PhD Thesis Physical work exposure and sex differences in ......Physical work exposure and sex...
186
PhD Thesis Thomas Heilskov-Hansen 2014 Physical work exposure and sex differences in work-related musculoskeletal disorders
Transcript of PhD Thesis Physical work exposure and sex differences in ......Physical work exposure and sex...
PhD ThesisThomas Heilskov-Hansen
2014
Physical work exposure and sex differences in work-related musculoskeletal disorders
Physical work exposure and sex differences in w
ork-related musculoskeletal disorders
Thomas Heilskov-Hansen – PhD thesis
ISBN 978-87-996042-6-5
UDGIVNE RAPPORTER
1. Rapport om Reproduktionsskader ved arbejde med Frisørkemikalier, 2013
2. Rapport om Arbejdsbetinget Eksem, 2013
3. Rapport om Gravides arbejdsmiljø. Udsættelse for Methyl metakrylat (MMA) blandt tandteknikere og operationspersonale, 2013
4. Report: Risk of disease following occupational exposure to Polychlorinated Biphenyls, 2013
5. Workplace Bullying, Depression and Cortisol – Maria Gullander, PhD thesis, 2014
6. Is retirement bene�icial for mental and cardiovascular health?Kasper Olesen, PhD thesis, 2014
7. Physical work exposure and sex differences in work-related musculoskeletal disordersThomas Heilskov-Hansen, PhD thesis 2014
heilskov_omslag.indd 1 04/11/14 13.10
This thesis has been submitted to the Graduate School of The Faculty of Health and Medical Sci-
ences, University of Copenhagen
I am not young enough to know everything
- Oscar Wilde
Title: Physical work exposure and sex differences in work-related musculoskeletal disorders
Author: Thomas Heilskov-Hansen, MHSc. Department of Occupational and Environmental Medicine Bispebjerg University Hospital, Copenhagen, Denmark Email: [email protected]
Supervisors: Jane Frølund Thomsen, MD, Ph.D.
Department of Occupational and Environmental Medicine, Bispebjerg University Hospital, Copenhagen, Denmark
Sigurd Mikkelsen, MD, DMSc.
Department of Occupational and Environmental Medicine, Bispebjerg University Hospital, Copenhagen, Denmark Susanne Wulff Svendsen, MD, Ph.D. Danish Ramazzini Centre, Department of Occupational Medicine, Regional Hospital West Jutland - University Research Clinic, Herning, Denmark Gert-Åke Hansson. MSc. Ph.D. Occupational and Environmental Medicine, Lund University, and University and Regional Laboratories Region Scania, Lund, Sweden
Opponents: Professo Michael Kjær (Chair), DMSc.
Department of Clinical Medicine, University of Copenhagen, Denmark
Professor Pascal Madeleine, Ph.D.
Department of Health Science and Technology, University of Aalborg, Denmark Professor Keith T. Palmer, BM BCh, DM, MA, MSc, FFOM, FRCP, MRGP. MRC Lifecourse Epidemiology Unit, University of Southampton, United Kingdom
Submitted: August 3rd 2014 Defended: November 21st 2014 ISBN: 978-87-996042-6-5
The work presented in this Ph.D.-thesis was the result of the SHARM-project
which was conducted between October 2010 and August 2014, at the Depart-
ment of Occupational and Environmental Medicine at Bispebjerg University
Hospital. The study was supported by grants from The Danish Working Envi-
ronment Research Fund (grant #: 43-2010-03) and The Danish Rheumatism
Association (grant #: R104-A2251).
During the completion of my master thesis entitled: “Comparison of two 3D
gait analyses protocols – supported by EMG”, I first started thinking about
conducting a Ph.D. I applied for two and was offered both – on the same day. I
chose the SHARM-project in part because of its multidisciplinary character,
allowing me to develop new skills within: questionnaires, biomechanical and
physiological measurements and register epidemiology.
After completing most of my data collection, a year and a half in to my Ph.D.-
study, I was lucky to have the opportunity to be a visiting scientist at Harvard
School of Public Health in collaboration with Liberty Mutual Research Center,
in Boston, USA. The purpose of my three month stay was to gain knowledge of
novel methods for assessing exposures of physical exertion during work. I
took part in several studies including a validation study of a wearable sensor
system for assessing spinal loading during manual materials handling tasks.
This experience gave me a lot of new perspectives, ideas and inspiration on
how to develop thorough measurements of physical exposure with methods
applicable at work sites.
In parallel with working on my thesis I have since 2010 given lectures on ap-
plying methods for measurement of physical activity, at Metropolitan Univer-
sity College. During the last year this has been intensified by giving lectures in
basic epidemiology and research methods at University College Capital. At this
institution I have also been principal supervisor for nine bachelor students of
Physiotherapy since 2011. Recently I have been appointed to be external ex-
aminer for the bachelor-exams at the Danish Educations for Physiotherapy.
Altogether I have developed many new skills on both the personal and profes-
sional level.
Birkerød, August 2014 Thomas Heilskov-Hansen
This thesis is based on the following original papers which can be found in the
contents under “Original papers”. Throughout the thesis the papers will be
referred to in roman numerals I-III.
Paper I: Sex differences in muscular load among house painters per-
forming identical work tasks.
(Eur J Appl Physiol 2014;1-11)
Paper II: Sex differences in task distribution and task exposures among
Danish house painters: An observational study combining
questionnaire data with biomechanical measurements.
(Submitted to PLOS ONE)
Paper III: Exposure-response relationship between postures and move-
ments of the wrist and carpal tunnel syndrome among male
and female house painters: a retrospective cohort study.
(Draft)
Abbreviations listed in alphabetical order.
AL Action Limits
ACGIH American Conference of Governmental Industrial Hygienists
BMI Body Mass Index
CRS Danish Civil Registration System
CTS Carpal Tunnel Syndrome
DNPR Danish National Patient Register
EDT Electro Diagnostic Test
EMG Electromyography
ICD-10 International Classification of Diseases, Version 10
IRR Incidence Rate Ratios
JEM Job Exposure Matrix
MSD MusculoSkeletal Disorder
MVC Maximal Voluntary Contractions
NCSP-D Nomesco Classification of Surgical Procedures- Danish version
NBII National Board of Industrial Injuries
NHSR National Health Service Register
PUD Painters Union in Denmark
RMS Root Mean Square
SHARM Shoulder, Hand, ARM-project
TEM Task Exposure Matrix
TLV Threshold Limit Values
WMSD Work-Related Musculoskeletal Disorder
1 List of papers ...................................................................................................................... 5
2 List of abbreviations ........................................................................................................ 6
3 Introduction ........................................................................................................................ 9
4 Aims..................................................................................................................................... 12
5 Background ...................................................................................................................... 13
5.1 Sex and gender terminology ......................................................................... 13
5.2 Musculoskeletal Disorders ............................................................................ 13
5.3 Carpal Tunnel Syndrome ............................................................................... 15
5.4 Exposure assessment ...................................................................................... 17
5.4.1 Exposure matrices ...................................................................................... 19
6 Methods ............................................................................................................................. 22
6.1 Study designs ...................................................................................................... 22
6.1.1 Paper I.............................................................................................................. 22
6.1.2 Paper II ............................................................................................................ 22
6.1.3 Paper III .......................................................................................................... 22
6.2 Study populations ............................................................................................. 23
6.2.1 Paper I.............................................................................................................. 23
6.2.2 Paper II ............................................................................................................ 23
6.2.3 Paper III .......................................................................................................... 23
6.3 Electromyography ............................................................................................ 24
6.3.1 Preparations and equipment ................................................................. 24
6.3.2 Maximal voluntary contractions (MVC) ............................................ 24
6.3.3 EMG-to-force calibration ......................................................................... 24
6.3.4 Measurements .............................................................................................. 24
6.3.5 Data processing ........................................................................................... 25
6.4 Borg CR-10 ........................................................................................................... 25
6.5 Questionnaire ..................................................................................................... 25
6.6 Goniometry and inclinometry measurements ...................................... 26
6.7 Log book ................................................................................................................ 26
6.8 Job and Task exposure matrices ................................................................. 26
6.9 Danish registers ................................................................................................. 26
6.10 Data preparation ............................................................................................... 27
6.11 Statistical analyses ............................................................................................ 27
6.11.1 Paper I.............................................................................................................. 27
6.11.2 Paper II ............................................................................................................ 27
6.11.3 Paper III .......................................................................................................... 27
6.12 Ethics statement and approvals .................................................................. 28
7 Results ................................................................................................................................ 29
7.1 Male and female strength (Paper I) ........................................................... 29
7.2 Time expenditure (Paper I) .......................................................................... 29
7.3 Relative muscular load (Paper I) ................................................................ 29
7.4 Exerted forces (Paper I) ................................................................................. 30
7.5 Borg CR-10 scale (Paper I) ............................................................................ 31
7.6 Sex specific task distributions (Paper II) ................................................ 31
7.7 Sex specific exposure assessment of postures and movements .... 32
7.7.1 TEM and JEM for the right wrist ........................................................... 32
7.7.2 TEMs and JEMS for the head and right shoulder ........................... 33
7.8 CTS cases (Paper III) ........................................................................................ 35
7.9 CTS prevalence and incidence rate in cohort (Paper III) ................. 35
7.10 Exposure response relationship (Paper III) .......................................... 36
7.10.1 Median velocity for flexion extension of the wrist ........................ 36
7.10.2 Mean power frequency ............................................................................. 39
7.10.3 Non-neutral postures ................................................................................ 39
8 Discussion ......................................................................................................................... 42
8.1 Methodological considerations ................................................................... 42
8.2 Discussion of findings ..................................................................................... 49
8.3 Clinical relevance of findings ....................................................................... 54
9 Conclusions ...................................................................................................................... 55
10 Perspectives ..................................................................................................................... 55
11 Summary ........................................................................................................................... 60
12 Dansk resumé (Danish summary) .......................................................................... 61
13 Acknowledgements ...................................................................................................... 62
14 References ........................................................................................................................ 64
15 Appendices ....................................................................................................................... 83
15.1 Appendix I: Borg CR-10 scale ....................................................................... 84
15.2 Appendix II: Log-book for biomechanical measurements ............... 85
15.3 Appendix III: Task- and job exposure matrices for the left side .... 86
15.4 Appendix IV: SHARM questionnaire (in Danish) ................................. 87
16 Original papers ............................................................................................................ 113
In 2009 MD Rolf Petersen from the Department of Occupational Medicine in
Slagelse, Denmark, observed a seemingly high sex difference in the number of
patients from the house painters profession who contacted the department
with work related muscular skeletal disorders (WMSDs) of the upper extremi-
ty. In order to investigate if this difference was entirely observed by chance, he
contacted The National Board of Industrial Injuries (NBII) in Denmark which
is the authority to whom workers report claims regarding occupational dis-
eases. The NBII extracted national incidence data of notified WMSD-cases
among male and female house painters in the period from 1998-2007. The
data showed a substantial sex difference (figure 1).
Figure 1: Incidence rates by sex of WMSDs among house painters, reported to the National Board of
Industrial Injuries in the period 1998-2007
Simple inspection of data shows that women painters had approximately
twice as high incidence rates of WMSD claims as men, and this ratio was quite
stable across calendar years. When dividing the reported cases among house
painters into specific anatomical regions, some of the highest incidence rates
and sex differences were located in the upper extremity (figure 2). The largest
sex difference was found for the wrist.
This data is in good accordance with the literature. Higher reporting of muscu-
loskeletal pain, complaints or WMSDs by women is well documented (1-8),
and is especially pronounced for the upper extremity (5;8-10).
Figure 2: Incidence rates by anatomical region of WMSDs among house painters, reported to the
National Board of Industrial Injuries in the period 1998-2007
Despite constituting about half the working population in most industrialized
countries, women are still underrepresented in research of WMSDs (11-13).
Sex differences in WMSDs are often explained by the following main hypothe-
ses:
1. Women have lower thresholds for reporting musculoskeletal pain, ei-
ther in a psychosocial or physiological sense (3;14-19).
2. Sex segregation in occupations and task segregation within profes-
sions results in different exposures for men and women (5;6;8;20-25).
3. Women are more vulnerable at the same exposure. Several physiologi-
cal differences between men and women ranging from hormonal
changes to heterogeneous muscular strength can influence the impact
of an exposure (6;8;14;26).
4. Sex differences in work strategies, techniques and procedures, ex-
pressed by a diverse composition of postures and movements, even if
the task is the same (27).
