smaller rotation velocity and acceleration valreduction, especially with a rigid or- ues during walking. Both case studies thotic (Figure 7). Although the inflat- indicate that maximum transverse tibial able insole arch providedsthe greatest rotation velocity and acceleration is control, the subject in Case 1 reported reduced by the use of shoes or an acarch pain after wearing the insole for commodative orthotic combined with more than 2 hours. The results of this shoes. The inflatable insole pump recase study again revealed that the use duced transverse tibial rotation, veloc- z 10. 4 X 0 Barefoot Shoes Soft Orthosis Rigid Orthosis Experimental Condition B T Shoes Soft Orthosis Rigid Orthosis Experimental Condition Barefoot Shoes Soft Orthosis Rigid Orthosis Experimental Condition FIGURE 7. Kinematic analysis of Case 2 while walking during each of the experimental conditions. A) Mean maximum transverse tibial rotation angle measured during the stance phase. 6) Mean maximum transverse tibial rotation velocity obtained during the first 30% of the stance phase. C) Mean maximum transverse tibial rotation acceleration obtained during the first 30% of the stance phase. Error bars indicate standard errors. JOSFT Volume 21 Number 6 June 1995

---- -.--- -.,,-- - - ----- -.----.. -- -- ---.------.- -" - ity, and acceleration the greatest. An interesting result from Case 2 is that the use of a rigid orthotic device can actually increase the amount of maximum transverse tibial rotation acceleration and, to a lesser extent, the amount of transverse tibial motion. Because force is the product of mass times acceleration, this increase in maximum acceleration of the tibia with a rigid orthotic indicates that there is a greater amount of force present. This may indicate that the use of a rigid orthotic device would be contraindicated for those individuals with compromised shock absorption capabilities and with excessively mobile feet. Further research is certainly warranted to substantiate all of the above findings. The results of the two case presentations have a number of potential clinical implications for anyone prescribing various footwear or insole orthotic devices designed to control pronatory foot patterns during walking. In addition, the results indicate that transverse tibial rotation measurement is reliable, and with further study, may prove to be a useful method in the assessment of footwear or foot orthotic effectiveness. Finally, as illustrated by the two case reports, it may be useful to look at other variables such as angular velocity or acceleration in addition to the traditional angular displacement values. CONCLUSIONS The results of this study indicate that rearfoot and transverse tibial rotation during the stance phase of gait are significantly related to each other. In addition, the case studies presented in this paper illustrate that the tibial pointer device can be used effectively to look at the efficacy of footwear as well as foot orthotic devices within a shoe. JOSPT REFERENCES Clarke TE, Frederick EC, Hamill C: The study of rearfoot movement in running. In: Frederick EC (ed), Sport Shoes and Playing Surfaces, pp 766-1 89. Champaign, IL: Human Kinetics Publishers, 1984 Clarke TE, Frederick EC, Hamill CL: The effects of shoe design parameters on rearfoot control in running. Med Sci Sports Exerc 15:376-38 1, 1983 Cornwall MW, McPoil TG: Reducing 2-dimensional rearfoot motion variability during walking. J Am Podiatr Assoc 83:394-397, 1993 Edington J, Frederick EC, Cavanagh PR: Rearfoot motion in distance running. In: Cavanagh PR (ed), Biomechanics of Distance Running, pp 135-164. Champaign, IL: Human Kinetics Publishers, 1990 lnman VT: The influence of the footankle complex on the proximal skeletal structures. Artif Limbs 13:59-65, 1969 Johanson MA, Donatelli R, Wooden MJ, Andrew PO, Cummings GS: Effects of three different posting methods on controlling abnormal subtalar pronation. Phys Ther 74:149-158, 1994 Kernozek TW, Ricard MD: Foot placement angle and arch type: Effect on rearfoot motion. Arch Phys Med Rehabil 71 :988-991, 1990 Landis JR, Koch GG: The measurement of observer agreement for categorical data. Biometrics 33:159-174, 1977 Lattanza L, Gray GW, Kantner RM: Closed versus open kinematic chain measurements of subtalar joint eversion: Implications for clinical practice. J Orchop Spom Ph ys Ther 9:310-3 14, 1988 Levens AS, lnman VT, Blosser ]A: Transverse rotation of the segments of the lower extremity in locomotion. J Bone Joint Surg 30A:850-872, 1948 McCulloch MU, Brunt 0, Vander Linden 0 : The effect of foot orthotics and gait velocity on lower limb kinematics and temporal events of stance. J Orthop Sports Phys Ther 17:2-10, 1993 McPoil TG, Brocato RS: The foot and ankle: Biomechanical evaluation and treatment. In: Could ]A, Davies GA (eds), Orthopedic and Sports Physical Therapy, pp 3 13-34 1. St. Louis, MO: C. V. Mosb y Company, 1985 McPoil TG, Cornwall M W: Relation- ship between rearfoot motion andplantar surface loading of the foot during walking. Eur J Phys Med Rehabil2:22-26, 1992 14. Nigg BM: Experiri~ental techniques used in running shoe research. In: Nigg BM (ed), Biomechanics of Running Shoes, pp 27-62. Champaign, IL: Human Kinetics Publishers, 1986 15. Nigg BM, Morlock M: The influence of lateral heel flare of running shoes on pronation and impact forces. Med Sci Sports Exerc 19:294-302, 1987 16. Phillips RD, Reczek DM, Fountain 0, Renner 1, Park DB: Modification of high-heeled shoes to decrease pronation during gait. J Am Podiatr Assoc 81:215-219, 1991 17. Reinschmidt C, Stacoff A, Stussi E: Heel movement within a court shoe. Med Sci Sports Exerc 24:1390-1395, 1992 18. Rogers MM, LeVeau BF: Effectiveness of foot orthotic devices used to modify pronation in runners. J Orthop Sports Phys Ther 4:86-90, 1982 19. Root ML, Orien WP, Weed JH: Clinical Biomechanics: Normal and Abnormal Function of the Foot (Volume 21, pp 175-178. Los Angeles: Clinical Biomechanics Corp., 1977 20. Rothstein RM: Measurement and clinical practice: Theory and application. In: Rothstein RM (ed), Measurement in Physical Therapy, pp 1-46. New York: Churchill Livingstone Inc., 1985 2 1. Shrout PE, Fleiss JL: Intraclrrss correlations: Uses in assessing rater reliability. Psycho1 Bull 86:420-428, 1979 22. Simpson KJ, Shewokis PA, Alduwaisan S, Reeves KT: Factors influencing rearfoot kinematics during a rapid lateral braking movement. Med Sci Spom Exerc 24586-594, 1992 23. Stacoff A, Kaelin X: Technological and biomechanical criteria of the court shoe. In: Segesser B, Pfoerringer W (eds), The Shoe in Sport, pp 47-52. London: Wolf Publishing, 1989 24. Stacoff A, Reinschmidt C, Stussi E: The movement of the heel within a running shoe. Med Sci Sports Exerc 24:695-701, 1992 25. Sutherland DH, Hagy JL: Measurement of gait movements from motion picture film. J Bone Joint Surg 54A:787-797, 1972 26. Tristant S, Blake RL: The myth of running limb varus. ] Am Podiatr Assoc 8 1 :325-327, 1991 Volume 21 Number 6 June 1995 JOSPT