Content uploaded by Leonard O'Sullivan
Author content
All content in this area was uploaded by Leonard O'Sullivan on Jan 12, 2018
Content may be subject to copyright.
Shoulder muscle loading and task performance for overhead work on
ladders versus Mobile Elevated Work Platforms
Denis Phelan
a
, Leonard O’Sullivan
b
,
*
a
Jacobs Engineering, Ireland
b
Ergonomics and Product Design Research Group, Enterprise Research Centre, University of Limerick, Castletroy, Limerick, Ireland
article info
Article history:
Received 16 July 2012
Accepted 25 March 2014
Available online 24 April 2014
Keywords:
Electrical work
Overhead
Shoulder loading
abstract
A high incidence of Musculoskeletal Disorders (MSDs) has been reported in the construction sector. The
use of ladders in the workplace has long been identified as a significant risk that can lead to workplace
accidents. However, it is unclear if platform types have an effect on the physical risk factors for MSDs in
overhead work. The aim of this study is to perform a pilot study on the effects of hand activity on both
shoulder muscle loading and task performance while working on ladders versus Mobile Elevated
Working Platforms (MEWPs). It is hypothesised that work on ladders would result in greater muscle
loading demands, increased levels of discomfort, and reduced performance due to the restrictions on
postures that could be adopted. A field study (n¼19) of experienced electricians on a construction site
found that workers spent approximately 28% of their working time on ladders versus 6% on MEWPs.
However, the durations of individual tasks were higher on MEWPs (153 s) than on ladders (73 s).
Additionally, maximum levels of perceived discomfort (on a VAS 0e100) were reported for the shoulders
(27), neck (23), and lower regions of the body (22). A simulated study (n¼12) found that task perfor-
mance and discomfort were not significantly different between platform types (ladder vs. MEWP) when
completing either of three tasks: cabling, assembly and drilling. However, platform and task had sig-
nificant effects (p<0.05) on median electromyographic (EMG) activity of the anterior deltoid and upper
trapezius. EMG amplitudes were higher for the deltoid than the upper trapezius. For the deltoid, the peak
amplitudes were, on average, higher for ladder work over MEWP work for the hand intensive cabling (32
vs. 27% Maximal Voluntary Exertion (MVE)) and the assembly task (19 vs. 6% MVE). Conversely, for
drilling, the peak EMG amplitudes were marginally lower for ladder compared to the MEWP (3.9 vs. 5.1%
MVE). The general implication was that working on the MEWP involved lower shoulder muscle
loading for cabling and assembly task. A difference due to platform type was not present for drilling
work.
Ó2014 Elsevier Ltd and The Ergonomics Society. All rights reserved.
1. Introduction
Musculoskeletal Disorders (MSDs) describe a wide range of in-
flammatory and degenerative diseases and disorders which can
result in pain and functional impairment of the neck, shoulders,
elbows, forearms, wrists and hands (Buckle and Devereux, 2002).
The European Foundation for Living and Working Conditions report
that MSDs are the most common occupational disease suffered by
European workers (EU-FOUND, 2007). A recent report from the
European Survey on Working Conditions highlighted that 24.7% of
European workers complain of backache as a result of performing
work, with a further 22% complaining of muscular pains (Eurostat,
2010). An EU study on risk factors associated with MSDs concluded
that construction workers were more likely to be exposed to a
number of risk factors including work at high speeds, repetitive
hand movement, carrying heavy loads, standing or walking, painful
or tiring positions, or vibration, when compared to other sector
workers (Eurostat, 2010).
Brenner and Ahearn (2010) report data on numbers of con-
struction trade workers that retired due to ill health over a period
from 1981 to 1996 in Ireland. Sheet metal workers, floor layers and
electricians represented the more frequent occupations of retirees
under fifty years of age. During this period, it was estimated that
24,428 years of working lives were lost due to premature
*Corresponding author. Tel.: þ353 61 234249; fax: þ353 61 202913.
E-mail address: Leonard.osullivan@ul.ie (L. O’Sullivan).
Contents lists available at ScienceDirect
Applied Ergonomics
journal homepage: www.elsevier.com/locate/apergo
http://dx.doi.org/10.1016/j.apergo.2014.03.007
0003-6870/Ó2014 Elsevier Ltd and The Ergonomics Society. All rights reserved.
Applied Ergonomics 45 (2014) 1384e1391
retirement from the industry. However, the report only considers
employees who are members of the regulated environment of the
Irish Construction Industry Federation pension scheme.
Anderson (1988) details trade worker activities that may be
associated with MSDs. Many of these activities include typical
construction activities such as sanding, grinding, hammering,
carpentry, overhead work, turning screws, soldering, welding, use
of hand tools, wiring, use of pliers, polishing, sawing, operating
finger triggers, jack hammering, use of vibrating tools, and working
in a cold environment. It is not surprising that the incidence of
MSDs in this industry is particularly high (Schneider, 1997). Albers
et al. (2004), in a study of ergonomic interventions to reduce
musculoskeletal loading in building installation tasks, identified a
number of high risk tasks associated with the electrical and me-
chanical trades. These included pulling cables/wires, attaching
raceway to ceilings, positioning fixtures, and connecting wires.