Many uncertainties resulting from one or more of these hypotheses could be
controlled by conducting a precise sex specific exposure assessment, but this
task is unresolved in most studies.
This knowledge initiated the planning of the SHARM-project (Shoulder, Hand,
ARM-project), trying to apply a systematic approach for a precise exposure
assessment, to be used in the evaluation of sex differences in WMSDs. This
scientific approach should help identify potential sex differences in develop-
ment of WMSDs and thereby ensure that preventive and rehabilitative
measures favor both sexes equally.
Based on the prevailing hypotheses for explaining sex differences in develop-
ment of WMSDs we wanted to establish a systematic approach which clarified
each of the hypotheses one at a time. The aims of this thesis were:
Within a seemingly homogenous profession to establish a precise ex-
posure assessment, examining potential sex differences in forces, load,
task distributions, postures and movements (Papers I and II).
To explore if an exposure-response relationship can be established be-
tween three different exposure variables of the wrist, and carpal tun-
nel syndrome defined through diagnoses and surgery reported to the
Danish national registers (Paper III).
To investigate whether the risk of carpal tunnel syndrome is different
for men and women with comparable occupational physical exposure
(Paper III).
The terminology regarding gender and sex research can be confusing and it is
often based on tradition or habit within certain fields of research. Traditional-
ly the term gender is used to describe social aspects related to subjective
properties i.e. identity, whereas sex usually refers to biological properties
(14;28). However in reality the two expressions tend to be used interchangea-
bly and attention has been drawn to the fact that they very rarely can be ex-
cluded completely from each other. Some have taken the consequence of this
and consistently use the term gender/sex irrespective of the properties in
question (10). In this thesis differences between men and women are referred
to as sex differences disregarding their nature.
The term Musculoskeletal disorder (MSD) is used to describe a wide variety of
conditions affecting the muscles, nerves, tendons, bones, ligaments or joints.
MSDs are usually caused by inflammation or degeneration, resulting in pain
and physical constraints. MSDs can have an acute or accumulative nature, but
traumas resulting in fractures are usually not considered as MSDs (29). MSDs
are very common and the annual costs are consequently very high (30;31).
They are more prevalent among women than men and even more so in the
upper extremity (29;32;33). Even within individuals performing identical
work, women have more WMSDs than men (8;34). Since occupational risk
factors are highly related to industrial work, WMSDs are more prevalent in the
lower social classes compared to high social classes (1;32;35-37)
Risk factors for MSD are often divided into non-occupational and occupational
nature. Common non-occupational or individual risk factors include: age, gen-
der, obesity, leisure time activities, smoking, strength, socioeconomic status
and ethnicity (10;37-40).
Occupational risk factors for WMSDs are many and they can be differentiated
by anatomical region(35). Some of the most common risk factors for WMSDs
are: repetitive motion, whole body or segmental vibration, forceful manual
exertion, heavy lifting, non-neutral body postures, high velocity and other
ergonomic stressors (7;10;29;35-42).
Occupational ergonomic stressors have been described by Punnet et al. (36)as
a key element in the occurrence of WMSDs. Examples of frequently occurring
MSDs include: Rotator cuff syndrome, CTS, lateral epicondylitis, low back pain
and hip and knee osteoarthritis (29;30;37).
Many risk factors for MSDs and WMSDs are associated with physiological or
psychological aspects and are therefore prone to differences between sexes
herein.
Numerous studies have investigated physiological differences between men
and women i.e. in relation to: Perception of pain (43), fatigability (44;45), ten-
don properties (46-49), hormonal differences (46;49), anthropometry and
muscular entities (50;51). Regarding the latter it has been well established
that the average man is approximately 50 % stronger than the average woman
(26;52;53). This muscular difference will cause women to use a higher level of
relative force if doing the same force demanding tasks as their male colleagues
(8;54). It has then been suggested that this difference accumulated over time
can contribute to the development of WMSDs (55). Others have reported that
women have an alternate muscular recruitment pattern or motor strategy
than men (50;51), and a different composition of muscle fibres, with women
having a higher proportion of type 1 muscle fibres compared to men(45;56).
These type 1 muscle fibres have been described by Hägg (57)as “Cinderella
fibres”, the name owing to their property of having the lowest recruitments
threshold and, in addition, staying activated for prolonged periods of low in-
tensity use. If working in repetitive low force tasks this could potentially lead
to an overload, resulting in a WMSD.
There is a growing body of evidence for psychosocial characteristics being risk
factors for WMSD (35-37;39;58;59). This evidence mainly addresses the modi-
fying role of psychosocial factors while the etiologic pathway is less estab-
lished (37). The most common reported risk factors include: high perceived
job stress, high job demands, non-work-related stress, low social support of
co-workers, low job control, low decision authority and low job satisfaction
(58-60). None of these studies reported any sex differences.
Gaining knowledge on sex differences in risk factors for WMSD could poten-
tially be helpful in limiting new cases of WMSDs or in the development of pre-
ventive measures.
Carpal tunnel syndrome (CTS) is a very common nerve entrapment (61;62). It
is caused by pressure on the median nerve in the carpal tunnel, resulting in
constant or recurring symptoms of numbness, burning, tingling or pain in one
or more of the first three digits and the radial part of the fourth. In some cases
there will be decreased grip force due to an atrophic abductor pollicis brevis
muscle (63;64). CTS can be treated with wrist splints, anti-inflammatory
drugs, corticoid steroid injections or in the final stage by surgery. In open or
endoscopic carpal tunnel release the transverse carpal ligament is cut in order
to relieve pressure from the median nerve (65-68) . CTS has been studied in-
tensively during the last 20-30 years (69;70) and is one of the upper extremity
MSDs with highest healthcare costs (71;72). Within epidemiological research
several different case definitions have been used when studying CTS. Studies
using only CTS symptoms for defining cases have reported higher CTS preva-
lence and incidence rates (70;73) and there is a widespread agreement that
case definitions influence prevalence and incidence rates (63;74-77). Some
studies have used dual case definitions and have divided cases into “probable
CTS” and “possible CTS” depending on which diagnostic criteria were met
(77;78). Many studies rely on what is referred to as “physician diagnosed”.
This is defined by clinical symptoms and a physical examination at minimum,
and usually includes an electro-physiological examination (79).
Reports of CTS prevalence span from 1.9 % in the general population (80) to,
for example, 16.6 % among dairy workers (81). In general there is a pro-
nounced variation in CTS prevalence and incidence rates reported in studies.
Considerable differences are seen among un-exposed control groups in stud-
ies of occupational factors (82-87). In general, the prevalence is higher in stud-
ies of occupational factors compared to studies of the general population
(61;80;88-91). This may in part be related to different case definitions as men-
tioned above, but actual differences between study populations may also have
an impact (74;79;92;93).
Several personal as well as occupational risk factors for CTS have been report-
ed. Some of the personal risk factors agreed upon are: female sex, age, high
body mass index (BMI), previous fractures near the wrist, co-morbidities (e.g.
hypothyroidism, diabetes, rheumatoid arthritis, gout and connective tissue
disorders), low height and family predisposition (16;63;94-100). The use of
oral contraceptives has been investigated by a few studies, but the results are
contradictive (95;101). Similar inconclusive findings have been made for
smoking (95;102;103). Studies of general populations have shown a higher
CTS prevalence in women (16;61;96;97). Bland and Rudolfer (97) have shown
a bi-modal age distribution of CTS for both men and women, with peaks
around 45-55 years of age and 75-85 years of age.
Roquelaure et al. (104) showed a higher incidence of CTS in the working
population compared to the non-working population. This corresponds with
the growing scientific literature where occupational risk factors for CTS are
documented in industrial settings (38;78;79;92;105;106). Tasks including
exposure to vibrations have been reported by many as one of the most im-
portant risk factors for CTS although the isolated effect of vibrations has been
difficult to distinguish from the concomitant effects of force and repetition
(79;92;93;105;107-115). Repetitive work and highly repeated flex-
ion/extension of the wrist (73;79;92;105;106;114-121) as well as forceful use
of the hand (69;79;92;98;105;116;120;121) are also frequently reported. Vio-
lante et al. (98) showed a 3-fold increase in CTS risk when exposed to unac-
ceptable overload, compared to acceptable load, according to the action limit
(AL) and threshold limit values (TLV) recommended by the American Confer-
ence of Governmental Industrial Hygienists (ACGIH). Others have also shown
an increased risk of CTS when exposed to forces exceeding the TLV (94;122-
124).
Combined effects of vibration, repetitious flexion/extension and force have
been shown to increase the risk of CTS even more (69;116;120;125). A combi-
nation of force and repetitive movements is the most commonly reported
(69;78;79;92;99;105) but others have also been reported e.g. the combination
of force and posture (92;105). Few have reported results on non-neutral pos-
tures, and none of these have shown an effect on the risk of CTS (38;42;126).
Computer work has been extensively investigated as a potential risk factor for
CTS (93;106;113;114;127-131), but the vast majority, including a recent me-
ta-analysis (129), have found no effect. Andersen et al. (127) and Ali et al.
(132) did however find an elevated risk of CTS when using a computer mouse
and doing combined work respectively. Work related psychosocial character-
istics such as low social support, job stress and high job strain has been re-
ported (although not statistically significant) as risk factors for CTS by some
(60;114;133;134), while others have found no effect (126;127). Harris-
Adamson et al. (135) showed a protective effect of experiencing social sup-
port, and an increased risk with high job strain. It is not obvious what the rel-
evant mechanisms could be and the evidence is conflicting. Perhaps psychoso-
cial stressors could be a proxy for job functions characterised by a high degree
of manual handling and thus a result of confounding not fully accounted for.
Many studies have shown an association between certain professions and the
risk of CTS. Examples of these professions are: Cashiers (73), Industrial work-
ers (122), meat and fish processing (82), Electronics assembly (136), slaugh-
terhouse workers (87) and dental hygienist (137). Typically, these professions
can be characterised as being either subjected to a high degree of occupational
vibration, repetitiveness or force requirements or any combination of these
(79;92;93).
Several studies have found a higher CTS prevalence in women
(32;78;96;97;100;138), but literature addressing a potential modifying effect
of sex on exposures associated with an increased risk of CTS is very scarce.
Some studies have reported differences in the task distribution and argue that
an uncontrolled sex difference in task composition can result in sex acting as a
proxy for exposure (123;139;140). Most studies addressing sex differences
have only reported minor if any sex difference in CTS risk using sex stratified
analyses (78;113;120;123;141).
In occupational research physical exposure has been assessed using observa-
tion, expert ratings, self-reports and direct technical measurements. Expo-
sures have most commonly been assigned as a group mean from the entire
study population or a sub-sample (142) and rarely as individually assessed
exposure , let alone sex specific (8).
In many studies that have examined the effect of a work related exposure on
the incidence rate of CTS, the researchers have used job title to assess the ex-
posure (92). In a recent meta-analysis 28 out of 37 studies had used job title
for assigning exposure (92). However, self-reported exposure assessments are
also widely used. Studies that have compared self-reported exposure assess-
ments with technical measurements have concluded that technical measure-
ments are superior in precision (143-147). Furthermore, self-reports may be
biased by a higher reporting of exposure in individuals experiencing pain
(147;148) or systematically higher reporting of tasks that are experienced as
hard (143;147;149;150). Using self-reported exposure assessment for evalu-
ating possible dose-response relations will therefore introduce the risk of am-
plification bias of the estimates.
Assessments of biomechanical exposures should preferably include measures
of intensity, duration and frequency (151;152) in order to provide a precise
and comprehensive unit of measure. Many studies have reported exposure on
a categorical scale (94;98;122-124) typically dividing exposures into two or
three groups using arbitrary threshold limits. This will hinder the possibility
of making comparisons between studies using different thresholds as well as
limit the determination of a precise dose-response estimate. Reporting on a
continuous scale enables both the ability of comparison with other studies and
the dose-response estimate. Precise measurements are required to do report-
ing on a continuous scale and it will therefore be associated with higher eco-
nomic costs(142).
Regardless of the applied methodology, assessing physical exposure will en-
compass a big challenge in accounting for the variability. When doing direct
measurements the optimal strategy would be to perform three or four meas-
urements on each individual on separate days, and preferably distributed
across the calendar year to capture potential seasonal changes (142;153-155).
Liv et al. (154) conclude that a better measurement precision is achieved by
using more data and suggests sampling of large proportions of the day per
subject.
