Increased muscular challenge in older adults during obstructed gait
Michael E. Hahn1, Heng-Ju Lee, Li-Shan Chou*
Motion Analysis Laboratory, Department of Human Physiology, 1240 University of Oregon, Eugene, OR 97403, USA
Accepted 21 November 2004
Skeletal muscle strength is known to decline with age. Although lower extremity (LE) muscle strength is critical to maintaining dynamic
stability, few studies have investigated lower extremity muscle challenge during activities of daily living. The purpose of this study was to
investigate the effects of age and obstructed gait on relative lower extremity muscular challenge, with respect to available joint strength.
Fifteen healthy young and fifteen healthy older adults were asked to walk over level ground and step over obstacles. Pre-amplified surface
electrodes were used to measure bilateral muscular activation of the gluteus medius (GM), vastus lateralis (VL), and gastrocnemius (GA).
Muscle activation signals were normalized to peak magnitudes collected during maximal manual muscle testing (MMT). Normalized
magnitudes were analyzed during the double-support phase for gluteus medius and vastus lateralis and during the single-support phase for
gastrocnemius. A two-factor ANOVAwas used to test for age group effect, with repeated measure of obstacle height. In general, older adults
demonstrated greater relative activation levels compared to young adults. Gluteus medius activity was significantly greater in the elderly as
compared to young during periods of double-support (weight transfer). Increased obstacle height resulted in greater relative activation in all
muscles, confirming the increased challenge to the musculo-skeletal system. While healthy elderly adults were able to successfully negotiate
obstacles of different heights during walking, their muscular strength capacity was significantly lower than young adults, resulting in
relatively higher muscular demands. The resulting potential for muscular fatigue during locomotion may place individuals at higher risk for
trips and/or falls.
# 2004 Elsevier B.V. All rights reserved.
Keywords: Elderly; EMG; Joint strength; Obstacle crossing; Balance control
Skeletal muscle strength is known to decline with age
[1,2]. As an intrinsic risk factor contributing to falls, lower
extremity (LE) muscle strength is a critical component in
limiting an elderly individual’s dynamic stability [3–5]. It
has been reported that fallers demonstrated only 37% of the
knee extensor strength, and 10% of the ankle plantar flexor
strength exhibited by their non-falling peers . Stepping
over obstacles has been shown to require greater muscle
force than level walking , and recent results have shown
significant associations between isometric strength and the
ability of elderly individuals to cross obstacles .
Involvement in a long-term lower extremity resistance
training regimen resulted in substantial strength gains
among older adults (197–285% increase), along with
significant improvement in obstructed gait function (speed
of crossing stride, increased obstacle clearance, etc.) .
Additionally, it has been demonstrated that lower extremity
joint strength affects stepping speed and toe trajectory
during early swing .
Quantification of the neuromuscular challenge encoun-
tered by lower extremity muscles of healthy elderly
individuals can provide baseline information against which
strength declines or strength training interventions may be
compared. It is interesting to note a lack of studies
addressing the validity of using normalized electromyo-
graphy (N-EMG) to identify age-related differences in
relative levels of muscular challenge encountered during
activities of daily living. This may be due in part to the non-
linear EMG/force relationships reported by various groups
Gait & Posture 22 (2005) 356–361
* Corresponding author. Tel.: +1 541 346 3391; fax: +1 541 346 2841.
E-mail address: firstname.lastname@example.org (L.-S. Chou).
1Present address: Department of Health and Human Development,
Montana State University, Bozeman, MT 59717, USA.
0966-6362/$ – see front matter # 2004 Elsevier B.V. All rights reserved.
[11–13]. These studies reported non-linear patterns in the
EMG/force ratio for isometric contractions, but did not test
EMG/force relationships during anisometric contractions.
One recent study demonstrated non-linear relationships
between force and EMG during anisometric contractions for
the first dorsal inter-osseus muscle . Their findings
indicate that less voluntary muscular activation is needed
during eccentric contractions as compared to concentric
contractions to produce the same force.
