Clinical Biomechanics 15 (2000) 441±448

www.elsevier.com/locate/clinbiomech

Arm motion and load analysis of sit-to-stand, stand-to-sit, cane

walking and lifting
Carolyn Anglin a,b,*, Urs P. Wyss a,b
a
Clinical Mechanics Group, Queen's University, Kingston, Canada
b
Sulzer Orthopedics Ltd., P.O. Box 65, CH-8404 Winterthur, Switzerland
Received 9 September 1999; accepted 30 November 1999

Abstract
Objective. To conduct a pilot study to characterize the hand loads, arm joint angles and external moments corresponding to ®ve
activities of daily living demanding of the shoulder, for healthy subjects over 50 years of age.
Design. The tasks were sit-to-stand, stand-to-sit, cane walking, lifting a 5 kg box with both hands, and lifting a 10 kg suitcase.
Background. Arm motion and loading have not been previously studied for functional daily-living tasks involving substantial
external loads.
Methods. Motion was tracked using an optoelectronic system. Loads were measured using an instrumented chair arm, a force
plate, and gravitational and acceleration loads. Six healthy volunteers (3 male, 3 female), mean age 55, with no history of shoulder
problems participated in the study.
Results. Average peak external moments ranged from 12.3 N m for sitting down into a chair to 27.9 N m for lifting a suitcase.
Except for lifting the box, which had much lower loads, average peak hand loads varied from 16% to 19% of body weight (114±134
N). The arcs of motion were larger than for seated activities of daily-living studied previously.
Conclusions. The ®ve tasks studied are commonly performed, yet involve large external moments. Lifting represents the greatest
potential loading at the shoulder as it resulted in the highest external moments; furthermore, loads larger than those used in this
study might be commonly lifted.

Relevance
External moments at the shoulder should not be underestimated, even for activities of daily living. Ó 2000 Elsevier Science Ltd.
All rights reserved.

Keywords: Activities of daily living; Motion analysis; Arm; Shoulder; Loading

1. Introduction the present study provides useful information on the

Few groups have studied whole-arm motion during
functional tasks [1,2]. Of the tasks that have been
studied, most exert only a small external moment;
loading at the shoulder is therefore generally underes-
timated. Although athletic tasks exert a large moment
[3,4], they are assumed not to be performed by older
people. We hypothesized that loading at the shoulder
could be substantial even for activities of daily living
(ADL). Whereas the motivation for performing this
study was to provide input to a biomechanical model in
order to calculate the contact forces at the shoulder [2],
*
Corresponding author.
E-mail address: carolyn.anglin@sulzer.com (C. Anglin).
0268-0033/00/$ - see front matter Ó 2000 Elsevier Science Ltd. All rights reserved.
PII: S 0 2 6 8 - 0 0 3 3 ( 9 9 ) 0 0 0 9 3 - 5
angles, loads and moments used to perform demanding
ADL. The goal was not to reach statistical conclusions
about these parameters, but rather to provide an indi-
cation of the expected values and variation in people of
a similar age to those who receive a shoulder prosthesis
[5].
There were ®ve research questions for this study:
1. What are the trunk angles, arm angles, and hand
loads at the time of the peak external moments for
the given tasks? This information is relevant to fur-
ther modelling of the arm.
2. What are the ranges of motion of the joints for these
tasks and how do they compare between tasks and to
other tasks? This is relevant for the design of orthoses
and prostheses for which the joint ranges of motion
and workspace must be de®ned.
442 C. Anglin, U.P. Wyss / Clinical Biomechanics 15 (2000) 441±448

