Patellofemoral pain (PFP) is defined as pain around or behind the patella that is typically aggravated by weight bearing activities (1). Patellofemoral pain is a prominent musculoskeletal complaint in runners (2). Excessive patellofemoral joint (PFJ) stress is thought to contribute to the development and progression of PFP (3–5) and treatments are often targeted at reducing this stress. Individuals with PFP also have altered lower-limb coordination variability during running (6–8). Lower-limb variability can be related to overuse injuries such as PFP as reduced variability may induce repetitive tissue stress (9). The manner in which lower-limb variability is effected by treatments aimed to reduce PFJ stress (10) are unknown. Changes to running cadence or footwear have both been shown to reduce PFJ stress in those with PFP (10). Specifically, a 10% increase in running cadence and running in a minimalist shoe resulted in comparable (~15%) reductions in PFJ stress in those with PFP (10).
Lower-limb variability can be assessed using various approaches (9). One approach is to examine the coordination variability of joint or segment couplings to determine how consistently or inconsistently the motion at one segment (or joint) influences the motion of another segment (or joint) (9). Coordination variability can discriminate between novices and experts (11) and between those with and without injury, including those with PFP (6–8). Less experienced runners demonstrate lower coordination variability compared with experienced runners (12), and individuals with PFP have reduced coordination variability compared with individuals without PFP during running (6,8), albeit inconsistently (7). An alternative approach to examine lower-limb variability is approximate entropy (ApEn) (13). Approximate entropy provides a nonlinear measure of the randomness of time-series data by examining the regularity of patterns over time (14). Approximate entropy differs to coordination variability as it examines the regularity of individual segmental time series data rather than examining the variability of the interaction between body segments. In the gait context, ApEn has revealed decreased variability in those with anterior cruciate ligament deficient knees (15) and those with cerebral palsy compared with healthy individuals (16).
In healthy individuals, the immediate effects of increasing running cadence by 10% generally resulted in reduced lower-limb coordination variability (17). This reduction suggests that the motor control system may constrain the number of available lower-limb movement patterns to achieve the new cadence. This explanation is consistent with reports of reduced coordination variability in less experienced compared with experienced runners (12) and during rehearsal of a new task (11,18). Footwear has also been shown to influence movement variability. Runners who adopted a midfoot strike in minimalist shoes demonstrated decreased stride time variability magnitude and a more random variability pattern that was suggestive of a less flexible motor system (19). In contrast, another study found running in a minimalist shoe did not change lower-limb coordination variability in trained runners (20). Notably, the minimalist shoe used in that study had similar cushioning properties to traditional footwear and previous studies have shown partial minimalist shoes have only small effects on running biomechanics (21,22).
Both footwear and cadence changes can be incorporated into a gait retraining intervention, which has been advocated for in the management of PFP (23). However, the effect of these interventions on lower-limb variability has not been examined in those with PFP. Knowledge of these effects is important because footwear and gait modifications can influence movement variability, which may have important implications for pain and injury persistence. Therefore, the primary aim of this study was to determine the effect of running with an increased cadence and in a minimalist shoe on lower-limb variability in runners with PFP. We hypothesized that running in these conditions would reduce lower-limb coordination variability and ApEn segment variability, due to the novelty of the task.
METHODS
Participants
Fifteen recreational runners with PFP participated in this study (Table 1). The diagnosis of PFP was made based upon an initial telephone screening followed by a physical examination by an experienced physiotherapist in accordance with previous clinical trials (24,25). Participants were included if they were: (i) age 18–40 yr; (ii) running at least 10 km·wk−1; (iii) had nontraumatic retropatellar pain of greater than 6 wk duration that was aggravated by at least two of: running, hopping, squatting, prolonged sitting, or kneeling; (iv) worst pain over the previous week ≥30/100 mm on a visual analog scale (where 0 is no pain and 100 is worst pain imaginable); (v) tender on palpation over the patella facet; and (vi) pain during a double squat or step-down from a 25-cm step. Participants were excluded if they had: (i) a history of lower-limb surgery; (ii) concomitant injury or pathology of other knee structures; (iii) any foot condition that precluded the use of a minimalist shoe; and (iv) pain or injury in the hip, pelvis, or lumbar spine. All participants used a rearfoot footfall (observed during treadmill running during the physical screening) and self-reported a history of running in standard cushioned shoes. The University’s Research Ethics Board approved all study procedures, and all participants provided written informed consent.