Previous studies have reported that older adults adopt a
conservative strategy when crossing obstacles [15,16], as
indicated by slower crossing speed, shorter step length,
shorter obstacle–heel-strike distance , and reduced
anterior/posterior separation between whole body center of
mass (COM) and the center of pressure (COP) . The
selection of a conservative obstacle-crossing strategy may
be related to natural age-related strength loss. If so, then it
follows that the relative challenges of obstacle crossing (and
other functional tasks) may tax the available joint strength to
a point where a person’s ability to control balance
dynamically could be seriously compromised. Quantifica-
tion of the relative challenge imposed on lower extremity
muscles would be beneficial to understanding the thresholds
of joint strength that are needed to allow adequate
negotiation of obstacles in daily life.
The purpose of this study was, therefore, to investigate
the effects of age and obstacle height on relative lower
extremity muscular challenge, with respect to available joint
strength. It was hypothesized that healthy elderly adults
would require a greater percentage of their neuromuscular
capacity (as measured by increased N-EMG values) during
level walking and while crossing an obstacle. We further
hypothesized that N-EMG levels increase as obstacle height
increases; indicating the task-specific challenge imposed on
lower extremity muscles during this functional activity.
Fifteen young adults (8 males/7 females; 24.5 ? 3.6
years,172.3 ? 6.8 cm,72.5 ? 10.1 kg)and15elderlyadults
(9 males/6 females; 72.6 ? 5.5 years, 168.3 ? 9.5 cm,
75.8 ? 12.2 kg) were recruited for this study from the
University of Oregon campus and the surrounding commu-
nity, within the guidelines of the Institutional Review Board.
Inclusion criteria required that subjects had no histories of
significant head trauma, neurological disease (e.g. Parkin-
son’s, post-polio syndrome, diabetic neuropathy), visual
impairment not correctable with lenses, musculo-skeletal
impairments (e.g. amputation, joint replacement, joint
fusions, joint deformity due to rheumatoid arthritis), or
persistent symptoms of vertigo, light-headedness, unsteadi-
ness. All of the subjects were community-dwelling
individuals. Elderly subjects were noted to be active
community members, with many of them currently involved
in recreational sporting activities.
over the bellies of the gluteus medius (GM), vastus lateralis
(VL), and medial head of the gastrocnemius (GA). These
muscle groups were previously shown to be substantially
challenged when stepping over obstacles [17,18]. Activation
magnitude of each muscle during gait was normalized to
values taken during maximal effort manual muscle testing
(MMT). Maximal GM activation was tested in 308 of hip
abduction, while side lying. For VL maximum, subjects
were seated with the knee in 458 of flexion. Maximal GA
activation was tested in neutral ankle position, with the
subject fixed to a table in prone position. MMT procedures
were performed by one examiner for each muscle group,
bilaterally. Subjects were verbally encouraged to ensure
Subjects were then asked to walk at a self-selected pace
during level and obstructed gait trials. Level walking trials
were performed first, followed by obstacle crossing trials. A
single obstacle consisting of two upright standards and a
light-weight crossbar was set to four height conditions;
2.5%, 5%, 10%, and 15% body height (BH). The relative
difficulty of stepping over obstacles was thus normalized to
individual body height, accounting for variation within the
sample. This obstacle crossing protocol has been accepted
previously [19–21]. Obstacle heights were randomized with
three trials collected for each condition. The leading limb
was defined as the first limb to cross the obstacle. Crossing
stride was defined as the trailing limb heel-strike before the
obstacle to heel-strike of the same limb after crossing the
For all MMT and gait trials, raw EMG signals were
collected at 960 Hz using the MA-300TMsystem (Motion
Lab Systems, Inc., Baton Rouge, LA), band-pass filtered
(20–350 Hz), full wave rectified, and passed through a linear
envelope at 10 Hz for final interpretation. Filtered signals
from the gait trials were then normalized to the MMT signal
maximum for each muscle to indicate relative activation
levels. In this way, the relative activation values were
recorded as a percentage of the maximum activation
available to each individual muscle (N-EMG). Fig. 1
demonstrates the data-processing steps leading from full-
