Available online at www.sciencedirect.

com

Gait & Posture 28 (2008) 8692

www.elsevier.com/locate/gaitpost

Prolonged standing as a precursor for the development of low

back discomfort: An investigation of possible mechanisms
Diane E. Gregory, Jack P. Callaghan *
University of Waterloo, Waterloo, Ontario, Canada
Received 30 May 2007; received in revised form 14 August 2007; accepted 14 October 2007

Abstract
Low back discomfort (LBD) has been associated with prolonged periods of standing, yet research has shown that the magnitude of spinal
loading during standing is relatively minimal. Therefore, the mechanism of this discomfort is not fully understood. Research has monitored
numerous variables during prolonged periods of standing; however the focus of this work has been primarily on the comparison of the effect of
different floor surfaces on these variables. No study to date has made an attempt to relate these changes to the development of LBD. The
purpose of this study was therefore to determine possible mechanisms for the development of LBD during standing by monitoring biological
variables. It was hypothesized that during a prolonged standing period, LBD would develop and the measured variables would change over
time. Sixteen individuals stood for 2 h while activation of torso and hip muscles, lumbar spine posture, back extensor muscle oxygenation,
torso skin temperature, and centre of pressure changes under the feet were monitored over time. Thirteen out of sixteen individuals developed
LBD as a result of the prolonged standing period, which significantly increased over the 2-h period ( p < 0.0001). Only three of the 37
variables measured were significantly altered over time. However, a generated regression model incorporating 15 of the 16 individuals (which
incorporated how each individual stood in the first 15 min) explained 78% of the variance in LBD at the end of the 2-h standing period.
Prolonged standing resulted in LBD, yet few significant changes in the measured variables were observed over time. It is possible that LBD is
not linked with alterations in standing over time, but rather associated with how an individual initially stands.
# 2007 Elsevier B.V. All rights reserved.
Keywords: Spine posture; Prolonged standing; Low back pain; Discomfort; EMG; Regression

1. Introduction
Research has shown that the actual compressive load on
the low back during standing is minimal [1], however
prolonged static standing has been associated with the
reporting of low back pain [28]. This is especially evident
in, but not limited to, the automotive industry where
automation has greatly reduced the prevalence of heavy
lifting and awkward postures experienced by the worker,
often resulting in reduced trunk motion and increased
repetitive upper extremity tasks. Jobs such as cashiers, bank
* Corresponding author at: Department of Kinesiology, Faculty of
Applied Health Sciences, University of Waterloo, 200 University Avenue
West Waterloo, Ontario, Canada N2L-3G1. Tel.: +1 519 888 4567x7080;
fax: +1 519 746 6776.
E-mail address: callagha@healthy.uwaterloo.ca (J.P. Callaghan).
0966-6362/$ see front matter # 2007 Elsevier B.V. All rights reserved.
doi:10.1016/j.gaitpost.2007.10.005

tellers, and casino dealers also spend the majority of their

workday standing.
Previous research has monitored numerous variables
during prolonged periods of standing; however the focus of
this work has been primarily focused on the comparison of
the effect of different floor surfaces on these variables.
Larger centre of pressure displacements while standing on a
hard surface as compared to a soft surface, which
participants rated as more comfortable, and larger displacements following experimental pain elicitation have been
previously observed [9]. Further increases in the lower limb
skin temperature while standing on the hard surfaces and a
decrease in temperature while on the soft surfaces have also
been reported [9]. Higher perceived whole body and lower
limb fatigue [10] as well as increased back extensor
muscular fatigue [11] have been documented while standing
on a bare floor surface as compared to various footwear

