Gait characteristics, balance performance and falls in ambulant adults with cerebral palsy an observational study

Accepted Manuscript
Title: Gait characteristics, balance performance and falls in
ambulant adults with cerebral palsy: An observational study
Author: P. Morgan A. Murphy A. Opheim J. McGinley
PII:
DOI:
Reference:

S0966-6362(16)30094-7
/>GAIPOS 4802

To appear in:

Gait & Posture

Received date:
Revised date:
Accepted date:

21-12-2015
7-4-2016
11-6-2016

Please cite this article as: Morgan P, Murphy A, Opheim A, McGinley J.Gait
characteristics, balance performance and falls in ambulant adults with cerebral palsy: An
observational study.Gait and Posture />This is a PDF file of an unedited manuscript that has been accepted for publication.
As a service to our customers we are providing this early version of the manuscript.
The manuscript will undergo copyediting, typesetting, and review of the resulting proof
before it is published in its final form. Please note that during the production process
errors may be discovered which could affect the content, and all legal disclaimers that
apply to the journal pertain.

Gait characteristics, balance performance and falls in ambulant adults with cerebral palsy: an
observational study

Morgan Pa, Murphy Ab, Opheim Ac,d, McGinley Je

a

Department of Physiotherapy, Monash University, Australia


b

Clinical Research Centre for Movement Disorders & Gait, Monash Health, Australia


c

Sunnaas Rehabilitation Hospital, Nesoddtangen, Norway,

d

University of Gothenburg, Institution of Neuroscience and Physiology, Rehabilitation Medicine,

Gothenburg, Sweden

e

Physiotherapy, Melbourne School of Health Sciences, University of Melbourne, Australia

Corresponding author:
Dr Prue Morgan
Physiotherapy Department
School of Primary Health Care
Faculty of Medicine, Nursing and Health Science
Monash University
PO Box 527 Frankston Vic 3199 Australia


Phone: 61 3 9904 4826


Highlights


Adults with CP demonstrate slow gait speeds and short stride length



Adults with CP demonstrate relatively high cadence to optimise gait speed



Those with a falls history took shorter strides at preferred and fast speeds



Faster gait speed was associated with better performance on BESTest total


Gait characteristics, balance performance and falls in ambulant adults with cerebral palsy: an
observational study

The relationship between spatiotemporal gait parameters, balance performance and falls history
was investigated in ambulant adults with cerebral palsy (CP). Participants completed a single
assessment of gait using an instrumented walkway at preferred and fast speeds, balance testing
(Balance Evaluation Systems Test; BESTest), and reported falls history. Seventeen ambulatory adults
with CP, mean age 37 years, participated. Gait speed was typically slow at both preferred and fast
speeds (mean 0.97 and 1.21 m/sec, respectively), with short stride length and high cadence relative
to speed. There was a significant, large positive relationship between preferred gait speed and
BESTest total score (rho=0.573; p<0.05) and fast gait speed and BESTest total score (rho=0.647,
p<0.01). The stride lengths of fallers at both preferred and fast speeds differed significantly from
non-fallers (p=0.032 and p=0.025, respectively), with those with a prior history of falls taking shorter
strides. Faster gait speed was associated with better performance on tests of anticipatory and
postural response components of the BESTest, suggesting potential therapeutic training targets to
address either gait speed or balance performance. Future exploration of the implications of slow
walking speed and reduced stride length on falls and community engagement, and the potential
prognostic value of stride length on identifying falls risk is recommended.

Keywords
Cerebral palsy
Gait
Gait analysis
Balance
Falls

Gait characteristics, balance performance and falls in ambulant adults with cerebral palsy: an
observational study

Information regarding mobility and balance performance of people with a range of health conditions
is valuable in determining level of function, change in function over time and health outcomes.
Measures of gait performance are suited for this purpose as they have been shown to be particularly
sensitive to change [1]. In older adults, slower preferred gait speed has been associated with an
increased falls risk [2] and overall survival [3]. A reported gait speed of 1.2 to 1.3 m/second is
considered necessary to be a successful community ambulator [4]. A usual gait speed of less than 1
m/second has identified persons at high risk of negative health related outcomes such as imminent
hospitalisation in well-functioning older people [3]. In specific populations such as persons with
stroke, people who cannot walk faster than 0.84 m/second six months post stroke, are unlikely to be
walking independently in the community [5]. Gait speed can provide useful information about health
outcomes and potential community participation.