;)
TABLE 1: Participant characteristics described as mean (SD) unless otherwise stated.
Data collection
Before data collection participants performed 4 min of treadmill running (Bertec, Columbus, OH) in a standard shoe and a minimalist shoe to determine preferred cadence. Preferred cadence (steps per minute) was determined from sagittal plane video footage (Casio Exilim; Casio, Japan) during the final minute of the run. The standard shoe was an Asics Gel-Cumulus 16, with a weight of 345 g, stack height of 31 mm and an 11-mm heel–toe offset. The minimalist shoe was a Vibram Seeya, with a weight of 136 g, stack height of 5 mm, and a 0-mm heel–toe offset. Participants then performed 5 min of running on an instrumented treadmill (Bertec) in three randomly ordered conditions: 1) standard shoe at preferred cadence, 2) standard shoe +10% cadence, 3) minimalist shoe at preferred cadence. A metronome (Seiko DM51; Seiko Instruments Inc, Japan) was used to control cadence during the +10% condition and sagittal plane video footage was captured and used to measure running cadence in all conditions. Trials were only accepted if participants achieved the desired cadence. A 5-min rest period between conditions was provided. Three-dimensional joint kinematics of the lower limb were captured using an eight-camera VICON motion analysis system (Oxford Metrics, Oxford, UK) sampling at 250 Hz. Ground reaction force (GRF) data were collected in synchrony with the motion capture data from the instrumented treadmill sampling at 1500 Hz. Thirty-two 14-mm retroflective markers were attached to anatomical land-marks in accordance with an established model (26). Markers were placed bilaterally on the iliac crests, anterior and posterior iliac spines, greater trochanters, anterior and lateral thighs and shanks, medial and lateral epicondyles of the femur, medial and lateral malleoli, calcanei, and the base of the third and fifth metatarsals.
Data analysis
Kinematic and GRF data were processed within Visual 3D software (C-Motion, Rockville, MD). Marker trajectories and GRF data were low pass filtered with a 20-Hz cutoff frequency. The cut-off frequency was determined via a residual analysis and visual inspection of the resulting kinematic and GRF data. Net internal joint moments were obtained by using a conventional Newton–Euler inverse dynamics approach. Joint moment data were normalized by body mass and reported in newton-meters per kilogram. Gait events were identified from a 60 N threshold of the vertical GRF. Data were extracted for the most affected limb for 20 strides during the final minute of the 5-min run in each condition using a customized MATLAB (Mathworks Inc, Natick, MA) program.
Coordination variability was assessed on time normalized (i.e., 0%–100%) stride-to-stride data using a modified version of vector coding as per Stock et al. (27). The segment angles between adjacent points (i.e., from one sample point to the next) were plotted on an X–Y scatter plot across the 20 strides for the segment couplings of interest. An ellipse was fitted to these points using previously described equations (28), with the size scaling adjusted according to the χ2 value (29). The area of the ellipse at each data point represented a bivariate measure of coordination variability, whereby a larger ellipse area indicated greater coordination variability (see Fig. 1 for example). Ellipse area was calculated at each point across time-normalized gait data to give a measure of continuous coordination variability across the gait cycle. Coordination variability was calculated for segment and joint couplings previously linked to PFP (6–8) or cadence interventions (17): 1) thigh flexion/extension versus tibial rotation; 2) thigh adduction/abduction versus tibial rotation; 3) thigh rotation versus tibial rotation; 4) tibial rotation versus ankle eversion/inversion; 5) thigh flexion/extension versus tibial flexion/extension; 6) thigh adduction/abduction versus ankle eversion/inversion; and 7) tibial rotation versus ankle adduction/abduction.
FIGURE 1: Example of ellipse area used to calculate coordination variability for joint coupling. The solid line represents a point in the gait cycle with a small ellipse area (low coordination variability), and the dashed line represents a point with a large ellipse area. Ellipse area was calculated at each point of the time normalized gait cycle.
