Sagittal plane momentum control during walking in elderly fallers

Gait & Posture 45 (2016) 121–126

Contents lists available at ScienceDirect

Gait & Posture journal homepage: www.elsevier.com/locate/gaitpost

Full length Article

Sagittal plane momentum control during walking in elderly fallers Masahiro Fujimoto a,b, Li-Shan Chou b,* a b

College of Sport and Health Science, Ritsumeikan University, Kusatsu, Shiga 525-8577, Japan Department of Human Physiology, University of Oregon, Eugene, OR 97403, USA

A R T I C L E I N F O

A B S T R A C T

Article history: Received 19 August 2015 Received in revised form 1 December 2015 Accepted 11 January 2016

Objective: The purpose of this study was to examine sagittal plane momentum control during walking with the use of center of mass (COM) velocity and acceleration. Methods: COM control in the antero-posterior direction during walking of healthy young and elderly adults, and elderly fallers (n = 15/group) was examined. Using a single-link-plus-foot inverted pendulum model, boundaries for the region of stability were determined based on the COM position at toe-off and its instantaneous velocity or the peak acceleration prior to toe-off (ROSv or ROSa, respectively). Results: Although no significant difference in forward COM velocity was detected between healthy young and elderly subjects, the peak forward COM acceleration differed significantly, suggesting agerelated differences in momentum control during walking. Elderly fallers demonstrated significantly slower forward COM velocities and accelerations and placed their COM significantly more anterior than healthy young and elderly subjects at toe-off, which resulted in their COM position-velocity combination located within the ROSv. Similar results were obtained in the ROSa, where elderly fallers demonstrated a larger stability margin than healthy young and elderly subjects. Interpretations: Significantly slower peak COM accelerations could be indicative of a poor momentum control ability, which was more pronounced in elderly fallers. Examining COM acceleration, in addition to its velocity, would provide a greater understanding of person’s momentum control, which would allow us to better reveal underlying mechanisms of gait imbalance or falls. ß 2016 Elsevier B.V. All rights reserved.

Keywords: Gait Balance control Inverted pendulum model Center of mass

1. Introduction Most falls resulting in fatal physical injuries in the elderly occur while walking [1]. A better understanding of the mechanisms underlying gait imbalance would enhance impairment assessment and implementation of fall prevention intervention. The occurrence of imbalance has been traditionally regarded as a consequence of the whole body center of mass (COM) going outside the boundaries of the base of support (BOS). The BOS boundaries have been considered as stability limits since the BOS provides a possible area for center of pressure (COP) movement, where balance is maintained by regulating the COP to keep the COM within the BOS [2–4]. The COM-BOS position relationship alone, however, has been demonstrated to inadequately describe the control of whole body dynamic equilibrium during movements [3]. The instantaneous

* Corresponding author at: Department of Human Physiology, 1240 University of Oregon, Eugene, OR 97403, USA. Tel.: +1 541 346 3391; fax: +1 541 346 2841. E-mail address: [email protected] (L.-S. Chou). http://dx.doi.org/10.1016/j.gaitpost.2016.01.009 0966-6362/ß 2016 Elsevier B.V. All rights reserved.

COM velocity needs to be considered when examining dynamic balance maintenance, where success of protective steps to avoid falling depends on locations of the BOS and COM as well as COM velocity [3,5,6]. Time-to-Contact, which uses the COM instantaneous velocity to predict when the COM will reach the BOS boundary, has been used to characterize dynamic stability [7]. Pai and Patton [6] used a biomechanical model to determine the allowable range of COM velocities at a given COM position that would permit successful movement termination [6,8,9]. Hof et al. also suggested the importance of COM velocity in the examination of gait stability and derived the extrapolated center of mass that accounts for the COM velocity [2,10]. Many studies have utilized such position–velocity relationship between COM and BOS to quantify dynamic stability during gait [2,3,6,11]. Although COM velocity, which reflects momentum, describes an instantaneous state of motion, it does not provide information on how such a momentum is controlled by skeletal muscles to maintain balance. As muscle forces produce joint torques and accelerations, the COM acceleration is directly regulated and therefore reflects active control of COM momentum. Elderly adults may exhibit difficulties in the regulation of COM momentum due