Spatiotemporal gait variables and, more specifically, measures of gait variability, have been
proposed as independent contributors to balance dysfunction and falls risk in older adults [6]. For
example, increased intra-individual variability in step length and double support phase have been
associated with increased risk of multiple falls in older adults [7] and extreme step width variability
has been associated with falls history in older persons [8]. There have been inconsistent findings
regarding significance of associations between spatiotemporal gait variables and balance
performance in people with neurological dysfunction. In those with early Parkinson’s disease a
reduction in gait speed and stride length has been associated with impaired dynamic balance
performance [9], however no significant association was found between gait speed or stride length
and measures of upright stability in Friedrich’s ataxia [11]. In those with Parkinson’s disease who fall,
stride-to-stride variability is increased [10], yet a recent review article exploring gait variability in


those with multiple sclerosis was unable to demonstrate a link between measures of gait variability
and falls in this population [12]. Evidence of significant associations between spatiotemporal gait

variables, gait variability and balance performance in adults with cerebral palsy (CP) has the
potential to direct future clinical management of this population.

With the recent increased attention on the lifespan health care needs of those ageing with
developmental disability, there has been greater focus on falls experienced by adults with CP. Of
concern, evidence suggests that this population fall frequently, with reports of more than 60% of
cohorts falling per year [13, 14]. However, research on the relationships between gait
characteristics, balance performance and falls history are only just emerging. A preliminary study
found better balance performance, assessed with the Berg Balance Scale, to be associated with
higher levels of gross motor function in ambulant adults with CP [13], however no significant
relationship was evident between balance performance and falls. Further research has indicated
high variability in balance performance [14] and kinematic gait variables [15] in ambulant adults with
CP who frequently fall, however associations between balance and falls, as well as between gait
variables and falls were not examined. Given the established value of gait measures in describing
falls risk in other populations, the aims of this study were to explore the relationship between gait
characteristics, balance performance and falls history in ambulant adults with CP.

Method
Ethical approval was gained (12206B) and registration with the Australian & New Zealand Clinical
Trials Registry (ACTRN12613000166774). All participants provided written informed consent.

Participants:


Participants were recruited through advertisements placed in outpatient clinics and organisations
that support people with disability, and through previous contact (prior five years) with Clinical Gait
Analysis Service at Monash Health, Australia, as part of a randomised controlled trial exploring
efficacy of a balance training intervention [16]. Participants had a diagnosis of any subtype of CP
[17], were 18 years or older, and ambulant with or without a walking aid (Gross Motor Function
Classification Scale – Extended and Revised Levels I-III; GMFCS-E&R [18]).

Outcome measures:
Participants were characterised according to age, gender, GMFCS-E&R Level, and CP subtype.
Gait measurements were recorded by a computerized walkway system, the GAITRite® (CIR Systems
Inc., USA). The GAITRite® electronic walkway contains 27 sensor pads with an active area of 61 cm
(wide) by 792 cm (long). Data were collected at a 120 Hz sampling frequency. The GAITRite® is valid
and reliable in measuring gait in people with neurological conditions and normative comparison
groups [19, 20]. Measures of gait speed, stride length (right and left step length), step width and
cadence were determined at both self-selected preferred and fast walking speeds, directly from
footfalls on the instrumented walking surface. Stride length was calculated from the geometrical
centre of the heel of one footprint to the next ipsilateral footprint. Step length was calculated along
the horizontal axis from the geometric centre of one heel to the next contralateral footprint. Step
width was calculated from the line perpendicular to the line of progression between the geometrical
centres of the heel of two consecutive footprints.

Participants started and finished walking 2m before and after the mat to allow for acceleration and
deceleration. Any partial footfalls that did not have clearly defined beginning and ending or were in
contact with the edge of the instrumented walking surface were removed. After a practice trial to


become accustomed to the walking surface and testing environment, each participant completed
three consecutive preferred and three consecutive fast-paced walking trials. For the fast trials,
participants were instructed to walk as fast as they safely could.