Approximate entropy was calculated to assess the continuous consistency (i.e., variability) of 20 strides of hip, knee, and ankle kinematic and kinetic data. Approximate entropy evaluates the likelihood that similar patterns in a time series (e.g., a continuous kinematic waveform) will exist in a later period (14). The ApEn calculations result in a score from 0 to 2, with a value closer to 0, indicating the time series had a more consistent pattern (i.e., reduced variability), whereas a value closer to 2 indicates a less consistent pattern (i.e., increased variability) (14). The ApEn values were calculated as per the equation outlined in Arpin et al. (14):
where N was the number of data points in the time series, m was the number of points compared, r was the similarity criterion based on the SD of the data series, and C was the number of self-similar vectors defined by m points based on the r criterion. Similar to previous studies (13,14,30), m was set at 2, and r was set at 20% of the SD of the time series.
Statistical analysis
A waveform analysis, using one-dimensional statistical parametric mapping (SPM1D) (31,32) was used to assess differences in coordination variability over the gait cycle across conditions. Specifically, SPM1D repeated-measures ANOVA with planned post hoc comparisons were performed on the continuous coordination variability data extracted from the vector coding process. Where a statistically significant ANOVA was present, post hoc comparisons between the standard shoe +10% cadence and minimalist shoe versus the standard shoe were undertaken using SPM1D paired t-tests. A type I familywise error rate of alpha = 0.05 was retained by calculating a Sidak corrected threshold based on the total number of comparisons being made (i.e., 2 comparisons × 7 joint couplings), resulting in an alpha of 0.0037 being used for these analyses.
Traditional repeated-measures ANOVA with planned post hoc comparisons were used on the discrete cadence and ApEn values. A type I familywise error rate of alpha = 0.05 was also retained for these analysis on ApEn values by calculating a Sidak corrected threshold based on the total number of comparisons being made (i.e., 2 comparisons × 9 kinematic and kinetic variables), resulting in an alpha of 0.0014 being used. Cohen’s d effect sizes were calculated for post hoc comparisons between ApEn values, with the interpretation guided by Cohen (33): 0.2–0.49 = small effect; 0.5–0.79 = medium effect; and ≥0.8 = large effect.
RESULTS
The mean (SD) running cadences in the standard shoe, standard shoe +10% cadence, and minimalist shoe were 165.4 (13.3), 181.3 (14.9), and 168 (13.4) steps per minute, respectively. There was a significant difference in running cadence between conditions (P < 0.001). Post hoc tests revealed cadence in the standard shoe +10% cadence was higher than preferred cadence in the standard shoe (mean difference, 15.8; 95% confidence interval [CI], 1–18.7; P < 0.001) and minimalist shoe (mean difference, 13.3; 95% CI, 7.3–19.5; P < 0.001). No difference in preferred running cadence was observed between the standard shoe and minimalist shoe (P = 0.215).
Repeated-measures ANOVA revealed significant differences between conditions for hip sagittal and frontal plane ApEn kinematic variability, and ankle sagittal plane ApEn kinematic variability (P < 0.0001 for all). Post hoc tests revealed higher ApEn values (i.e., greater variability) in the standard shoe +10% cadence compared with the standard shoe at preferred cadence for hip flexion/extension (mean difference, 0.018; 95% CI, 0.007–0.028; P < 0.001; d = 1.12), hip adduction/abduction (mean difference, 0.036; 95% CI, 0.016–0.056; P < 0.001; d = 0.99), and ankle dorsiflexion/plantarflexion (mean difference, 0.049; 95% CI, 0.023–0.075; P < 0.001; d = 1.37) (Fig. 2). No statistically significant differences (P > 0.0014) were observed between the minimalist shoe and standard shoe at preferred cadence for joint kinematics.
FIGURE 2: ApEn values (mean ± SD) across 20 gait cycles for each condition. FLEX, flexion; EXT, extension; ADD, adduction; ABD, abduction; IR, internal rotation; ER, external rotation; DF, dorsiflexion; PF, plantarflexion; INV, inversion; EVE, eversion. *Statistically significant difference (P < 0.0014).
