This article was downloaded by: [University of Western Ontario]

On: 11 November 2014, At: 03:56

Publisher: Taylor & Francis
Informa Ltd Registered in England and Wales Registered Number: 1072954 Registered office: Mortimer House,
37-41 Mortimer Street, London W1T 3JH, UK

Ergonomics
Publication details, including instructions for authors and subscription information:
http://www.tandfonline.com/loi/terg20

The impact of work configuration, target angle and

hand force direction on upper extremity muscle
activity during sub-maximal overhead work
a a a
Jaclyn N. Chopp , Steven L. Fischer & Clark R. Dickerson
a
Department of Kinesiology , University of Waterloo, 200 University Avenue W , Waterloo,
ON N2L 3G1, Canada
Published online: 12 Jan 2010.

To cite this article: Jaclyn N. Chopp , Steven L. Fischer & Clark R. Dickerson (2010) The impact of work configuration, target
angle and hand force direction on upper extremity muscle activity during sub-maximal overhead work, Ergonomics, 53:1,
83-91

To link to this article: http://dx.doi.org/10.1080/00140130903323232

PLEASE SCROLL DOWN FOR ARTICLE

Taylor & Francis makes every effort to ensure the accuracy of all the information (the “Content”) contained
in the publications on our platform. However, Taylor & Francis, our agents, and our licensors make no
representations or warranties whatsoever as to the accuracy, completeness, or suitability for any purpose of the
Content. Any opinions and views expressed in this publication are the opinions and views of the authors, and
are not the views of or endorsed by Taylor & Francis. The accuracy of the Content should not be relied upon and
should be independently verified with primary sources of information. Taylor and Francis shall not be liable for
any losses, actions, claims, proceedings, demands, costs, expenses, damages, and other liabilities whatsoever
or howsoever caused arising directly or indirectly in connection with, in relation to or arising out of the use of
the Content.
This article may be used for research, teaching, and private study purposes. Any substantial or systematic
reproduction, redistribution, reselling, loan, sub-licensing, systematic supply, or distribution in any
form to anyone is expressly forbidden. Terms & Conditions of access and use can be found at http://
www.tandfonline.com/page/terms-and-conditions
 Ergonomics
Vol. 53, No. 1, January 2010, 83–91

The impact of work configuration, target angle and hand force direction on upper extremity
muscle activity during sub-maximal overhead work
Jaclyn N. Chopp, Steven L. Fischer and Clark R. Dickerson*
Department of Kinesiology, University of Waterloo, 200 University Avenue W, Waterloo, ON N2L 3G1, Canada
(Received 28 July 2008; final version received 5 September 2009)

Overhead work has established links to upper extremity discomfort and disorders. As many jobs incorporate
working overhead, this study aimed to identify working conditions requiring relatively lower muscular shoulder
load. Eleven upper extremity muscles were monitored with electromyography during laboratory simulations of
Downloaded by [University of Western Ontario] at 03:56 11 November 2014

overhead work tasks. Tasks were defined with three criteria: work configuration (fixed, stature-specific); target angle
(7158, 08, 158, 308 from vertical); direction of applied hand force (pulling backwards, pushing forwards,
downwards, sideways, upwards). Normalised electromyographic activity was greater for fixed configurations,
particularly when pulling in a backward direction (total activity ¼ 108.3% maximum voluntary exertion (MVE))
compared to pushing down or forward (total activity ranging from 10.5 to 17.3%MVE). Further, pulling backwards
at angles of –158 and 08 showed the highest muscular demand (p 5 0.05). These results suggest that, if possible,
positioning overhead work in front of the body with exertions directed forwards will result in the lowest upper
extremity muscle demand.

Statement of Relevance: Overhead work pervades occupational settings and is associated with risk of upper extremity
musculoskeletal disorders. The muscular intensity associated with performing overhead work was assessed in
several combinations of work placement and hand force direction. These findings should have utility for designing
overhead work tasks that reduce muscular exposure.
Keywords: electromyography; overhead work; work design; muscle demand