wave rectified EMG to smoothed EMG data (time-normal-
ized to 100% crossing stride). The mean value (within-trial)
for each support phase in the gait cycle (double-support and
single-support) was calculated for the leading and trailing
Initial statistical assessment required screening of out-
liers using inter-quartile range. If a case was found to be
more than three times the inter-quartile range away from the
median, that case was removed from further analysis. Mean
N-EMG values were analyzed for the effects of age
group and obstacle height (two-factor ANOVA with
repeated measures of obstacle height). Significance level
was set at a = 0.05 for all tests. Statistical analyses were
conducted with SYSTAT (Version 9, SPSS Inc., Chicago,
M.E. Hahn et al./Gait & Posture 22 (2005) 356–361 357
Initial screening indicated that eight data points were
outside reasonable variability (?3 inter-quartile ranges). Of
the 2700 total data points in the analysis (6 variables ? 30
subjects ? 5 conditions ? 3 trials), the outlying data
accounted for only 0.3% of the total data collected. Removal
of these data from the analysis was, therefore, felt to be
All subjects were able to complete the required measures
of MMT and gait analysis trials. No incidents of tripping
were observed. The testing session typically required 2 h for
completion. Subjects did not indicate discomfort during any
of the testing conditions, nor did they express any sense of
fatigue at the end of the testing session.
No significant age group differences were found for any
of the temporal-distance parameters (Table 1). As obstacle
height increased so did stride time (p < 0.001) and stride
length (p = 0.005). Gait velocity was found to decrease
linearly with increased obstacle height (p < 0.001). Since
the gait velocity showed no significant difference between
age groups (p = 0.182), it was not entered as a covariate in
the analysis of following N-EMG values.
Healthy elderly adults
activation levels in the leading and trailing limbs, compared
to young adults (Table 2). During double-support, weight
transfer and acceptance occurs laterally as well as anteriorly.
In this phase, the GM of the healthy elderly was activated
up to an average of 46% of their maximum capacity,
compared to 23% in the young for all testing conditions.
Similarly, the VL of the healthy elderly was activated up to
an average of 35% of their capacity in double-support
phase, compared to 25% in the young. Maintenance of
dynamic stability and forward progression is required
during the single-support phase of gait. During single-
support, healthy elderly GA activity reached 45% of MMT
M.E. Hahn et al./Gait & Posture 22 (2005) 356–361358
Fig. 1. Representative level walking condition showing vertical ground reaction force patterns from consecutive foot strikes, rectified EMG activation of the
GM, and smoothed GM activation (after normalization to MMT). The time scale of the normalized EMG patterns is with respect to the crossing stride. Support
and swing phases of gait are indicated: DS – double-support, SS – single-support, SW – swing. Fzrepresents the vertical ground reaction force.
Temporal-distance variables compared between groups and across the obstacle heights: group mean (S.D.)
Young ElderlyYoungElderlyYoung ElderlyYoungElderly Young Elderly
aPhrepresents height effect. Pgrepresents group effect.
maximum for all testing conditions, while the young
required 36% of their capacity.
For the trailing limb, there was a significant age effect on
the normalized activation levels of the GM (p = 0.003;
Fig. 2), but not the VL (p = 0.053) or the GA (p = 0.360). In
the leading limb, significant age effects were found for the
GM (p < 0.001) and VL (p = 0.042), but not for the GA
(p = 0.154). Increased obstacle height resulted in an
increased relative activation of all muscles of both the
leading and trailing limb (p ? 0.018). Results of the
ANOVA revealed no significant interactions between the
factors of group and obstacle height.
As elderly adults cross obstacles, they encounter
essentially the same mechanical challenges as young adults
of similar stature (assuming similar gait velocities). As there
were no significant differences in gait velocity between the
two groups (p = 0.182), it can be assumed that the
mechanical challenges encountered by the joints were
similarfortheyoungand elderly adults. Consideringthatthe
maximum available strength has been shown to be lower in
elderly adults [1,2], it follows that the elderly will use a
greater percentage of their neuromuscular capacity to
successfully ambulate and safely cross over obstacles.