D.E. Gregory, J.P. Callaghan / Gait & Posture 28 (2008) 8692

insoles and mats. Participants found mats and insoles to be

more comfortable during the standing period. Muscle
activation of the back and lower limbs has also been
examined during a prolonged period of standing, specifically
the tibialis anterior and the lumbar paraspinal muscles
during a 2-h period of standing on a concrete floor and a floor
mat [5]. This study found no differences in the mean muscle
activity of these muscles between the two floor types. It is
evident that extensive work has been conducted to assess the
effect of various floor surfaces on standing. However, to our
knowledge, there has been no reported attempt to associate
these changes with the development of low back discomfort
(LBD) over time. Therefore, the purpose of this study was
to examine variables that potentially contribute to the
mechanism of this discomfort as well as to determine if
trunk muscle activation, lumbar spine posture, centre of
pressure changes under the feet, skin temperature in the back
region, and erector spinae muscle oxygenation correlate
with the development of LBD during prolonged standing. It
was hypothesized that low back discomfort would increase
over the 2-h period of standing. It was also hypothesized that
the previously mentioned variables would change over time
as low back discomfort develops. Specifically it was
hypothesized that during the prolonged standing period,
the displacement/variability of centre of pressure, skin
temperature, and perceived fatigue and discomfort would
increase. These hypotheses were considered for several
reasons. It was hypothesized that due to the static nature of
prolonged standing, the discomfort perceived by the
individuals in the study would be a result of metabolite
buildup in the low back muscles. This accumulation of
metabolites has been shown to be a potential source of
perceived discomfort, which individuals may try to alleviate
via increased movement. Further, due to the static nature of
standing, an increase in blood pooling in the low back, and
thus an increase in metabolite buildup, may be observed as
increased temperature at the skin.

2. Materials and methods

2.1. Participants
Sixteen individuals (Table 1) were recruited from the general
university population, eight males and eight females. All participants were required to be free of LBD for at least 12 months prior to
data collection. Each participant was required to review and
consent to an outline of the experiment, approved by the University
Office of Research.
Table 1
Summary of participant parameters (age, height, mass) for males and
females

Age (years)
Height (m)
Mass (kg)

Male (n = 8)

Female (n = 8)

25.1 (2.1)
1.83 (0.05)
79.7 (9.3)

23.5 (2.3)
1.72(0.1)
69.7 (12.3)

87

2.2. Data collection and analysis

Individuals were required to stand in a confined working space
(0.50 m  0.46 m, surface area of a forceplate) for 2 h while
performing four precision tasks, presented in random order, that
resemble jobs that often require periods of prolonged standing.
These tasks included small object assembly (retractable pens) to
mimic assembly line workers; currency sorting to mimic a bank
teller; grocery store checkout to mimic a cashier; and card dealing
to mimic a casino dealer. No rest was given between the four tasks
such that each individual maintained the standing posture for two
straight hours. Individuals were not permitted to use anti-fatigue
devices (mats or footrests) but were allowed to adjust their posture
within the constrained space and rest their forearms on the worktable without supporting their body weight. All participants were
required to wear their own athletic shoes. Fig. 1a and b depicts the
experimental and participant set-up.
2.3. Electromyography
Muscle activation levels from 10 muscle sites were measured
using surface EMG with disposable pre-gelled EMG AgAgCl
electrodes (Blue Sensor, Medicotest, Inc., lstykke, Denmark).
These muscles included, bilaterally, thoracic erector spinae (TES),
lumbar erector spinae (LES), rectus abdominis (RA), external oblique (EO), and gluteus medius (GM). Electrode locations for the trunk
muscles were determined as per McGill [12], and the gluteus medius
electrodes were placed 15 cm inferior and 5 cm posterior of each iliac
crest. Raw EMG was collected at 2048 samples/s using a 16 bit A/D
card with a 2.5 V range. Systematic bias was removed from the raw
EMG signal prior to signal full-wave rectification. The signal was
then passed through a second-order Butterworth filter, with a 2.5 Hz
cut-off frequency [13], to produce a linear envelope for each of the 10
muscle groups recorded and then normalized to a maximum voluntary isometric contraction (MVC). Briefly, to perform a back extensor MVC, the participants were asked to extend against resistance,
while their torso was suspended over the edge of a bench. To carry out
an abdominal MVC, the participants were asked to perform a
modified sit-up against resistance, while twisting about the waist
to ensure maximal contraction of the abdominal muscles. To carry
out each hip abductor MVC, the participants were asked to lie on their
side (opposite of the leg performing the MVC) and abduct their leg
against resistance.
The total number of gaps (period of time when EMG levels drop
below 0.5% MVC for longer than 0.2 s [14]) and average activation
amplitudes for each 15 min time block were obtained for each
muscle. The number of shifts in EMG was also calculated for each
15-min block in order to determine if the activation of EMG
became more static or dynamic over the 2-h standing period. Shifts
were defined as a point in time when the level of muscle activation
exceeded a pre-determined threshold. The threshold was set as two
standard deviations above and below the mean, which was determined based on the first minute of standing. It was assumed that
individuals were standing quietly and were not fatigued during that
first minute of standing. The number of times the amplitude of
activation crossed the threshold, termed shift (either above or
below the mean) was tallied and compared between time blocks to
determine if the level of muscle activation either became more
static or dynamic throughout the standing period. The mean about
which the thresholds were bounded was recalculated for each 15min block (while maintaining the same standard deviation) such