3. What are the peak external moments acting at the
shoulder for these tasks? Not only do the moments
give an indication of the forces experienced at the
shoulder, but how closely the moments approach a
person's strength is relevant to therapists.
4. How do the external moments compare between
tasks and to tasks studied previously? This informa-
tion is necessary to place these tasks in the context
of both less-demanding ADL and more-demanding
wheelchair and lifting activities.
5. How consistent are the angles, loads, moments and
ranges of motion between subjects? This is necessary
for both modelling and design.
As older people lose strength and stability in their
legs, they rely more on the assistance of their arms. Knee
forces were shown to decrease by at least 18% by using
the arms to aid in rising from a chair [6±8]. Those unable
to get out of a chair without the use of armrests exerted
a greater hand force, ¯exed their trunks more and took
longer to rise than able subjects [9,10]. Okuno and co-
workers [11] found that using a cane reduced peak bone-
on-bone forces at the knee by 20±30%. Brand and
Crowninshield [12] calculated a 40% reduction in hip
contact force when using a cane. A cane has also been
shown to aid balance, thereby preventing falling [13].
Lifting is clearly demanding of the shoulder, but despite
the extensive literature on lifting, the arm angles and
moment arms have not been studied.
The sit-to-stand and stand-to-sit tasks were separated
into two activities so that they could be clearly de®ned
[14]. Cane height, adjustable every 25 mm, was set for
comfort because Okuno and coworkers [11] had shown
that knee loading was not a€ected by cane height. The
toolbox had dimensions of 42 cm ´ 15 cm ´ 16 cm high;
those of the suitcase were 65 cm ´ 18.5 cm ´ 51 cm high
with an additional 5.5 cm for the handle. A custom-built
three-degree-of-freedom strain-gauge load transducer
was built into the chair arm, the hand landing on top of
the transducer to minimize applied moments (con®rmed
in photographs). Calibrations showed accuracy within 2
N and minimal cross-talk. For the cane task, subjects
were positioned following practice strides, without their
knowledge, such that the cane landed on an A M T I OR6-
5-1 force platform (A M T I , Watertown, MA, USA). The
load of the 10 kg suitcase was calculated from gravita-
tional and vertical acceleration forces [15]. The load of
the 5 kg box on one hand was assumed to be 2.5 kg; no
account was taken of acceleration since at the time of
greatest external moment accelerations were negligible.
Eight infrared light-emitting diodes (IREDs) were
positioned to track the subjectÕs movements, ®ve on the
right arm and three on the trunk as displayed in Fig. 1.
The three trunk markers were placed on an 8 cm square
plastic extension with a 4 cm wide base perpendicular to
the spine, at approximately the ®rst three thoracic
vertebrae; two markers (1 and 2) were aligned in the
vertical direction when the subjectÕs trunk was visually

2. Methods

The tasks chosen were: standing up from and sitting

down into a chair using the arms, walking with a cane,
lifting a 5.0 kg box with both hands from the ¯oor to
shoulder height, and lifting and walking with a 10.0 kg
suitcase. Six healthy volunteers were recruited who were
required to be at least 50 years of age with no history of
shoulder problems. Healthy subjects were chosen for
this pilot study because the sessions were demanding; a
larger group would be necessary to investigate the
variations in those with a shoulder prosthesis. This
would be appropriate in a second study.
The subjects were evenly split between men and
women. Age varied from 51 to 64 years (mean, 55 years),
weight from 52 to 89 kg (mean, 73 kg) and height from
152 to 187 cm (mean, 168 cm); all but one subject
were right-hand dominant. Five trials were conducted
for each task, all in the same session. There was a rest
between tasks but not between trials. The QueenÕs
University Research Ethics Board approved the experi-
ments. Full experimental details are provided in [2].
Fig. 1. Marker positions and coordinate systems. See text for a de-
A standard chair was used: the height of the seat was
scription of the marker positions; see Table 1 for the coordinate system
44 cm and that of the armrests 66 cm; the armrests were de®nitions. The X-axes are out of the plane of the page. The markers
centred 49 cm apart. Seat height was not adjusted to the were translated to the joint centres to establish the origins of the co-
individual since this is more typical of daily experience. ordinate systems.
 C. Anglin, U.P. Wyss / Clinical Biomechanics 15 (2000) 441±448 443