Balance performance was assessed using the Balance Evaluations Systems Test (BESTest) [21]. The
BESTest is a clinical balance assessment tool, previously used to provide insight into balance
performance in ambulant adults with CP [14]. It aims to identify balance dysfunction in six different
components of balance control: (I) biomechanical constraints; (II) stability limits/verticality; (III)
anticipatory postural adjustments; (IV) postural responses; (V) sensory orientation; and (VI) stability
in gait. Balance performance is recorded on a 4-level ordinal scale (0-3, 0=cannot perform, 3=normal

performance) for each component. A total sum is calculated, converted to percentage score (0100%), with higher scores reflecting better balance performance and lesser balance dysfunction.
Sub-scores are also recorded for each of the six components.

Falls recalled in the previous 12 months was collected in an interview with the participant and/or
carer, and categorised into non-faller (0 falls) or faller (1 or more falls).

Data analysis:
The SPSS statistical software version 22.0 (SPSS Inc, Chicago, Illinois) was used for all quantitative
analysis. All continuous variables were analysed for distribution and skewness. Mean values for
walking speed (m/second), stride length (m), step width (m), and cadence (steps/min) were
described for both preferred and fast gait speed conditions. Double support time as a percentage of
gait cycle, and differences between preferred and fast gait speed conditions were calculated. Due to
potential asymmetry in step lengths, steps were categorised within individuals as either the longer
step or shorter step. Variability in step length, step width and double support time were calculated


using the standard deviation (SD) of all steps for the respective measure from the three preferred
and three fast gait speed trials. Pearson correlation coefficients (r) were calculated to determine the
relationship between preferred and fast gait speeds. Spearman rank order correlations (rho) were
calculated to determine the relationship between gait variables and measures of balance (BESTest).
The strength of any relationships was described as small (0.10-0.29), medium (0.30-0.49) or large
(0.50-1.0)[22]. Independent t-tests were used to examine differences between fallers and non-fallers
on balance and gait variables.

Results
Seventeen adults (10 males) with CP participated, mean age 37 years (range 19-53 years). Fourteen
were spastic CP subtype (5 unilateral and 9 bilateral) and 3 had other subtypes (2 mixed, 1 ataxic).
Two participants were GMFCS-E&R Level I, ten were GMFCS-E&R Level II, and five were GMFCS-E&R
Level III. Seven participants used a walking aid; two used four-wheeled walkers, four used forearm
crutches and one used a single point stick.

Gait characteristics at preferred and fast speeds are reported in Table 1. The mean stride length at
preferred speed was 0.69 m (0.28), increasing only slightly at fast speed to 0.77 m (0.33). The mean
cadence was more than 100 steps/min at both preferred and fast speeds.

<<<<<<<<<<<<<<<<<<<<<<<<<<<<Insert Table 1 around here >>>>>>>>>>>>>>>>>>>>>>>>>>>>


The scatter plot of preferred versus fast gait speeds for all participants is shown in Figure 1. Ten
participants walked at a preferred speed of less than 1.0 m/second. All but one participant were able
to increase gait speed, with a mean increase of 0.24 m/second, ranging from -0.01 to 0.48 m/second
(-2% to 41% increase). Fast speed was highly correlated to preferred gait speed (r=0.977, p<0.001).
There was also a significant large correlation between amount of speed increase and preferred
speed (r= 0.807, p<0.001), that is, faster walkers increased walking speed more than slower walkers.

<<<<<<<<<<<<<<<<<<<<<<<<Insert Figure 1 around here >>>>>>>>>>>>>>>>>>>>>>>>>>>>

Relationship between gait and balance variables
Relationships between gait variables and balance, as assessed by BESTest were explored (Table 2). A
significant large positive relationship between fast gait speed and BESTest total score was evident
(rho=0.647; p<0.01). There were also significant large positive correlations between both gait speed
conditions and all individual BESTest components, with exception of components II and V. Similarly,
there were significant large negative correlations between double support time and BESTest total
and all components with exception of components II and V. A large negative significant correlation
between double support time variability at preferred speed and BESTest component I was
demonstrated (rho=-0.620, p=0.008). Significant large negative correlations between double support
time variability and BESTest components I, III and VI were also found at fast speed (p=0.042,
p=0.047, p=0.043 respectively). Significant large positive correlations between amount of speed
increase and all BESTest components and total score were also evident (Table 2). All correlations

between step length variability (shorter or longer) at either gait speed and BESTest total or
component scores, and step width variability at either speed and BESTest total or component scores,
were non-significant.