Results from this study revealed that there were age-
dependent increases in the percentage of muscular capacity
used to cross obstacles. Specifically, the gluteus medius and
the vastus lateralis were activated to a significantly greater
percentage of maximum capacity in the elderly. Further-
more, as obstacle height increased, the relative activation
increased for each muscle tested, indicating substantial
challenge encountered by the neuromuscular system in
maintaining balance during the dynamic task of stepping
over obstacles. These findings are supported by joint
strength data further compiled from the two subject groups
(see Fig. 3). Isometric strength testing on a KinCom
dynamometer (Rehab World, Hixson, TN) revealed a
significant age-related reduction of strength (normalized
to individual body weight) during HA, KE and APF (one-
tailed t-test; p < 0.001, 0.001, and 0.002, respectively).
It was unexpected that walking speed would reveal no
significant group differences. However, walking speeds of
older adults in the present study compare favorably with
those reported by McFadyen and Prince , and the young
adult group demonstrated speeds comparable to those
reported by Chen et al. . Inter-laboratory variation might
M.E. Hahn et al./Gait & Posture 22 (2005) 356–361359
Normalized EMG activation percentages during double-support phases for GM and VL, single-support phases for GA: group mean (S.D.)
Limb/muscle Age groupObstacle height (% BH)
Level 2.5 5.010.0 15.0Effect
Ld. GM Elderly
Ld. VL Elderly
Tr. GA Elderly
Ld. GA Elderly
Tr.: trailing limb; Ld.: leading limb.
aSignificant age group effect (p < 0.05).
bSignificant obstacle height effect (p < 0.05).
Fig. 2. Normalized EMG activation percentages of the GM during double-
support phases of the crossing stride. Elderly adults required significantly
greater percentages of their capacity during level walking and obstacle
crossing tasks. Activation levels increased linearly as obstacle height
be used to explain the relative equalizing of gait velocities
between the two age groups. Another explanation for the
non-significant velocity differences might be the relatively
vigorous activity level of our older adult sample. The fact
that their gait velocities were not significantly lower than
younger adults may indicate that they represent a more able-
bodied section of the broader elderly population. It is quite
possible that a sample of less active elderly individuals
would demonstrate significantly slower walking speeds.
Age-related increases in relative activation of the gluteus
medius occurred during the weight acceptance and transfer
phase of gait, indicating the critical role played by hip
center of mass shifts rapidly between each foot. The timing
of EMG activation peaks was in agreement with previous
work [23,24]. Magnitudes of our N-EMG data were also in
agreement with Dubo and colleagues’  values for the VL
(?25% and 35% for young and elderly adults, respectively),
however, our GA magnitudes (?36% and 45% for young
and elderly adults, respectively) were noticeably lower than
Dubo’s (63%). As we measured activation of the medial
head and Dubo measured activation of the lateral head (or
difficult to make because of the broad age range of their
sample (8–72 years).
It was recently reported that elderly adults with balance
disorders displayed greater ranges of motion and higher
velocities in M-L motion of the center of mass while
stepping over obstacles . Increases in M-L COM motion
would certainly increase the demand on hip abductors
during weight transfer. Recent results showed that healthy
elderly adults demonstrate slightly greater displacements
and velocities in the frontal plane (as compared to young
adults), however, these differences were not significant .
This indicates that while healthy elderly adults may not
allow critically greater M-L COM motion, they may be
showing the effects of increased challenge in the task of
balance control, indicated by higher demand on the
neuromuscular capacity of the hip abductors.
One potential limitation of this study is the variable
nature of MMT collection. Although the same examiner
performed MMT on each subject, no simultaneous
measurement was made of the force applied by the
examiner, removing the possibility for quantifying variation
during MMT testing. However, it was noted that EMG
signals consistently leveled off during maximal contrac-
tions,indicatingthat maximum effortwas being madebythe
subjects. A second potential limitation exists in elderly
subjects having more adipose tissue over the GM muscle
tissue. Care was taken to accurately place the GM electrode
and its ability to pick up GM muscle activation was
rigorously checked prior to MMT and dynamic trials.