1. Introduction that awkward postures, that is, deviation from a

Upper extremity discomfort resulting from working neutral posture in extension, flexion and abduction
overhead is a common concern in the modern work- greater than 458, require a high level of muscle activity
place (Bernard 1997). As it may not be possible to (Herberts et al. 1984). Furthermore, Punnett et al.
modify or design all jobs to eliminate overhead (2000) identified an association between shoulder
working components, it may be beneficial to identify flexion and abduction greater than 908 (for 410% of
preferable work designs that minimise shoulder mus- the work cycle) and shoulder disorder development.
cular demands. This may in turn decrease shoulder Sporrong et al. (1998) earlier noted this association for
discomfort and risk of pathology. Over the past 10 overhead precision tasks. Additionally, humeral
years, an annual average of 5700 shoulder-related elevation above 608 is known to decrease the
injury claims were reported to the Workplace Safety subacromial space, potentially leading to development
and Insurance Board of Ontario (2006). Furthermore, of supraspinatus and biceps tendon impingement
Silverstein et al. (1998) examined the incidence of (Flatow et al. 1994, McFarland et al. 1999, Bey et al.
work-related upper extremity disorders in the US state 2007). Several risk factors may contribute to upper
of Washington from 1987 to 1995 and found that extremity discomfort, including task repetition, high
shoulder disorders constituted 43.2% of lost time hand force, awkward postures, direct pressure,
claims per year, with a mean of 6146 claims per year. vibration and prolonged constrained postures (Rempel
This highlights the need for more shoulder-focused et al. 1992). Although these risk factors may exist in
workplace design. the absence of elevated arm posture, their presence, in
Overhead work is strongly associated with the addition to working overhead, may place the upper
development of upper extremity discomfort and dis- extremity at elevated risk. Specifically, prolonged
orders (Bernard 1997, Punnett et al. 2000, Grieve and activity in overhead working postures creates strain
Dickerson 2008). Specifically, past research indicates and fatigue on shoulder muscles (Herberts and

*Corresponding author. Email: [email protected]

ISSN 0014-0139 print/ISSN 1366-5847 online

Ó 2010 Taylor & Francis
DOI: 10.1080/00140130903323232
http://www.informaworld.com
 84 J.N. Chopp et al.

Kadefors 1976). A strong relationship also exists working postures minimise muscle activity, encourage
between muscle fatigue and increased ratings of safer job design and assist with injury prevention in
discomfort (Oberg et al. 1994). Furthermore, postural jobs that require overhead work. The goal is to
discomfort is known to occur when the arms are determine which of the set of overhead working
required to work overhead (Wiker et al. 1989). It conditions minimises shoulder muscular load.
follows that workers performing jobs with overhead
components are at an increased risk of developing
2. Method
shoulder disorders (Svendsen et al. 2004, Miranda
et al. 2005). 2.1. Participants
Despite the strong association between overhead In total, 14 right-hand dominant, university-aged
work and musculoskeletal disorders, it is sometimes male students (aged 22 + 2 years) participated in the
difficult to avoid these postures in practice. However, study. Male participants were selected for ease of
in industrial tasks that require arm elevation above comparison with earlier published results and subject
shoulder height, flexibility in the location of the availability. Participants had a mean height and mass
task with regard to the body and the required hand of 1.85 + 0.08 m and 85.6 + 7.5 kg respectively.
Downloaded by [University of Western Ontario] at 03:56 11 November 2014

force direction may exist. Increased muscular demand Participants were excluded from participation if they
as documented through electromyography has been self-reported any upper extremity disorders within
linked to shoulder pain and discomfort (Wiker the past year. The research protocol was approved
et al. 1989). Thus, if muscle activity required to by the university ethics review board.
produce the same target force is decreased at a specific
work configuration or with a specific direction of
force application, then that condition lowers muscular 2.2. Equipment
demand and can be considered more optimal in terms Exerted hand force and muscle activation were
of lowering injury risk. measured in this study. Force was quantified using a
Working configurations, target angles and hand multiaxis loadcell (Figure 1) (MC3A; AMTI,
force directions have all been examined individually in Watertown MA, USA). Visual force feedback was
the literature; however, limited research examines how provided through a custom designed program written
these factors interact. Extensive research exists to with Labview software (National Instruments, Austin
relate specific arm angles, working heights and upper Texas, USA), which provided visual feedback to the
extremity discomfort (Flatow et al. 1994, Sporrong participant on the level of force applied to the force
et al. 1998, McFarland et al. 1999, Palmerud et al. transducer. Feedback was displayed on three bar
2000, Punnett et al. 2000, Anton et al. 2001, Garg et al. graphs corresponding to the direction of force
2002, Svendsen et al. 2004, Okunribido and Haslegrave application; forward/backward (X), upward/
2008). Further, existing research generally suggests downward (Y), medial (Z). Each graph had a 30 N
eliminating certain types of overhead work by recom- threshold line. Subjects were instructed to exert the
mending specific arm angles and postures to avoid; for specified force level on the transducer until the
example, arm flexion or abduction over 908 (Bernard
1997, Punnett et al. 2000). However, few studies have
evaluated the implications of different hand force
directions (a notable exception is Haslegrave et al.
1997) in conjunction with work height. During
sustained work, postures or force application direc-
tions that require low-level muscle activation may be
less likely to trigger the development of a musculoske-
letal disorder than those that require higher activa-
tions. Higher muscle activation may also accelerate
fatigue development, which is associated with an
increased probability of postural discomfort (Wiker
et al. 1990).
The aim of this research is to quantify shoulder
muscle activity for several overhead working config-
urations, target angles and hand force directions to Figure 1. Diagram showing the five hand force directions:
pulling backwards and pushing upwards, downwards,
determine how differences in overhead work condi- forwards and sideways (A–E). The exertion is an
tions influence specific muscle activation. The results open-handed push or pull with the palm of the hand
from this research will help in establishing which centred on the opposing face of the force transducer.
 Ergonomics 85