122

M. Fujimoto, L.-S. Chou / Gait & Posture 45 (2016) 121–126

to age-related declines in muscle functions [12–15]. Poor momentum control due to an inappropriate COM acceleration generation could lead to gait imbalance. However, it remains unclear how the COM acceleration is generated to regulate momentum and maintain gait stability, and whether elderly adults, especially those who experienced falls, would differ from young adults in such control. Thus, an examination of COM acceleration, in addition to its velocity, could enhance our understanding on how balance is controlled during gait, which could better reveal the mechanisms underlying falls during walking. We have recently proposed that examining COM acceleration could provide further insights into balance control during sit-tostand movement and established the region of stability using COM position and its acceleration [16,17]. Our findings demonstrated that COM acceleration could more sensitively differentiate individuals with different balance control abilities, as COM acceleration differed significantly between young and elderly adults although no detectable differences in COM velocity were found. The objective of this study was, therefore, to expand our investigation to COM acceleration during walking to examine differences in dynamic momentum control in the antero-posterior (AP) direction during walking among healthy young and elderly adults, and elderly fallers. It was hypothesized that elderly fallers would adapt a strategy with a slower forward COM velocity and acceleration, indicating their reduced momentum control ability.

2.1.1. ROSv The boundaries of the ROSv were defined using the following equation derived based on the concept of XcoM [2]. X˜ TO X˜˙ TO 1X˜ TO

(1)

where X˜ TO and X˜˙ TO are normalized COM position and velocity at  pffiffiffiffiffiffiffiffi ¼ X˙ =ðL v Þ v ¼ g=l; TO, defined as X˜ ¼ ðX X Þ=L ; X˜˙ TO

TO

h

f

TO

TO

f

0

0

Lf ¼ X t X h : foot lengthÞ. 2.1.2. ROSa The ROSa was defined as the region confined by peak COM acceleration needed to be generated prior to TO. The COM acceleration was modeled as a triangle-shape prior to TO. The initial conditions of COM position and velocity were obtained when the COM velocity reached its minimum prior to TO (Fig. 1b). The boundaries of the ROSa were derived using the following equation: ðX˜ TO þ X˙ i =ðv0 Lf ÞÞðX˜ TO X˙ i =ðv0 Lf ÞÞ X˜ TO X˜ i < X˜¨ p <

ð1X˜ TO þ X˙ i =ðv0 Lf ÞÞð1X˜ TO X˙ i =ðv0 Lf ÞÞ X˜ TO X˜ i

(2)

where X˜ TO and X˜ i are the normalized COM position at TO and initial COM position defined as X˜ TO ¼ X TO =Lf and X˜ i ¼ X i =Lf , respectively X˜¨ p is the normalized peak COM acceleration prior to TO defined as X˜¨ p ¼ X¨ p =v20 Lf .

2. Methods 2.2. Experimental protocol 2.1. Derivation of regions of stability The regions of stability in the AP direction were derived in two ways: one using COM velocity (ROSv), and the other using COM acceleration (ROSa). A single-link-plus-foot inverted pendulum model in the sagittal plane was used to define stability boundaries at toe-off (TO), which is the beginning of the single-limb support phase (Fig. 1a). Detailed derivation of the boundaries is presented in Appendix A.