88

D.E. Gregory, J.P. Callaghan / Gait & Posture 28 (2008) 8692

Fig. 1. Experimental and participant set-up including Optotrak marker, electrode, NIRS (muscle oxygen sensor), and skin temperature sensor locations: (a)
posterior view and (b) sagittal view.

that if the EMG signals were to gradually increase/decrease over

time, it would not be perceived as one long shift.

Philadelphia, PA, USA), and skin temperature (2100 Tele-Thermometer, Yellow Springs Instruments, Yellow Springs, OH, USA) was
also measured over the right TES and right LES.

2.4. Kinematics
2.6. Rating of perceived low back discomfort
Kinematics were measured using an active motion analysis
system (Optotrak Certus, NDI, Waterloo, Ont., Canada). IRED
markers were adhered to the skin at nine anatomical landmarks as
well as adhered to three rigid spine fins (additional nine markers)
shown in Fig. 1b. Three-dimensional lumbar spine posture was
calculated from marker locations on rigid fins (adhered over the
spinous processes of C7/T1, T12/L1, and L4/L5) using custom
software, 3DBack (University of Waterloo, Waterloo, Ont.,
Canada). Average flexionextension (FE), lateral bend (LB), and
axial twist (AT) angles were determined for each 15-min block and
were compared over time. Similar to the calculation of shifts in
EMG, shifts in lumbar posture about each axis were determined.
Joint coordinates were used to calculate reaction forces and the net
joint moment at L4/L5 using a two-dimensional rigid link segment
model (GOBER, University of Waterloo, Waterloo, Ont., Canada).
With the use of a simplified single muscle equivalent model, boneon-bone forces were determined at L4/L5. The single muscle
equivalent model incorporated a single extensor muscle equivalent
vector with a 5.38 angle of pull in the posterior direction with a
6 cm moment arm [15] and a single flexor muscle equivalent vector
anterior to the L4/L5 joint acting parallel to the joint compression
axis with a moment arm of 4.5 cm [16].
Centre of pressure under the feet in the mediallateral (CoPML)
and anteriorposterior (CoPAP) directions were determined from
forceplate outputs (Advanced Mechanical Technologies Inc., Newton, MA, USA), which were sampled at 1024 samples/s. Individuals were permitted to change the location of their foot placement
providing both feet maintained contact with the forceplate.
2.5. Back extensor muscle oxygenation and skin temperature
Muscle oxygenation of the right TES was measured using nearinfrared spectroscopy (NIRS) (RunmanTM CWS-2000, NIM Inc.,

Participants were required to rate their level of perceived LBD

using a 100 mm visual analog scale (with end point anchors of no
discomfort and worst discomfort imaginable). Ratings of discomfort were conducted at the start of the 2-h standing period and
every 15 min until the end of the collection period (total of nine
ratings of LBD). The magnitude of LBD was determined by
measuring the distance from the hatch mark on the scale to the
origin (in mm).
2.7. Statistical analysis
Ratings of perceived LBD, average muscle activation, average
gap number, shifts in EMG for each muscle, average lumbar
posture, shifts in lumbar spine posture, average L4/L5 joint loading, shifts in CoP, muscle oxygenation, and average thoracic and
lumbar skin temperature were determined for each 15-min block. A
one-way repeated measure analyses of variance with blocks of time
was used to determine if any variable changed significantly over the
2 h duration. Tukeys post hoc multiple comparisons were used to
examine any significant main effect finding.
Pearsons correlations were used to determine the strength of the
relationship between each of the previously mentioned variables
and the ratings of LBD. Three regression analyses were performed
with the ratings of LBD at the end of the 2-h standing period as the
dependent variable and the magnitude of the 12 most highly
correlated variables with LBD, measured in only the first 15 min
of standing, as the independent variables. These were performed in
order to determine if the level of LBD could be predicted by how
the participant stood during the first 15 min of the standing period.
The three regression models included one which incorporated all 16
participants; one which incorporated only the individuals who
developed LBD; and one which included 15 of the 16 individuals.