moving bar on the screen met the threshold line. Off- et al. 2006). It was also low enough to constitute an
axis force levels were monitored and kept below 5 N or overhead work location for the range of tested
the trial was repeated. They were given ample practice statures. In the stature scaled condition, the transducer
before commencing trials and collected force values spatial location was anthropometrically scaled to the
indicated that participants were capable of staying participant and was placed at a distance of 70% of the
within +5 N for each trial. Surface electromyographic participant’s maximum overhead reach, measured
(EMG) signals were collected using the Noraxon vertically from the greater trochanter to the centre of
T2000 telemetered system (Noraxon, Scottsdale, the palm. This distance was maintained along all target
Arizona, USA). Force and electromyography were angles (Figure 2).
measured at 1500 Hz, synchronised through the Vicon Within each of the two work configurations, four
Nexus 1.2 software (Oxford, UK). Target angles were target angles were examined: 158; 08; 158; 308. These
confirmed using a goniometer. angles were defined as the angle (with a positive
EMG signals were collected from 11 different sites, angle being a clockwise rotation) between vertically
overlying muscles on the right side of the upper upright and a line connecting the centre of the
extremity using bi-polar Ag-AgCl Noraxon dual sur- participant’s pelvis (the 3-D centre of the greater
Downloaded by [University of Western Ontario] at 03:56 11 November 2014

face electrodes with a fixed 2 cm inter-electrode trochanters) to the centre of the force transducer,
spacing (Noraxon). Specifically, electrodes were placed while the participant was seated (Figure 2). Thus, a
on the biceps brachii, triceps brachii, anterior deltoid, 08 target angle resulted in the transducer being
middle deltoid, posterior deltoid, upper trapezius, located directly overhead and positive angles creating
lower trapezius, latissimus dorsi, infraspinatus and an overhead reach in front of the body and negative
pectoralis major (both clavicular and sternal inser- angles behind the body. Within each target angle
tions) using published placements (Cram and Kasman participants exerted 30 N of force onto the force
1998). Prior to electrode placement, the skin overlaying transducer, in each of five globally defined hand
the muscle was shaved and cleansed with alcohol to force directions: pushing forward, sideways (medial),
minimise impedance. downward, upward, and pulling backward (Figure 1).
Each combination (work configuration-target
angle-hand force direction) lasted 5 s and was repeated
three times; all combinations were randomised. The
2.3. Experimental procedures
completion of the protocol entailed 120 sub-maximal
2.3.1. Experimental protocol exertions: two work configurations 6 four target
Muscle specific maximal voluntary exertions (MVEs) angles 6 five hand force directions 6 three
were performed for the 11 muscles monitored using repetitions. Participants were given a minimum of
the protocol recommended by Cram and Kasman 1 min rest between trials, with the option of more rest
(1998). Each exertion was 5 s in duration, during at their discretion.
which participants ramped up to their maximum (1 s),
sustained their maximum (3 s) and then ramped down
(1 s). Participants performed three repetitions of each 2.3.2. Data analysis
MVE in order to increase the reliability of the results Raw EMG signals were band pass filtered from
(Fischer et al. 2009). In addition, there was a minimum 10–500 Hz and differentially amplified (common-mode
of 2 min rest between each exertion to minimise the rejection ratio 4100 dB at 60 Hz, input impedance
likelihood of fatigue and promote recovery (Mathias- 100 MO) to generate maximum signal (in the range of
sen et al. 1995). the A/D board). EMG signals were A/D converted
Following instrumentation set-up and the comple- using a 16 bit A/D card with a +3.5 V range.
tion of the MVE protocol, participants were seated Following collection, EMG signals were full wave
under the force transducer. Three overhead work rectified and filtered at 6 Hz (Dickerson et al. 2008)
parameters were manipulated: working configuration; using a fourth order low pass Butterworth filter to
target angle; hand force direction. The work config- produce a linear enveloped EMG response, then
uration variable was defined by either ‘fixed’ or ‘stature normalised to MVEs outlined in the experimental
scaled’. In the fixed condition, the transducer was protocol. Within-trial means in normalised
placed at 120 cm from the stool; independent of the electromyography were calculated for each muscle.
other two variables, the height of the force transducer Then, the sum of all mean muscle activities during
remained the same for all trials. This height was chosen each combination was calculated to give an estimate of
as it reflects the maximum reach distance of a 5th total shoulder effort for each trial. The rationale for
percentile male and thus the study findings would be this calculation is the assumption that lower total
germane to the entire working population (Chaffin muscle activity may prevent potential injury and,
 86 J.N. Chopp et al.
Downloaded by [University of Western Ontario] at 03:56 11 November 2014