However, no reference is made to the variety and effects of the
many access platforms used in the industry.
The use of ladders in the workplace has long been identified as a
significant risk factor leading to workplace accidents. The safety
concerns relating to ladders have been well documented and
include the absence of fall protection for the user and the difficulty
in securing fall restraints in the event of a fall. In the US, it was
reported that 133 fatal ladder related falls occurred in 2004, with
ladders accounting for 16% of workplace fall related fatalities
(Lombard et al., 2011). Previous ergonomic studies on ladders have
focused on climbing/handling of ladders (Bloswick and Chaffin,
1990; Imbeau et al., 1998; Hoozemans et al., 2005). Ladder
handling was identified as a significant risk hazard for MSDs within
the telecommunication sector (Imbeau et al., 1998). This included
loading/unloading of ladders from vehicles and the positioning of
ladders onto the shoulder. Overexertion was considered a risk
factor in the handling of ladders (24e31 kgs). However, no refer-
ence is made to ergonomic risks of work on ladders. Bloswick and
Chaffin (1990) examined the ergonomic implications of ladder
climbing activities using EMG of the erector spinae muscle group
with biomechanical modelling of compressive and shear forces at
L5/S1. They concluded that fast climbing resulted in 35% greater
EMG activity than slow climbing for the torso muscles, and that
erector spinae EMG activity almost reached the maximum during a
fast climb, but no data were presented on shoulder muscle activity.
Hoozemans et al. (2005) examined the effect of differing rung
separation on perceived exertion, discomfort, safety and mechan-
ical loading of the lower joints during ladder ascending and
descending, but the study did not include the effects of performing
tasks using the ladder as a platform.
The stepladder is a very common feature on the majority of
construction sites in Ireland, used by craftspersons when per-
forming work at height. The popularity of the stepladder is likely to
be influenced by its versatility and low relative economic cost. The
use of Mobile Elevated Work Platforms (MEWPs) on construction
sites is also very common. MEWPs are self propelled machines that
are capable of raising a working platform to the desired working
height. Irish health and safety legislation (HSA, 2007a) recognises
that work at height can be performed safely using a wide range of
work equipment, but guidance favours the use of MEWPs over
ladders (HSA, 2007b). On MEWPs, the risk of an operator falling is
minimised as the operator is contained within a double handrail.
Additionally, in the event of a fall, the operator is secured if a body
harness is worn. It is clear that MEWPs are a preferred method for
accessing work at height compared to ladders due to their safety
features in preventing falls from height. However, it is unclear if
MEWPs are also preferable in preventing MSDs during overhead
work. A review of the literature indicated no previous studies
comparing the use of MEWPs to ladders on risks of MSDs for
overhead work. The MEWP provides a working platform, typically
about 1.0 2.0 m, thus allowing the user a stable base to move
while performing construction tasks. Additionally, the height of the
platform can be adjusted with ease as per the user’s preference.
The purpose of this study was to compare work on a ladder
versus a MEWP for three different tasks on shoulder muscle loading
and discomfort, and also task performance. Part I reports on plat-
form type usage on a commercial construction site at a point in
time. Part II was a simulated study of commercial electrical work to
compare the effect of working on ladders versus MEWPs on
shoulder EMG, discomfort, and task performance. The hypothesis of
this study was that work on ladders would lead to greater muscle
loading demands, increased discomfort and lower levels of per-
formance than on a MEWP, due to both the posture and balance
constraints required for ladder work and the stable platform and
reduced balance requirements on MEWPs.
2. Method
The research methods were approved by the University of
Limerick Research Ethics Committee.
2.1. Part I: field study of discomfort
The survey was performed at a construction site of a large
pharmaceutical plant in Ireland. Site management approved the
study. Labour levels peaked at approximately 400 personnel with
up to 60 electricians at any one time. The electricians on site (all
male) were presented with details of the survey during their
morning meetings and invited to participate. Twenty electricians
agreed to participate in the one day study from a workforce of 30 on
the day of sampling. One had to leave the site during the day giving
a sample size of 19.
The survey comprised a simple checklist of broad work activities
and location by time segment (15 min) which the participants
completed as the day progressed. The work activities on the
checklist were as follows:
Working on ladder
Working on a MEWP
Working on other access platform
Electrical work from ground
Completing safety documentation/retrieving materials
Other (please state)
A modified version of the Corlett and Bishop (1976) body part
discomfort rating method was used to record perceived discomfort
at the end of the working day in the neck, shoulders, upper arms,
lower arms, lower trunk and upper trunk. The main difference was
the use of a 100 mm visual analogue scale with anchors of “no
discomfort”on the left and “extreme discomfort”on the right, as
per Carey and Gallwey (2002, 2005).
2.2. Part II EMG study of overhead work
2.2.1. Treatment details and experimental design
Twelve qualified electricians (all male) with a minimum of four
years work experience in the company volunteered to participate in
the experiment, none of which participated in Part I of this study.
The mean age was 43 yrs (SD 6.11), mean body mass 88.8 kg (SD
5.80) and stature 1.81 m (SD 0.05). There were no female electri-
cians available in the company at the time of the study to partici-
pate in the experiment.