There were significant differences between conditions for hip transverse plane ApEn kinetic variability (P < 0.0001), knee sagittal and frontal (P < 0.0001), and transverse plane ApEn kinetic variability (P = 0.0006). Post hoc tests revealed higher ApEn values in the standard shoe +10% cadence compared with the standard shoe at preferred cadence for knee flexion/extension moments (mean difference, 0.041; 95% CI, 0.016–0.067; P < 0.001; d = 0.93) (Fig. 3). Higher ApEn values were also observed in the minimalist shoe compared with the standard shoe at preferred cadence for hip internal/external rotation moments (mean difference, 0.111; 95% CI, 0.051–0.172; P < 0.001; d = 0.76), knee adduction/abduction moments (mean difference, 0.076; 95% CI, 0.028–0.124; P < 0.001; d = 0.51), and knee internal/external rotation moments (mean difference, 0.082; 95% CI, 0.018–0.146; P < 0.001; d = 1.02) (Fig. 3). One-dimensional statistical parametric mapping repeated-measures ANOVA revealed no statistically significant differences (P > 0.0037) between conditions for coordination variability across any of the joint couplings examined (Fig. 4).
FIGURE 3: ApEn values (mean ± SD) across 20 gait cycles for each condition Mom., moment. *Statistically significant difference (P < 0.0014).
FIGURE 4: Mean continuous coordination variability of the select joint couplings examined for each condition. Greater values are indicative of higher coordination variability.
DISCUSSION
The primary purpose of this study was to investigate whether increased cadence or minimalist footwear alters lower-limb variability during running in people with PFP. We hypothesized that increased cadence and minimalist footwear would reduce movement variability due to the novelty of the task. In contrast to this hypothesis, running in minimalist footwear or with an increased cadence increased ApEn segment kinematic and kinetic variability and had no effect on coordination variability in those with PFP. These findings generate the hypothesis that increased ApEn segment variability, in addition to previously reported reductions in PFJ stress may: (i) mediate beneficial effects of commonly prescribed treatments for those with PFP and (ii) be potentially relevant to consider in the design of interventions used to treat PFP during running.
To our knowledge, the effect of minimalist footwear on movement variability in people with PFP was previously not known. Notably, we found that minimalist footwear had no effect on coordination variability at segment or joint couplings that have been previously linked to PFP (6–8). This null finding is consistent with a previous study involving healthy, trained runners who did not change lower-limb coordination variability in the sagittal plane while wearing minimalist footwear (20). Though different minimalist footwear (NikeFree (20) v Vibram Seeya) were used in these studies, the findings were similar. Minimalist footwear also had no effect on ApEn kinematic variability but did cause greater ApEn variability in hip and knee joint moments in the transverse and frontal planes, with medium to large effect sizes. Women with PFP have been shown to have greater frontal knee joint moments, and the impulse of the moment has been associated with greater PFJ stress (34). Although it cannot be determined from this study, greater variability in the frontal plane knee joint moment may reduce cumulative localized loading at the PFJ by creating variety in loading patterns and thereby assisting redistribution of cartilage pressure. Further research is required to determine if greater variability in transverse and frontal plane moments of the hip and knee underpins a reduction in PFJ stress and symptoms. A previous study in healthy individuals also found decreased stride time variability magnitude and a more random variability pattern only when running with a midfoot (not rearfoot) strike in minimalist shoes (19). Hence, the clinician using minimalist shoes as part of a management plan for PFP is advised to consider that the response may differ dependent upon their footfall pattern.