Figure 2. Diagram showing both the fixed (top) and stature scaled (bottom) work configurations with target angles of 308, 158,
08, –15 (left to right). In the fixed configuration condition, the height of the force transducer remains consistent at 120 cm
from the stool. In the stature-specific condition, the cube is moved vertically and horizontally to maintain the hip–hand distance
at each specified angle (note the varying vertical location of the transducer relative to the participant).

thus, total activity gives a holistic representation of [F (1, 506) ¼ 28.5], target angle [F (3, 506) ¼ 11.4] and
muscular effort. direction [F (4, 506) ¼ 289.1]. Also, two-way
A 2 6 4 6 5 repeated measures ANOVA was interaction effects were found between work
used to determine the effects of work configuration configuration and direction [F (4, 506) ¼ 19.4] and
(fixed and stature-scaled), target angle (7158, 08, 158, target angle and direction [F (12, 506) ¼ 8.7]. The two-
308) and force application direction (forward, back- way interaction between work configuration and target
ward, upward, downward, sideways) on total normal- angle and the three-way interaction between work
ised EMG activity. A p-value of 0.05 was used configuration, target angle and direction were not
to determine significance. Pairwise comparisons were found to be overall significant (Table 1). Thus,
examined post hoc using a Tukey with Bonferroni hereafter only interaction effects between work
adjustments to ensure a strict p value, and reduce the configuration and direction, and target angle and
risk of type 1 error. Effect size was calculated using direction will be discussed.
the method outlined by Cohen (1988). All statistical
analysis was done using JMP software (SAS Institute,
Cary, NC, USA). 3.1.1. Work configuration * direction interaction
A statistically significant (p 5 0.05) interaction existed
between the two work configurations (fixed, stature
3. Results scaled) and five hand force directions (backwards,
Significant influences of the tested factors, both main forwards, downwards, sideways, upwards). The
effects and interactions, were found that included all combination with the highest total normalised
three manipulated condition levels (work configura- electromyography was pulling backwards at a fixed
tion, target angle and hand force direction) for both configuration (total muscle activity ¼ 108.3%MVE)
total and individual muscle activity. (Figure 3). A distinct pattern existed for work
configuration–direction combinations. The level of
activity from highest to lowest was first dictated by
3.1. Total muscle effects direction (backward, sideways, upwards, downwards,
Work configuration, target angle and hand force forwards), but within each direction (for backward,
direction all independently and in conjunction influ- sideways and upwards) the fixed condition had
enced total muscle activity. Main effects of each of the significantly higher activity than the stature scaled.
individual factors were present in work configuration In contrast, the directions ‘downward’ and ‘forward’
 Ergonomics 87

Angle * Direction
0.59
0.86
1.35
0.29
1.12
0.30
0.63
0.31
0.54
0.60

F(12,506) ¼ 1.08
Configuration *

¼
¼
¼
¼
¼
¼
¼
¼
¼
¼
F(12,506)
F(12,506)
F(12,506)
F(12,506)
F(12,506)
F(12,506)
F(12,506)
F(12,506)
F(12,506)
F(12,506)
Angle * Direction

1.08

F(12,506) ¼ 1.18
7.42
5.56
1.81
1.95
7.29
3.07
5.11
4.27
6.65
¼
¼
¼
¼
¼
¼
¼
¼
¼
¼
F(12,506)
F(12,506)
F(12,506)
F(12,506)
F(12,506)
F(12,506)
F(12,506)
F(12,506)
F(12,506)
F(12,506)

Figure 3. Comparison of total muscle activity (11 muscles)

required to maintain 30 N of force between different work
F(4,506) ¼ 22.35

F(4,506) ¼ 21.63

F(4,506) ¼ 11.58
Configuration *

F(4,506) ¼ 2.19

F(4,506) ¼ 1.83

F(4,506) ¼ 1.75

F(4,506) ¼ 1.02
F(4,506) ¼ 5.59

F(4,506) ¼ 3.43
F(4,506) ¼ 3.86

F(4,506) ¼ 2.40
Downloaded by [University of Western Ontario] at 03:56 11 November 2014

configurations * directions of hand force interactions;

Direction

significantly different conditions (p 5 0.05) are indicated

by different letters. MVE ¼ maximum voluntary exertion.

showed no significant difference in activity between

each other for either work configuration (total
muscle activity ranged from 10.5 to 17.3%MVE for
F(3, 506) ¼ 0.86
F(3,506) ¼ 0.73)
Configuration *

F(3,506) ¼ 0.27

F(3,506) ¼ 0.26

F(3,506) ¼ 0.50
F(3,506) ¼ 0.63
F(3,506) ¼ 0.16

F(3,506) ¼ 0.89
F(3,506) ¼ 3.45

F(3,506) ¼ 2.87

F(3,506) ¼ 3.70

fixed-forward to stature-downward respectively).

ANOVA results summary of all 11 muscles demonstrating main effects and interactions.