Normalizing EMG signals from adipose-rich tissue should
be similar to signals from other tissue, as sub-maximal
signal magnitudes are simply divided into the signal
maximum. The strength of normalization is the ability to
report data with respect to present condition.
The N-EMG data presented in this study may be used to
quantify the level of challenge on the neuromuscular system
when interpreted in light of available joint strength.
Although surface EMG has been known to exhibit high
variability, it appears that with careful normalization to
maximal voluntary activation levels, these proportional
values provide a reasonable representation of the control
system’s response to the challenge posed by daily activities.
Furthermore, the magnitude of demand/capacity ratio
presented here appears more reasonable than joint moment
percentages reported by Bus et al. . Their results
indicated that several activities of daily living required
greater than 100% of the available torque production in a
joint (per maximal voluntary contraction, using a dynam-
ometer). The reason for these high values may arise from the
necessary assumptions made in joint torque measurement
and estimation techniques . Further study is certainly
necessary to resolve discrepancies between measured
dynamometric torque and estimated joint moments. Addi-
tionally, comparison between these studies suggests that
further research is necessary to compare N-EMG values
calculated using MMT with those which may be calculated
from normalizing to maximal dynamometric contractions.
As EMG isadirectmeasure ofmuscleactivation,perhapsits
use is more reliable for representing musculo-skeletal
challenge than methods relying on dynamometric joint
torque and joint moment estimation techniques, which
require many assumptions to be met .
The results of this study indicate that healthy elderly
adults do require greater percentages of their neuromuscular
capacity during level walking and obstacle crossing tasks
than young adults. Specifically, the gluteus medius was
significantly challenged when the older adults stepped over
obstacles. Bus et al.  reported significantly greater hip
abductor moments in active elderly adults during stair
ascent, stair descent and level walking, compared to young
M.E. Hahn et al./Gait & Posture 22 (2005) 356–361360
Fig. 3. A comparison of isometric maximal joint strengths between the
young and older adult groups (normalized to body weight). Older adults
demonstrated significantly lower strength values for each of the joint
motions tested (*p ? 0.002).
adults. Our results support their suggestion that hip
abductors should be considered in the functional assessment
of elderly adults, and targeted as a critical component of
resistance strength training in the older adult population.
As stepping over a higher obstacle poses a greater
challengetodynamicbalance control, itmay be inferred that
population at greater risk for falls. These findings are in
agreement with recent work by Lamoureux et al. ,
showing strong association between lower extremity
isometric strength and elderly individuals’ ability to
negotiate obstacles. Combined with the additional findings
of Lamoureux et al. , the present results add emphasis to
the need for muscle strengthening as a preventative
intervention, thereby providing improved function in the
ambulatory tasks of daily living.
In conclusion, the N-EMG data reported in this study
were used to quantify the level of muscular challenge
imposed on an individual in relation to their strength
capacity. While healthy elderly adults were able to
successfully negotiate obstacles of different heights during
walking, their muscular strength capacity was significantly
lower than young adults, resulting in higher relative
muscular demands. Higher demands placed on lower
extremity muscles while stepping over obstacles may
increase the likelihood of muscle fatigue during daily
ambulation, thus placing elderly individuals at higher risk
for trips and/or falls. These findings provide further
justification for including lower extremity resistance
training in the list of preventative interventions for the
This work was supported by the Oregon Medical
Research Foundation, National Institutes of Health (AG
022204-01; HD 042039-01A1) and the International Society
of Biomechanics Dissertation Matching Grant. We grate-
fully acknowledge the contributions of Marisa Hastie and
Sentaro Koshida in data collection and analysis for this
 Grimby G, Saltin B. Mini-review: the aging muscle. Clin Physiol
 Fiatarone MA, Evans WJ. The etiology and reversibility of muscle
dysfunction in the aged. J Gerontol 1993;48S:77–83.
JL. Circumstances and consequences of falls experienced by a com-
munity population 70 years and over during a prospective study. Age
 Buchner DM, de Lateur BJ. The importance of skeletal muscle
strength to physical function in older adults. Ann Behav Med
 Brown M, SinacoreD, Host H.The relationship ofstrength to function
in the older adult. J Gerontol 1995;50A:55–9.