The experimental design was full factorial. The independent
variables were Platform (2 levels: MEWP and ladder) and Task (3
D. Phelan, L. O’Sullivan / Applied Ergonomics 45 (2014) 1384e1391 1385
levels: drilling, cabling and assembly) combined giving six treat-
ments. The dependent variables were discomfort, task performance
and EMG amplitude during the task (50th and 90th percentile
MVE). Orders were balanced using a 6 6 Latin Square
(Montgomery, 1991). Participants 7e12 did the reverse treatment
orders of the first six participants.
The treatments were simulated tasks comprising combinations
of three electrical construction tasks (drilling, cabling and assem-
bly), with both access platforms (ladders and MEWP) in a workshop
environment. The tasks selected were identified as routine elec-
trical activities during the field study. Participants were instructed
to perform each task for a period of 5 min followed by a rest period
of 5 min between tasks, as per Carey and Gallwey (2002, 2005) and
Mukhopadhyay et al. (2007). In order to represent the actual
practice as performed on construction sites, the participants were
instructed to set the ladder/MEWP position to suit their individual
preference. Fig. 1 is a photograph of a participant on the MEWP.
2.2.1.1. Drilling task. The drilling task required participants to drill
a 100 mm medium density concrete block using an electric drill and
a 14 mm masonry drill bit. Performance was measured from the
amount drilled in mm. A new concrete block was used for each
participant.
2.2.1.2. Cabling task. The cable task involved feeding a 10 mm
diameter power cable through a series of brackets mounted on a
wall. The length of cable pulled (metres) during 5 min was the
measure of performance.
2.2.1.3. Assembly task. A 10 mm threaded bar was fixed to the wall
for the assembly task. Participants used their dominant hand to
move a nut from the outside edge of the bar towards the wall. The
distance each nut was threaded (mm) during the 5 min was the
measure of performance.
2.2.2. EMG equipment and data reduction
Muscle activity from the upper trapezius and anterior deltoid
muscles was measured using bipolar surface EMG. A Nexus 10
portable EMG system (Mind Media) with Bluetooth connectivity
recorded the EMG signals. The sampling frequency was set at
1024 Hz with a band pass filter set to 20e500 Hz in the software.
The sampling window was set at 0.5 s (512 samples) with a 25%
overlap. The Root Mean Square (RMS) calculation was applied to
the raw EMG to determine signal amplitude.
Maximal Voluntary Exertions (MVEs) were performed for both
muscles at the start of the experiment and signal amplitudes from
these recordings were used to convert the signals from experi-
mental treatments to percentages of MVE. Percentiles (50th and
90th) of the MVE data from the Amplitude Probability Distribution
Function (APDF) were determined for each recording, as per
Vasseljen and Westgaard (1995). The 50th percentile is indicative of
the median EMG activity amplitude during the treatment whereas
the 90th percentile is indicative of peak EMG activity.
2.2.3. Procedure
The participants completed the informed consent form and
were briefed on the nature of the experiment and what it involved.
The EMG electrodes were attached over the upper trapezius and
anterior deltoid muscles on their dominant side. The guidance of
SENIAM (1999) was followed in preparing the skin and the guid-
ance of Delagi et al. (1980) was followed in positioning the elec-
trodes. The signal quality was tested and EMG signals were
recorded for MVE exertions of both muscles. The participants
familiarised themselves with the tasks prior to commencement,
including operation of the MEWP and positioning of the platform as
per the task. Participants self selected their preferred working
height on both platforms. It was necessary to control the pace of the
simulated work as it is known to affect discomfort for the upper
limb in simulated tasks (Finneran and O’Sullivan, 2013). As such,
the participants in repetitive tasks were instructed to “perform the
task at a pace they felt they could maintain for an 8 h day”(as per
Finneran and O’Sullivan, 2010). This was expected toapproximately
result in an activity equating to a Methods-Time Measurement
(MTM) pace rating of 100%.
Each treatment lasted 5 min, followed by a 5 min break. At the
end of each treatment discomfort was rated using the same scale as
in Part I above. Performance was measured using a tape measure
for the drilling and assembly task, and a measuring wheel for the
cabling task.
2.2.4. Statistical analysis
The Statistical Package for the Social Sciences (SPSS V16) was
used for the statistical analysis. A paired sample t-test was used to
compare average cycle times on ladders versus MEWPs for the field
study data. In the simulated study the discomfort data were not
normally distributed so the Wilcoxan Signed Ranks test was used to
Fig. 1. Photo of a participant on the MEWP cabling.
Fig. 2. Mean discomfort (þ1SD) by body region.
D. Phelan, L. O’Sullivan / Applied Ergonomics 45 (2014) 1384e13911386
test effects for Platform (2 levels) and the Friedman test used to test
effects of Task (3 levels). A two way ANOVA (Platform Participant)
was used to analyse the performance data for each task separately.
ANOVA was used to analyse the median and peak percentile MVE
data (recorded from the treatments) for both the deltoid and upper
trapezius data. Hence there were four separate analyses: two per
muscle. The ANOVA main effects were Platform, Task and Partici-
pant. Due to insufficient degrees of freedom, the three way inter-
action was not included in the analysis.