D.E. Gregory, J.P. Callaghan / Gait & Posture 28 (2008) 8692

The removed individual in this regression model did not develop

LBD during the standing period but responded quite differently to
the other non-LBD individuals and was considered an outlier for
the third regression analysis.

3. Results
3.1. The effect of standing on LBD
Thirteen of the sixteen participants had some level of
LBD at the end of the 2 h of standing. When individuals
were examined based on whether they developed LBD, no
significant differences were observed in any single measured
variable between the two groups (those who developed LBD
and those who did not).
The rating of perceived discomfort in the low back was
found to be significantly affected by time ( p < 0.0001)
shown in Fig. 2 with a trend of increasing LBD over time. A
noteworthy point is that all participants had no perceived
LBD at the start of the standing period, but had an average
rating of 3.06 mm after the first 15 min and 19.00 mm after
the full 2 h. It is interesting to note the increased variability
in the perceived discomfort rating later in the standing
period as compared to at the start. While pain perception to
controlled stimuli has been shown to result in large interindividual pain rankings [17] there is evidence that the intraindividual perception has strong testretest reliability [18].
The subjective nature of pain perception would therefore
yield variability as the level of pain develops, yet has good
repeatability within individuals.
3.2. The effect of time
Only three variables significantly changed over the 2-h
period; the degree of lumbar spine FE, the magnitude of L4/
L5 joint shear, and skin temperature over the right TES site.
The degree of FE was significantly affected by time
( p = 0.0156), such that as time increased, individuals tended
to increase the degree of flexion in their lumbar spine from

89

an average of 0.358 (standard deviation (S.D.) 1.068) of

flexion after the first 15 min to 1.768 (S.D. 0.748) flexion
after the 2 h of standing. Time was also found to affect the
magnitude of bone on bone AP shear force at L4/L5
( p < 0.0001), such that posterior shear of L4 with respect to
L5 had a tendency to decrease over time. The shear force was
calculated to be an average of 22.6 N (L4 posterior with
respect to L5) (S.D. 35.5 N) after the first 15 min, and 14.6 N
(S.D. 29.0 N) in the posterior direction after the full 2 h.
Last, skin temperature over the right TES was also affected
by time ( p < 0.0001). The skin temperature significantly
decreased over the 2 h from an average of 31.1 8C (S.D.
1.30 8C) after 15 min to 30.7 8C (S.D. 1.21 8C) after the 2 h.
3.3. Regression analysispredicting the magnitude of
LBD
Pearsons correlations were used to determine the
strength of the relationship between the magnitude of each
variable in the first 15 min of standing and the final rating of
LBD. Three step-wise regression analyses were performed
with the rating of LBD at the end of the 2-h standing period
as the dependent variable and the magnitude of the 12
highest correlated variables (determined using Pearsons
correlations) in the first 15 min of standing as the
independent variables. These regressions were performed
in order to determine if the level of LBD after 2 h could be
predicted from how an individual stood in the first 15 min.
Model A incorporated all 16 participants (R2 = 0.59), model
B incorporated 15 of the 16 participants; with one participant
(who did not develop LBD) removed (R2 = 0.78), and model
C incorporated the 13 participants who developed LBD
(R2 = 0.85). Model B could accurately predict the magnitude
of LBD at the end of 2 h of standing for all participants who
developed some level of LBD as well as predict minimal
LBD for two of the three participants who did not develop
LBD. Model B was thought to be the most functional as it
was able to distinguish between those who eventually
developed LBD from those who did not, and is therefore the
focus of the discussion. Coefficients for each variable and
the regression equation for each of the three models can be
found in Table 2.