There was no effect of increased cadence on measures of coordination variability as assessed in the current study. Our observations contrast previous research in healthy individuals, where measures of coordination variability decreased across several joint couplings and phases of gait with an increase in cadence (17). Although the selected joint couplings were comparable between the studies and our approach followed published recommendations (27), there were differences in the vector coding method used to calculate coordination variability. This may partly explain the contrasting findings. Alternatively, it is highly plausible that people with PFP respond differently to increased cadence compared with people without pathology, given PFP has been associated with altered biomechanics and coordination variability compared with healthy individuals during running (6–8,35). Increased cadence was associated with increased ApEn kinematic variability at the hip (sagittal and frontal plane) and ankle (frontal plane) as well as increased ApEn variability in the sagittal plane knee joint moment. These changes occurred without concurrent changes to coordination variability, which suggests that the coupling strategies between joints (or segments) were preserved despite greater variations in individual joint kinematics across strides. Prospective evidence implicates greater frontal hip plane motion in the development of PFP in female runners (36). However, the connection between excessive frontal plane hip joint motion and greater variability in frontal plane kinematics remains to be resolved. The knee extension moment contributes directly to PFJ forces and can explain the elevated PFJ stress in runners with PFP compared with healthy runners (5). Imaging studies suggest that PFP may arise from localized cartilage pressure or subchondral bone damage (37,38). Thus, greater variability in this moment of force may be beneficial to distribute force patterns at the PFJ in those with PFP.
Clinical interventions used to treat PFP in runners often attempt to alter patellofemoral biomechanics (1). Two common goals are to reduce PFJ load and/or normalize patellofemoral kinematics. Previously, we have demonstrated that running in a minimalist shoe at preferred cadence and running in a cushioned shoe with +10% cadence reduced PFJ stress by 15% to 16% and joint reaction force by 17% to 19%, compared to a standard shoe condition (10). Mechanisms underpinning pain in those with PFP are not well understood, and pain may be related to elevated patellar bone metabolic activity in this population (37) attributable to elevated and localized cartilage stress during movement. We speculate that an increase in segment variability in response to minimalist footwear and increased cadence is potentially conducive to pain relief by distributing the apparent localized patellofemoral cartilage stress.
Movement variability in joint or limb segment coordination has been suggested to be inherent within a healthy motor control strategy (9). From a dynamical systems approach, the optimal levels of movement variability are unknown (9) and are indeed likely patient specific and dependent on type and duration of pathology, as well as symptom severity. Although the evidence is inconsistent, individuals with PFP reportedly have lower movement variability compared with their healthy counterparts (6,8). This is hypothesized to reflect a reduction in available movement patterns and may contribute to the persistence of PFP through cumulative use of the same movement strategy. Thus, scope to increase movement variability is plausible in people with PFP. Moreover, research is needed to determine if changes in variability induced by gait retraining interventions are sustainable over time to reduce symptoms.
Several limitations of this study warrant consideration. Firstly, we examined the immediate effect of cadence and footwear on lower-limb variability and it is not known if these adaptations are retained over time. It is possible that variability may revert to baseline with continued exposure to these interventions and longitudinal studies are required to better understand the time course changes in variability and symptoms. A proportion of our sample was diagnosed with bilateral PFP (33%) which may influence measures of coordination variability acquired from the most symptomatic side. Unilateral injury can impair bilateral neuromuscular control, as demonstrated previously in people with knee pathology (39). Standardized footwear was used in this study but we did not account for foot type, foot strike pattern or usual footwear. Therefore, findings should be generalized with caution beyond the footwear and +10% cadence evaluated in this study. Despite the relatively large effect sizes observed, the CI bounding the mean differences are reasonably large. This may reflect individual variation in response to the interventions evaluated.
In conclusion, both increased cadence and minimalist footwear induced alterations in segment variability during running. Increased cadence resulted in primarily greater hip kinematic variability and sagittal plane knee kinetic variability, whereas running in minimalist footwear increased hip and knee joint kinetic variability. This greater segmental variability may be beneficial to distribute loading patterns at the PFJ. Future research is needed to understand the clinical relevance of modifying variability on clinical symptoms and PFJ loading.
This research did not receive any specific grant from funding agencies in the public, commercial or not-for-profit sectors. M. H. is supported by the Sir Randal Heymanson Research Fellowship from the University of Melbourne.
J. T. F. has previously received funding from ASICS Oceania to attend a sports medicine conference and donations of footwear to facilitate the completion of separate research. The authors have no professional relationships with companies or manufacturers who will benefit from the results of the present study.
Disclosure of Funding: No external funding was received for the undertaking of this study or production of this manuscript. The authors declare that the results of this study are presented clearly, honestly, and without fabrication, falsification, or inappropriate data manipulation. The results of this study do not constitute endorsement by the American College of Sports Medicine.