3. Results
3.1. Part I discomfort and work activities
As illustrated in Fig. 2, a number of body regions had high
average levels of discomfort. The neck, shoulders and lower trunk
values were in excess of 20 (on the 0e100 scale). Highest discom-
fort was recorded for the shoulders (27.9), with the upper trunk
(7.6) recording least discomfort. Mean levels of discomfort were
almost 50% greater for the right shoulder (16.6) than for the left
(11.3). Additionally, discomfort was greater for the right arm and leg
(58.5) versus the left arm and leg (23.7).
A breakdown of the work activity and work locations performed
by the electricians is shown in Table 1. The average length of time
on MEWPs was 153.3s versus 73.7 s for ladders. Hence the duration
of tasks on MEWPs was on average more than double the time
spent on ladders on any one instance but ladders were used more
frequently. These differences were statistically significant (paired
sample t-test p<0.05).
Generally, the data indicate that the electricians performed
approximately 37% of their typical day working at height. On
average, 28% of the work day was spent working on ladders, with
6% on MEWP’s. Over 35% of each electrician’s working day was
spent completing electrical work on the ground. The data also
showed that 3% of their time involved using other access platforms
excluding MEWPs and ladders. This would include fixed scaffolds,
mobile aluminium towers and hop-ups.
3.2. Part II
3.2.1. Effects of platform and task on discomfort
Mean discomfort for the upper body was slightly higher for
ladders (25.2) than for MEWP (24.3). These differences were not
significant (Wilcoxon signed rank p¼0.693). Ratings of upper body
discomfort were significantly different between tasks (Friedman
test, p<0.001). The mean ranks showed least discomfort for the
assembly task (1.4), compared to the drilling (2.3) and cabling (2.3).
3.2.2. Task performance data
ANOVA performed on the data showed that Platform did not
have a significant effect on task performance for the drilling, ca-
bling and assembly tasks (p¼0.052, p¼0.070 & p¼0.070
respectively). While not significant, the majority of participants
were better able to perform the drilling task on the ladders than the
MEWP.
3.2.3. Effects of platform and task on EMG data: anterior deltoid
The log calculation was applied to the EMG data to transform it
to a normal distribution. Average raw (R) and log transformed (T)
EMG activity data for the anterior deltoid are show in Table 2.
ANOVA performed on the median percentile EMG activity data for
the anterior deltoid, identified significant effects for Task
(p<0.001), Platform (p¼0.007), Participant (p¼0.02) and for a
two way interaction between Task Platform (p¼0.004). For the
peak percentile EMG activity data the same factors were significant
and at similar levels, with one exception which was Platform
(p¼0.07).
Fig. 3 shows a plot of the average transformed median EMG data
where anterior deltoid activity was highest overall for the cabling
task on both platforms, with values higher for Ladder than MEWP
(Ladder T0.93, R 8.51%, MEWP T0.78, R6.03%). Assembly task
muscle activity was also higher for Ladder than MEWP (Ladder
T0.86, R7.24%, MEWP 0.21, R 1.65%). Across all conditions, the
treatment with the lowest overall activity was for the assembly task
on the MEWP (0.21, R 1.65%). The EMG activity data for drilling
followed an opposite trend as for cabling and the assembly task
with muscle activity slightly higher for work on a MEWP (T0.42,
R2.63%) than on a Ladder (T0.35, 2.24%). This illustrates the sig-
nificant two way interaction for Task Platform (p¼0.004).
For the peak percentile data (Fig. 4), the same general trend was
present as for the median values. The data were highest for cabling,
although the differences between MEWP and Ladder were less than
for the assembly task. Again drilling followed an opposite trend
with values lower for Ladder than MEWP. Overall the lowest values
were for drilling on a ladder.
3.2.4. Effects of platform and task on EMG data: upper trapezius
ANOVA on the upper trapezius median percentile EMG data
indicated significant effects for Participant (p¼0.005), Platform
(p¼0.047) and Task (p<0.05). Median EMG activity was higher for
all tasks completed on ladders compared to MEWPs. Interactions
between Platform Participant, Task Platform, &
Task Participant were not significant (P>0.05). Table 3 shows
higher EMG activity for the cabling task compared to the other
tasks for all participants.
ANOVA on the peak EMG data revealed that Task was significant
(p<0.01) but that Platform was not (P¼0.526). Participant was
significant (p<0.05). No significant interactions between
Participant Task (p¼0.404), and Participant Platform
(p¼0.680) were recorded.
Median and Peak EMG activity (Figs. 5 and 6) were higher for
cabling and assembly tasks performed for ladders, compared to
MEWP. However, drilling recorded lower peak EMG activity for
ladder than MEWP .
4. Discussion
4.1. Working platform types and durations
Approximately 37% of the electricians work activities surveyed
on site were performed above ground level on some type of access
platform with overall duration of work on ladders (28%) being most
common. Work on MEWPs accounted for only 6% of time in this
study. Individual cycle times were longer on MEWPs (153 s) than on
ladders (73 s) but ultimately the electricians used the ladders more
frequently. The extensive requirement to perform work at height
can be expected for electrical installation work as many services are
installed on ceilings and walls. It was noted that the use of ladders
Table 1
Percentage time on working platforms and other activities (n¼19).