4. Discussion

Fig. 2. The effect of time on the rating of perceived low back discomfort
over the 2-h standing period. Standard error bars are shown. Values grouped
under the same horizontal bar are not significantly different from each other.

As originally hypothesized, the 2-h standing period

elicited low back discomfort in 13 out of 16 individuals. This
is a substantial finding as these individuals who developed
LBD during the 2-h period of standing had no history of
chronic LBD and were completely asymptomatic at the start
of the study.
It was hypothesized that the displacement in centre of
pressure, skin temperature, and perceived fatigue/discomfort would increase during the standing period. The results,
however, showed that few variables consistently changed

90

D.E. Gregory, J.P. Callaghan / Gait & Posture 28 (2008) 8692

Table 2
Regression equations for each of the three models (coefficients for each variable are shown in parenthesis)
Model participants

R2

Regression equation

All 16 participants

0.59

15 of 16 participants

0.78

13 LBD participants

0.85

Low back discomfort = (0.028  number of gaps in LRA activation) +

( 0.055  number of shifts in angle of axial twist) + (14.532)
Low back discomfort = (0.062  number of shifts in CoP (AP)) + (0.113 
number of gaps in LGM activation) + ( 1.678  degree of axial twist (8)) + ( 0.186)
Low back discomfort = (0.063  number of shifts in CoP (AP)) + (0.095  number
of gaps in LGM activation) + ( 1.424  degree of axial twist (8)) + (1.978)

Note that the dependent variable is the magnitude of low back discomfort at the end of 2 h in mm using a 100 mm visual analog scale.

over the 2-h period. The variables that did change over time
were the degree of lumbar flexion, L4/L5 joint shear, and
skin temperature over the upper back region. While the
drop in skin temperature may be directly due to the
participants back being exposed during the duration of the
collection, the moderate change in spine flexion and L4/L5
joint shear may be associated with the development of low
back discomfort. The moderate spine flexion developed
over the 2 h may have changed the facet separation and
ligament lengths. These passive tissues of the intervertebral
joint have been shown to be sensitive to length and pressure
changes and are highly innervated with type III and IV
nociceptors [19]. This altered loading of the passive
structures may have been one potential source of the
reported discomfort. Further, stadiometry research has
shown that prolonged standing results in disc height loss
[20,21]. This likely also occurred in the current study,
which creates several possible pain generating pathways.
The creep associated with the disc height loss in severe
cases may have altered the morphology of the neural
foramina resulting in nerve impingement, which has been
shown in an animal model to initiate pain with nerve root
compression of magnitudes as low as 8.4% [22]. More
moderate cases of creep changes could create strains in the
passive tissues in the intervertebral joint such as the disc or
articular capsule sufficient to generate pain from compressive loading [23]. It should be noted that the significant
decrease in posterior L4/L5 joint shear over the 2 h was
expected as the model used to determine the joint loading
required lumbar flexion as an input. Given that the actually
observed change in spine flexion and L4/L5 joint shear was
small, these variables were not believed to be the source of
the discomfort. More likely the discomfort was related to
the initial postural and motor control conditions that each
individual self-selected at the start of the standing task, as
shown by the regression results. Based on these results, it
was possible to predict (R2 = 0.78) the level of LBD after
2 h of standing in a model composed of 15 of the 16
participants by examining the magnitude of three variables
during the first 15 min of standing.
4.1. The ability to predict low back discomfort
As previously mentioned, regressions were performed
using the rating of LBD at the end of the 2-h standing period as