vertical. The third marker (3), placed posteriorly, de- tion for these purposes, and that the markers were
®ned the perpendicular plane of the trunk (Yt in Fig. 1). placed ± following mathematical translation ± at the
The markers on the arm were placed on the skin directly joint centres. The large distance between markers im-
lateral to the glenohumeral joint centre (4), just distal to proved accuracy.
the bony lateral epicondyle of the elbow (5), at the wrist The joint angles were calculated using a program
midway between the radial and ulnar styloids (6), on an developed by the author for a previous study [18,19],
extension above the wrist (7) and on the third meta- with the addition of trunk movement. The technique
carpalphalangeal knuckle (8). Markers were always ap- used a direct geometric calculation as opposed to Euler
plied by the same investigator. The direction of the wrist angles to avoid inconsistencies [2] ± recently, more
and knuckle markers was changed between the ®rst sophisticated matching methods have been developed
three and the last two tasks so that the markers re- using least-squares techniques to minimize these incon-
mained visible to the cameras at all times. An O P T O T - sistencies and the noise encountered in the rigid body
R A K 3020 optoelectronic tracking system (Northern de®nitions [20,21].
Digital, Waterloo, Canada), consisting of an array of The position of the arm at any time was described
three one-dimensional cameras, was used to record the relative to the trunk by a sequence of Euler rotations
three-dimensional marker positions to within an accu- (see Table 2). The conceptual starting position was with
racy of 0.45 mm. Data were sampled at 50 Hz and ®l- the arm elevated 0°. The arm is ®rst rotated to de®ne the
tered at 6 Hz. Costigan and coworkers [16] describe the plane of elevation, then elevated, then internally/exter-
data processing. nally rotated about the humeral axis. The plane of ele-
Following attachment of the markers, the distances vation has also been called horizontal ¯exion/extension
from the markers to the approximate bone joint centres or azimuth. The humeral rotations were followed by the
were measured such that they could be later translated elbow and wrist rotations. The pro/supination angle
mathematically to the joint centres. The glenohumeral reported in Table 5 is the pro/supination angle derived
distance was based on Wang [17], namely 37 mm infe- from the rotation sequence plus the value of the carrying
rior, 14 mm lateral and 8 mm anterior to the acromion. angle in order to agree with the anatomical de®nition
The elbow joint centre was taken midway between the [18,19,22]. The accuracy resulting from the tracking
medial and lateral epicondyles. The wrist joint centre system, marker positions, translation to the joint centre
was assumed to be at half the wristÕs thickness. The and skin movement artefacts was estimated to be within
carrying angle (angle between the extension of the upper 6° for all joint angles.
arm and the forearm when the elbow is extended in the
anatomical position) was measured using a goniometer.
The body-segment coordinate systems were each de- Table 2
®ned from three non-collinear markers (see Table 1). Angle de®nitions
Humeral rotations were relative to the thorax. Since
Angle Rotation and sign de®nition
markers were sometimes located on an adjacent seg-
ment, it was assumed both that the joints acted as Trunk ¯exiona About Xg , positive forwards
hinges, which is a common and reasonable approxima- Lateral bendinga About Yg , positive right
Plane of elevation About Zt (or Xu ), starting from
90° abduction, positive forwards
from the frontal plane
Table 1 Elevation 90° minus angle about Yu0 ,
Coordinate systemsa positive outwards
Internal/external rotation About Zu00 , internal rotation
Body X Y Z positive
segment Elbow Ab/adduction About Yf , abduction positive
(carrying angle)
Thorax Xt ˆ 13 ´ 12 Yt ˆ Zt ´ Xt Zt ˆ 21
Upper armb Xu ˆ Yu ´ Zu Yu ˆ 50 60 Zu ˆ 50 40 Elbow ¯exion About Xf0 , 0° when arm extended
Forearm Xf ˆ 60 70 ´ 60 50 Yf ˆ Zf ´ Xf Zf ˆ 60 50 Pro/supination About Xf00 , corrected to be about
Hand Xh ˆ 60 70 ´ 80 60 Yh ˆ Zh ´ Xh Zh ˆ 80 60 the ulnar axis, pronation positive
Wrist ¯exion About Xh , palmar direction
a
The coordinate systems were ®rst de®ned from the marker positions positive
shown in Fig. 1. Markers shown with 0 were translated mathematically Radial/ulnar deviation About Yh0 , radial deviation
to the joint centre along the appropriate coordinate axis and the co- positive
ordinate system recalculated.
b a
The upper arm rotation was based initially on the wrist marker, Trunk angles were initially calculating in an Euler sequence, but for a
however elbow ¯exion and internal/external rotation were adjusted better intuitive presentation, the absolute angle about the two global
geometrically to account for the apparent rotation due to the carrying axes are presented; twisting of the trunk could not be distinguished
angle and to account for the fact that the pro/supination axis passes from rotation of the whole body since there were no markers on the
through the ulnar head at the wrist (see [18] for details). lower limb.
444 C. Anglin, U.P. Wyss / Clinical Biomechanics 15 (2000) 441±448