Fifteen healthy young adults [Young: 7 men; mean age 22.1  1.9 years, mean height 170.4  11.0 cm, mean mass 68.2  14.6 kg], 15 healthy elderly adults [Elderly: 6 men; mean age 70.0  3.2 years, mean height 170.1  8.7 cm, mean mass 79.1  18.3 kg], and 15 elderly adults with a history of falls [Fallers: 3 men; mean age 71.9  4.3 years, mean height 164.2  8.6 cm, mean mass 83.1  20.1 kg] participated in this study. The criterion for inclusion in the Fallers group was a self-report of two or more falls

Fig. 1. (a) A single-link-plus-foot inverted pendulum model in the sagittal plane. X indicates the COM position in the AP direction. Xh and Xt indicate the heel and toe positions. m, l and M are whole body mass, pendulum length (distance from the ankle to the COM), and ankle joint moment. (b) Representative time–history plot of COM velocity and acceleration with modeled COM acceleration profile. LHS/RHS and LTO/RTO indicate left/right heel-strike and toe-off instants.

M. Fujimoto, L.-S. Chou / Gait & Posture 45 (2016) 121–126

2.5

Peak COM velocity [m/s] / acceleration [m/s2]

within the year prior to the study [18]. All participants did not have a history or clinical evidence of neurological, musculoskeletal or other medical conditions. All participants reported no history of neurological pathology, head trauma, cerebrovascular accident, vestibular dysfunction, or visual impairment uncorrectable by lenses. The experimental protocol was approved by the Institutional Review Board. Written and verbal instructions of testing procedures were provided, and written consent was obtained from each subject prior to testing. Subjects were instructed to walk barefoot at a self-selected comfortable pace along a 10-m walkway. Data from six walking trials were collected for each subject after he/she became familiar with the laboratory setting by performing a few practice trials. Whole body motion was captured with an eight-camera motion analysis system (Motion Analysis Corp., Santa Rosa, CA). A total of 29 reflective markers were placed on the subject [19]. Threedimensional marker trajectories were collected at 60 Hz and smoothed using a fourth-order Butterworth filter with a cut-off frequency of 8 Hz. The whole-body COM was calculated as the weighted sum of 13 body segments, including head and neck, trunk, pelvis, two upper arms, two forearms with hands, two thighs, two shanks, and two feet [19]. Anthropometric reference data were adopted from the initial work of Dempster [20]. The COM velocity and acceleration were calculated using Woltring’s generalized cross validated spline algorithm from COM positions, which has been used in previous studies as a validated method to estimate smoothed velocities and accelerations [19,21–24]. TO and HS instants were detected based on the vertical velocity of the midfoot [25,26]. In addition, knee extensor strength of the dominant leg, the leg used to kick a ball, was determined as the peak torque during isometric maximum voluntary contraction and measured with a dynamometer (Biodex Medical Systems, NY). The torque was recorded in a seated position with 608 of knee flexion and was normalized to body mass. Dynamic control of COM motion during walking was examined during the right leg single stance period. The COM position was referenced to locations of the right heel and normalized to the right foot length. Normalized COM position and velocity at left TO and normalized peak COM acceleration prior to left TO were used to construct the ROSv and ROSa. The stability margin was defined as the shortest distance from the experimental data to the forward boundary of the ROSv and ROSa, respectively [9]. Peak forward COM velocity and acceleration of the gait cycle were also obtained. A one-way ANOVA was used to detect group differences among Young, Elderly and Fallers groups. Post hoc analyses used t-tests with Tukey’s HSD method. Statistical analysis was performed using SPSS (Chicago, IL). Significance level was set at a = 0.05.

2

123

∗

Young Elderly Fallers

∗

†

‡

∗

1.5

1

0.5

0 Peak_COMvel

Peak_COMacc

Fig. 2. Peak forward COM velocity (COMvel) and acceleration (COMacc) for Young, Elderly, and Fallers. Values are mean  SD. (*p < 0.001, yp = 0.011, zp = 0.042).