<<<<<<<<<<<<<<<<<<<<<<<<Insert Table 2 around here >>>>>>>>>>>>>>>>>>>>>>>>>>>>>>

Relationship between balance, gait variables and falls history
Nine of seventeen participants reported one or more falls in the past twelve months. There were
fewer GMFCS E&R Level I participants (n=0) and more GMFCS E&R Level III participants (n=4) in the
group reporting falls in comparison to the no falls group (n= 2 and n=1, respectively) (Table 3). The
relationship between balance, selected gait variables and falls history was explored. There was a
significant difference between stride lengths at preferred speed for fallers and non-fallers (0.56 m vs
0.85 m; p=0.032), and significant difference between stride lengths at fast speed for fallers and nonfallers (0.61 m vs 0.97 m; p=0.025), with fallers taking shorter strides.

<<<<<<<<<<<<<<<<<<<<<<<<Insert Table 3 around here >>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>


Discussion
Measures of gait variables of adults ageing with developmental disability such as CP is useful for
facilitating greater understanding of gait in this population, and potentially prospective prediction of
falls and falls risk. This study examined the relationship between gait variables, balance performance
and falls history in ambulant adults with CP, and demonstrated slow gait speeds, short stride length,
and relatively high cadence to optimise gait speed. Faster gait speed was associated with better
performance on BESTest total, and those with prior history of falls took shorter strides at both
preferred and fast gait speeds.

A key finding in the current study was that preferred gait speed in this cohort was typically slow
(<1.0 m/second). This was slow in comparison to normative age-matched data that suggests typical
gait speed for 30-39 year old adults is around 1.3-1.4 m/second [23], and supports previous evidence

describing slow preferred gait speed in adults with CP [24]. A reported normal walking speed of 1.2
to 1.3 m/second is considered necessary for successful community ambulation [4], with a recent
systematic review proposing that speeds of up to 1.3 m/second are required for safe road crossing
at timed traffic lights [25]. This speed was not achieved by 15 of 17 participants, suggesting
preferred gait speed of ambulant adults with CP may not reach acceptable values for community
engagement, potentially limiting societal activity and participation. Five of 15 participants who
walked more slowly than 1.2 m/second at preferred speed were briefly able to increase speed to
>1.2 m/second for approximately 10 meters of testing. However their ability to sustain this increased
speed to enable safe community access, such as being able to successfully cross a road, is unknown.

Interestingly, in this study mean cadence was similar to normal range at both preferred and fast gait
speeds (>100 steps/min)[23] , but stride length was considerably shorter, with a mean of 0.7 m at
preferred speed in comparison to normative values of >1.2 m [23]. A selected gait pattern at any
preferred speed typically reflects strategies to optimise energy expenditure, balance, between-step


variability, and attentional demand [26]. Higher than anticipated cadence and shorter stride lengths
at low gait speeds have previously been described in hemiparetic patients following stroke [27] and
in those with Parkinsons’ disease [28]. It is possible that the relatively higher cadence proportional to
stride length is a strategy to accommodate spasticity and/or contracture-induced restrictions in
stride length in ambulant adults with CP, whilst concurrently keeping walking speed acceptable.
Greater cadence is achieved through a range of biomechanical work production strategies,
potentially resulting in higher energy expenditure to achieve selected speed. Increasing fatigue has
been reported as a consequence of ageing with CP [29]. Effortful gait strategies may contribute to
experienced fatigue. Considerable individual variation in muscle activation strategies to modulate
gait speed has been described in patients with stroke [27], and in a cohort of ambulant adults with
CP [15]. This warrants further exploration in this population.

This study demonstrated a positive correlation between balance as measured by the BESTest total
and gait speed. Although a strong relationship between gait speed and balance measures in the

general population has not consistently been demonstrated [30], this relationship is more evident in
those with acquired neurological dysfunction such as Parkinson’s disease and stroke [9, 31]. In the
current study, those with better balance performance typically walked faster at preferred speeds,
with shorter double support time, reinforcing the notion of enhanced motor control and dynamic
balance being integral elements of gait. The mean double support time as percentage of the gait
cycle, which if increased has been shown to be indicative of poorer balance ability, was in this study
31%, much higher than 22-25% reported in an older adult population [32]. This is probably indicative
of the generally poorer balance in this group. It was interesting that no significant correlations were
identified between stride length and BESTest scores. In older adults, gait becomes slower and strides
shorten with ageing [21]; both are thought to improve stability against balance challenges. For those
ageing with CP, there may be limited adaptability in impairments contributing to their gait pattern to
accommodate balance decline.