Activity Mean Std deviation
Working on a ladder 27.9% 13.5
Working on a MEWP 5.8% 8.5
Working on other access platform 3.3% 7.0
Working on ground 37.0% 14.9
Completing safety documentation 10.4% 3.4
Other (retrieving materials etc) 15.6% 7.5
D. Phelan, L. O’Sullivan / Applied Ergonomics 45 (2014) 1384e1391 1387
varies during the lifecycle of a construction project. It is the expe-
rience of one of the current authors (DP) from working on com-
mercial construction sites, that as a project nears completion, it is
often difficult to use MEWPs due to reduced access, particularly
when partitions are erected. It is also the experience that ladders
are often favoured during the latter stages of projects because they
are less likely to cause damage to the finished structure.
The average length of task duration on ladders was on average
half that of work on MEWPs. The general inference is that often for
short duration activities, the ladder was preferred for short dura-
tion tasks over using a MEWP. Other factors that may have influ-
enced the access method used include availability, training
requirements, access difficulties, and costs. In practice, when an
electrician was required to perform a task at height, the decision to
use a MEWP required that it will be available, that the operator had
adequate training, and that the MEWP could gain access to where it
was required. The cost difference between platform types is sub-
stantial with ladders costing V200e300 and MEWPs up to several
thousand Euros.
NIOSH (1997) highlights the strong evidence of a relationship
between posture and MSDs of the neck and neck/shoulder region.
Higher incidence of shoulder elevations was likely to be more
evident on ladders as the worker commenced work straight away
once they ascended the ladder. Ladder work required the electri-
cian to carry all materials or tools. On MEWP work, materials can be
carried in the MEWP allowing the worker more time to assess the
task. MEWP work also allowed the user to raise the working height
above the task height thus reducing the incidence of neck
extension. During the field study, it was noted that workers were
more likely to look down on a task when in a MEWP.
This survey approach involved workers logging their platform
use throughout the working day. A more precise way to estimate
the frequencies would be to use systematic work-study sampling
methods by an independent observer. But the construction site
used in this case study was very large, making it particularly diffi-
cult to locate workers. As such, work-study sampling was consid-
ered unpractical.
4.2. EMG and discomfort study of overhead work on ladders and
MEWPs
4.2.1. Platform and task effects on upper limb discomfort
It was expected that the use of ladders would result in greater
levels of discomfort and reduced performance due to the restrictive
posture associated with ladder use. MEWP platforms provide for
greater flexibility allowing the user to adopt a variety of postures to
suit the task being performed. However, the results indicated that
platform type did not have a significant effect on upper limb
discomfort in this study. It was anticipated that the MEWP platform
would provide a more stable base to allow the user to perform the
work in a variety of positions, as advocated by Kroemer (2009).
While Platform did not have a significant effect on the discomfort
data, it did on the median EMG data. The effects were more pro-
nounced for the Deltoid. If the treatments were of a longer duration
than those tested (5 min), then fatigue differences and increasing
Table 2
Average raw and transformed % MVE data (50th & 90th percentiles) for the Anterior Deltoid.
Raw Log transformed
50th %ile 90th %ile 50th %ile 90th %ile
Ladders MEWP Ladders MEWP Ladders MEWP Ladders MEWP
Drilling task 2.24 2.63 3.98 5.13 0.35 0.42 0.6 0.71
Cable pulling task 8.51 6.03 32.36 27.54 0.93 0.78 1.51 1.44
Assembly task 7.24 1.62 19.05 6.17 0.86 0.21 1.28 0.79
Mean 6.00 3.43 18.47 12.95 0.71 0.47 1.13 0.98
Fig. 3. Transformed average 50th percentile EMG activity for the Anterior Deltoid for
Platform versus Task. Task, Platform and Task Platform were significant at min
p<0.01.
Fig. 4. Transformed average 90th percentile EMG activity for the Anterior Deltoid for
Platform versus Task. Task and Task Platform were significant at p<0.01, Platform
was not significant p¼0.07.
D. Phelan, L. O’Sullivan / Applied Ergonomics 45 (2014) 1384e13911388
discomfort may have been more pronounced. However, it is not
possible to confirm this in the absence of a trial with more par-
ticipants and longer duration treatments.
Task had a significant effect on upper limb discomfort.
Discomfort ratings were highest for the cabling task followed by
the drilling and assembly tasks. This could be expected given the
differing physical demands of the tasks. The cabling task required
vigorous movement of the arms to pull the cable, an explanation
supported by the higher muscle activity levels for the cabling task.
Methta and Agnew (2010) performed a laboratory study on the
effects of a drilling task on muscle activity. It found that fatigue was
shown to influence discomfort ratings for all body parts measured.
This included the hand, lower and upper arm, and upper and lower
back.
4.2.2. Platform effects on task performance
It was expected that the MEWP would allow the participant to
adopt a wider range of postures to suit the physical demands of
each task, thus leading to higher task performance. However,
Platform did not have a significant effect on performance. While not
significant, performance improved when on ladders. This indica-
tion was counter to what was expected. One explanation is that
participants felt more stable and comfortable on the MEWP to take
micro breaks between work cycles, whereas on the ladder they
were focused on completing the task so as to return to ground level.