Nordander et al. (8) investigated measurement errors and individual varia-
tion, when using inclinometry and goniometry to do technical exposure as-
sessments within homogeneous work tasks. Supported by others (156-160),
they concluded that the measurement errors were small, and as a result of this
20 subjects in each group would be sufficient to demonstrate potential differ-
ences in exposure between groups.
Some have suggested that men and women doing the same task will have the
same MSD risk (141;161;162), while others have proved differently
(6;34;163). Uncertainties about the impact of physical exposure in reported
sex differences of WMSDs are still very much an issue. Locke et al. (20) con-
cluded that uncontrolled sex differences in task distribution may result in ex-
posure misclassification, leading to erroneous risk estimates, and recom-
mended subject specific exposure assessments on task level. Kennedy et al.
(21) concluded in a review that sex matters when assessing physical exposure,
but at the same time they notice that the direction and magnitude not neces-
sarily can be predicted a priori. This illustrates the importance of having as
precise an exposure measure as possible combined with a valid and reliable
outcome, but many existing studies have deficiencies in at least one of these.
An exposure matrix is basically a cross tabulation of values for different expo-
sure variables combined with tasks, occupations or industries. A third axis can
be included providing data on seasonal variations (142). Many exposure ma-
trices are constructed on a job level (8), typically using categorical classifica-
tion (164;165). The exposures of interest included in a typical job exposure
matrix (JEM) are often expert assessed exclusively or in combination with
measurements. Expert assessed job exposure matrices usually don’t distin-
guish between sex within the same occupation and often professions with
similar exposure are even grouped (165-167). A serious shortcoming of JEMs
in general is the inability to capture exposure variation within a profession
(142), thereby preventing comparisons between high and low exposed sub-
jects.
This distinction is possible in a task exposure matrix (TEM). It will also allow
an identification of a specific task that might contain elements which could
potentially cause a high risk of WMSDs. In the same manner, the TEM will al-
low comparisons of task characteristics of men and women, identifying poten-
tial sex differences in exposure between and within professions. A study com-
paring cumulative exposures found significant differences between the meth-
odology of the JEM and the TEM (168).
Although more precise, the TEM is also more costly to establish than the JEM.
Therefore some authors recommend a careful consideration of the purpose
and need, before constructing a TEM (169;170).
Based on previous studies it seems that recommendations for setting up a
high quality study of sex specific exposure-response relationships should at
best include a valid and precise definition of an outcome, assessed in an objec-
tive manner preferably using diagnostic testing, combined with an individual-
ly objectively and precisely measured assessment of exposure, reported on a
continuous scale, determining the aspects of intensity, duration and frequen-
cy.
With the SHARM-project we try to initiate a systematic approach by clarifying
aspects of the existing hypotheses for sex differences in the development of
WMSDs one at a time. This is accomplished using a five step set-up as illus-
trated below (Table 1).
Table 1. Common potential determinants for different incidence rates of work-related musculoskeletal disorders among men
and women, and how they are clarified in the studies.
Step 1 Step 2 Step 3 Step 4 Step 5
Potential
determinants
for sex differences in
WMSD
Different
jobs?
Different
muscular
load?
Different
tasks?
Different
physical
exposure?
Different
response?
Clarified by: Design Paper I Paper II Paper II Paper III
In steps 1 and 3 we will investigate the hypothesis that sex-segregated profes-
sions and tasks may influence the occurrence of CTS, one of the frequent
WMSDs, by determining potential sex differences within our study population.
In step 2 we will investigate aspects of the hypothesis of physiological dispari-
ties clarifying the size of sex differences in muscular entities within our study
population.
In step 4 we will investigate the hypothesis of different work strategies, tech-
niques and procedures of men and women even within the same profession,
doing the same tasks.
In step 5 we will investigate the hypothesis that women may be more vulner-
able than men at the same absolute level of exposure.
By choosing CTS as the health outcome we try to limit the influence of poten-
tial sex differences in reporting since we rely on an objective diagnostic crite-
rion.
In Denmark one third of all house painters are women and tasks are supposed
to be equally distributed between sexes. This makes the profession well suited
as material for the proposed research project. Through the Painters’ Union in
Denmark (PUD) we had access to data on practically all house painters in
Denmark. Those matching the study criteria were invited to participate.
This study was an observational study. In a laboratory setting, male and fe-
male painters had electromyography (EMG) recorded from four muscles while
performing nine predetermined common house painter tasks on a defined
area. After each task the participants were asked about their perceived exer-
tion on the Borg CR10 scale. Comparisons of absolute forces, relative load and
perceived exertion were made between men and women.
In addition to what is included in paper I, inclinometry and goniometry was
simultaneously recorded as described for paper II. This will allow future com-
parisons between laboratory and field measurements, which can enable the
use of force measurements in the field measured JEM.
This observational study consisted of two parts. Part 1: Questionnaire data
reporting task distributions for a common week in 12 predetermined tasks,
were collected from members of the PUD. Comparisons of task distributions
were made between sexes.
Part 2: In a work place setting, inclinometry measurements were made of pos-
tures and movements of the upper arms and head. Goniometry measurements
were made for postures and movements of the wrists, These were used to
construct TEMs and JEMs, comparing several variables for postures and
movements between sexes.
This was a retrospective cohort study including members of the PUD. Expo-
sure response relationships were analysed for three different exposure varia-
bles for the right wrist (individually assessed by combining task distributions
and TEMs) and an outcome of first time diagnoses of, and surgeries for, CTS
identified in the Danish registers. Effects were tested for any modification by
sex.
A power calculation showed that at a double-sided significance level of 5%
there would be an 80% probability of detecting a 15% sex difference in EMG
measurements with 32 participants divided into two groups. On the basis of
this, 16 male (mean age 25.5) and 16 female (mean age 28.3) house painters
were recruited through an advertising spot on the web side of the PUD. Only
right handed subjects without current disorders or complaints in the upper
extremity were included.
Questionnaire: 9364 members of the PUD born 1940 or later who were still
alive on March 1st 2011 were contacted by mail in April 2011 and asked to fill
out and return a questionnaire. 4957 (53 %) responded, 3124 men (50 % of
all men, mean age 49.7) and 1833 women (59 % of all women, mean age 35.2).
Measurements of postures and movements: All house painter companies in the
capital region (Region Hovedstaden) of Denmark were identified in “The Cen-
tral Business Register” which is the Danish national register containing prima-
ry data on all businesses in Denmark. They were contacted in a randomized
order and asked to participate by each company providing between one and
four painters (preferably the same amount of men and women) for personal
measurements of postures and movements of the upper extremity during a
full work day. 22 companies agreed to participate and 50 full work day meas-
urements consisting of 25 men (mean age 45) and 25 women (mean age 32)
were performed on ordinary random work days (Fridays excluded). Only right
handed subjects without current disorders or complaints in the upper extrem-
ity were included.
A cohort consisting of all members of the PUD born 1940 or latter who were
still alive on March 1st 2011. 9364 individuals were included of which 6236
(66.6 %) were men and 3128 (33.4 %) were women. Some covariates were
obtained for all cohort members from the Danish registers. Other covariates
provided by the questionnaire described for paper II were only assessable for
the responding part of the cohort (n=4957).
Standard regimes for applying EMG-electrodes were followed including shav-
ing the skin and wiping it with alcohol. In a unilateral setup for the right side,
disposable surface electrodes (Multi Bio Sensors, TX, USA) were placed in a bi-
polar configuration on m. trapezius, m extensor carpi radialis and m. flexor
carpi radialis according to recommendations by Perotto (171). Using hypo-
dermic needles two fine wire electrodes (Spes Medica, Battipaglia(SA), Italy)
were inserted into the m supraspinatus according to recommendations by
Rudroff (172). The signals were transmitted wirelessly by a Bluetooth trans-
mitter (MQ16, Marq-Medical, Farum, Denmark) to a PC were they were sam-
pled at 2048 Hz. For a complete detailed method see paper I.
Using two different standardised test positions for each muscle, MVCs were
recorded in all subjects (see Table 2 for sex specific mean values).
EMG and signals from a force transducer (Hottinger Baldwin Messtechnik,
Darmstadt GmbH, Germany) were simultaneously recorded and EMG ampli-
tude for each muscle was calibrated to an external isometric force. This was
achieved using the increasing ramp contraction methodology described by
Jonsson (173). A linear relationship was determined up to 30 % MVC.
In collaboration with the PUD a list was made covering the most common
tasks within the house painters trade. Due to some restrictions in the labora-
tory, only 9 tasks were possible for the participants to perform. These were
done in the following order: 1: Sanding (by hand) 2: painting (brush), 3:
mounting glass-felt, 4: painting wall (roll), 5: painting ceiling (roll), 6: full lev-
elling wall, 7: full levelling ceiling, 8: sanding wall (“Giraffe” dry-wall-sander),
9: sanding ceiling (“Giraffe” dry-wall-sander)(Figure 3). All tasks were done in
a predetermined area. See Figure 3
for video recordings of tasks.
Figure 3. QR-code for
link to video recordings
of measurement ex-
amples.
EMG data was filtered and visually inspected. Root mean square (RMS) ampli-
tudes were calculated and amplitude probability distribution functions
(APDF) were constructed (Figure 4) by sorting the EMG measurements in as-
cending order. For the statistical analyses three percentiles were selected,
representing different load intensities: The 90th percentile (p90), the 50th per-
centile (p50) and the 10th percentile (p10).
The EMG-to-force calibration was used to obtain a measure exerted force in
Newtons (N). The EMG amplitudes were normalised to EMGmax, describing a
relative load.
Figure 4. APDF curves of a typical subject in one of the harder tasks (#8 paper I). Light blue lines
indicate the 10th
, 50th
and 90th
percentiles.
After the completion of each task (Paper I) the participants were asked to rate
their perceived exertion using the Borg CR-10 scale (appendix I).
A questionnaire was constructed (appendix IV). This consisted of 100 items,
including questions on task distributions within a typical week since 1990, for
p10
p50
p90
the 12 most common tasks within the house painters trade. Additionally it had
questions on covariates thought to be possible confounders. To increase num-
bers of responders a news article in the magazine of the PUD promoting the
survey was made. Also a prize was offered to be drawn among responders and
two friendly reminders were sent out.
Measurements of postures and movements for the wrists were made using
biaxial goniometers (SG75, Biometrics Ltd, Newport, UK). Postures and
movements for the head and upper arms were measured using triaxial incli-
nometers (Logger Teknologi HB, Åkarp, Sweden). The signals were recorded
by person worn data loggers (Logger Teknologi HB, Åkarp, Sweden) sampling
at 20 Hz. Data was recorded for a full work day and analyses were performed
for both sides. The technical measurements were performed using the meth-
odology described by the group of Hansson et al. (156-159;174-176).
A log book was constructed covering the 12 most common tasks in the house
painters trade (appendix II). In the log book participants in goniometry and
inclinometry measurements wrote down the clock time for changes between
tasks. They were given a clock with a large digital display in order to secure a
precise indication for changes in tasks.
Several variables were computed for the 12 tasks individually (TEM) and for
total work and pause (JEM)(Tables 5 and 6).
One of the keystones in Danish register research is the Danish Civil Registra-
tion System (CRS). Since 1968 every person with residency in Denmark has
been assigned a unique ten digit personal identification number (PIN) consist-
ing of data on date of birth (first 6 digits) followed by a sex-specific random
number (last 4 digits). This unique individual number enables linkage of regis-
ter information on multiple aspects (177). Data on diagnoses and surgeries for
CTS was obtained from the Danish National Patient Register (DNPR) and the
Danish National Health Service Register (NHSR).
A measure of work duration in a given year was constructed by combining
data from the Danish Register for Evaluation of Marginalisation (DREAM) with
the Integrated Database for Labour Market Affiliation for Persons (IDAP).
Information on births was obtained from the DNPR.
In preparation for the Poisson regression analyses the data was tabulated into
blocks of risk time using a public accessible macro (178) based on the princi-
ples of a Lexis diagram (179).
In all three studies, the level of significance was set to 5 %. All analysis was
performed in SAS 9.2 and 9.3 (SAS Institute Inc., Cary, NC, USA).
Sex differences in relative muscular load and exerted forces were analysed
using a two factor mixed model. This examined the task*sex interaction and
dependencies of sex and tasks. Analysis for sex differences in ratings of per-
ceived exertion (Borg CR-10) was made using an un-paired double sided t-test
with unequal variance. Pearson’s correlations coefficient was used to test for
correlation between Borg ratings and %EMGmax.