Angle

3.1.2. Target angle * direction interaction

A statistically significant (p 5 0.05) interaction existed
between the four angles (7158, 08, 158, 308) and five
F(4,506) ¼ 227.93
F(4,506) ¼ 104.54

F(4,506) ¼ 156.11

F(4,506) ¼ 232.61

Note: F ratios shown in bold indicate the main effects and interactions that were significant.

hand force directions (backwards, forwards,

F(4,506) ¼ 90.10
F(4,506) ¼ 31.31
F(4,506) ¼ 53.56

F(4,506) ¼ 68.11

F(4,506) ¼ 17.59

F(4,506) ¼ 1.62
F(4,506) ¼ 3.00

downwards, sideways, upwards). Pulling backwards at

Direction

angles of –158 and 08 showed the highest muscular

demand and were not significantly different from one
another (Figure 4). Further, higher muscle demand
was present when pulling backwards at the two
remaining angles than for any other hand force
direction. Within each angle, backward pulling and
F(3,506) ¼ 18.46
F (3,506) ¼ 4.31

F(3,506) ¼ 2.58
F(3,506) ¼ 0.41
F(3,506) ¼ 1.25

F(3,506) ¼ 2.48
F(3,506) ¼ 1.34

F(3,506) ¼ 0.56
F(3,506) ¼ 4.72

F(3,506) ¼ 3.49
F(3,506) ¼ 9.27

sideways pushing produced higher muscle activity than

Angle

other directions. An exception to this was at a 308

target angle. At this angle, pushing upwards had
significantly higher activity (43.4%MVE) than
sideways (29.8%MVE).
F(1,506) ¼ 69.47

F(1,506) ¼ 53.87

F(1,506) ¼ 12.68
F(1,506) ¼ 2.18

F(1,506) ¼ 2.67
F(1,506) ¼ 0.42

F(1,506) ¼ 0.30

F(1,506) ¼ 1.23

F(1,506) ¼ 1.09
F(1,506) ¼ 4.07

F(1,506) ¼ 4.65
Configuration

3.2. Individual muscle contributions

Specific muscles contributed more toward achieving
30 N of hand force than others, with some having a
combination of significant main effects and
interactions (Table 1). Further, the activity elicited
from each muscle during work configuration–hand
Posterior Deltoid

force direction interactions and arm angle–hand force

Latissimus Dorsi

Lower Trapezius
Upper Trapezius
Anterior Deltoid
Middle Deltoid

direction interactions varied considerably. Specifically,

Infraspinatus

Major (C)

Major (S)

the work configuration–hand force direction

Pectoralis

Pectoralis
Table 1.

combination that resulted in the largest level of total

Muscles

Triceps
Biceps

muscle activity was the ‘fixed configuration–backward

hand force direction’ combination. In this
 88 J.N. Chopp et al.

combination, certain muscles were far more activated activity and ‘easy’ as the combination requiring the
than in the combination that resulted in the lowest lowest level of muscle activity. The remaining muscles
level of total muscle activity (‘fixed configuration– showed relatively minimal change in activity level
forward hand force direction’) (Figure 5). Anterior during different combinations.
deltoid, middle deltoid, biceps, lower trapezius and The target angle–hand force direction combination
infraspinatus had 90% higher activation during the found to require the highest activation was ‘–158 target
more ‘difficult’ work configuration–force combination angle–backward hand force direction’ and lowest
than the ‘easier’ combination; ‘difficult’ was defined as activation was ‘308 target angle–forwards hand force
the combination requiring the highest level of muscle direction’. Similar to the work configuration–hand
force interaction, anterior deltoid, middle deltoid,
biceps, lower trapezius and infraspinatus showed the
most variety between combinations, requiring the
highest activation compared to lowest activation,
with highest activation having 95% higher activity
(Figure 6).
Downloaded by [University of Western Ontario] at 03:56 11 November 2014

4. Discussion
This research quantified shoulder muscle activity for a
variety of overhead working conditions to determine
how differences in overhead work configurations,
target angles and hand force directions influence
muscular demands. The magnitude of normalised
EMG activity was greater for the fixed height
Figure 4. Comparison of total muscle activity (sum of % configurations (anthropometrics not considered) and
maximum voluntary exertion (MVE) for 11 muscles) when the target angle was –158 and when pulling
required to maintain 30 N of force between different target
angle * directions of hand force interactions; significant backwards (effect size ¼ 4.1–6.3). These results
differences between conditions (p 5 0.05) are indicated by a highlight the utility of job modification to reduce
difference in letters. shoulder muscular demands during overhead work,