For the drilling task, the result approached significance (p¼0.052).
If the task was of a longer duration or if more participants were
included then this result may have been significant.
Previous reports on task performance and muscle fatigue show
conflicting findings. Methta and Agnew (2010) found that task
performance in a simulated drilling task in a laboratory was not
influenced by task difficulty. The report compared a fatigue and a
non fatigue condition, and while the report did not observe any
effects of task difficultly on performance, the authors acknowledge
the level of difficulty may have not been sensitive to changes in
fatigue.
4.2.3. Platform and task effects on muscle activity
The Platform main effect was significant for median EMG ac-
tivity for both the anterior deltoid and upper trapezius. Generally,
shoulder muscle activity was greater on ladders compared to
MEWPs and as such the results would suggest that ladder use
required greater levels of force and thus increased biomechanical
demands on the user. While NIOSH (1997) concluded that there
was insufficient evidence to link force of exertion on its own and
MSDs of the shoulder, they also note that high shoulder muscle
requirement can increase muscle contraction activity, which may
lead to an increase in both muscle fatigue and tendon tension, and
impair microcirculation.
Veiersted et al. (2013) studied full day EMG recordings of the
upper trapezius which included a group of electricians (n¼14).
They reported mean 50th percentile APDF values of 5%MVE and
90th percentile APDF of 15.4% MVE which are comparable to the
values from the mean values across the three tasks in the current
study (50th percentile: 4.8% ladders, 6.8% MEWP, 90th percentile:
Ladders 8.7, MEWP 9.8). But the 90th percentile ADPF values were
Table 3
Average raw and transformed % MVE data (50
th
&90
th
percentiles) for the Upper Trapezius.
Raw Log transformed
50th %ile 90th %ile 50th %ile 90th %ile
Ladders MEWP Ladders MEWP Ladders MEWP Ladders MEWP
Drilling task 4.17 4.68 6.92 5.01 0.62 0.67 0.84 0.7
Cable pulling task 7.08 9.33 12.88 15.14 0.85 0.97 1.11 1.18
Assembly task 3.16 6.31 6.17 9.12 0.5 0.8 0.79 0.96
Mean 4.8 6.8 8.7 9.8 0.7 0.8 0.9 0.9
Fig. 5. Transformed average 50th percentile EMG activity for the Upper Trapezius for
Platform versus Task. Platform and Task were significant at p<0.05).
Fig. 6. Transformed average 90
th
percentile EMG activity for the Upper Trapezius for
Platform versus Task. Task was significant (p<0.01) but Platform was not (P¼0.5).
D. Phelan, L. O’Sullivan / Applied Ergonomics 45 (2014) 1384e1391 1389
highest for cabling on the MEWP (15.1% MVE) is very similar to the
same parameter reported for the electricians by Veiersted et al.
(2013).
This study has demonstrated that higher upper trapezius and
deltoid muscle activity was present when performing overhead
work on ladders compared to MEWPs. However, this study only
considered two muscles and they were studied separately. Previous
studies indicate patterns of redistribution of forces in shoulder
muscles due to fatigue or repetitive exertions. This may result in
recruitment of other muscles (Cote et al., 2002; Selen et al., 2007)in
the exertion. Antony and Keir (2009) concluded that performing a
simultaneous shoulder exertion and hand grip led to posture spe-
cific redistribution of shoulder muscle activity for both isometric
and dynamic conditions. The effects of hand activity, specifically
the use of tools, have been shown to influence shoulder muscle
activity. Antony and Keir (2009) examined the effects of loading on
shoulder muscle activity and concluded that a 0.5 kg load increased
shoulder muscle activity by 4% MVE. Sporrong et al. (1998) reported
an average increase in muscle activity of 22% of the resting state
when light precision work was introduced to a task. EMG activity
was higher for the cabling and assembly tasks on ladders but the
drilling task had lower peak EMG activity even though participants
had to hold and control the tool. In this experiment, the tasks
required low to moderate levels of precision. In practice, some work
by electricians, such as connecting wires in components, requires
high levels of precision. If high precision tasks are performed on a
ladder rather than a MEWP this may increase the differences in
shoulder muscle activity between the two platform types. A sepa-
rate study is necessary to verify this.
The interaction between Platform and Task was similar across
both muscles. Drilling resulted in lower peak EMG activity for both
muscles for ladder than MEWP. As such, platform could be
considered as having a marginal effect on the EMG data overall for
the drilling task. This may be due to the overall static nature of
drilling once the task commences. This was not the case for the
cabling and assembly tasks. As previously mentioned, these tasks
involved performing more dynamic movements. The assembly task
provided the greatest contrast with the largest increase in EMG
levels when completed on ladders. The significant effect for par-
ticipants is to be expected given individual strength differences.
4.3. Limitations
Part I of this study involved a survey of a single construction site
during a specific phase of the construction. As such, the results
indicate the discomfort and work practice by those workers at that
phase of the construction project only. Future research studies
should repeat this survey on other construction sites throughout
the lifecycle of the project to more accurately capture the work
practices, especially the ladder versus MEWP use patterns. The
actual postures used by the workers to perform their work were
also not recorded during the field study and this is also a limitation
of this work.