Sex differences in task proportions within each age-group were tested using
Cochran-Mantel-Haenszel statistics. An unpaired double sided t-test with post
hoc Bonferroni correction was used to test for sex differences in task expo-
sures. A paired double sided t-test was used to test for differences between
sides.
The effect of three exposure variables: median velocity, MPF (average fre-
quency of wrist movements used as a measure of repetitiveness) or non-
neutral postures (% time exceeding 45° flexion/extension or 20° ulnar radial
deviation), on CTS diagnoses and surgeries was analysed using a log-linear
Poisson regression model adjusted for potential confounders. We examined
the sex*exposure interaction for any modifying effects. The exposure intensi-
ty*duration interaction was also tested.
The effect of pregnancies was also tested in the models.
Analyses were performed both on questionnaire responders and on the full
cohort as well as stratified by sex.
Sensitivity analyses were made using only the outcome reported in DNPR.
All parts of the study were conducted in accordance with current international
ethical standards. Participants in measurements gave informed written con-
sent.
First part of the study (Paper I) was approved by the Regional Scientific Ethics
Committee (j.no.:H-3- 2012-157). Other parts of the project (paper II and III)
were assessed by the Regional Scientific Ethics Committee (j.no.:H-C-FSP-
2010-036), which concluded that these investigations were not notifiable ac-
cording to Danish laws in this field (Committee Act).
The Danish Data Protection Agency gave permission to store data regarding
every aspect of the project (j.no.:2010-41-5325).
The Danish Health and Medicines Authority approved the project for use of
data from the NHSR. (J.nr. 7-505-29-1947/1).
Men were significantly stronger than women in all measurements of absolute
force reported in Newton (N) (P < 0.001). On average men were 50-70 %
stronger than women.
No significant differences were found between men and women in EMGmax
reported in millivolts (Table 2)
The duration of the tasks varied between all participants, but no sex difference
was found in total duration of the nine tasks.
For all muscles, percentiles and task, women had a higher relative muscular
load (with the exception of task 9, percentile 10 for m. trapezius).
There was no significant interaction between sex and task.
Significant effects of sex adjusted for task were found in all three percentiles
for m. supraspinatus, m. extensor carpi radialis and m. flexor carpi radialis. No
-
109 21 66 15 <0.001 1.40 0.63 1.10 0.56 0.169
268 39 176 35 <0.001 1.44 0.66 1.14 0.59 0.186
467 120 299 137 <0.001 0.72 0.51 0.66 0.45 0.705
268 39 176 35 <0.001 0.99 0.59 1.10 0.60 0.595
203 56 125 34 <0.001 1.31 0.56 1.00 0.58 0.136
500 92 333 42 <0.001 1.00 0.59 0.80 0.46 0.280
230 51 135 42 <0.001 1.36 0.56 1.10 0.40 0.143
500 92 333 42 <0.001 0.58 0.18 0.51 0.20 0.273
significant effect was found in any percentiles for m. trapezius (Table 3). All
significant effects were caused by women being exposed to a higher load than
men.
187 b
(P=0.001)
131 to 266 156 b
(P=0.003)
118 to 207 131 b
(P=0.005)
109 to 158
119
(P=0.339)
83 to 171 120
(P=0.162)
93 to 155 127
(P=0.055)
99 to 162
180 b
(P=0.006)
120 to 270 164 b
(P=0.009)
114 to 237 149 a
(P=0.019)
107 to 207
159 a
(P=0.012)
112 to 225 140 a
(P=0.037)
102 to 193 136 a
(P=0.043)
101 to 183
a 0.010 < P < 0.050
b 0.001 < P < 0.010
The interaction between sex and task was significant for a single case of the
10th percentile for m. trapezius. Post hoc analysis revealed it to be in task 9
(sanding ceiling with the “Giraffe” dry-wall-sander).
Significant effects of sex adjusted for task were only found in the 50th percen-
tile for m. trapezius, (Table 4). The significant effect was caused by women
exerting less force than men.
111
(P=0.564)
78 to 158 95
(P=0.733)
70 to 129 80
(P=0.145)
62 to 103
70
(P=0.059)
48 to 101 70a
(P=0.043)
49 to 99 74 (P=0.113) 50 to 108
104
(P=0.831)
69 to 158 95
(P=0.786)
67 to 136 86
(P=0.379)
62 to 121
116
(P=0.438)
79 to 171 103
(P=0.888)
71 to 148 99
(P=0.968)
70 to 141
a 0.010 < P < 0.050
On average women rated their perceived exertion higher than men in all tasks.
In tasks 5-9 the sex difference was statistically significant (P < 0.05)(Figure 5).
Figure 5. Ratings of Borg CR-10 scale for perceived exertion. * indicates significant differences be-
tween men and women
Comparing male and female age specific task distributions ascertained from
the questionnaire, several differences were observed although none exceeded
± 4 %. Men had higher proportions than women in the tasks usually consid-
ered the hardest (figure 6).
Figure 6. Sex differences in mean task proportions, by age-group. Women constitute the reference-group.
* Indicates a statistically significant (p<0.05) sex difference in the specific age-group.
Removing wallpaper
Levelling
Sanding (hand)
Sanding (giraffe)
Painting (brush)
Painting (roll)
Spraying
Hanging wallpaper
CoveringPauses
DrivingOther
Sex
-diff
eren
ces
in %
-4
-2
0
2
417-25 years26-35 years36-45 years46-55 years56-70 years
0
2
4
6
8
10
12
1 2 3 4 5 6 7 8 9
female
male
*****
*
*****
*
****
*
** **
*
****
*
**
*
*****
*****
****
** **
* *
****
**
Table 5 shows the TEM for postures and movements of the right wrist for each
sex. “Total work” represents the JEM. No statistically significant sex differ-
ences were observed. For both sexes, there were clear differences in expo-
sures between tasks. For example, the median velocity for flexion/extension
during painting (brush) was approximately 4 °/s less than for sanding (hand)
for both men and women. Some measures seem to reflect the same task prop-
erties to a great extent. For example, the 50th percentiles for flexion/extension
and non-neutral postures showed the same difference between men and
women within tasks.
Table 5. Task exposure matrix for postures and movements of the right wrist for each sex. Data is displayed for the 7 tasks that constitute the work. Additionally, data is shown for total work and pause (job exposure matrix). For flexion/extension and ulnar/radial deviation, positive angles denote flexion and ulnar deviation, respectively, and negative angles extension and radial-deviation, respec-tively, [SD=standard deviation; MPF= mean power frequency]. Continues on next page. Levelling Sanding (hand) Painting (brush) Painting (roll)
Mean SD Mean SD Mean SD Mean SD Flexion/extension Percentile (°) 10th Men -57 16 -47 6 -50 12 -43 11
Women -46 8 -51 9 -52 10 -47 9 50th Men -26 10 -15 3 -23 10 -17 11 Women -18 8 -26 13 -25 7 -23 8 90th Men 0 10 10 2 3 9 8 13 Women 6 8 0 11 5 9 6 11 Range of
motion 5th-95th
Men 75 10 74 9 69 10 65 12 Women 67 4 65 11 74 14 70 15 Median velocity (°/s) Men 18.1 8 19.3 5 15.5 7 17.6 6 Women 21.2 7 19.1 4 15.3 5 16.3 5 Repetitiveness (MPF;
Hz) Men .28 .05 .29 .04 .26 .06 .30 .08
Women .33 .05 .28 .06 .25 .04 .28 .06 Ulnar/radial deviation Percentile (°) 10th Men -8 5 -19 5 -15 10 -18 9 Women -13 8 -21 16 -21 11 -22 9 50th Men 9 3 -4 5 1 11 -2 7 Women 3 6 -4 16 -5 10 -5 9 90th Men 27 5 11 4 18 12 17 10 Women 19 6 14 12 12 10 12 8 Range of
motion 5th-95th
Men 44 11 40 3 42 8 44 9 Women 41 8 44 7 44 5 43 6 Median velocity (°/s) Men 10.9 6 11.5 3 10.8 6 12.9 7 Women 12.3 5 15.3 6 9.9 3 13.1 6 Repetitiveness (MPF;
Hz) Men .29 .06 .33 .04 .28 .06 .30 .08
Women .33 .06 .30 .06 .27 .06 .31 .07 Combined wrist postures Non-neutral postures
(% time) Men 43 16 20 2 33 19 28 14
Women 26 9 39 21 36 13 32 13 Number of recordings Men 5 - 5 - 14 - 13 - Women 7 - 8 - 17 - 15 - Mean recording duration (minutes)
Men 82 - 76 - 162 - 138 - Women 127 - 51 - 149 - 102 -
Table 5 Continued. Task exposure matrix for postures and movements of the right wrist for each sex. Data is dis-played for the 7 tasks that constitute the work. Additionally, data is shown for total work and pause (job exposure matrix) . For flexion/extension and ulnar/radial deviation, positive angles denote flexion and ulnar deviation, re-spectively, and negative angles extension and radial-deviation, respectively, [SD=standard deviation; MPF= mean power frequency] Covering Driving Other Total work Pause
Mean SD Mean SD Mean SD Mean SD Mean SD Flexion/extension Percentile (°) 10th Men -42 14 -46 14 -45 18 -48 12 -44 17
Women -52 12 -47 13 -48 8 -50 8 -48 8 50th Men -17 18 -20 13 -20 14 -20 11 -15 12 Women -23 13 -13 9 -21 11 -22 8 -22 10 90th Men 10 15 5 13 5 12 7 11 12 15 Women 5 9 11 8 7 10 7 8 9 14 Range of
motion 5th-95th
Men 66 20 67 12 65 11 72 7 70 17 Women 72 13 76 14 74 9 75 10 74 11 Median velocity (°/s) Men 13.3 8 10.0 5 14.0 8 14.5 5 5.5 4 Women 14.2 8 7.9 3 14.5 6 14.6 4 4.8 4 Repetitiveness (MPF;
Hz) Men .29 .08 .28 .06 .29 .03 .27 .04 .29 .06
Women .26 .06 .24 .05 .28 .06 .27 .04 .20 .03 Ulnar/radial deviation Percentile (°) 10th Men -9 8 -9 6 -13 11 -15 9 -15 9 Women -20 12 -17 9 -20 11 -19 11 -18 10 50th Men 6 9 6 6 4 10 1 9 0 9 Women -4 8 -2 8 -4 11 -2 10 -3 10 90th Men 19 10 21 7 19 9 18 9 14 9 Women 13 8 13 7 11 10 14 9 12 10 Range of
motion 5th-95th
Men 36 10 38 9 42 7 42 8 37 8 Women 42 9 38 8 41 6 43 6 39 7 Median velocity (°/s) Men 8.1 4 5.7 3 8.8 4 9.0 3 3.6 3 Women 8.9 7 4.8 1 8.9 4 9.2 3 3.2 2 Repetitiveness (MPF;
Hz) Men .31 .06 .29 .08 .27 .03 .28 .05 .21 .06
Women .28 .07 .25 .06 .28 .06 .28 .06 .20 .05 Combined wrist postures Non-neutral postures
(% time) Men 24 22 29 18 31 18 30 15 22 14
Women 34 16 25 9 28 17 32 12 31 15 Number of recordings Men 12 - 8 - 10 - 24 - 23 - Women 16 - 8 - 15 - 25 - 25 - Mean recording duration (minutes)
Men 46 - 45 - 94 - 280 - 48 - Women 55 - 41 - 78 - 309 - 59 -
Table 6 shows the TEM for postures and movements of the head and right
upper arm for each sex; again, JEMs are presented as well. There were no sta-
tistically significant sex differences. Between-minute variation was higher for
“total work” than for any of the tasks that constituted the work. This shows
that, unlike the rest of the exposure measures, job exposures in terms of be-
tween-minute variation cannot even approximately be derived by a straight
forward time weighting of task exposures.
Job exposures differed statistically significantly (p<0.05) between left and
right wrists (see appendix III for left side TEMs and JEMS) .For flex-
ion/extension and ulnar/radial deviation, both men and women had a higher
median velocity and MPF on the right side; the same was present for median
velocity of shoulder elevation.