the dependent variable and the observed variables in the first

15 min of standing alone as the independent variables, in an
attempt to determine if the level of eventual LBD was related
to how that individual stood in the first 15 min. When only one
participant was removed from the complete participant pool
(model B), an R2 of 0.78 was observed. Moreover, this high R2
value suggests that this model can predict the rating of LBD at
the end of the 2 h in both individuals who did in fact develop
LBD and in those who did not, based on how they stood in the
first 15 min. The single participant, who did not report LBD in
the study, and who was excluded from model B was selected
as they were the only person who clearly did not fit the
eventual model. This was possibly due to either possessing a
much different strategy to aid in the prevention of LBD or an
altered perspective of perceived discomfort. When the
excluded participants variables were included in this model,
their predicted magnitude of LBD was 28.0 mm, with the
recorded value for this participant being 0 mm. Therefore, this
participant tends to stand similarly to those who develop a
substantial level of LBD at the end of 2 h, but did not report
LBD during the standing period.
The variables included in both model B and C were shifts
in the CoPAP, gaps in the activation of left GM, and the
degree of axial twist. The coefficient for the number of
shifts in CoPAP is positive, suggesting that a higher number
of shifts was associated with a higher rating of discomfort.
In addition, the coefficient for the number of gaps in left GM
activation is also positive, suggesting that a higher number
of gaps in activation was associated with a higher rating of
LBD. These findings contradict previous findings that state
that an increased number of rest periods/gaps in the muscle
and dynamic postures, tracked by increased shifts in CoP,
potentially reduce discomfort and possibly prevent injury
[24,25]. The regression analysis and consequently the
generated model, suggest the opposite, implying that a
more static posture and continuous muscle activation is
associated with a reduced risk of the development of
discomfort.
Two possible interpretations exist that may explain the
direction of the observed relationships with LBD and the
potential mechanisms for the development of LBD. The first
is that the perceived discomfort is a result of the dynamic
nature of the aforementioned variables (increased CoP shifts
and increased muscle gaps). It is possible, according to the
regression model, for an individual to move too much (for

D.E. Gregory, J.P. Callaghan / Gait & Posture 28 (2008) 8692

example awkward twisting, bending and/or reaching) in the

first 15 min of standing, which may be the cause of the
development of discomfort. An alternative explanation is
that those who demonstrated increased shifts did so in an
attempt to prevent or reduce the expected LBD. In other
words, individuals who are more susceptible to developing
LBD may tend to stand with increased movement (example,
increased number of CoP shifts) as a pre-emptive strategy.
The ability to potentially predict the development of
discomfort in the low back during standing based on the first
15 min alone is a substantial finding. Examining the
variables that have the largest effect on the generated
model provides information regarding which factors
potentially contribute to LBD. Based on this model, it
may be possible to decipher between individuals who will
develop LBD from those who will not based on how they
stand initially. Furthermore, understanding the characteristics that possibly aid in reducing LBD during prolonged
standing may help to determine how an individual who
generally develops LBD can adopt these characteristics.
Caution should be taken when applying a generated
regression model to an entire population. The current model
incorporated a small sample group (15 participants), and
future work with a new sample group is required to
determine if the current regression model can be generalized
to and validated on a larger population. Therefore, the main
importance of this regression model is not the ability to
predict LBD development during prolonged standing but
rather to increase the understanding of LBD and the
variables that are associated with its development.

5. Conclusions
This study confirms the likelihood of developing LBD
during a prolonged period of standing, as 13 of the 16
participants developed LBD. Of the variables measured,
very few revealed differences between individuals who
developed LBD during prolonged standing and those who
did not. Additionally, few changes in the measured variables
were observed over time, despite the steady development of
LBD over time. However, there is strong potential in the
ability to predict the magnitude of the LBD developed after a
prolonged period of standing in the individuals monitored in
the current study, based on the first 15 min alone. This
indicates that individuals may adopt initial standing postures
that relate to the magnitude of LBD that they will develop
over time, possibly in an effort to minimize their eventual
level of discomfort. This sheds new light on the understanding of the variables associated with the mechanisms of
LBD development.

Conflict of interest statement

None.

91

Acknowledgments
The authors wish to acknowledge AUTO21Network of
Centres of Excellence, Canadian Institute for the relief of
pain and disability, and Canadian Institutes for Heath
Research for financial support. Dr. Jack Callaghan is also
supported by a Canada Research Chair in Spine Biomechanics and Injury Prevention. The authors also wish to
acknowledge Erin Harvey, University of Waterloo, for
statistical consulting.