The sample size in the experiment (Part II) was relatively small
(n¼12) and this limits the generalisability of the results. It is
necessary to test a larger number of participants and for longer
durations to confirm if these findings would be representative of
the larger populations of electricians working for full day durations
on site.
The treatments in Part II were of a relatively short duration.
Others have completed laboratory based discomfort studies of
simulated tasks to study wrist, forearm and shoulder posture ef-
fects (Carey and Gallwey,1999,2002; O’Sullivan and Gallwey, 2005;
Mukhopadhyay et al., 2007; Khan et al., 2009). Further research
studies are necessary performing these types of simulated tasks for
longer durations than 5 min to ensure the findings reflect the actual
cumulative fatigue effects as experienced in the course of a working
day. Solutions include repeating the treatments as performed here
but for longer durations (up to 8 h with rest breaks) or performing
site surveys using posture analysis with EMG and discomfort
studies with workers performing individual tasks similar to the
treatments studied here for long durations.
Many upper limb muscles are active in performing the exertions
involved in the experiment, yet in this study, only two were eval-
uated. As such, the results only reflect loading on these two specific
muscles. This is a further limitation of the study.
5. Conclusions
Part I Field Study
High levels of perceived discomfort in the neck, shoulders and
lower body, were reported by the small group of electricians
surveyed.
Thirty seven percent of the electricians’days were spent work-
ing at height, using ladders, MEWPs or other access platforms.
Ladders were used much more extensively than MEWPs in this
study at 28% compared to 6% respectively.
Average cycle time durations were significantly longer on
MEWPs at 153 s compared to 74 s for ladders, but the electri-
cians used the ladders a lot more frequently than the MEWPs. A
larger survey involving a greater number of participants across
multiple sites is necessary to more accurately reflect the actual
practice of platform use in the construction sector.
Part II: Simulated Study
Platform type did not have a significant effect on performance or
on perceived discomfort. These findings must be taken in
context of the short duration treatments (5 min) and the small
sample size (n¼12).
Anterior deltoid and upper trapezius median (50th percentile)
EMG activity were significantly affected by Platform and Task,
and the Task by Platform two way interaction (each p<0.05).
For the peak (90th) EMG data, Task was significant but Platform
was not. The main finding was that muscle activity (50th and
90th percentile EMG data) was higher for hand intensive work
(cabling and assembly) on the ladder versus MEWP. But for
drilling, the only power tool treatment, the differences in plat-
form effects were considerably lower, and sometimes reversed
(compared to cabling and assembly) suggesting ladder work to
be more suitable.
Working on a MEWP was not found to increase task perfor-
mance or lower discomfort in this study. But the EMG data show
that MEWP work involved lower shoulder muscle loading for
the cabling and assembly task in particular. It is necessary to
study these conditions in longer duration treatments to better
understand fatigue effects as they would apply in an actual
workplace.
References
Albers, J., Estill, C., MacDonald, L., 2004. Identification of ergonomic interventions
used to reduce musculoskeletal loading for building installation tasks. Appl.
Ergon. 36, 427e439.
Anderson, V.P., 1988. Cumulative Trauma Disorders: a Manual for Musculoskeletal
Disease of the Upper Limbs. Taylor and Francis, London.
Antony, N., Keir, P., 2009. Effects of posture, movement and hand load on shoulder
muscle activity. J. Electromyogr. Kinesiol. 20, 191e198 .
Bloswick, D.S., Chaffin, D.B., 1990. An ergonomic analysis of the ladder climbing
activity. Int. J. Indus. Ergon. 6, 17e27.
Brenner, B., Ahearn, A., 2010. Sickness absence and early retirement on health
grounds in the construction industry in Ireland. Occup. Environ. Med. 57, 615e
616.
D. Phelan, L. O’Sullivan / Applied Ergonomics 45 (2014) 1384e13911390
Buckle, W.P., Devereux, J.J., 2002. The nature of work-related neck and upper limb
musculoskeletal disorders. Appl. Ergon. 33, 207e217.
Carey, E.J., Gallwey, T.J., 1999. Discomfort prediction from postural deviations of the
wrist. In: Hanson, M.A., Lovesey, E.J., Robertson, S.A. (Eds.), Contemporary Er-
gonomics., Taylor & Francis, London, pp. 296e300.
Carey, E.J., Gallwey, T.J., 2002. Effects of wrist posture, pace and exertion on
discomfort. Int. J. Indus. Ergon. 29, 85e94.
Carey, E.J., Gallwey, T.J., 2005. Wrist discomfort levels for combined movements at
constant force and repetition rate. Ergonomics 48, 171e186.
Corlett, E.N., Bishop, R.P., 1976. A technique for assessing postural discomfort. Er-
gonomics 19, 175e182.
Cote, J.N., Mathieu, P.A., Levin, M.F., Feldman, A.G., 2002. Movement reorganisation
to compensate for fatigue during sawing. Exp. Brain Res. 146, 394e398.