Table 6. Task exposure matrix for postures and movements of the head and right upper arm for each sex. Data is displayed for the 7 tasks that constitute the work. Additionally, data is shown for total work and pause. For flex-ion/extension, positive angles denote flexion and negative angles extension. [SD=standard deviation]. Continues beneath. Levelling Sanding (hand) Painting (brush) Painting (roll) Covering
Mean SD Mean SD Mean SD Mean SD Mean SD Head inclination Percentile (°) 1st Men -46 12 -39 14 -45 13 -53 12 -26 22
Women -40 6 -40 17 -45 14 -53 15 -23 15 50th Men 18 14 24 8 16 16 8 13 25 16 Women 16 5 18 11 15 11 7 22 27 12 90th Men 64 11 54 6 54 15 50 12 52 17 Women 55 8 55 11 53 14 54 12 54 12 Right upper arm elevation 99th percentile (°) Men 136 11 123 12 123 12 121 23 89 27 Women 131 14 124 30 127 18 126 21 95 22 >90° (% time) Men 15 8 6 2 12 9 8 7 2 2 Women 9 4 11 6 12 8 11 7 3 4 Within-minute varia-
tion (°)a Men 75 18 62 2 68 15 65 17 42 10
Women 73 16 75 21 70 20 68 16 47 11 Between-minute
variation (°)a Men 29 5 29 6 28 6 26 9 18 8
Women 31 5 26 12 29 7 27 7 19 8 Median velocity (°/s) Men 51.6 16 59.9 13 47.0 17 52.5 19 47.0 24 Women 59.9 20 66.4 25 43.2 14 50.8 13 48.4 25 Number of recordings Men 5 - 6 - 17 - 14 - 12 -
Women 7 - 8 - 17 - 15 - 16 - Mean recording duration (minutes)
Men 88 - 95 - 158 - 130 - 52 - Women 158 - 51 - 149 - 102 - 55 -
aThe measures of variation were calculated from the 5
th-95
th interpercentile range for each minute
Table 6. Continued. Task exposure matrix for postures and movements of the head and right upper arm for each sex. Data is displayed for the 7 tasks that constitute the work. Additionally, data is shown for total work and pause. For flexion/extension, positive angles denote flexion and negative angles extension. [SD=standard deviation] Driving Other Total work Pause Mean SD Mean SD Mean SD Mean SD
Head inclination Percentile (°) 1st Men -16 10 -38 13 -45 14 -22 14
Women -18 4 -40 12 -47 12 -21 14 50t
h Men
15 14 19 15 17 14 16 13
Women 13 9 17 11 17 8 13 12 90t
h Men
44 14 51 14 52 13 43 15
Women 40 11 51 11 53 10 36 14 Right upper arm elevation 99th percentile (°) Men 90 17 121 25 127 13 90 27 Women 94 23 126 15 128 15 84 23 >90° (% time) Men 4 8 6 5 9 4 2 3 Women 3 6 10 11 9 5 1 1 Within-minute varia-
tion (°)a Men 39 11 59 20 62 10 32 11
Women 46 16 62 13 62 11 30 15 Between-minute
variation (°)a Men 18 6 28 6 31 5 20 7
Women 19 3 30 5 32 6 22 7 Median velocity (°/s) Men 33.6 12 38.5 18 42.9 12 17.0 19 Women 29.4 11 43.9 18 43.7 13 10.1 9 Number of recordings Men 9 - 10 - 25 - 24 -
Women 8 - 15 - 25 - 25 - Mean recording dura-tion (minutes)
Men 51 - 124 - 313 - 52 - Women 41 - 78 - 318 - 61 -
aThe measures of variation were calculated from the 5th-95th interpercentile range for each minute
From the DNPR and the NHSR, cases of CTS outcomes of first time diagnoses
and first time CTS-surgery respectively were identified for use in the analyses
(Figure 7).
Figure 7. Flow chart of CTS diagnoses (1994-2011) and CTS-surgery (1996-2011) in the painters cohort (N=9364)
In the total population the female/male prevalence ratios of CTS diagnoses
and surgery were 2.6 and 2.8 respectively, and the corresponding incidence
rate ratios were 3.6 and 4.0 (Table 7).
Table 8 shows the results of survival analyses from models with the wrist ve-
locity as the measure of exposure intensity and work duration in the previous
year, sex, age, BMI, fractures near the wrist, comorbidity and questionnaire
response status as the other explaining factors. We omitted seniority as a co-
variate due to a high correlation with age (see below). Owing to missing values
among questionnaire responders, mainly to the task distribution question, the
analysis of questionnaire responders (model 2) is based on 4198 observations
(44.8% of the total material). Crude incidence rate ratios (IRR) were similar to
the IRR’s in the models with mutual adjustment except for sex and age. How-
ever, when these two covariates were mutually adjusted, their IRRs became
similar to those of the other models. This pattern was expected and reflects
the composition of the material with men being older and having less CTS than
women.
Increasing median velocity was associated with a statistically significantly
higher IRR for both CTS diagnoses and surgery. The IRR estimates were quite
stable across models, approximately 1.30 per 1 °/s, with the exception of
model 2a (men only) where it was a little lower (1.22) and not statistically
significant. Work proportion in the previous year had no significant effects.
Table 7. Study population characteristics based on register information. Total and for questionnaire responders.
Diagnoses of carpal tunnel syndrome* Surgery for carpal tunnel syndrome*
Number
of cases
Risk
Time
(years)
Incidence
rate per
10.000
years
Study
period
preva-
lence
(%)
Number
of cases
Risk
Time
(years)
Incidence
rate per
10.000
years
Study
period
preva-
lence
(%)
Total (n=9364) 230 104308 22.05 2.5 162 104792 15.46 1.7
Men (n=6236) 101 76694 13.17 1.6 66 76969 8.57 1.1
Women (n=3128)
((n=3128)
129 27614 46.72 4.1 96 27823 34.50 3.1
Questionnaire re-
sponders (n=4957)
162 60993 26.56 3.3 116 61332 18.91 2.3
Men (n=3124) 71 43310 16.39 2.3 48 43509 11.03 1.5
Women (n=1833) 91 17683 51.46 5.0 68 17823 38.15 3.7
*Records from the Danish National Patient Register and the Danish National Health Register during the study
period 1994-2011.
The effect of sex was highly statistically significant with IRR estimates for
women versus men ranging from 4.6-4.9 for CTS diagnoses and 6.0-6.1 for CTS
operations. The IRR increased significantly with increasing age. The IRR esti-
mates increased with BMI, wrist fractures and co-morbidity. The estimates of
BMI effects were of similar size and statistically significant for men and wom-
en and for the two CTS outcomes. The IRR estimates of wrist fractures and
comorbidity were less stable with scattered statistically significant effects.
There seemed to be different effects of wrist-near fractures and comorbidity
for men and women (models 2a and 2b).
The IRR estimates for questionnaire responders were higher than for non-
responders (model 1 and model 3).
The IRR estimates of work proportion, sex and age were similar for the total
material (model 3) and for questionnaire responders (model 2).
Table 8. Incidence rate ratios (IRR) of CTS diagnoses and CTS surgery. Models with median velocity of flexion/extension of the right
wrist, work proportion, sex, age, BMI, fractures near the wrist, comorbidity and questionnaire response as explaining factors, de-
pending on the model. Median velocity of
flexion/ extension
of the wrist
(Per 1 °/s)
Work
proportion
previous
year
Sex
(women/
men)
Age
(per 10
years)
Body mass
index
(per 5 units)
Fracture
near the
wrist
(yes/no)
Co-
morbidity
(yes/no)
Question-
naire
response
(yes/no)
Model 1. Crude associations (n=4420)
CTS diagnosis
IRR 1.351 0.752 3.552 0.992 1,251 1.471 1.371 1.692
95 % CI 1.10-1.64 0.54-1.05 2.73-4.60 0.90-1.09 1.09-1.43 0.97-2.24 0.96-1.96 1.27-2.25
P-value 0.0035 0.049 <0.0001- 0.89 0.0012 0.070 0.083 0.0003
CTS surgery
IRR 1.39 0.75 4.02 1.02 1.29 1.22 1.45 1,79
95 % CI 1.10-1.76 0.50-1.11 2.94-5,50 0.91-1.14 1.11-1.50 0.72-2.07 0.96-2.20 1.27-2.51
P-value 0.0051 0.15 <0.0001 0.79 0.0009 0.56 0.079 0.0009
Model 2. Mutually adjusted associations.
All questionnaire responders (n=4198)
CTS diagnosis
IRR 1.29 .80 4.64 1.25 1.27 1.57 1.40 -
95 % CI 1.07-1.56 0.52-1.22 3.21-6.71 1.08-1.44 1.12-1.45 1.03-2.40 0.97-2.03 -
P-value 0.0085 0.29 <0.0001 0.0030 0.0002 0.035 0.075 -
CTS surgery
IRR 1.32 0.86 6.03 1.41 1.32 1.33 1.42 -
95 % CI 1.06-1.65 0.52-1.41 3.89-9.34 1.18-1.67 1.14-1.52 0.78-2.25 0.92-2.19 -
P-value 0.014 0.55 <0.0001 0.0001 0.0002 0.30 0.11 -
Model 2a. Mutually adjusted associations.
Male questionnaire responders (n=2596)
CTS diagnosis
IRR 1.22 0.72 - 1.13 1.25 2.12 1.21 -
95 % CI 0.86-1.72 0.39-1.33 - 0.91-1.41 1.01-1.55 1.21-3.70 0.72-2.04 -
P-value 0.27 0.29 - 0.26 0.0394 0.0085 0.48 -
CTS surgery
IRR 1.30 0.80 - 1.33 1.33 1.47 1.10 -
95 % CI 0.85-1.99 0.38-1.68 - 1.00-1.76 1.06-1.66 0.69-3.13 0.58-2.08 -
P-value 0.23 0.56 - 0.048 0.015 0.33 0.77 -
Model 2b. Mutually adjusted associations.
Female questionnaire responders (n=1602)
CTS diagnosis
IRR 1.32 0.80 - 1.34 1.30 1.13 1.60 -
95 % CI 1.05-1.64 0.44-1.46 - 1.11-1.62 1.11-1.53 0.58-2.17 0.96-2.69 -
P-value 0.016 0.47 - 0.0026 0.0015 0.73 0.073 -
CTS surgery
IRR 1.32 0.85 - 1.45 1.31 1.22 1.78 -
95 % CI 1.02-1.71 0.43-1.71 - 1.16-1.80 1.09-1.58 0.58-2.55 1.003-3.16 -
P-value 0.037 0.65 - 0.0010 0.0041 0.60 0.049 -
Model 3. Mutually adjusted associations.
Total cohort (n=9364)
CTS diagnosis
IRR - 0.84 4.63 1.25 - - - 1.44
95 % CI - 0.59-1.20 3.41-6.29 1.11-1.41 - - - 1.08-1.92
P-value - 0.34 <0.0001 0.0002 - - - 0.0013
CTS surgery
IRR - 0.86 5.71 1.34 - - - 1.47
95 % CI - 0.57-1.38 3.95-8.24 1.16-1.54 - - - 1.04-2.09
P-value - 0.87 <0.0001 <0.0001 - - - 0.028
Bold highlights statistically significant IRRs (p<0.05). 1. Questionnaire responders (n=4420 for wrist velocity, n=4773 for BMI, n=4825 for fractures near
the wrist, n=4787 for co-morbidity). 2. Total material (n=9364)
Table 9 shows the results from the models with the other two exposure inten-
sity variables (MPF and non-neutral postures), analysed in the same models as
in Table 8. In Table 9 the results for the other covariates were omitted justi-
fied by a high similarity to the results presented in Table 8. Increasing MPF
was associated with statistically significantly higher IRR estimates for both
CTS outcomes. An increase in MPF by 0.01 Hz resulted in IRRs ranging from
1.30 to 1.40. The effect was statistically significant in all models except model
2a (male questionnaire responders).
Combined non-neutral postures did not have any statistically significant ef-
fects, either on CTS diagnoses or operations.
Table 9. Incidence rate ratios of CTS diagnoses and CTS surgery. Models with mean power frequency and non-neutral postures for the right wrist, work proportion, sex, age, BMI, fractures near the wrist, comorbidity and questionnaire response as explaining factors, depending on the model. Same models as in Table 8. The results for explaining factors other than the two exposure intensity measures were similar to those of Table 8 and therefore omitted from this table.