The load moment was calculated by multiplying the 3. Results

components of the load vector times the appropriate
perpendicular distances between the shoulder and hand;
3.1. Joint angles, hand positions and hand loads
this value was con®rmed by multiplying the three-
dimensional load vector times the perpendicular three-
The joint angles corresponding to the peak external
dimensional moment arm from the shoulder to the load
moment (Table 3) were similar between sit-to-stand
vector. For the moment calculations, the hand and el-
(standing) and stand-to-sit (sitting), although the arm
bow locations were based on the marker positions (8
tended to be farther behind the body (more horizontally
and 5 respectively) whereas the shoulder location was
extended) during sitting; the peak load was higher for
based on the marker position moved medially to the
standing than for sitting. Lifting a box was unlike the
joint centre (40 ) relative to each subject.
other four tasks in that the arm was forward and more
The total moment was the moment due to the hand
elevated, with minimal trunk rotation. Elbow ¯exion
load plus the moment due to the gravitational load of the
was greatest in the chair tasks.
arm (adding to or subtracting from the load vector de-
At the time of the peak external moment, the hand
pending on the respective moment directions). The upper
was positioned roughly below the shoulder for the
arm gravitational load was assumed to be 2.8% of body
standing, sitting and suitcase tasks, down and forwards
weight, lying at 43.6% of the distance from the shoulder
for the cane task and forwards for lifting a box. The
to the elbow; the forearm and hand were assumed to be
total distance ranged from 33.3 cm for standing from a
2.2% of body weight with a centre of gravity at 68.2% of
chair to 57.2 for lifting a suitcase (Table 4).
the distance between the elbow and hand [23]. No inertial
Average hand loads ranged from 3% body weight
components were assumed except when lifting a suitcase,
(BW) for lifting the 5 kg box to 19% BW for standing
since, except for the suitcase, the velocities at the peak
from a chair (Table 5). The loads were primarily verti-
moment were negligible. The same accelerations were
cal. Maximum hand loads reached as high as 30% BW
assumed for the arm segments as for the suitcase.
for cane walking.

Table 3
Average angles corresponding to the peak external momenta

Angle.(°) Standing Sitting Cane Box Suitcase

Trunk ¯exion 25.(9) 29.(13) 9.(5) )1.(6) 22.(12)

Lateral bending )6.(5) )3.(4) )1.(4) )3.(3) 3.(12)
Plane of elevation )26.(22) )40.(37) )10.(37) 60.(11) 20.(47)
Elevation 29.(8) 29.(8) 14.(8) 69.(8) 22.(9)
Internal/external rotation 29.(21) 23.(17) 24.(11) 16.(12) 12.(32)
Elbow ¯exion 91.(16) 91.(15) 57.(10) 59.(11) 33.(6)
Pro/supination 53.(12) 47.(11) 22.(20) )32.(10) 17.(20)
Wrist ¯exion )49.(22) )50.(21) )26.(12) )1.(5) )16.(11)
Radial/ulnar )3.(9) )2.(10) )19.(9) 33.(7) 11.(5)
a
See Table 2 for angle and sign descriptions. The schematics represent the position at the time of the peak hand load. Standard deviations are given in
brackets.

Table 4
Hand position relative to the shoulder at the peak external momenta

Task X-coordinate (forward) Y-coordinate (left) Z-coordinate (up) Total distance

(cm) (cm) (cm) (cm)

Sit-to-stand 4.3 (6.0) )5.6 (3.0) )32.9 (7.2) 33.3 (2.9)

Stand-to-sit 2.5 (7.0) )5.2 (1.2) )32.5 (5.8) 36.5 (5.6)
Cane walking 12.8 (8.8) )4.5 (2.4) )49.5 (9.4) 52.3 (8.1)
Lifting box 52.4 (7.2) )4.8 (5.6) )1.6 (4.4) 53.1 (7.1)
Lifting suitcase 3.3 (14.0) )11.4 (2.9) )54.1 (7.6) 57.2 (7.1)
a
Given in the global coordinate system: mean (standard deviation).
 C. Anglin, U.P. Wyss / Clinical Biomechanics 15 (2000) 441±448 445