80% (12 out of 15 subjects) and 60% (9 out of 15 subjects) from Elderly and Fallers groups, respectively (Fig. 3). Similarly, the data from all Young subjects were located outside the forward boundary of the ROSa, so were 93% (14 out of 15 subjects) and 67% (10 out of 15 subjects) from Elderly and Fallers groups, respectively (Fig. 4). Mean normalized COM position at TO for Fallers group (0.30  0.11) was found to be significantly anterior to those of Young and Elderly groups (p  0.001; Young: 0.49  0.10; Elderly: 0.47  0.13; Fig. 3). No significant difference was detected between Young and Elderly groups. Mean normalized COM velocity at TO for Fallers (1.29  0.16) was also significantly slower than those for Young and Elderly groups (p < 0.001; Young: 1.76  0.19; Elderly: 1.61  0.20; Fig. 3). Mean normalized peak COM acceleration prior to TO differed significantly among all three groups (p  0.030; Young: 0.59  0.11; Elderly: 0.49  0.12; Fallers: 0.38  0.11; Fig. 4). Stability margins based on the ROSv and ROSa both differed significantly among all three groups (p  0.028 for ROSv; p  0.004 for ROSa; Fig. 5), where Fallers demonstrated the largest stability margins, followed by Elderly and Young groups. 4. Discussion

3. Results No significant differences were found in the body height, weight, and foot length and width for both sides among Young, Elderly, and Fallers groups (p  0.069). Elderly and Fallers groups demonstrated significantly weaker knee extensor strengths than Young group (p  0.002; Young: 1.55  0.42 Nm/kg; Elderly: 0.98  0.38 Nm/kg; Fallers: 1.06  0.23 Nm/kg). Fallers demonstrated a significantly smaller peak forward COM velocity than both Young and Elderly groups (p  0.001; Fig. 2), consistent with the average COM velocity (p  0.001; Young: 1.39  0.12 m/s; Elderly: 1.29  0.17 m/s; Fallers: 1.03  0.11 m/s). Peak COM forward acceleration differed significantly among all three groups (p  0.042; Fig. 2). Regions of stability defined with normalized COM velocity and position at TO (ROSv) and with normalized peak COM acceleration prior to TO and COM position at TO (ROSa) were constructed (Figs. 3 and 4, respectively). Data from all Young subjects were located outside the forward boundary of the ROSv, so were

This study examined sagittal plane momentum control during walking with the use of regions of stability derived by COM velocity (ROSv) and acceleration (ROSa). Subjects in the Fallers group walked with a significantly decreased forward COM velocity and acceleration, and a shortened distance between the COM and BOS at toe-off as compared to subjects in the Young and Elderly groups. In addition, our results revealed that there was a significant difference in the peak forward COM acceleration between healthy young and elderly subjects although no significant difference was found in the forward COM velocities. Elderly fallers showed significantly slower forward COM velocities than both healthy young and elderly subjects, while no significant differences were detected between the healthy groups, in agreement with previous reports [19,26,27]. Although no significant differences were found in COM velocities between the Young and Elderly groups, the peak forward COM acceleration differed significantly between them. This is consistent with our previous findings for COM control during sit-to-stand movement,

124

M. Fujimoto, L.-S. Chou / Gait & Posture 45 (2016) 121–126

2.5

Young

Normalized COM velocity (AP direction)

∗ ∗∗

Elderly Fallers

2

Young(mean) Elderly(mean)

∗∗

Fallers(mean)

1.5

ROSv_Forward ROSv_Backward 1

0.5

0 -0.8

-0.7

-0.6

-0.5

-0.4

-0.3

-0.2

-0.1

0

Normalized COM position (AP direction) Fig. 3. Normalized COM velocity at TO with respect to normalized COM position at TO for Young, Elderly, and Fallers. Each plot indicates the mean data for each subject. Mean  SD for each group is also indicated. Black and gray solid lines indicate the forward and backward boundaries of the ROSv, respectively. (*p < 0.001, **p = 0.001).