Mean power frequency (Hz)
(per units of 0.010 Hz*)
Non-neutral wrist postures (% time)
Model 1. Crude associations (n=4420)
CTS diagnosis
IRR 1.38 .98
95 % CI 1.12-1.70 0.88-1.08
P-value 0.0026 0.67
CTS surgery
IRR 1.33 0.97
95 % CI 1.04-1.71 0.86-1.10
P-value 0.025 0.62
Model 2. Mutually adjusted associa-tions. All questionnaire responders (n=4198)
CTS diagnosis
IRR 1.39 0.98
95 % CI 1.13-1.71 0.88-1.08
P-value 0.0022 0.66
CTS surgery
IRR 1.34 0.97
95 % CI 1.04-1.72 0.86-1.09
P-value 0.023 0.59
Model 2a. Mutually adjusted associa-tions. Male questionnaire responders (n=2596)
CTS diagnosis
IRR 1.33 1.07
95 % CI 0.57-3.13 0.94-1.21
P-value 0.51 0.31
CTS surgery
IRR 1.30 1.02
95 % CI 0.46-3.66 0.87-1.20
P-value 0.62 0.80
Model 2b. Mutually adjusted associa-tions. Female questionnaire responders (n=1602)
CTS diagnosis
IRR 1.40 0.86
95 % CI 1.13-1.74 0.74-1.002
P-value 0.0024 0.053
CTS surgery
IRR 1.35 0.91
95 % CI 1.04-1.75 0.77-1.08
P-value 0.025 0.28
* Almost equal to the interquartile range. Bold highlights statistically significant IRRs (p<0.05)
All of the models presented in Table 8 and Table 9 were also examined using
work proportions accumulated over two and five years instead of only the
previous year. The results of these analyses showed very similar effect esti-
mates and confidence limits as the analyses with work proportion in the pre-
vious year only.
There were no statistically significant interaction terms between exposure
intensity or duration variables and sex. Neither were the interaction terms
between the variables of exposure intensity and duration.
Seniority was not included in the models (Tables 8 and 9) since it had a very
high correlation (0.83) with age. The correlations were similar among men
(0.82) and women (0.77). When seniority was included in the models instead
of age, the effects of seniority were insignificant and the estimates were close
to 1. When age and seniority were included in the same model, the effect was
statistically significant for age but not for seniority in all models except model
3 for wrist velocity diagnoses (P=0.04). The estimates increased for age and
decreased for seniority when both variables were included in the models.
Sensitivity analyses were applied, limiting the outcome to diagnoses and sur-
gery listed only in the DNPR, slightly increased the IRRs and level of signifi-
cance (data not shown).
This thesis aims to establish a precise exposure assessment examining sex
differences in load, force, task distributions, postures and movements of the
upper extremity using a systematic approach. Also, it explores whether an
exposure-response relationship is present between exposures of the wrist and
CTS, and to what degree this will be influenced by sex. The first study showed
that the relative muscular load was significantly higher in women compared to
men and these objectively measured differences corresponded well to subjec-
tive ratings of physical exertion. Minimal sex differences were found in exert-
ed force, with men using more force than women. In the second study, self-
reported task distributions only showed minor sex differences and no signifi-
cant differences were found between the sexes in upper extremity postures
and movements. The third study found an exposure-response relationship
between median wrist velocity and CTS, and MPF and CTS, but not between
non-neutral postures and CTS. There was no significant effect of work propor-
tion accumulated over 1, 2 or 5 years prior to a CTS event. These results imply
that un-accumulated median velocity and MPF may be work related risk fac-
tors of CTS.
The risk of CTS was significantly different between men and women with
comparable exposures. However, the effect of the exposure was not modified
by sex.
Using the term sex as a common denominator for all properties concerning
men and women has not been widely accepted. However, at a recent symposi-
um on gender work and health (OBEL Summer School, Montreal, Canada)
there was widespread agreement among international researchers working
within the field of WMSDs and sex/gender for the need of an overall term in-
corporating both sex and gender, since aspects of one or the other are usually
mutually influenced. Until such a term has been agreed upon, it should be rec-
ommended to clearly define the preconception.
EMG has been used extensively to assess intrinsic exposure in work settings.
The quantification of muscular load has traditionally been performed using
%EMGmax (173;180-183). However, this measure of relative muscular load
does not express the exerted forces being applied during work. For this reason
we applied an EMG-to-force calibration comparing a certain level of relative
muscular load to absolute forces. It has been argued that the relationship be-
tween EMG and forces only can be assumed to be linear in the first 30% of the
MVC (184). This did not cause a problem in the present study since the vast
majority of monitored tasks did not exceed 30% of EMGmax (Paper I). The re-
ported forces in Newtons should be interpreted with caution since the actual
forces exerted in a task are composed of more elements than the muscle we
measured. It should, however, be considered valid for use when comparing
men and women doing identical tasks.
Among questionnaire responders men had higher age and seniority compared
to women. Hence different trends of task composition in certain time periods
could potentially introduce bias between men and women. To control for this
we stratified by age, which also diminished the risk of bias as a result of age
difference between responders and non-responders. The retrospective nature
of a questionnaire will always per se introduce a potential risk of recall bias. It
is well described how demanding tasks are often overestimated in self-
reports, especially in combination with complaints (143;147;149;150). We
assumed that any potential recall bias would be equal among sexes in the re-
spective age groups, which would still allow for valid sex comparisons of task
distributions (Paper II). Constructing the questionnaire we were aware not to
use suggestive phrases indicating our interest in CTS, since this can influence
symptomatic participants (148).
The methods applied for the technical measurements in this thesis have been
shown to be both valid and reliable (156-159;174-176).
The tasks performed during the field measurements were completely random.
Because of the relatively few individuals in each group (25 males and 25 fe-
males) this resulted in some tasks having less than the recommended mini-
mum of five recordings (174). These tasks were pooled together in the task
“other”. The optimal strategy would have been to keep doing measurements
until the desired number had been acquired, but that was beyond the re-
sources of the study. Duration and occurrence of tasks in the measurements
were self-reported using a log book, a method which has been criticised for
being imprecise (185). In order to adjust for potential overflow we trimmed
the measurements of each task by two minutes at each end. Only minor
changes occurred as a consequence of this, indicating an overall precise re-
porting.
Both the technical measurements and CTS case definitions were assessed us-
ing recommended valid methods, the resulting data in the TEMs and JEMs, as
well as the CTS diagnoses, can be considered generic, which enables compari-
sons with others studies evaluating the potential effect of postures and
movements on CTS using the same or equally valid and reliable methods. The
need for similar designs is highlighted by several authors, arguing that this
will allow a more valid pooling of data (100;186;187).
The scope of the study was to have an individually assessed task distribution
(Paper II) as a prerequisite for being included in the exposure response anal-
yses (Paper III). For these reasons only current members of the PUD, who we
could contact, were included in the cohort. This introduces the possibility of a
potential healthy worker bias if individuals that are more susceptible to CTS
have left the profession as a result of their disorder. However, the prevalence
and incidence rates reported in our study resemble those reported by others
(80;88;91;138;188). This indicates that many have continued working as
house painters despite CTS. Unfortunately we had no information on profes-
sion changes due to CTS. Atroshi et al. (61) reported a CTS incidence rate of
18.2 for men and 42.8 for women (per 10.000 person years) in the general
population. In comparison, we found 16.4 in men and 51.5 in women. These
results could indicate healthy worker bias at least for men. Painters with CTS
who had left the profession before our investigation will contribute to an at-
tenuation of prevalence and incidence rates, and potentially bias effect esti-
mates toward unity.
There is a large variation in the quality of studies that have investigated CTS.
Many studies are restricted by design using cross sectional or case control
methodology. Even though some studies have used a prospective design
(9;62;94;100;103;113;122;127;133) van Rijn et al. (79) demonstrated in a
review that the quality of studies had not improved over two decades. Out of
37 studies included in a meta-analysis investigating the association between
exposures in the work place and CTS, Barcenilla et al. (92) reported 28 cross-
sectional studies, 5 case-control studies and 4 cohort studies. Even though
exposure was assessed retrospectively in our design, regarding outcome we
followed the cohort in the Danish registers which are of high quality and gen-
erally thought of as valid. Therefore the registration of CTS incidences was
prospective.
A potential limitation of our study is that cases had had their outcome before
rating their task distribution which is the basis for the individual exposure
assessment. This enables the risk of differential misclassification, since it is
well established that persons with complaints have higher reporting of de-
manding tasks (143;147;149). However, comparing the task distribution of
CTS cases with that of the remaining responders did not show a consistent
increasing pattern in the most demanding tasks (Paper III). Also, it is unlikely
that questionnaire responders would be able to know which tasks had been
measured to be the most demanding in terms of wrist velocity, MPF and non-
neutral postures and, finally, the cases in the cohort were scattered over the
entire study period which meant that the majority of cases would most likely
have been symptom free at the time of the questionnaire.
The drawback of this study is that it has been very costly both in terms of time
and money. This dilemma has been the focus of some researchers who com-
pared expenditures of exposure assessments using inclinometry or direct ob-
servation in data collection and data analyses, respectively. Inclinometers per-
formed consistently better than observation in both data steps, but inclinome-
try is an expensive way of collecting data (189-191). Novel biomechanical
systems that can be applied to the participants, who then wear it for a period
of time and return it by postal mail, may revolutionise physical exposure as-
sessments if used as a common inclinometry and goniometry set-up (192-
194).
Likewise, substantial expenses were spent on the printing, postal distribution
and scanning of the questionnaires. Online distribution of the questionnaire
could have been a cost efficient alternative. This possibility was discussed
with the PUD who advised against it, since it was their experience that only
half of their members were confident internet users.
It has been argued that too many resources may be used compared to what
may be gained (142;169), while others advocate for a precise task based expo-
sure assessment as possible, due to a potentially high risk of misclassification
when only using job titles (20).
Patil et al. (81) showed a higher CTS prevalence among dairy workers doing
one kind of task compared to dairy workers doing other tasks. This illustrates
the need for precise task based exposure assessments even within the same
profession.
Some have suggested that increasing precision for exposures on a continuous
scale leads to a decrease in observed sex differences in the workplace, insinu-
ating that most observed sex differences in occupational settings are caused
by an imprecise exposure assessment (93;141). In contrast to this belief we
have, along with others (8), shown that sex differences in CTS persist in spite
of a precise exposure assessment on a continuous scale.
Many studies have only reported exposures in hours per day or week, spent
on the movements, postures or tasks in question, per profession (79). This
methodology only gives an assessment of frequency and not intensity. There-
fore, when applied to individuals working in different professions, with differ-
ent tasks, or even with different techniques or strategies this may introduce a
bias resulting in attenuation of estimates. Different thresholds of exposures
between studies (i.e. > 1 hour, > 2 hours or > 3 hours) and reporting on a cate-
gorical scale adds to the inter-study heterogeneity and may contribute to
some of the observed differences (36). To solve these challenges a common
standard has been proposed by the ACGIH using the AL and TLV based on
hand activity level and normalized peak force (94;98;122-124;195). However,
studies using very wide categories or dichotomisation when analysing sex
differences may also be at risk of interclass confounding, for example if there
is an offset between women and men within a given exposure category (33).
As well as using subjectively assessed distributions of common tasks as a
measure of frequency, we applied a precise biomechanical measurement for
each of the tasks described, reported on a continuous scale (Paper II). Hence, a
precise continuous measure of exposure intensity was included, adding to the
overall precision of the exposure assessment.
As described earlier, CTS case definitions vary in epidemiologic research. Simi-
lar to other recent studies (61) the CTS cases included in our study were phy-
sician diagnosed. We did not have any data on whether or not cases had an
electro diagnostic test (EDT) made, but every case had a clinical interview and
a physical examination, sufficient for decision regarding treatment. In Den-
mark it is recommended that EDT is performed prior to treatment but in very
obvious cases this might be omitted. Atroshi et al. (96) demonstrated differ-
ences in CTS prevalence in the general population depending on which diag-
nostic criteria were used. For symptoms and EDT measurement it was 4.9 %;
physical examination and symptoms 3.8 %; and physical examination, symp-
toms and electro physical examination 2.7 %. Others have shown similar re-
sults (74;133). Some have hypothesised that symptoms and physical examina-
tions may capture other aspects of CTS than EDT, due to different mechanisms
(63;196;197). If the majority of our cases did not have an EDT this could po-
tentially introduce an over-reporting of CTS, but in Paper III we found CTS
prevalence and incidence rates comparable to those reported by others
(61;74;80;88;91;96;133;138;188). This indicates we used precise diagnostic
criteria. The use of symptoms and physical examination, complemented by a
high quality EDT, is recommended in the literature also for epidemiological
research (63;64;77;91).