References
[1] Nachemson A. Disc pressure measurements. Spine 1981;6:937.
[2] Magora A. Investigation of relation between low back pain and
occupation: medical history and symptoms. Scand J Rehabil Med
1974;6:818.
[3] Biering-Sorensen F, Thomsen CE, Hilden J. Risk indicators for lowback trouble. Scand J Rehabil Med 1989;21:1517.
[4] Ryan GA. The prevalence of musculoskeletal symptoms in supermarket workers. Ergonomics 1989;32:35971.
[5] Cook J, Branch TP, Baranowski TJ, Hutton WC. The effect of surgical
floor mats in prolonged standingan EMG study of the lumbar paraspinal and anterior tibialis muscles. J Biomed Eng 1993;15:24750.
[6] Veniza N, Geoffrion L, Chatigny C, Messing K. A manual materials
handling job: symptoms and working conditions among supermarket
cashiers. Chronic Dis Can 1994;15:1722.
[7] Macfarlane GJ, Thomas E, Papageorgiou AC, Croft PR, Jayson MIV,
Silman AJ. Employment and physical work activities as predictors of
future low back pain. Spine 1997;22:11439.
[8] Xu Y, Bach E, Orhede E. Work environment and low back pain: the
influence of occupational activities. Occup Environ Med 1997;54:
7415.
[9] Madeleine P, Voigt M, Arendt-Nielsen L. Subjective, physiological
and biomechanical responses to prolonged manual work performed
standing on hard and soft surfaces. Eur J Appl Physiol 1998;77:19.
[10] King PM. A comparison of the effects of floor mats and shoe in-soles
on standing fatigue. Appl Ergon 2002;33:47784.
[11] Kim JY, Stuartbuttle C, Marras WS. The effects of mats on back and
leg fatigue. Appl Ergon 1994;25:2934.
[12] McGill SM. Electromyographic activity of the abdominal and low
back musculature during the generation of isometric and dynamic
axial trunk torque: implications for lumbar mechanics. J Orthop Res
1991;9:91103.
[13] Brereton LC, McGill SM. Frequency response of spine extensors
during rapid isometric contractions: effects of muscle length and
tension. J Electromyogr Kinesiol 1998;8:22732.
[14] Veiersted KB. Sustained muscle tension as a risk factor for trapezius
myalgia. Int J Ind Ergon 1994;14:3339.
[15] McGill SM, Norman RW. Effects of an anatomically detailed erector
spinae model on L4/L5 disc compression and shear. J Biomech
1987;20:591600.
[16] McGill SM, Norman RW, Cholewicki J. A simple polynomial that
predicts low-back compression during complex 3D tasks. Ergonomics
1996;39:110718.
[17] Mader TJ, Blank FSJ, Smithiline HA, Wolfe JM. How reliable are pain
scores? A pilot study of 20 healthy volunteers. J Emerg Nurs
2003;29:3225.
[18] Bijur PE, Latimer CT, Gallagher J. Validation of a verbally administered numerical rating scale of acute pain for use in the emergency
department. Acad Emerg Med 2003;10:3902.
[19] Cavanaugh JM. Neural mechanisms of lumbar pain. Spine 1995;20:
18049.

92

D.E. Gregory, J.P. Callaghan / Gait & Posture 28 (2008) 8692

[20] McGill SM, van Wijk MJ, Axler CT, Gletsu M. Studies of spinal
shrinkage to evaluate low-back loading in the workplace. Ergonomics
1996;39:92102.
[21] Beynon C, Reilly T. Spinal shrinkage during a seated break and
standing break during nursing tasks. Appl Ergon 2001;32:61722.
[22] Winkelstein BA, DeLeo JA. Mechanical thresholds for initiation
and persistence of pain following nerve root injury: mechanical
and chemical contributions at injury. J Biomech Eng 2004;126:
25863.

zaktay AC, Yamashita HT, King AI. Lumbar facet

[23] Cavanaugh JM, O
pain: biomechanics neuroanatomy and neurophysiology. J Biomech
1996;29:111729.
[24] Veiersted KB, Westgaard RH, Andersen P. Pattern of muscle activity
during stereotyped work and its relation to muscle pain. Int Arch
Occup Environ Health 1990;62:3141.
[25] Videman T, Nurminen M, Troup JDG. Lumbar spinal pathology in
cadaveric material in relation to history of back pain, occupation, and
physical loading. Spine 1990;15:72840.