Table 5
Peak hand loads

Task Average peak Average peak Average forward/lateral Maximum load Maximum
load (N) load (´BW)a load angle (°)b (N) load (´BW)a

Sit-to-stand 134 (37) 0.19 (0.06) 8.0 (9.0)/)0.7 (3.1) 154 0.24
Stand-to-sit 114 (29) 0.16 (0.07) )5.1 (32.3)/3.7 (11.6) 179 0.27
Cane walking 130 (46) 0.18 (0.05) 3.9 (5.3)/1.8 (2.5) 225 0.30
Lifting box )25 (0) 0.03 (0.01) 0 25 0.05
Lifting suitcase )126 (5) 0.18 (0.03) 0 137 0.26
a
BW ± body weight; standard deviations in brackets.
b
Forward is the angle from the vertical in the global Yg Zg plane; lateral is the angle from the vertical in the global Xg Zg plane. The load vector is the
reaction force of the chair arm acting upwards against the hand.

Table 6
Peak external moments

Task Load moment Total moment Coordinates of total moment Maximum total moment
(N m)a (N m)a; b (N m)c (N m)

Sit-to-stand 18.2 (6.5) 16.2 (7.6) (0.76, )0.53, 0.09) 31.5

Stand-to-sit 15.6 (9.9) 12.3 (9.7) (0.23, )0.97, 0.06) 47.7
Cane walking 23.5 (11.9) 24.0 (11.3) (0.84, 0.53, )0.05) 45.6
Lifting box 13.3 (1.7) 21.8 (4.0) ()1.00, 0.06, 0) 27.1
Lifting suitcase 26.5 (4.4) 27.9 (4.2) (0.16, 0.99, 0) 38.0
a
Average and standard deviation of peak values for subjects.
b
Total moment ˆ load moment + gravitational moment of upper arm, forearm and hand.
c
Coordinates ± average normalized coordinates of moment vector about the Xg -, Yg -, Zg -axes in the global system (see Fig. 1) ± (right, forward, up).

3.2. Ranges of motion similarity among subjects. Trunk movement, elevation,

The tasks required di€erent ranges of rotation for
each joint, and the degree of variability varied with each
task and joint. Interested readers are referred to [2] for
®gures of the ranges of motion, angle-time sequence and
load-time sequence for each task.
elbow ¯exion and radial-ulnar deviation were the most
consistent among subjects. Plane of elevation, internal/
external rotation, pro/supination and wrist ¯exion were
less consistent, implying that di€erent strategies were
being used by di€erent subjects.