Differences between CTS incidence rates and prevalence in different popula-
tions may, besides different case definitions and exposure measurements,
partly reflect variations in underlying non-occupational risk factors i.e. co-
morbidities. A recent meta-analysis observed a significant heterogeneity
among studies (92). In a meta-regression analysis they identified several sig-
nificant determinants, such as study design, case definitions, risk of bias score
and country.
In a review by van Rijn et al. (79) attention was directed to the discrepancy
between the proportion of studies that had only used questionnaire data to
estimate CTS (14 %) and studies that had only used questionnaires to assess
the exposure (66 %). They argue that the heterogeneity between studies
therefore should be higher for the exposure assessments than for the CTS as-
certainment. This is supported by the meta-analysis of Barcenilla et al. (92)
that showed a significant heterogeneity in exposure assessment methods be-
tween studies.
In the present study, several Danish registers were used to supply information
on diagnoses of CTS, surgery for CTS, work proportion and pregnancies. Data
reported to both the DNPR and NHSR is primarily used to settle payments
with health care providers and may have limitations. An underreporting of 5
% was reported in a 2008 estimate of surgeries to the DNPR. In relation to our
study this should be considered as non-differential misclassification which
may attenuate findings using CTS surgery as outcome. The service code used
for classifying CTS in the NHSR is not unique. We performed sensitivity analy-
sis, excluding cases from the NHSR. This revealed higher estimates, indicating
some misclassification. However, since the misclassification would be non-
differential, the NHSR was kept in the model.
Most studies that have investigated occupational risk factors for CTS have not
included sex specific analyses. Knowing that female sex is a well-established
risk factor for CTS, sex should at least be included as a confounder.
It has, in recent years, been suggested to include sex stratification in the anal-
yses of exposure response relationships instead of only adjusting for sex
(37;198-200). However this recommendation may primarily be applicable in
studies where there is uncertainty as to whether or not men and women are
equally exposed or in studies where there are large sex differences in expo-
sure. Comparable exposures between sexes assessed by objective and valid
methods should, in most settings, allow for simple adjustment. On the other
hand, it can be argued that estimates of potential covariates may be influenced
in different ways depending on sex (123;141;201;202). This effect could be
masked if only a multivariate model is used. Stratified analyses will potentially
highlight differences which can be associated to specific physiological or psy-
chosocial features of the sexes. We included sex-stratified analyses for expo-
sure response relationship (Paper III), and these resulted in higher estimates
in women for age and co-morbidities whereas men had higher estimates for
fractures near the wrist. Multilevel and cluster analyses have also been sug-
gested as suitable methods for comparing sex specific risks (203).
A Poisson log linear regression model was applied in the investigation of pos-
sible exposure response relations (Paper III). The Poisson model was chosen
since it is recommended for regression analyses of time dependent continuous
variables. Some have proposed to classify subjects in groups by anthropome-
try instead of sex (10;54). Won et al. (54) found stronger differences between
anthropometric groups than sex. A possible way to test this hypothesis would
be to apply a latent class analysis (204). This could potentially direct attention
to the most influential effect modifiers.
Seniority was not included as a confounder in the final multivariate model
(Paper III). It was initially included but it had a high correlation with age, was
insignificant and changed direction after adjustment for age. Also, based on
the literature, we did not find any indication to keep it in the model since sen-
iority is not a well-established risk factor for CTS. This fits the reports of only
recent exposure having an effect on CTS (78).
Household chores and leisure time activity have been hypothesized to influ-
ence the observed sex difference in prevalence of WMSD (4;8;199;205), alt-
hough with some conflicting evidence regarding CTS (201). We did not include
any variables regarding this issue. From our questionnaire data we learnt that
the distributions of household tasks were very traditional among the house-
painters with women doing more kitchen and cleaning chores and men doing
more versatile tasks. In leisure time activities we did not find any systematic
difference between men and women. In order to incorporate these aspects
into the exposure assessments we would have needed precise biomechanical
measurements of all the tasks in question. This would have proved too de-
manding in terms of time and finances.
Since the measurements of loads and forces were done in a laboratory setup,
the concentrated nature could suggest that the values obtained might be high-
er than what would be expected in real world settings. Hence, in order to add
this to the constructed JEMs and TEMs it would be necessary to weigh these
measurements by potential differences between goniometry and inclinometry
data obtained in the two settings.
The relative differences in muscular load could partly explain why women
could be more vulnerable than men, doing identical tasks at the same extrinsic
physical exposure. Nordander et al. (8) investigated sex differences in workers
with identical tasks. They found similar postures and movements but higher
relative muscular load in women compared to men, and a higher prevalence of
MSDs in women. They suggested that the higher risk of MSD in women could
be partly explained by the higher relative muscular activity. Assuming that
individuals are more susceptible to MSDs if a threshold of relative muscular
load is exceeded this could be a reasonable explanation for sex differences in
MSDs being caused by higher relative loads in women compared to men.
Some have suggested that anthropometrics or the design of tools (which are
usually designed to fit the average man) may be part of the explanation why
MSDs are more prevalent among women than men (4;29;54;206). This hy-
pothesis assumes that men and women exert different levels of force on the
tools. In contrast, our results show that men and women apply the same de-
gree of force when performing common house painter tasks, indicating that
the tools are not causing the sex difference in MSDs.
We found a linear relationship between Borg CR10 ratings and relative mus-
cular load (Paper I). The results in figure 5 indicate that the difference be-
tween sexes is more pronounced in the more strenuous tasks. We also found
that women experience a higher relative load when applying the same abso-
lute force in a given task. Therefore, if a task is considered high load, women
will use a higher proportion of their EMGmax compared to men, and the higher
the load, the bigger the difference in perceived exertion. Burt et al. (123)have
showed that high peak work ratings on Borg CR10 were a risk factor of CTS
and others too have used it for exposure assessment (36). This indicates that
the Borg CR10 could be used as a subjective measure of differences in per-
ceived physical exertion.
Though statistically significant, only minor sex differences were found in self-
reported task distributions, and the age stratified patterns within each task
were similar for men and women. This does not support the reports of women
having a more strenuous task composition as a result of sex segregation, lead-
ing to the development of MSDs (6;8;20;24;25;199;207). If anything, we found
the men do more of the strenuous tasks.
Our results in Paper III suggest that exposure intensity rather than exposure
duration might have an impact on the development of WMSDs. However, since
the individual measure of intensity is based on a lifetime estimate of task dis-
tributions, corresponding to a constant frequency of tasks, we cannot make
any conclusions regarding limited periods of high intensity work leading up to
a CTS incident. In practice this would require a prospective study design with
regular registrations of task frequency. Some have used company data as a
way to collect information on worker exposure (208;209). However, with the
exception of some professions with clearly defined and monotonous tasks (i.e.
truck drivers and baggage handlers), the level of detail in this data is often
limited to work time (152) or other information comparable to what can be
required from the Danish registers (210-213). Therefore they will primarily
be a measure of the work duration. Among house painters, a substantial dif-
ference in task intensity would be expected between individuals working on a
fixed scheme being paid by the hour, and workers doing piecework contracts.
The latter usually work faster in order to earn more money, so having a higher
proportion of these workers in a population will most likely increase exposure
variables related to working speed. From our questionnaire data we know
that approximately 20 % of the population was employed on piecework con-
tracts most of the time. Since the data collection in this study was carried out
during a period of recession, it was our experience that many workers had
gone from piecework contracts to being paid by the hour. This may have influ-
enced some respondents’ composition of work tasks, but since they were
asked about a typical week in their work life, we do not think this has influ-
enced the data to a greater extent. However, the technical measurements were
not done on any workers doing piecework contracts. This may have biased the
results, primarily due to the different work pace. Hence the external validity
may have been reduced, but sex differences in postures and movements
should not be influenced since the potential bias must be assumed to be equal
among men and women.
Multivariate analyses in Paper III showed a significant effect of exposure in-
tensity but not duration in terms of work proportion. This led us to conclude
that a potential decline in high risk exposure, as a preventive measure, could
result in a decreased risk of CTS. Shiri et al. (78) supports this notion by only
reporting a risk of CTS for exposure in the most recent job. In spite of having
only 45% of the potential material available for the full analyses (Model 2 in
Tables 8 and 9), very stable estimates of IRR were found when comparing the
effects of the covariates (sex, age and work proportion) to the full cohort.
Thus, we do not think that non-response would likely have introduced a selec-
tion bias.
Excessive extension was the main contributor to non-neutral postures in our
measurements. The rationale for this is to position the hand so the flexor mus-
cles can exert more power without being affected by active insufficiency. This
wrist position has been shown to produce elevated carpal tunnel pressures
which can potentially lead to CTS (214-216) but in contrast to some studies
(36;217) we did not find any effect of non-neutral postures on CTS. This could
possibly be explained by the definition of non-neutral postures (Paper II)
where our limits for deviations were in the high end of what is reported in
normative material (218). However, our findings are supported by other stud-
ies that found no relationship between non-neutral postures and increased
risk of MSDs (38;42;92;126).
Another explanation could be the choice of exposure measure. Direct meas-
urements are usually considered superior to self-reports (142) and Spielholz
et al. (145) showed a substantial difference between self-reports and direct
measurements of extreme posture duration. This discrepancy offers an expla-
nation for the inconsistency in reported findings.
We did not conduct any specific investigation on the hypothesis of women
having an overall lower threshold for reporting complaints. Nathan et al. (102)
have postulated that CTS symptoms are perceived more finely by women than
men but using the recommended diagnostic criteria for CTS as we did in our
study, the effects of this should be very limited because of objective examina-
tions. This is supported by Mondelli et al. (16) who found similar results for
men and women in clinical and electro-physical severity of CTS.
No modifying effect of sex was found for the exposure intensity and exposure
duration variables on CTS outcomes. This indicates that men and women are
equally affected by the exposure. In a study of work related risk factors for
musculoskeletal symptoms, Hooftman et al. (33) found a modifying effect of
sex. Contrary to their own expectations they found a higher vulnerability to
exposures in men. However, they could not offer any explanation for this find-
ing and although the study was prospective, both the exposure and the out-
come were subjectively assessed in a questionnaire.
The results of the Poisson regression are reported on a logarithmic IRR-scale.
When back-transformed the incidence rate at zero exposure will determine
the increase in incidence rates. Hence, as a result of higher incidence rates due
to the independent effect of sex, women will have steeper incidence rate
curves than men. This corresponds with the observed incidence rates by ter-
tiles, and similar patterns are found for both diagnoses and surgery for the
exposure variables median velocity and MPF (figures 8 and 9). However, this
sex difference in incidence rate curves is only observed, and supplementary
analyses testing for statistical significance using additive models should be
considered. The differences in exposure-response curves between men and
women may be interpreted as a difference in vulnerability in response to ef-
fects of physical exposures on CTS. These results may, however, result from
the nature of the multiplicative model we used and no firm conclusion can
therefore be made.
Figure 8. Observed incidence rates of CTS-outcomes by tertiles of wrist velocity (grey and black columns) with estimated inci-
dence rates overlaid. Estimates are based on crude associations (Model 1 in table 8)
CTS-diagnoses incidence rates by sex and tertiles of median wrist velocity
Median wrist velocity (deg./s)
8 10 12 14 16 18 20 22
CT
S-in
cide
nce
per
1000
per
son
year
s at
ris
k
0
5
10
15
20
25
FemaleMaleMaleFemale
CTS-operations incidence rates by sex and tertiles of median wrist velocity
Median wrist velocity (deg./s)
8 10 12 14 16 18 20 22
CT
S-in
cide
nce
per
1000
per
son
year
s at
ris
k
0
2
4
6
8
10
12
14
16
18
20
FemaleMaleMaleFemale
Figure 9. Observed incidence rates of CTS-outcomes by tertiles of MPF (grey and black columns) with estimated rates overlaid.