Of particular interest were significant relationships between gait speed and anticipatory postural
adjustments (III), reactive postural responses (IV), and stability in gait (VI) components of the
BESTest in this study. Whereas the strong association between stability in gait (VI) and gait speed
was not surprising as this component includes measures of speed-sensitive gait performance, the
strong relationship between gait speed and anticipatory and reactive postural responses confirms
construct validity of the BESTest in this population. Opheim and colleagues [14] previously found
anticipatory (III) and reactive (IV) postural responses to score lowest from the BESTest in a group of
adults with spastic bilateral CP. These findings may suggest that anticipatory and reactive postural
responses are more sensitive measures of balance than other BESTest components in this
population, and a potential target for therapeutic intervention to improve balance [16].

Nine of seventeen participants reported one or more falls in the past twelve months, in line with
previous reports of frequent falls in this population [13]. In this study, fallers took shorter strides at
both preferred and fast gait speeds than non-fallers, possibly reflective of more impaired lower limb
function. Reduced stride length has been proposed as a strategy to reduce trip risk in older adults
[33). Importantly, reduced stride length per se has not consistently been demonstrated to

independently relate to falls frequency in older adults [34]. The link between shorter stride length
and falls may reflect longstanding physical features in lower limbs of adults with CP such as
spasticity, contractures, and altered skeletal alignment, making them more vulnerable to balance
disturbances and foot clearance errors, for example. Of note, there was a trend towards greater step
length variability in the shorter step at faster gait speeds in those with falls history (p=0.061),
supporting the notion of increased demand on an impaired motor output with increasing speed
generation. There was also a trend towards lower performance on BESTest components I and VI in
those who have fallen in this study. The significance of stride length outcome, step length variability


and BESTest performance require further investigation to assist in prospective identification of falls
risk in this population.

There was a trend towards slower gait speeds in fallers in this study. In older adults, slower gait
speed has been associated with an increased risk of falls [2]. However in the stroke population, a ‘U’
shaped curve has been proposed suggesting that those who walk at preferred speeds either very
slowly (<0.6m/second) or fast (>1.3m/second) are at greater risk of falls [5]. Given overall low
preferred gait speeds, and few persons walking at fast speeds evident in ambulant adults with CP,
further investigation of this relationship could be warranted.

There were several limitations associated with this study. Firstly, this study attracted a small
convenience sample of adults with CP who likely had prior concern regarding mobility and/or falls.
Further, 12 month falls history was obtained using recall rather than prospective falls diary
methodology. However, we would suggest that under-reporting is prevalent in both methodologies,
and dichotomising of data into fall/no fall simplified falls recall process. Limited significant findings
were returned regarding measures of gait variability and BESTest, and gait variability and falls
history, possibly impacted by large standard deviations in some measures. Despite these limitations,
this study has provided useful baseline information to inform future investigation into predictive
value of gait speed, stride length, step length variability and BESTest on falls risk identification in this
population.

In conclusion, this study examined the relationship between gait variables, balance performance and
falls history. Ambulant adults with CP typically demonstrated slow gait speeds, short stride length
likely predetermined by characteristics of lifelong neuromuscular dysfunction, and relatively high
cadence to optimise gait speed. Those with prior history of falls took shorter strides at both


preferred and fast gait speed. Faster gait speed was associated with better performance on BESTest
total and anticipatory and postural responses components, suggesting both better gross motor
function in those walking faster, and potential therapeutic training targets either focused on
optimising balance responses or enhancing gait speed. Future exploration of implications of low
walking speed and reduced stride length on falls and community engagement, and predictive value
of stride length for falls risk identification is recommended.

Acknowledgement
This study was supported by a 2013 Lions John Cockayne Memorial Fellowship Trust Fund.

Conflict of Interest Statement
The authors report no conflict of interest


References
[1].

Studenski S, Perera S, Wallace D, Chandler JM, Duncan PW, Rooney E, et al. Physical

Performance Measures in the Clinical Setting. J Am Geriatr Soc 2003;51(3):314-22.
[2].

Verghese J, Holtzer R, Lipton RB, Wang C. Quantitative Gait Markers and Incident Fall Risk in

Older Adults. J Gerontol Series A: Biol Sci Med Sci 2009;64A(8):896-901.
[3].