 Whipple RH, WolfsonLI, Amerman PM.The relationship of knee and
ankle weakness to falls in nursing home residents: an isokinetic study.
J Am Geriatr Soc 1987;35:13–20.
 Chou L-S, Draganich LF. Stepping over an obstacle increases the
motions and moments of the joints of the trailing limb in young adults.
J Biomech 1997;30:331–7.
 Lamoureux EL, Sparrow WA, Murphy A, Newton RU. The relation-
ship between lower strength and obstructed gait in community-dwell-
ing older adults. J Am Geriatr Soc 2002;50:468–73.
 Lamoureux EL, Sparrow WA, Murphy A, Newton RU. The effects of
improved strength on obstacle negotiation in community-dwelling
older adults. Gait Posture 2003;17:273–83.
 Berg WP, Blasi ER. Stepping performance during obstacle clearance
in women: age differences and the association with lower extremity
strength in older women. J Am Geriatr Soc 2000;48:1414–23.
 Barry DT, Gordon KE, Hinton GG. Acoustic and surface EMG
diagnosis of pediatric muscle disease. Muscle Nerve 1990;13:286–90.
 Ebersole KT, Housh TH, Johnson GO, Evetovitch TK, Smith DB,
Perry SR. MMG and EMG responses of the superficial quadriceps
femoris muscle. J Electromyogr Kinesiol 1999;9:219–27.
 Maton B, Petitjean M, Cnockaert JC. Phonomyogram and electro-
J Appl Physiol 1990;60:194–201.
 Madeleine P, Bajaj P, Sogaard K, Arendt-Nielsen L. Mechanomyo-
graphy and electromyography force relationships during concentric,
isometric and eccentric contractions. J Electromyogr Kinesiol
 Chen H-C, Aston-Miller JA, Alexander NB, Schultz AB. Stepping
over obstacles: gait patterns of healthy young and old adults. J
 Hahn ME, Chou LS. Age-related reduction in sagittal plane center
of mass motion during obstacle crossing. J Biomech 2004;37:
 Chou LS, Draganich LF. Placing the trailing foot closer to an obstacle
reduces flexion of the hip, knee, and ankle to increase the risk of
tripping. J Biomech 1998;31:685–91.
 Chou LS, Draganich LF. The effects of increasing obstacle height and
decreasingtoe-obstacledistanceon trailinglimbkinetics.Gait Posture
 Chou LS, Kaufman KR, Brey RH, Draganich LF. Motion of the whole
Gait Posture 2001;13:17–26.
 Chou LS, Kaufman KR, Hahn ME, Brey RH. Medio-lateral motion of
the center of mass during obstacle crossing distinguishes elderly
individuals with imbalance. Gait Posture 2003;18:125–33.
 Hahn ME, Chou LS. Can motion of individual body segments
identify dynamic instability in the elderly? Clin Biomech 2003;18:
 McFadyen BJ, Prince F. Avoidance and accommodation of surface
height changes by healthy, community-dwelling, young and elderly
men. J Gerontol 2002;57A:B166–74.
 SutherlandDH. Electromyographic study of plantar flexors of ankle in
normal walking on level. J Bone Joint Surg 1966;48:66–71.
 Dubo HIC, Peat M, Winter DA, Quanbury AO, HobsonDA, SteinkeT,
et al. Electromyographic temporal analysis of gait: normal human
locomotion. Arch Phys Med Rehabil 1976;57:415–20.
 Bus SA, Row BS, Okita N, Cavanagh PR. Lower extremity strength
requirements in young and older female adults during locomotor
activities of daily living. In: Proceedings of the 25th Annual Meeting
of the American Society of Biomechanics; 2001. p. 95–6.
 Arampatzis A, Karamanidis K, De Monte G, Stafilidis S, Morey-
Klapsing G, Bruggemann G-P. Differences between measured
and resultant joint moments during voluntary and artificially
elicited isometric knee extension contractions. Clin Biomech
M.E. Hahn et al./Gait & Posture 22 (2005) 356–361 361