3.3. External moments 4. Discussion

Average external moments ranged from 12.3 N m for
sitting down to 27.9 N m for lifting a suitcase; maximum
moments were substantially higher (Table 6). The mo-
ments for the cane and box tasks occurred primarily
about the Xg axis; for the suitcase the moment was
mostly about the Yg axis, although the direction did
vary; the moment directions for the chair tasks were
mainly in the Xg Yg plane, but varied within the plane
among subjects (Table 6).
3.4. Consistency of angles, loads, moments and ranges of
motion
All of the parameters showed considerable variabili-
ty, as evidenced by the standard deviations in the re-
spective tables. In general, lifting a box with both hands
to shoulder height showed the greatest repeatability and
4.1. Joint angles, hand positions and hand loads
In contrast to most daily living tasks, in which a light
load is manipulated in front of the subject from a seated
position [1,18,19], in the ®ve tasks studied, the hand was
positioned farther back, higher loads were experienced
and standing positions were included. For example, the
plane of elevation angle of the humerus was small or
negative (i.e. to the side or behind the body) for the chair
and cane tasks. Lifting the box to shoulder height re-
quired more elevation and the chair and lifting tasks
required greater elbow extension than is typical for
other ADL [1,19].
Hand loads for the sit-to-stand task (134 N or 19%
body weight) were less than but comparable to those
reported by Alexander and coworkers (144 N/21% for
healthy older subjects, and 159 N/29% for those unable
446 C. Anglin, U.P. Wyss / Clinical Biomechanics 15 (2000) 441±448
to get up without arm assistance). On the other hand,
the load angles in the present study were distinctly
more vertical (i.e. 8° versus approximately 40°) than
those reported by Alexander and coworkers, implying
that our subjects were pushing themselves more up
than forward. An upper limit on the hand load is
provided by Runciman [3] who recorded a hand load of
40% body weight at 10° when subjects pushed them-
selves up from a chair with straightened legs. Cane
loading in the present study was at the higher end of
the range reported by Brand and Crowninshield [12]
and by Ely and Smidt [24] for walking, but lower than
that reported by Ashton-Miller and coworkers [13] to
prevent falling.
4.2. Ranges of motion
The tasks studied covered larger ranges of motion
in elevation, plane of elevation and elbow ¯exion than
has been reported in previous studies [1]. If these tasks
are desired, then their ranges of motion should be
taken into account in the design of orthoses and
prostheses.
The ranges of motion compared well to previous
studies of the sit-to-stand task, which was the only task
for which some angles were available. Trunk ¯exion
ranged up to 36°; for comparison, Alexander and co-
workers [9] reported average ¯exion angles of 33° for
able subjects and 42° for those who were unable to get
out of the chair without arm support; Wheeler and co-
workers [10] reported a forward lean of only 13° for an
older group getting out of a standard chair, but 23° for
those getting out of a more comfortable chair. The
ranges of elbow ¯exion, wrist ¯exion and radial/ulnar
deviation for sit-to-stand were similar to previous
studies [10,14,25±29], di€erences being perhaps due to
the chair used.
The lifting technique was deliberately not speci®ed;
most of the subjects chose to bend the trunk to lift the
box rather than bending the knees. Although the shelf
was at shoulder height, an average elevation of only 69°
was required by the subjects due to elbow ¯exion.
4.3. External moments
Considerable loading of the shoulder can occur
during ADL. The chair tasks, even as the lowest among
the ®ve, resulted in external moments of 16.2 and 12.3
N m. By comparison, external moments for the ma-
jority of daily living tasks that have been studied would
be below, often well below, 10 N m due to the low
elevations, ¯exed elbows and low loads carried [18,19].
The two lifting tasks, already having the highest ex-
ternal moments (21.8 and 27.9 N m), would be even
higher if higher loads were lifted. For example, Arbo-
relius and coworkers [30] recorded maximum moments
about the shoulder of from 32 to 50 N m (depending
on the lifting technique) for lifting a 12.8 kg object.
Van der Helm and Veeger [4] reported external mo-
ments up to 50 N m for wheelchair propulsion while
applying 40% of the subject's maximum voluntary