REFERENCES
1. Crossley KM, Stefanik JJ, Selfe J, et al. 2016 Patellofemoral pain consensus statement from the 4th International Patellofemoral Pain Research Retreat, Manchester. Part 1: terminology, definitions, clinical examination, natural history, patellofemoral osteoarthritis and patient-reported outcome measures.
Br J Sports Med. 2016;50(14):839–43.
2. Taunton JE, Ryan MB, Clement DB, McKenzie DC, Lloyd-Smith DR, Zumbo BD. A retrospective case-control analysis of 2002 running injuries.
Br J Sports Med. 2002;36(2):95–101.
3. Farrokhi S, Keyak JH, Powers CM. Individuals with patellofemoral pain exhibit greater patellofemoral joint stress: a finite element analysis study.
Osteoarthr Cartil. 2011;19(3):287–94.
4. Fulkerson JP, Shea KP. Mechanical basis for patellofemoral pain and cartilage breakdown. In: Ewing J, editor.
Articular Cartilage and Knee Joint Function: Basic Science and Arthroscopy. New York: Raven Press; 1990. pp. 93–101.
5. Liao TC, Keyak JH, Powers CM. Runners with patellofemoral pain exhibit greater peak patella cartilage stress compared with pain-free runners.
J Appl Biomech. 2018;34(4):298–305.
6. Hamill J, van Emmerik RE, Heiderscheit BC, Li L. A dynamical systems approach to lower extremity running injuries.
Clin Biomech (Bristol, Avon). 1999;14(5):297–308.
7. Cunningham TJ, Mullineaux DR, Noehren B, Shapiro R, Uhl TL. Coupling angle variability in healthy and patellofemoral pain runners.
Clin Biomech (Bristol, Avon). 2014;29(3):317–22.
8. Heiderscheit BC, Hamill J, Van Emmerik RE. Variability of stride characteristics and joint coordination among individuals with unilateral patellofemoral pain.
J Appl Biomech. 2002;18(2):110–21.
9. Hamill J, Palmer C, Van Emmerik RE. Coordinative variability and overuse injury.
Sports Med Arthrosc Rehabil Ther Technol. 2012;4(1):45.
10. Bonacci J, Hall M, Fox A, Saunders N, Shipsides T, Vicenzino B. The influence of cadence and shoes on patellofemoral joint kinetics in runners with patellofemoral pain.
J Sci Med Sport. 2018;21(6):574–8.
11. Bernstein E.
The Co-ordination and Regulation of Movements. 1st ed. Oxford, NY: Pergamon Press; 1967. p. 196.
12. Hafer JF, Peacock J, Zernicke RF, Agresta CE. Segment coordination variability differs by years of running experience.
Med Sci Sports Exerc. 2019;51(7):1438–43.
13. Pincus SM. Approximate entropy as a measure of system complexity.
Proc Natl Acad Sci U S A. 1991;88(6):2297–301.
14. Arpin DJ, Stuberg W, Stergiou N, Kurz MJ. Motor control of the lower extremity musculature in children with cerebral palsy.
Res Dev Disabil. 2013;34(4):1134–43.
15. Georgoulis AD, Moraiti C, Ristanis S, Stergiou N. A novel approach to measure variability in the anterior cruciate ligament deficient knee during walking: the use of the approximate entropy in orthopaedics.
J Clin Monit Comput. 2006;20(1):11–8.
16. Kaipust JP, Huisinga JM, Filipi M, Stergiou N. Gait variability measures reveal differences between multiple sclerosis patients and healthy controls.
Motor Control. 2012;16(2):229–44.
17. Hafer JF, Freedman Silvernail J, Hillstrom HJ, Boyer KA. Changes in coordination and its variability with an increase in running cadence.
J Sports Sci. 2016;34(15):1388–95.
18. Arutyunyan GA, Gurfinkel V, Mirskii ML. Organisation of movements on execution by man of an exact postural task.
Biofizica. 1969;14(6):1103–10.
19. Fuller JT, Amado A, Emmerik RE, et al. The effect of footwear and footfall pattern on running stride interval long-range correlations and distributional variability.
Gait Posture. 2016;44:137–42.