Figure 5. Comparison of individual muscle activity between Figure 6. Comparison of individual muscle activity
‘work configuration * hand force direction’ combination between ‘target angle * hand force direction’ combination
requiring highest level of total muscle activity (fixed requiring highest level of total muscle activity
configuration–backward hand force direction) and (7158 target angle–backward hand force direction)
combination requiring lowest level of total muscle activity and combination requiring lowest level of total muscle
(fixed configuration–forward hand force direction) to activity (308 target angle–forward hand force direction)
maintain 30 N of force. MVE ¼ maximum voluntary to maintain 30 N of force. MVE ¼ maximum
exertion. voluntary exertion.
 Ergonomics 89

particularly in cases when overhead work is not provided contrasting results with those examining
entirely removable. other biomechanical factors. Garg et al. (2002)
examined endurance times while holding various
weights statically at varied overhead arm angles. Their
4.1. Work configuration in space research supports job design directed at minimising
Although work configuration did not influence activity the external shoulder moment, allowing for greater
to the same degree as hand force direction, the data endurance times. This is accomplished by reducing the
suggested a stature scaled (anthropometric) work- applied force at the hand or reducing the perpendicular
station can reduce muscle activity for overhead work distance between the shoulder and the applied hand
compared to a fixed working configuration (effect load. In their case, the applied hand load acted
size ¼ 0.79). This finding is particularly pronounced vertically downwards due to gravity; therefore, the
with backward or sideways hand force directions lowest perpendicular distance and resulting lowest
(Figure 3). The anthropometrically scaled condition moments occurred when the arm was either placed
eliminated some excessive reaches and brought the task parallel to the torso at the side or directly overhead
to an equal distance from the worker (Figure 2). These (08). Their argument has limited application, however,
Downloaded by [University of Western Ontario] at 03:56 11 November 2014

results coincide with Sood et al. (2007), who found that as it lacks information about specific muscles.
an increase in muscle activity is required (for anterior Although the moment created by external weights is
and middle deltoid) when the working height is moved greater when the arm is farther from an overhead
from low to high. Further research showed that position (in the case presented here at 308), this study
moving the task closer to the worker (moving up a found less total muscle activation was required in these
step on a ladder) reduced muscle activity (Anton et al. forward reaches compared to when the target was
2001). Other research has shown that by reducing placed directly overhead (08).
workstation dimensions, that is, by having a half reach The results from this study showed the target angle
instead of a full reach and lower height (elbow of –158 (158 backwards) required the most muscle
compared to shoulder), muscle activity was 3.5–3 activity (effect size ¼ 0.49–1.08), particularly when the
times lower, respectively (Habes and Grant 1997). hand force direction was also pulling backwards
The present results support this previous research, (Figure 4). Haslegrave et al. (1997) also examined the
which encourage the use of a scaled work design or a location of overhead tasks: 158 forward, 158
reduced reach distance in order to decrease muscle backwards, 158 medially and 158 laterally with regard
demands. to the vertical. Although forward and either side were
not significantly different from one another, overall,
participants had the lowest force capability in the
4.2. Angle of work from directly overhead backward direction. The results from this study
As a main effect, target angle was found to be compare with those of Haslegrave et al. (1997), as it is
significant. However, results show that target angle likely that the capacity to produce a greater force in
interacts with hand force direction, in that certain the rearward angle position is less due to these
target angles yield higher muscle activity dependent findings that there is less muscle activation in reserve
on hand force direction. Overhead arm angle has to increase force output.
been previously examined by Anton et al. (2001),
Haslegrave et al. (1997) and Garg et al. (2002). The
present findings are similar to Anton et al. (2001), 4.3. Applied hand force direction
when the current data are limited to only the upward Hand force direction was the most influential factor
push force direction at close (08) and far (308) target affecting muscle activity during the overhead working
angles (Figure 4). They only examined the effects of task examined. Most individual muscles had larger
overhead arm angle, using an upward drilling task and mean EMG recordings for the backward direction
were therefore not able to show the interactions than for the forward and downward directions
discovered in the current study when using alternate (Figures 5, 6). Further, pulling backwards had a higher
force application directions at each of the angles. These level of total activity than all other hand force
interactions are important for job design to provide directions (effect size ¼ 4.1–6.3). This may be due to
additional options for modification when the task is in some gravitational assistance in pushing both forward
a specified overhead location, but the worker is not and downward. Simply resting or leaning the hand
constrained to a single posture. on the force transducer in these directions may have
Analysis of muscle activity while reaching to a contributed to producing the 30 N of required hand
variety of overhead target angles provided insight into force (thus eliciting minimal shoulder muscle
the demand on the shoulder. This quantification demands). Conversely, in the backward pull, the
 90 J.N. Chopp et al.