moment.
These average moments represent up to half the
available shoulder strength for women [31,32]; the
maximum moments represent a substantially higher
proportion of available strength. The strength of those
with a repair of the rotator cu€ muscles, common in
shoulder arthroplasty patients, is further reduced
[33,34]. Whereas it appears that the demands of lifting a
10 kg suitcase exceed abduction strength, extra strength
was probably achieved by supporting the arm on the
body.
4.4. Consistency of angles, loads, moments and ranges of
motion
Di€erences in elbow ¯exion among subjects could
partly be explained by di€ering anthropometrics, i.e.
since the chair height remained the same, each subject
required di€erent kinematics. This was intentional since
the majority of chairs are not adjustable. Changes in
internal/external rotation were linked to changes in the
plane of elevation: due to the mathematical sequence of
rotations, a change in one necessitates a change in the
other to keep the hand pointing in the same direction.
Furthermore, when elevation was zero (all tasks ap-
proached this at some point), the plane of elevation and
internal/external rotations were indistinguishable (called
gimbal-lock).
4.5. Limitations
The main limitation of this study was that only six
subjects were studied. Although they represented a large
range of weights and heights as well as both sexes, the
results cannot be extrapolated to the general population.
Rather, this study was planned as a pilot study to pro-
vide initial information about the angles, loads and
moments for demanding ADL for a healthy older pop-
ulation since such information was previously not
available. A further limitation was that the subjects at
times found the actions awkward, especially walking
with a cane and sitting down into the chair. There was
no attempt to vary the method of performing each task
since our interest was in studying the most natural
method for the subject. While there was some uncer-
tainty (estimated to be less than 6°) in the angular data
due to, for example, skin movement artefacts and sim-
plifying assumptions, the load data were accurate to
within a few Newtons.
 C. Anglin, U.P. Wyss / Clinical Biomechanics 15 (2000) 441±448
5. Conclusions
The tasks studied were common but demanding
functional activities. Whereas sit-to-stand and stand-to-
sit had the lowest average external moments, their
maximum moments were large. Walking with a cane and
lifting a 5 kg box to shoulder height with both hands
had similar average moments, but the former had much
higher maximum moments. The highest average external
moment was for lifting a 10 kg suitcase (27.9 N m). Since
lifted loads are only limited by a personÕs strength, even
higher external moments are possible at the shoulder.
Acknowledgements
We wish to thank David Siu for assisting in all of the
motion analysis sessions, Gerry Saunders for con-
structing the load transducer for the chair arm, and Dr.
Dave Pichora for discussions during the project, all of
whom are from the Clinical Mechanics Group at
Queen's University. Thanks are also due to the six
subjects who volunteered to participate in this study, for
their enthusiasm and patience. Financial assistance was
provided by Sulzer Orthopedics Ltd., Queen's Graduate
School and the Ontario government.
References
[1] Buckley MA, Yardley A, Johnson GR, Carus DA. Dynamics of
the upper limb during performance of the tasks of everyday living
± a review of the current knowledge base. In: Proceedings of the
Institution of Mechanical Engineers, Part H. J Eng Med
1996;210:241±47.
[2] Anglin C. Shoulder prosthesis testing. Ph.D. Thesis, Queen's
University, Kingston, 1999. (Including: Anglin C, Wyss UP,
Pichora DR. Glenohumeral contact forces. In: Proceedings of the
Institution of Mechanical Engineers, Part H. J Eng Med (in press)
and Anglin C, Wyss UP. Review of arm motion analyses. In:
Proceedings of the Institution of Mechanical Engineers, Part 4.).
[3] Runciman RJ. Biomechanical model of the shoulder joint. Ph.D.
Thesis, University of Strathclyde, Glasgow, 1993.
[4] Van der Helm FCT, Veeger HEJ. Quasi-static analysis of muscle
forces in the shoulder mechanism during wheelchair propulsion.
J Biomech 1996;29:39±52.
[5] Boyd AD, Thomas WH, Scott RD, Sledge CB, Thornhill TS.
Total shoulder arthroplasty versus hemiarthroplasty: indications
for glenoid resurfacing. J Arthroplasty 1990;5:329±36.
[6] Schultz AB, Alexander NB, Ashton-Miller JA. Biomechanical