Delagi, E.F., Perotto, A., Iazzetti, J., Morrison, D., 1980. Anatomical Guide for the
Electromyographer. Charles C Thomas, New York.
EU FOUND (European Foundation for the Improvement of Living and Working Con-
ditions), 2007. Managing Musculoskeletal Disorders. http://www.eurofound.
europa.eu/ewco/studies/tn0611018s/tn0 611018s.htm (accessed 07.01.14.).
Eurostat, 2010. Statistics in Focus Population and Social Conditions. http://epp.
eurostat.ec.europa.eu/cache/ITY_OFFPUB/KS-SF-09-063/EN/KS-SF-09-063-EN.
PDF (accessed 07.01.14.).
Finneran, A., O’Sullivan, L.W., 2010. On the effects of physical risk factor induced
discomfort on productivity in a simulated repetitive task. Int. J. Indus. Ergon. 40,
257e266.
Finneran, A., O’Sullivan, L.W., 2013. Effects of grip type and wrist posture on forearm
EMG activity, endurance time, and movement accuracy. Int. J. Indus. Ergon. 43,
91e99.
Health and Safety Authority, 2007a. Guide to the Safety, Health and Welfare at
Work (General Application) Regulations, 2007 Part 4, Work at Height. Health
and Safety Authority The Metropolitan Building, James Joyce Street, Dublin 1,
Ireland.
Health and Safety Authority, 2007b. Using Ladders Safely. http://www.hsa.ie/eng/
Publications_and_Forms/Publications/Safety_and_Health_Management/Using_
Ladders_Safely_Information_Sheet.html (accessed 07.01.14.).
Hoozemans, M.J.M., de Looze, M.P., Kingma, I., Reijneveld, K.C.N., de Korte, E.M., van
der Grinten, M.P., van Dieën, J.P., 2005. Workload of window cleaners using
ladders differing in rung separation. Appl. Ergon. 36, 275e282.
Imbeau, D., Montpetit, Y., Desjardins, L., Riel, P., Allan, J.D., 1998. Handling of
fibreglass extension ladders in the work of telephone technicians. Int. J. Indus.
Ergon. 22, 177e194.
Khan, A.A., O’Sullivan, L., Gallwey, T.J., 2009. Effects of combined wrist deviation and
forearm rotation on discomfort score. Ergonomics 52, 345e361.
Kroemer, K., 2009. Fitting the Human, Introduction to Ergonomics, sixth ed. CRC
Press, London.
Lombard, D., Smith, G., Courtney, T., Brennan, M., Kim, J., Perry, M., 2011. Work-
related Falls from ladders-a follow up study of US Emergency department cases.
Scand. J. Work Environ. Health 37, 525e532.
Methta, R.K., Agnew, M.J., 2010. Analysis of individual and occupational risk factors
on task performance and biomechanical demands for a simulated drilling task.
Int. J. Indus. Ergon. 40, 584e591.
Montgomery, D.C., 1991. Design and Analysis of Experiments. Wiley, New York.
Mukhopadhyay, P., O’Sullivan, L., Gallwey, T.J., 2007. Estimating upper limb
discomfort level due to intermittent isometric pronation torque with various
combinations of elbow angles, forearm rotation angles, force and frequency
with upper arm at 90
0
abduction. Int. J. Indus. Ergon. 37, 313e325.
National Institute for Occupational Safety and Health (NIOSH), 1997. Musculoskel-
etal Disorders and Workplace Factors: a Critical Review of Epidemiological
Factors for Work-related Musculoskeletal Disorders of the Neck, Upper Ex-
tremity, and Low Back. Department of Health and Human Services (DHHS),
Cincinnati, U.S.
O’Sullivan, L.W., Gallwey, T.J., 2005. Forearm torque strengths and discomfort pro-
files in pronation and supination. Ergonomics 48, 703e721.
Schneider, S., 1997. Musculoskeletal Injuries in Construction: Are They a Problem?.
In: Proceedings of the 13th Triennial Congress of the International Ergonomic
Association, June 29-July 4, Tampere, Finland, vol. 6, pp. 69e171.
Selen, L., Beek, P., Van Dieen, J., 2007. Fatigue-induced changes of impedance and
performance in target tracking. Exp. Brain Res. 181, 99e108.
SENIAM, 1999. European Recommendations for Surface ElectroMyoGraphy. Roes-
singh Research and Development, Enschede, The Netherlands.
Sporrong, H., Palmerud, G., Kadefors, R., Herberts, P.,1998. The effect of light manual
precision work on shoulder muscles-an EMG analysis. J. Electromyogr. Kinesiol.
8, 177e184 .
Vasseljen, O., Westgaard, R.H., 1995. Can stress-related shoulder and neck pain
develop independently of muscle activity? Pain 64, 221e230.
Veiersted, K.B., Forsman, M., Hansson, G.-K., Mathiassen, S.E., 2013. Assessment of
time patterns of activity and rest in full-shift recordings of trapezius muscle
activity - effects of the data processing procedure. J. Electromyogr. Kinesiol. 23,
540e547.
D. Phelan, L. O’Sullivan / Applied Ergonomics 45 (2014) 1384e1391 139 1