Normalized COM acceleration (AP direction)

1.2

Young Elderly Fallers Young(mean) Elderly(mean) Fallers(mean) ROSa_Forward(Young) ROSa_Forward(Elderly) ROSa_Forward(Fallers)

1

∗ ∗∗

0.8

†

‡

0.6

∗ 0.4

0.2

0 -0.8

-0.6

-0.4

-0.2

0 -0.2

-0.4

Normalized COM position (AP direction) Fig. 4. Normalized peak COM acceleration prior to TO with respect to normalized COM position at TO for Young, Elderly, and Fallers. Each plot indicates the mean data for each subject. Mean  SD for each group is also indicated. Solid and dashed curves indicate the forward boundaries of the ROSa for each subject group. Since the boundaries are dependent on the initial COM position varied between subjects and trials, those boundaries were averaged for each group. (*p < 0.001, **p = 0.001, yp = 0.030, zp = 0.024).

where COM acceleration differed significantly between young and elderly adults although there were no detectable differences in COM velocity [16,17]. These results suggest that even though similar momentum was observed during walking, the momentum could be controlled differently between individuals with different balance control abilities. Demonstrating larger stability margins, elderly fallers showed a more conservative gait balance strategy than healthy subjects. The COM positions of the Fallers group were located significantly more anterior than those for the Young and Elderly groups, with a significantly slower forward COM velocity at toe-off, which placed their mean COM position–velocity point inside the boundaries of the ROSv. In contrast, the mean COM position–velocity point for healthy subjects were located outside the forward boundary of the ROSv. These results imply that healthy subjects, on average, were

dynamically unstable at toe-off during walking, utilizing a forward momentum to maintain a forward progression of the whole body. On the other hand, 40% of subjects in the Fallers group were located inside the boundaries of the ROSv, indicating they were dynamically stable, which might result from adopting a protective strategy for potential falls. Similar to the results of the ROSv, the mean COM position– acceleration point for the Fallers group was significantly closer to the forward boundary of the ROSa, resulting in a significantly larger stability margin than the healthy groups. In agreement with our results of the peak COM velocity and acceleration, the normalized peak COM acceleration prior to toe-off differed significantly among all three groups, although no significant difference was found between the healthy groups in the normalized COM velocity at toeoff. Healthy young and elderly subjects both utilized a similar

M. Fujimoto, L.-S. Chou / Gait & Posture 45 (2016) 121–126

ROSv_Forward

ROSa_Forward

0.1

0

Stability Margin

-0.1

-0.2

-0.3

-0.4

-0.5

† ‡ ∗ Young Elderly Fallers

∗∗ ∗∗∗ ∗

Fig. 5. Stability margins based on the ROSv and ROSa for Young, Elderly, and Fallers. Stability margin was defined as the shortest distance from the experimental data to the forward boundary of the ROSv and ROSa. Values are mean  SD. (*p < 0.001, **p = 0.003, ***p = 0.004, yp = 0.028, zp = 0.011).

momentum to propel their body forward at toe-off, but elderly subjects demonstrated a significantly smaller COM acceleration prior to toe-off, possibly due to age-related decline in muscle functions. Elderly adults are known to have decreased muscular strength and difficulties in generating muscle torques at a higher rate [12–15,28]. In fact, subjects in the Elderly and Fallers groups demonstrated significantly weaker knee extensor strength than the Young group. Significantly smaller peak COM accelerations demonstrated by our elderly subjects could be indicative of their age-related decline in muscle strength. Such smaller peak COM accelerations observed in elderly subjects could indicate a reduced ability of skeletal muscle function to accommodate momentum changes when balance is perturbed. A fall would be induced by a sudden change in momentum, such as trips or slips [29–32]. An inability to generate adequate acceleration to properly control sudden changes in momentum would result in imbalance and a greater risk of falling. Furthermore, differences in stability margins among all three groups were more pronounced when the ROSa was used, which are consistent with our previous results for sit-to-stand movement that the ROSa seems to differentiate individuals with different balance control abilities more sensitively than the ROSv [16,17]. These findings, taken together, suggest that COM acceleration could allow us to detect age-related differences in momentum control during gait and to grade individuals with different gait balance control abilities. The boundaries of regions of stability were derived with the use of several modeling assumptions, which is the limitation of this study. A single segment connecting the COM with the ankle was used to represent the human body. Hip muscles also play an important role of dynamic balance of the upper body during walking [4]. This would reduce the predictive capability of the stability boundaries. However, despite using a simplified pendulum motion from the ankle, the empirically measured COM positions and velocities during walking were adequately illustrated by the feasible stability region (FSR), which is similar to our ROSv, and the model predictions from the FSR were consistent with experimental observations during walking [6,9,33,34]. The model used to define the ROSa also assumed that the COM only undergoes forward acceleration prior to toe-off. This assumption would only provide a conservative estimation of the boundaries, where the