muscles would need to not only contribute to both Further, lower apparent muscular requirements for a
the 30 N force, but also act to support the arm against stature-specific working condition support the
gravity. notion of population scalability in work design, a
Haslegrave et al. (1997) examined isometric stance historically advocated by ergonomists
strength capability for six different directions of force (Nussbaum 2001). Positioning the task further in front
exertion (push, pull, medial, lateral, up, down) and of the body (up to 308 forward from vertical)
found participants were most capable in the vertical required lower shoulder muscle activity than directly
direction (up or down). The results from this study overhead or behind the worker in nearly all cases.
showed that a downward push required the least Thus, working in a target angle range of 15–308 from
muscle activity. Haslegrave et al. (1997) found that directly overhead is preferred when a forward or
force capability decreased as pushes moved from downward force application is used. If an upward push
vertical to horizontal (sagittal plane) and then to is required, then working directly above the head as
lateral (frontal plane). A similar pattern was found in close to the body as possible minimises shoulder
this study, where activity was lower for those directions muscular activity, but may pose issues with
that corresponded to the greatest strength capability, line-of-sight and neck fatigue.
Downloaded by [University of Western Ontario] at 03:56 11 November 2014

with the exception of the backwards (horizontal)

direction, which required the most activity overall.
4.6. Limitations
Despite attempts to minimise the likelihood of error
4.4. Relationship to fatigue with the study design, there were certain limitations
Determining a combination of working conditions that that should be addressed. Total muscle activity
lower muscular demand may, in turn, decrease the risk was used as a surrogate for shoulder muscular load
of musculoskeletal injury. In exceeding the recom- and thus combinations resulting in higher total muscle
mended arm angles (Bernard 1997) or exerting force activity were considered ‘better’ in terms of lowering
in unfavourable directions (i.e. backwards), one may the risk of upper extremity injury. There were,
increase required muscle activity and in turn increase however, trials in which total muscle activity was lower
the probability of fatigue development (Rohmert but an individual muscle or muscles showed a greater
1973). Sood et al. (2007) stated that localised muscle response. This discrepancy would make it difficult to
fatigue is a valid concern when determining injury risk distinguish between advantageous and potentially
and, therefore, attempts should be made to minimise dangerous combinations. Additionally, it was decided
muscle activity and resulting fatigue. The present to perform the maximal exertions according to
findings dovetail well with Sood et al. (2007), as specific recommendations (Cram and Kasman, 1998).
the stature scaled work placement required less This is somewhat limited as EMG measurements are
muscle activity, which would likely be less fatiguing. influenced by local body posture. However, to perform
Further, the current results agree with recommenda- maximal exertions in each posture tested was
tions that, if overhead work is unavoidable, all prohibitive for both fatigue generation and time
attempts should be made to keep the work as close considerations. Thus, all trial values were normalised
to shoulder height as possible to reduce the likelihood by the defined maximal exertions with the
of sustaining shoulder fatigue or discomfort (Wiker understanding of this limitation.
et al. 1989).

5. Conclusion
4.5. Did any overhead work combination minimise Certain tasks or jobs intrinsically demand overhead
shoulder muscular demand? work postures. Therefore, it is important to determine
When designing an overhead work task to minimise if an overhead working condition exists that lowers
muscular demands, many factors are important. In muscular demand. The present results indicated
order of greatest to least influence in the present that muscle demands in low-intensity overhead tasks
experiment, these were: 1) force direction; 2) position- are lowest when: 1) the required hand force is directed
ing of the work interface; 3) angular work placement. either forwards or downwards; 2) the task is scaled
This study identified a benefit in terms of muscular to individual anthropometry; 3) the task is moved
demand associated with generating forward or down- closer. These findings delineate the muscular impact of
ward compared to backward applied hand forces performing overhead work across a range of
(or horizontal and up). Consequently, adjusting the conditions and may therefore be useful as a part of job
direction of applied hand force appears to yield the design decisions where shoulder muscle demand is a
largest benefits in terms of reducing muscle activity. concern.
 Ergonomics 91

McFarland, E., et al., 1999. Internal impingement of the

Acknowledgements shoulder: A clinical and arthroscopic analysis. Journal of
Support for Jaclyn Chopp and Steven Fischer during Shoulder Elbow Surgery, 8 (5), 458–460.
completion of this study came from a Hallman Research Mathiassen, S.E., Winkel, J., and Hagg, G.M., 1995.
Fellowship and an NSERC doctoral award, respectively. Normalization of surface EMG amplitude from the
Additional project support came from an NSERC discovery upper trapezius muscle in ergonomic studies – A review.
grant held by Dr Clark Dickerson. Journal of Electromyography and Kinesiology, 5 (4), 197–
226.
Miranda, H., et al., 2005. A population study on
References differences in the determinants of a specific shoulder
Anton, D., et al., 2001. The effect of overhead drilling disorder versus nonspecific shoulder pain without
position on shoulder moment and electromyography. clinical findings. American Journal of Epidemiology, 161
Ergonomics, 44 (5), 489–501. (9), 847–855.
Bernard, B., 1997. Musculoskeletal disorders and workplace Nussbaum, M.A., 2001. Static and dynamic myoelectric
factors. DHHS (NIOSH) Publication No. 97–141. measures of shoulder muscle fatigue during intermittent
Cincinnati, OH: US Department of Health and Human dynamic exertions of low to moderate intensity. European
Services, NIOSH. Journal of Applied Physiology, 85 (3–4), 299–309.
Bey, M.J., et al., 2007. In vivo measurement of subacromial Oberg, T., Sandsjo, L., and Kadefors, R., 1994. Subjective
Downloaded by [University of Western Ontario] at 03:56 11 November 2014