Cesari M, Kritchevsky SB, Penninx BW, Nicklas BJ, Simonsick EM, Newman AB, et al.

Prognostic value of usual gait speed in well-functioning older people--results from the Health, Aging
and Body Composition Study. J Am Geriatr Soc 2005;53(10):1675-80. Epub 2005/09/27.
[4].

Brach JS, Berthold R, Craik R, VanSwearingen JM, Newman AB. Gait variability in community-

dwelling older adults. J Am Geriatr Soc 2001;49(12):1646-50. Epub 2002/02/15.
[5].

Quach L, Galica AM, Jones RN, Procter-Gray E, Manor B, Hannan MT, et al. The Non-linear

Relationship between Gait Speed and Falls: The MOBILIZE Boston Study. J Am Geriatr Soc
2011;59(6):1069-73.
[6].

Hausdorff J. Gait variability: methods, modeling and meaning. J NeuroEng Rehabil 2005;2:19

[7].

Callisaya J, Blizzard L, Schmidt M, Martin K, McGinley J, Sanders L, et al. Gait, gait variability

and the risk of multiple incident falls in older people: a population-based study. Age Ageing 2011;
40:481-487.
[8].

Brach J, Berlin J, VanSwearingen J, Newman A, Studenski S. Too much or too little step width

variability is associated with a fall history in older persons who walk at or normal gait speed. J
NeuroEng Rehabil 2005; 2:21
[9].

Yang Y-R, Lee Y-Y, Cheng S-J, Lin P-Y, Wang R-Y. Relationships between gait and dynamic

balance in early Parkinson's disease. Gait & Posture. 2008;27(4):611-5.
[10].

Hausdorff J. Gait dynamics, fractals and falls: Finding meaning in the stride-to-stride

fluctuations of human walking. Human Mov Sci 2007; 555–589.


[11].

Milne SC, Hocking DR, Georgiou-Karistianis N, Murphy A, Delatycki MB, Corben LA.

Sensitivity of spatiotemporal gait parameters in measuring disease severity in Friedreich ataxia.
Cerebellum (London, England). 2014;13(6):677-88. Epub 2014/07/16.
[12].

Socie M, Sosnoff J. Gait variability and multiple sclerosis. Multiple Sclerosis International

2013 />[13].

Morgan P, McGinley J. Performance of adults with cerebral palsy related to falls, balance and

function: a preliminary report. Dev Neurorehabil 2013;16(2):113-20. Epub 2013/03/13.
[14].

Opheim A, Jahnsen R, Olsson E, Stanghelle JK. Balance in Relation to Walking Deterioration

in Adults With Spastic Bilateral Cerebral Palsy. Phys Ther 2012;92(2):279-88.
[15].

Opheim A, McGinley JL, Olsson E, Stanghelle JK, Jahnsen R. Walking deterioration and gait

analysis in adults with spastic bilateral cerebral palsy. Gait & Posture. 2013;37(2):165-71.
[16].

Morgan P, Murphy A, Opheim A, Pogrebnoy D, Kravtsov S, McGinley J. The safety and

feasibility of an intervention to improve balance dysfunction in ambulant adults with cerebral palsy:
A pilot randomized controlled trial. Clin Rehabil 2014. Epub 2014/11/22.
[17].

SCPE Collaborative Group. Surveillance of cerebral palsy in Europe: a collaboration of

cerebral palsy surveys and registers. Surveillance of Cerebral Palsy in Europe (SCPE). Devl Med Child
Neurol 2000;42(12):816-24. Epub 2000/12/29.
[18].

Palisano R, Rosenbaum P, Bartlett D, Livingston M. Gross motor function classification

system - Expanded and Revised. 2007. Hamilton, Canada: CanChild Centre for Childhood Disability
Research. Retrieved from />[19].

Menz HB, Latt MD, Tiedemann A, Mun San Kwan M, Lord SR. Reliability of the GAITRite®

walkway system for the quantification of temporo-spatial parameters of gait in young and older
people. Gait & Posture. 2004;20(1):20-5.
[20].

Kuys SS, Brauer SG, Ada L. Test-retest reliability of the GAITRite system in people with stroke

undergoing rehabilitation. Disabil Rehabil 2011;33(19-20):1848-53.


[21].