125

subjects could have generated backward acceleration to reduce the momentum prior to toe-off. Although the significant group differences found in COM acceleration would not be affected by the model assumptions, a modeled acceleration profile which better represents the empirical data would provide a more sensitive estimation of the boundaries. Another limitation of this study is that the gait velocity was not controlled during testing. Significantly smaller peak COM accelerations demonstrated by elderly subjects could be due to their slower gait velocities. However, both Elderly and Fallers groups demonstrated significantly smaller COM acceleration than Young group even after adjustment was made for average COM velocities (p  0.027, oneway ANCOVA with LSD test). In conclusion, this study examined dynamic momentum control during walking, using regions of stability derived by COM velocity and acceleration. Healthy young and elderly subjects utilized a similar momentum to propel the body forward, but controlled the momentum differently. Elderly fallers adapted a strategy with a significantly slower forward COM velocity and acceleration, and a shorter COM-BOS separation at toe-off in the AP direction. This strategy may be indicative of their reduced momentum control ability and a protective strategy for potential falls as well as reduced muscular functions. Examining COM acceleration in addition to its velocity would provide a greater understanding of person’s momentum control, which would allow us to better understand the mechanisms underlying imbalance or falls. Acknowledgments The authors would like to thank Tzurei Chen for her help with data collection. Conflict of interest The authors have no conflicts of interest in relation to the work reported here.

Appendix A. Supplementary data Supplementary data associated with this article can be found, in the online version, at http://dx.doi.org/10.1016/j.gaitpost.2016.01. 009. References [1] Li W, Keegan TH, Sternfeld B, Sidney S, Quesenberry Jr CP, Kelsey JL. Outdoor falls among middle-aged and older adults: a neglected public health problem. Am J Public Health 2006;96:1192–200. [2] Hof AL, Gazendam MG, Sinke WE. The condition for dynamic stability. J Biomech 2005;38:1–8. [3] Pai YC, Rogers MW, Patton J, Cain TD, Hanke TA. Static versus dynamic predictions of protective stepping following waist-pull perturbations in young and older adults. J Biomech 1998;31:1111–8. [4] Winter DA. A.B.C of balance during standing and walking. Waterloo: Waterloo Biomechanics; 1995. [5] McIlroy WE, Maki BE. Age-related changes in compensatory stepping in response to unpredictable perturbations. J Gerontol A Biol Sci Med Sci 1996;51:M289–96. [6] Pai YC, Patton J. Center of mass velocity-position predictions for balance control. J Biomech 1997;30:347–54. [7] Hasson CJ, Van Emmerik RE, Caldwell GE. Predicting dynamic postural instability using center of mass time-to-contact information. J Biomech 2008;41:2121–9. [8] Pai YC, Iqbal K. Simulated movement termination for balance recovery: can movement strategies be sought to maintain stability in the presence of slipping or forced sliding. J Biomech 1999;32:779–86. [9] Yang F, Espy D, Pai YC. Feasible stability region in the frontal plane during human gait. Ann Biomed Eng 2009;37:2606–14. [10] The Hof AL. ‘Extrapolated center of mass’ concept suggests a simple control of balance in walking. Hum Mov Sci 2008;27:112–25.