analyses of rising from a chair. J Biomech 1992;25:1383±91.
[7] Burdett RG, Habasevich R, Pisciotta J, Simon SR. Biomechan-
ical comparison of rising from two types of chairs. Phys Therapy
1985;65:1177±83.
[8] Ellis MI, Seedhom BB, Wright V. Forces in the knee joint whilst
rising from a seated position. J Biomed Eng 1984;6:113±20.
[9] Alexander NB, Schultz AB, Warwick DN. Rising from a chair:
e€ects of age and functional ability on performance biomechan-
ics. J Gerontol: Med Sci 1991;46:M91±98.
447
[10] Wheeler J, Woodward C, Ucovich RL, Perry J, Walker JM.
Rising from a chair: in¯uence of age and chair design. Phys
Therapy 1985;65:22±6.
[11] Okuno M, Wyss UP, Costigan PA, Deluzio KJ, Chew R. Knee
joint forces during contra-lateral cane-assisted gait. Proc Int Soc
Biomech 1995;15:694±5.
[12] Brand RA, Crowninshield RD. The e€ect of cane use on hip
contact force. Clin Orthopaed Relat Res 1980;147:181±4.
[13] Ashton-Miller JA, Yeh MWL, Richardson JK, Galloway T. A
cane reduces loss of balance in patients with peripheral neurop-
athy results: from a challenging unipedal balance test. Arch Phys
Med Rehabilit 1996;77:446±52.
[14] Packer TL, Wyss UP, Costigan PA. Elbow kinematics during sit-
to-stand and stand-to-sit movements. Clin Biomech 1993;8:322±8.
[15] Kincaid D, Cheney W. Numerical analysis: mathematics of
scienti®c computing. Paci®c Grove (CA): Brooks/Cole, 1990
[Chapter 7].
[16] Costigan PA, Wyss UP, Deluzio KJ, Li J. Semiautomatic three-
dimensional knee motion assessment system. Med Biol Eng
Comput 1992;30:343±50.
[17] Wang XG. Construction of arm kinematic linkage from external
surface markers. In: Proceedings of the Fourth International
Symposium on 3D Analysis of Human Movement, Grenoble,
France, 1996.
[18] Anglin C. A functional task analysis and motion simulation for
the development of a powered upper-limb orthosis. MASc Thesis,
University of British Columbia, Vancouver, 1993.
[19] Romilly DP, Anglin C, Gosine RG, Hershler C, Raschke SU. A
functional task analysis and motion simulation for the develop-
ment of a powered upper-limb orthosis. IEEE Trans Rehabilit
Eng 1994;2:119±29.
[20] Challis JH. A procedure for determining rigid body transforma-
tion parameters. J Biomech 1995;28:733±7.
[21] Soderkvist I, Wedin P-A. Determining the movements of the
skeleton using well-con®gured markers. J Biomech 1993;26:1473±
7.
[22] Shiba R, Sorbie C, Siu D, Bryant JT, Cooke TDV, Wevers HW.
Geometry of the humeroulnar joint. J Orthopaed Res 1988;6:897±
906.
[23] Dempster WT. Space Requirements of the seated operator.
WADC-TR-55-159, Wright Patterson Air Force Base; 1955. In:
Winter DA, editor. Biomechanics and motor control of human
movement, second ed. New York: Wiley, 1990, p. 57.
[24] Ely DD, Smidt GL. E€ect of cane on variables of gait for patients
with hip disorders. Phys Therapy 1977;57:507±12.
[25] Morrey BF, Askew LJ, An KN, Chao EY. A biomechanical
study of normal functional elbow motion. J Bone Joint Surg Am
A 1981;63:872±7.
[26] Packer TL, Peat M, Wyss U, Sorbie C. Examining the elbow
during functional activities. Occupat Therapy J Res 1990;10:323±
33.
[27] Wheeler J, Woodward C, Ucovich RL, Perry J, Walker JM.
Rising from a chair: in¯uence of age and chair design. Phys
Therapy 1985;65:22±6.
[28] Brum®eld RH, Champoux JA. A biomechanical study of normal
functional wrist motion. Clin Orthopaed Relat Res 1984;187:23±
5.
[29] Ryu J, Cooney WP, Askew LJ, An K-N, Chao EYS. Functional
ranges of motion of the wrist joint. J Hand Surg Am A
1991;16:409±19.
[30] Arborelius UP, Ekholm J, Nemeth G, Svensson O, Nisell R.
Shoulder joint load and muscular activity during lifting. Scand J
Rehabilit Med 1986;18:71±82.
[31] Kuhlman JR, Iannoti JP, Kelly MJ, Reigler FX, Gevaert ML,
Ergin TM. Isokinetic and isometric measurement of strength of
external rotation and abduction of the shoulder. J Bone Joint
Surg Am A 1992;74A:1320±33.
448 C. Anglin, U.P. Wyss / Clinical Biomechanics 15 (2000) 441±448

[32] Murray MP, Gore DR, Gardner M, Mollinger LA. Shoulder
motion and muscle strength of normal men and women in two
age groups. Clin Orthoped 1985;192:268±73.
[33] Gore DR, Murray MP, Sepic SB, Gardner GM. Shoulder muscle
strength and range of motion following surgical repair of full-
thickness rotator cu€ tears. J Bone Joint Surg Am A
1986;68A:266±72.
[34] Kirschenbaum D, Coyle Jr. MP, Leddy JP, Katsaros P, Tan FJR,
Cody RP. Shoulder strength with rotator cu€ tears. Pre-and post-
operative analysis. Clin Orthoped 1993;288:174±8.