20. Frank NS, Callaghan JP, Prentice SD. Lower limb kinematic variability associated with minimal footwear during running.
Footwear Sci. 2013;5(3):171–7.
21. Bonacci J, Saunders PU, Hicks A, Rantalainen T, Vicenzino BG, Spratford W. Running in a minimalist and lightweight shoe is not the same as running barefoot: a biomechanical study.
Br J Sports Med. 2013;47(6):387–92.
22. Squadrone R, Rodano R, Hamill J, Preatoni E. Acute effect of different minimalist shoes on foot strike pattern and kinematics in rearfoot strikers during running.
J Sports Sci. 2015;33(11):1196–204.
23. Barton CJ, Lack S, Hemmings S, Tufail S, Morrissey D. The ‘Best Practice Guide to Conservative Management of Patellofemoral Pain’: incorporating level 1 evidence with expert clinical reasoning.
Br J Sports Med. 2015;49(14):923–34.
24. Collins N, Crossley K, Beller E, Darnell R, McPoil T, Vicenzino B. Foot orthoses and physiotherapy in the treatment of patellofemoral pain syndrome: randomised clinical trial.
BMJ. 2008;337:a1735.
25. Mills K, Blanch P, Dev P, Martin M, Vicenzino B. A randomised control trial of short term efficacy of in-shoe foot orthoses compared with a wait and see policy for anterior knee pain and the role of foot mobility.
Br J Sports Med. 2012;46(4):247–52.
26. McLean SG, Su A, van den Bogert AJ. Development and validation of a 3-D model to predict knee joint loading during dynamic movement.
J Biomech Eng. 2003;125(6):864–74.
27. Stock H, van Emmerik R, Wilson C, Preatoni E. Applying circular statistics can cause artefacts in the calculation of vector coding variability: a bivariate solution.
Gait Posture. 2018;65:51–6.
28. Duarte M, Zatsiorsky VM. Effects of body lean and visual information on the equilibrium maintenance during stance.
Exp Brain Res. 2002;146(1):60–9.
29. Mullineaux DR. CI2 for creating and comparing confidence-intervals for time-series bivariate plots.
Gait Posture. 2017;52:367–73.
30. Ivkovic V, Kurz MJ. Parkinson’s disease influences the structural variations present in the leg swing kinematics.
Motor Control. 2011;15(3):359–75.
31. Pataky TC. One-dimensional statistical parametric mapping in python.
Comput Methods Biomech Biomed Engin. 2012;15(3):295–301.
32. Pataky TC, Robinson MA, Vanrenterghem J. Vector field statistical analysis of kinematic and force trajectories.
J Biomech. 2013;46(14):2394–401.
33. Cohen J.
Statistical Power Analysis for the Behavioral Sciences. 2nd ed. Hillsdale, NJ: Lawrence Erlbaum Associates; 1988. p. 40.
34. Waiteman MC, Briani RV, Pazzinatto MF, et al. Relationship between knee abduction moment with patellofemoral joint reaction force, stress and self-reported pain during stair descent in women with patellofemoral pain.
Clin Biomech (Bristol, Avon). 2018;59:110–6.
35. Neal BS, Barton CJ, Gallie R, O’Halloran P, Morrissey D. Runners with patellofemoral pain have altered biomechanics which targeted interventions can modify: a systematic review and meta-analysis.
Gait Posture. 2016;45:69–82.
36. Noehren B, Hamill J, Davis I. Prospective evidence for a hip etiology in patellofemoral pain.
Med Sci Sports Exerc. 2013;45(6):1120–4.
37. Draper CE, Fredericson M, Gold GE, et al. Patients with patellofemoral pain exhibit elevated bone metabolic activity at the patellofemoral joint.
J Orthop Res. 2012;30(2):209–13.
38. Ho KY, Keyak JH, Powers CM. Comparison of patella bone strain between females with and without patellofemoral pain: a finite element analysis study.
J Biomech. 2014;47(1):230–6.
39. Urbach D, Nebelung W, Weiler HT, Awiszus F. Bilateral deficit of voluntary quadriceps muscle activation after unilateral ACL tear.
Med Sci Sports Exerc. 1999;31(12):1691–6.