126

M. Fujimoto, L.-S. Chou / Gait & Posture 45 (2016) 121–126

[11] Carty CP, Mills P, Barrett R. Recovery from forward loss of balance in young and older adults using the stepping strategy. Gait Posture 2011;33:261–7. [12] Aniansson A, Grimby G, Hedberg M, Rungren A, Sperling L. Muscle function in old age. Scand J Rehabil Med Suppl 1978;6:43–9. [13] Larsson L, Grimby G, Karlsson J. Muscle strength and speed of movement in relation to age and muscle morphology. J Appl Physiol 1979;46:451–6. [14] Murray MP, Duthie Jr EH, Gambert SR, Sepic SB, Mollinger LA. Age-related differences in knee muscle strength in normal women. J Gerontol 1985;40:275–80. [15] Young A, Stokes M, Crowe M. Size and strength of the quadriceps muscles of old and young women. Eur J Clin Invest 1984;14:282–7. [16] Fujimoto M, Chou LS. Dynamic balance control during sit-to-stand movement: an examination with the center of mass acceleration. J Biomech 2012;45:543– 8. [17] Fujimoto M, Chou LS. Region of stability derived by center of mass acceleration better identifies individuals with difficulty in sit-to-stand movement. Ann Biomed Eng 2014;42:733–41. [18] Shumway-Cook A, Baldwin M, Polissar NL, Gruber W. Predicting the probability for falls in community-dwelling older adults. Phys Ther 1997;77: 812–9. [19] Hahn ME, Chou LS. Age-related reduction in sagittal plane center of mass motion during obstacle crossing. J Biomech 2004;37:837–44. [20] Winter DA. Biomechanics and motor control of human movement. 2nd ed. New York: Wiley; 1990. [21] Woltring HJ. A FORTRAN package for generalized, cross-validatory spline smoothing and differentiation. Adv Eng Software 1986;8:104–13. [22] Giakas G, Baltzopoulos V. A comparison of automatic filtering techniques applied to biomechanical walking data. J Biomech 1997;30:847–50. [23] Heiderscheit BC, Hoerth DM, Chumanov ES, Swanson SC, Thelen BJ, Thelen DG. Identifying the time of occurrence of a hamstring strain injury during

[24]

[25] [26] [27] [28]

[29] [30]

[31]

[32]

[33] [34]

treadmill running: a case study. Clin Biomech (Bristol Avon). 2005;20:1072–8. Sjolander P, Michaelson P, Jaric S, Djupsjobacka M. Sensorimotor disturbances in chronic neck pain – range of motion, peak velocity, smoothness of movement, and repositioning acuity. Man Ther 2008;13:122–31. O’Connor CM, Thorpe SK, O’Malley MJ, Vaughan CL. Automatic detection of gait events using kinematic data. Gait Posture 2007;25:469–74. Lugade V, Lin V, Chou LS. Center of mass and base of support interaction during gait. Gait Posture 2011;33:406–11. Lee HJ, Chou LS. Detection of gait instability using the center of mass and center of pressure inclination angles. Arch Phys Med Rehabil 2006;87:569–75. Riley PO, Kerrigan DC. Torque action of two-joint muscles in the swing period of stiff-legged gait: a forward dynamic model analysis. J Biomech 1998;31:835–40. Tinetti ME, Speechley M. Prevention of falls among the elderly. N Engl J Med 1989;320:1055–9. Blake AJ, Morgan K, Bendall MJ, Dallosso H, Ebrahim SB, Arie TH, et al. Falls by elderly people at home: prevalence and associated factors. Age Ageing 1988;17:365–72. Campbell AJ, Borrie MJ, Spears GF, Jackson SL, Brown JS, Fitzgerald JL. Circumstances and consequences of falls experienced by a community population 70 years and over during a prospective study. Age Ageing 1990;19:136–41. Courtney TK, Sorock GS, Manning DP, Collins JW, Holbein-Jenny MA. Occupational slip, trip, and fall-related injuries – can the contribution of slipperiness be isolated. Ergonomics 2001;44:1118–37. Bhatt T, Wening JD, Pai YC. Adaptive control of gait stability in reducing sliprelated backward loss of balance. Exp Brain Res 2006;170:61–73. Yang F, Bhatt T, Pai YC. Role of stability and limb support in recovery against a fall following a novel slip induced in different daily activities. J Biomech 2009;42:1903–8.