Horak FB, Wrisley DM, Frank J. The Balance Evaluation Systems Test (BESTest) to

Differentiate Balance Deficits. Phys Ther 2009;89(5):484-98.
[22].

Cohen J. Statistic power analysis for the behavioural sciences. 2nd ed. Hillsdale, NJ:

Lawrence Erlbaum Associates; 1988.
[23].

Oberg T, Karsznia A, Oberg K. Basic gait parameters: reference data for normal subjects, 10-

79 years of age. J Rehabil Res Dev. 1993;30(2):210-23. Epub 1993/01/01.
[24].

Taylor NF, Dodd KJ, Baker RJ, Willoughby K, Thomason P, Graham HK. Progressive resistance

training and mobility-related function in young people with cerebral palsy: a randomized controlled
trial. Dev Med Child Neurol 2013;55(9):806-12. Epub 2013/06/25.
[25].

Salbach NM, O'Brien K, Brooks D, Irvin E, Martino R, Takhar P, et al. Speed and distance

requirements for community ambulation: a systematic review. Arch Phys Med Rehabil
2014;95(1):117-28 e11. Epub 2013/07/04.
[26].

Kuo AD, Donelan JM. Dynamic Principles of Gait and Their Clinical Implications. Phys Ther

2010;90(2):157-74.
[27].

Jonsdottir J, Recalcati M, Rabuffetti M, Casiraghi A, Boccardi S, Ferrarin M. Functional

resources to increase gait speed in people with stroke: Strategies adopted compared to healthy
controls. Gait & Posture. 2009;29(3):355-9.
[28].

Morris M, Iansek R, McGinley J, Matyas T, Huxham F. Three-dimensional gait biomechanics

in Parkinson's disease: Evidence for a centrally mediated amplitude regulation disorder. Mov Disord
2005;20(1):40-50.
[29].

Jahnsen R, Villien L, Stanghelle JK, Holm I. Fatigue in adults with cerebral palsy in Norway

compared with the general population. Dev Med Child Neurol 2003;45(05):296-303.
[30].

Ringsberg K, Gerdhem P, Johansson J, Obrant KJ. Is there a relationship between balance,

gait performance and muscular strength in 75-year-old women? Age Ageing 1999;28(3):289-93.


[31].

Patterson SL, Forrester LW, Rodgers MM, Ryan AS, Ivey FM, Sorkin JD, et al. Determinants of

walking function after stroke: differences by deficit severity. Arch Phys Med Rehabil 2007;88(1):1159. Epub 2007/01/09.
[32].

Callisaya ML, Blizzard L, Schmidt MD, McGinley JL, Srikanth VK. Ageing and gait variability—a

population-based study of older people. Age Ageing 2010;39(2):191-7.
[33].

Espy DD, Yang F, Bhatt T, Pai YC. Independent influence of gait speed and step length on

stability and fall risk. Gait & Posture. 2010;32(3):378-82.
[34].

Thaler-Kall K, Peters A, Thorand B, Grill E, Autenrieth CS, Horsch A, et al. Description of

spatio-temporal gait parameters in elderly people and their association with history of falls: results
of the population-based cross-sectional KORA-Age study. BMC Geriatrics. 2015;15:32. Epub
2015/04/17.

Figure 1: Scatter-plot of participants’ preferred gait speed (X-axis) vs. fast gait speed (Y-axis). The
solid line indicates no change between preferred and fast gait speeds. The dashed lines indicate a
threshold walking speed of 1.0 m/second [3].


Fast gait speed m/sec

Preferred gait speed m/sec


Table 1: Gait characteristics at preferred and fast speed
Preferred gait speed

Fast gait speed

(n=17)

(n-16)*

Mean (SD)

range

Mean (SD)

range

Gait speed (m/sec)

0.97 (0.26)

0.45-1.48

1.21 (0.37)

0.44-1.93

Stride length (m)

0.69 (0.28)

0.35-1.38

0.77 (0.33)

0.39-1.47

Step length – longer step (m)

0.59 (0.14)

0.32-0.83

0.63 (0.15)

0.36-0.97

Step length – shorter step (m)

0.54 (0.14)

0.33-0.80

0.59 (0.15)

0.35-0.93

Step width (m)

0.14 (0.07)

0.02-0.29

0.14 (0.06)

0.02-0.28

Cadence (steps/min)

102.8 (21.2)

47-140

117.5 (27.3)