space width during shoulder elevation: Technique and and objective evaluation of shoulder muscle fatigue.
preliminary results in patients following unilateral rotator Ergonomics, 37 (8), 1323–1333.
cuff repair. Clinical Biomechanics, 22 (7), 767–773. Okunribido, O.O. and Haslegrave, C.M., 2008. Ready steady
Chaffin, D.B., Andersson, G.B.J., and Martin, B.J., 2006. push – a study of the role of arm posture in manual
Occupational biomechanics, 4th ed. Hoboken, New exertions. Ergonomics, 51 (2), 192–216.
Jersey: John Wiley & Sons, Inc. Punnett, L., et al., 2000. Shoulder disorders and postural
Cohen, J., 1988. Statistical power analysis for the behavioral stress in automobile assembly work. Scandinavian
sciences, 2nd edition. Hillsdale, NJ: Erlbaum. Journal of Work, Environment and Health, 26 (4), 283–
Cram, J.R. and Kasman, G.S., 1998. Introduction to surface 291.
electromyography. Gaithersburg, Maryland: Aspen 459 Rempel, D.M., Harrison, R.J., and Barnhart, S., 1992.
Publications. Work-related cumulative trauma disorders of the upper
Dickerson, C.R., Hughes, R.E., and Chaffin, D.B., 2008. extremity. Journal of the American Medical Association,
Experimental evaluation of a computational musculos- 267 (6), 838–842.
keletal model. Clinical Biomechanics, 23, 886–894. Rohmert, W., 1973. Problems in determining rest
Fischer, S.L., Belbeck, A.L., and Dickerson, C.R., 2009. The allowances. 1. Use of modern methods to evaluate
influence of providing feedback on force production and stress and strain in static muscular work. Applied
within-participant reproducibility during maximal volun- Ergonomics, 4 (2), 91–95.
tary exertions for the anterior deltoid, middle deltoid, Silverstein, B., et al., 1998. Claims incidence of work-related
and infraspinatus. Journal of Electromyography and disorders of the upper extremities: Washington State,
Kinesiology. Published online 24 February 2009. 1987 through 1995. American Journal of Public Health,
doi:10.1016/j.jelekin.2009.01.007. 88 (12), 1827–1833.
Flatow, E.L., et al., 1994. Excursion of the rotator cuff under Sood, D., Nussbaum, M.A., and Hager, K., 2007. Fatigue
the acromion: patterns of subacromial contact. American during prolonged intermittent overhead work: Reliability
Journal of Sports Medicine, 22 (6), 779–788. of measures and effects of working height. Ergonomics,
Garg, A., et al., 2002. The effect of maximum voluntary con- 50 (4), 497–513.
traction on endurance times for the shoulder girdle. Inter- Sporrong, H., et al., 1998. The effect of light manual precision
national Journal of Industrial Ergonomics, 30 (2), 103–113. work on shoulder muscles – An EMG analysis. Journal of
Grieve, J.R. and Dickerson, C.R., 2008. Overhead work: Electromyography and Kinesiology, 8 (3), 177–184.
Identification of evidence-based exposure guidelines. Svendsen, S.W., et al., 2004. Work above shoulder level and
Occupational Ergonomics, 8 (1), 53–66. degenerative alterations of the rotator cuff tendons –
Habes, D.J. and Grant, K.A., 1997. An electromyographic A magnetic resonance imaging study. Arthritis &
study of maximum torques and upper extremity Rheumatism, 50 (10), 3314–3322.
muscle activity in simulated screwdriving tasks. Interna- Wiker, S.F., Chaffin, D.B., and Langolf, G.D., 1989.
tional Journal of Industrial Ergonomics, 20 (4), 339–346. Shoulder posture and localized muscle fatigue and
Haslegrave, C.M., Tracy, M.F., and Corlett, E.N., 1997. discomfort. Ergonomics, 32 (2), 211–237.
Force exertion in awkward working postures–strength Wiker, S.F., Chaffin, D.B., and Langolf, G.D., 1990.
capability while twisting or working overhead. Ergo- Shoulder postural fatigue and discomfort. International
nomics, 40 (12), 1335–1362. Journal of Industrial Ergonomics, 5 (2), 133–146.
Herberts, P. and Kadefors, R., 1976. Study of painful Workers Safety and Insurance Board of Ontario, 2006.
shoulder in welders. Acta Orthopaedic Scandinavica, 47 Statistical supplement to the annual report 2006.
(4), 381–387. Hamilton, ON: Workers Safety and Insurance Board of
Herberts, P., et al., 1984. Shoulder pain and heavy manual Ontario.
labor. Clinical Orthopaedics and Related Research, 191,
166–178.