47-168

Double support time (% of

31.4 (8.0)

21.5-46.1

28.4 (9.4)

17.3-52.0

0.01-0.07

0.03 (0.02)

0.01-0.08

0.03 (0.01)

0.01-0.05

0.03 (0.01)

0.01-0.06

Step width variability (m)

0.02 (0.01)

0.01-0.04

0.02 (0.01)

0.01-0.03

Double support time

3.2 (2.1)

1.5 – 10.9

2.9 (2.0)

1.2 – 9.1

gait cycle)
Step length variability – longer 0.03 (0.02)
step (m)
Step length variability –
shorter step (m)

variability (% of gait cycle)
*data from one participant unavailable


Table 2: Correlations between gait parameters and BESTest scores
Gait speed (m/sec)

Stride length (m)

Step width (m)

Cadence

Double support time

Double support time

Speed

(steps/min)

(% gait cycle)

variability

increase

Pref

Fast

Pref

Fast

Pref

Fast

Pref

Fast

Pref

Fast

Pref

Fast

(m/sec)

BESTest

n=17

n=16

n=17

n=16

n=17

n=16

n=17

n=16

n=17

n=16

n=17

n=16

n=16

Total

0.573*

0.647**

0.255

0.269

-0.081

-0.021

0.275

0.485

-0.608**

-0.671**

-0.474

-0.495

0.776**

I biomechanical constraints

0.528*

0.561*

0.322

0.348

0.052

0.057

0.307

0.491

-0.646**

-0.712**

-0.620**

-0.514*

0.671**

II stability limits/verticality

0.297

0.404

0.014

-0.041

-0.171

0.050

0.314

0.449

-0.425

-0.388

-0.383

-0.345

0.588*

III anticipatory postural adjustments

0.648**

0.735**

0.440

0.426

-0.088

0.040

0.238

0.456

-0.670**

-0.683**

-0.423

-0.503*

0.822**

IV postural responses

0.611**

0.638**

0.258

0.290

-0.066

-0.026

0.217

0.406

-0.622**

-0.709**

-0.398

-0.445

0.628**

V sensory orientation

0.245

0.368

0.027

0.020

-0.001

-0.077

0.246

0.286

-0.323

-0.469

-0.459

-0.455

0.576*

VI stability in gait

0.832**

0.886**

0.255

0.269

-0.143

0.019

0.332

0.485

-0.822**

-0.788**

-0.444

-0.511*

0.845**

*Correlation is significant at the 0.05 level; **correlation is significant at the 0.01 level


Table 3: The relationship between balance, gait variables and past year falls history

No falls

≥1 fall

N=8

N= 9

Mean (SD)

Mean (SD)

Level I

n=2

n=0

Level II

n=5

n=5

Level III

n=1

n=4

BESTest total, %, mean (SD)

57.9 (29.6)

46.9 (15.3)

0.343

I biomechanical constraints

59.2 (27.0)

43.7 (18.9)

0.187

II stability limits/verticality

73.2 (19.2)

73.5 (17.0)

0.970

III anticipatory postural adjustments

50.7 (30.8)

37.0 (12.7)

0.240

IV postural responses

48.6 (41.0)

26.5 (29.4)

0.218

V sensory orientation

57.5 (36.5)

60.0 (27.9)

0.875

VI stability in gait

58.3 (33.5)

40.7 (14.3)

0.171

Preferred

1.08 (0.26)

0.89 (0.21)

0.114

Fast

1.36 (0.41)

1.10 (0.31)

0.154

Preferred

0.14 (0.05)

0.13 (0.08)

0.745

Fast

0.15 (0.04)

0.13 (0.08)

0.565

Preferred

0.85 (0.34)

0.56 (0.13)

0.032*

Fast

0.97 (0.39)

0.61 (0.16)

0.025*

Preferred

98.3 (11.2)

106.7 (27.5)

0.429

Fast

113.8 (18.2)

120.4 (33.6)

0.648

Double support time

Preferred

28.8 (7.5)

34.0 (7.8)

0.150

(% of gait cycle)

Fast

25.1 (7.8)

31.0 (10.1)

0.221

0.28 (0.15)

0.21 (0.11)

0.267

GMFCS E&R Level

Gait speed (m/sec)

Step width (m)

Stride length (m)

Cadence (steps/min)

Speed increase (m/sec)

p-value

NT

24