Estimates are based on crude associations (Model 1 in Table9)
CTS-diagnoses incidence rates by sex and tertiles of mean power frequency
MPF (Hz)
0.24 0.26 0.28 0.30 0.32 0.34
CT
S-in
cide
nce
per
1000
per
son
year
s at
ris
k
0
5
10
15
20
25
30
35
FemaleMaleMaleFemale
CTS-operations incidence rates by sex and tertiles of mean power frequency
MPF (Hz)
0.24 0.26 0.28 0.30 0.32 0.34
CT
S-in
cide
nce
per
1000
per
son
year
s at
ris
k
0
5
10
15
20
25
FemaleMaleMaleFemale
Many studies have claimed that sex differences in WMSDs may be caused in
part by uncontrolled sex differences in exposure (32;83;207;219). In contrast
to this belief we have by means of a precise sex specific exposure assessment,
shown that an increased risk of a common work-related upper extremity dis-
order in women compared to men persists, despite a comparable physical
exposure.
Our main strength in this study is the use of a physician-diagnosed outcome
reported independently from the technical assessed exposure.
Based on current results, individual or sex specific exposure assessment
should be recommended in order to minimise misclassification caused by un-
controlled differences in tasks.
Results obtained from a task-based exposure assessment may prove useful in
developing preventive measures due to its ability to distinguish between the
impacts of potential risk factors within a profession. Given that the absolute
incidence rates for women increased at a steeper rate with increasing expo-
sures than the incidence rates for men did, a larger potential for prevention
would exist for women than for men.
Since our population was limited to Danish house painters, interpretations of
results should be made with caution if applied to other professions.
Within the Danish house painting trade, women had a higher relative load
than men, without exerting more force. Only minor sex differences were
found in task distributions and postures and movements of the upper extremi-
ty. A systematic approach resulted in a precise assessment of physical expo-
sure especially for intensity and duration and somewhat for frequency. Infor-
mation on physician diagnosed CTS and surgery for CTS was obtained from
valid Danish registers. The IRR estimates for CTS increased significantly for
wrist velocity, and for repetitive use of the wrist, but not for non-neutral pos-
tures. Female sex had a significantly higher risk of CTS, but sex did not have
any modifying effect on the exposure variables in the applied models. Howev-
er, it is not clear to what extent these results reflect that women may be
more vulnerable to these exposures than men and to what extent they reflect
the basic assumptions of the statistical model used for the analyses.
When trying to reduce risk factors for CTS in the workplace, caution is given
regarding reduction of one hazardous element of a task without paying atten-
tion to the interrelated elements that constitute that task. Instead the focus
should be on reducing time spent on high risk tasks (8;220).
Due to a relatively low response proportion in the questionnaire, the use of
covariates in the full models was limited to half of the potential population.
This complete data analysis does not necessarily introduce bias if the data is
assumed to be missing completely at random. However, the missing data re-
duces the potential power of the analysis. A common way to resolve this is to
perform imputations. Imputation of missing data can be performed in several
ways. A very basic method is to impute the missing values with the mean of
the population or subsets thereof. More sophisticated methods can also be
applied, for example multiple imputations by chained equations. In this meth-
od numerous multiple imputed datasets have a regression analysis performed
and the estimates of these are then pooled, taking advantage of the variation
that has been generated (221). This method could be tested on our data to see
if some of the insignificant results would change if the statistical power was
increased.
The extensive data collection we obtained on Danish house painters, will allow
a similar study on shoulder disorders as the one performed on CTS. Svendsen
et al. (170) have previously showed that there are significant differences in
postures and movements of the upper arm between the tasks of Danish male
house painters. A large exposure contrast between tasks will increase the pos-
sibility of detecting any sex difference in a TEM, assuming the same contrast is
present among female house painters. Since shoulder disorders are the most
commonly reported WMSD to the NBII by both male and female house paint-
ers this would be very relevant and sex specific risk factors could potentially
be detected, allowing for prevention in both men and women.
As part of the SHARM-project we obtained full work day measurements of
postures and movements in the upper extremity in several other professions
believed to have either repetitive or strenuous tasks (Table 10). These meas-
urements were made in collaboration with Annett Dalbøge from the Danish
Ramazzini Centre, Department of Occupational Medicine, Aarhus University
Hospital, Aarhus, Denmark.
As displayed in Table 10 we obtained measurements on both men and women
in additional six professions. For all the professions listed we also have infor-
mation on diagnoses and surgery from the DNPR. Therefore we will be able to
investigate if sex differences in prevalence and incidence rates of CTS within
these professions are comparable to those observed in Danish house painters.
Prevalence and incidence rates should also be determined for the professions
where we only have information on one of the sexes. However, we do not have
any questionnaire information on these supplementary professions. Therefore
we will have to rely on register information on job titles, age, sex, seniority
and work proportions. Regarding the work day measurements for these addi-
tional professions we do not have detailed information on separate tasks per-
formed. If testing for an exposure response relationship we would therefore
be restricted to use a JEM instead of a TEM.
As recommended in the thesis tests for sex differences can only be perform if
the task distribution can be assumed to be comparable between men and
women. Therefor samples should be made testing for homogeneity in tasks.
Table 10. Listing of the number of individuals in each profession who have had whole day measurements made using goniometry and inclinometry.
Women Men Total
House painter 25 25 50
Laundry worker 13 10 23
Car mechanic - 11 11
Paper industry worker 10 10 20
Electronics worker 11 10 21
Truck driver - 10 10
Construction worker - 10 10
Storage worker 10 10 20
Postal worker 10 10 20
Kitchen assistant 10 - 10
Health care assistant 10 - 10
Scaffolder - 10 10
Bank clerk - 10 10
Dustman - 11 11
Carpenter - 10 10
Insulator - 10 10
Plumber - 11 11
Gardener 9 11 20
Smith - 12 12
Nurse 10 - 10
Bricklayer - 10 10
Wood industry worker - 10 10
Farmer - 10 10
Total number of full day measurements
118 221 339
Figures 10 and 11 illustrate the distribution of professions for the exposure
variables median velocity and MPF. As we reported in Paper II it shows that
male and female house painters have very similar exposures even compared
to other professions. It would be very interesting to see if the effect of the ex-
posure variables on CTS is consistent across professions.
Figure 10. Median velocity (°/s) of the right wrist. Distribution of professions.
Figure 11. MPF (Hz) of the right wrist. Distribution of exposure in professions .
Many studies have showed a higher prevalence and incidence rate of work-
related musculoskeletal disorders in women compared to men, especially in
the upper extremity. However, many studies have relied on self-reported ex-
posure and/or outcome and in many cases the exposure assessments have
had several methodological deficiencies.
In this study a systematic approach was applied trying to establish a precise
sex specific exposure assessment examining sex differences in relative muscu-
lar load, exerted forces, perceived exertion, task distributions and postures
and movements of the upper extremity. For the data collection we used elec-
tromyography, Borg CR10 scale, questionnaires, Danish registers, goniometry
and inclinometry. The Danish house painters profession was studied since it
has a high proportion of women (one third) and a supposedly homogeneous
task distribution.
Postures movements and task distributions were combined in a task exposure
matrix which was used to explore the exposure response relationship be-
tween exposures of the wrist and physician diagnosed carpal tunnel syndrome
(CTS) obtained from the Danish registers. This was tested for modification by
sex. The relative muscular load was significantly higher in women compared
to men and these objectively measured differences corresponded well to sub-
jective ratings of physical exertion. Minimal sex differences were found in ex-
erted force, by men using more force than women. Self-reported task distribu-
tions only showed minor sex differences and no significant differences were
found between the sexes in upper extremity postures and movements. An
exposure-response relationship was found between median wrist velocity and
CTS, and mean power frequency and CTS, but not between non-neutral pos-
tures and CTS. There was no significant effect of work proportion accumulated
over 1, 2 or 5 years prior to a CTS event. These results imply that un-
accumulated median velocity and MPF may be work related risk factors of
CTS.
The risk of CTS was significantly higher in women than in men with compara-
ble exposures, but the effect of the exposure was not modified by sex.
However, it is not clear to what extent these results reflect that women may be
more vulnerable to these exposures than men and to what extent they reflect
the basic assumptions of the statistical model used for the analyses.
Mange undersøgelser har vist en højere prævalens og incidensrate af arbejds-
relaterede sygdomme i det øvre bevægeapparat hos kvinder end hos mænd.
Mange undersøgelser har anvendt selvrapporterede vurderinger af ekspone-
ring og/eller udfald, som har haft flere metodologiske mangler.
I denne undersøgelse blev en systematisk tilgang brugt til at etablere en præ-
cis kønsspecifik eksponeringsvurdering til brug i analyserne af kønsforskelle i
relativ muskulær belastning, anvendt styrke, subjektivt bedømt anstrengelse,
opgave fordeling og arbejdsstillinger og bevægelser i overekstremiteterne. Til
dataindsamlingen anvendtes elektromyografi, Borg CR10 skala, spørgeskema,
danske registre, gonio- og inklinometri. Danske malere blev undersøgt, da de
har en høj andel af kvinder og angiveligt en homogen opgavefordeling.
Arbejdsstillinger, bevægelser og opgavesammensætninger blev kombineret i
en opgave-eksponeringsmatrice, som blev brugt til at undersøge dosis-
respons sammenhængen mellem eksponeringsvariable for håndleddet og læ-
ge diagnosticeret karpaltunnelsyndrom (KTS) rapporteret til de danske regi-
stre. Dette blev testet for modifikation af køn. Den relative muskulære belast-
ning var signifikant højere hos kvinder end hos mænd, og dette korrelerede
med subjektive vurderinger af fysisk anstrengelse. Der blev fundet minimale
kønsforskelle i anvendt styrke, med højeste værdier hos mænd. Der blev kun
fundet mindre kønsforskelle i opgavefordelingen, og der blev ikke fundet no-
gen signifikante forskelle mellem kønnene for arbejdsstillinger og bevægelser
i det øvre bevægeapparat. Der blev fundet en dosis-respons sammenhæng
mellem middelhastigheden for håndleddet og KTS, og et mål for repetivitet og
KTS, men ikke mellem ikke-neutrale håndledsstillinger og KTS. Der var ingen
signifikant effekt af kumuleret belastning mellem 1 og 5 år forud for et KTS
tilfælde. Disse resultater antyder, at middelhastighed og repetivitet kan være
arbejdsrelaterede risikofaktorer for KTS uden at være kumuleret over længe-
re tid. Risikoen for KTS var signifikant højere hos kvinder end hos mænd med
sammenlignelige eksponeringer, men effekten af eksponeringen blev ikke
modificeret af køn.
Det er imidlertid ikke klart, i hvilket omfang disse resultater afspejler, at kvin-
der kan være mere sårbare over for disse eksponeringer end mænd, og i hvil-
ket omfang de afspejler de grundlæggende antagelser i den statistiske model,
der anvendes til analyserne.
I owe thanks to many people for contributing to the completion of this thesis:
First of all I want to thank all of my supervisors for always being very commit-
ted in working with the SHARM-project. I want to express my huge apprecia-
tion to Jane Frølund Thomsen and Sigurd Mikkelsen for always being support-
ive of my ideas and providing an encouraging environment where I could ben-
efit from their vast knowledge within the field of occupational epidemiology.
To Susanne Wulff Svendsen for always providing constructive criticisms and
linguistic expertise which without a doubt have increased the quality of my
thesis.
To Gert-Åke Hansson for a very thorough training and supervision in the use
of biomechanical measurements, and for numerous telephone conversations,
discussing the analyses and findings. In this context I would also like to ex-
press my thankfulness to Lothy Granquist for being a big help in the analyses
of our measurements and for always providing a comfortable environment in
Lund.
To Rolf Petersen for compiling the data that initiated the SHARM-project
To Erik B. Simonsen who inspired me to choose “the road of research” and in
cooperation with Tine Alkjær was a big help in setting up the EMG-study.
To Henrik Koblauch for always being in the other end of the phone, when I
needed the help for any kind of computer software.
To Mark Lidegaard for assisting in the collection of biomechanical measure-
ments.
To Jacob Meyland for joining the project at a very short notice, and producing
high quality results in an important part of the study.
To Nils Fallentin and Jack T. Dennerlein for enabling my stay as a visiting sci-
entist in Boston.
A huge thank you to my good Australian friend Jennifer Verner, for linguistic
and grammatically assistance during the completion of the thesis. It was much
appreciated.
And of course to all my co-workers in the research unit at the Department of
Occupational and Environmental Medicine at Bispebjerg University Hospital
for providing a very special environment with room for both academically and
socially companionship. You will all be missed.
A special thanks to the Painters Union in Denmark and their members and all
the other companies and employees who participated in the study.
Finally I wish to thank: “Knud Højgaards fond”, “Augustinusfonden” and “Else
and Mogens Wedell-Wedellsborgs fond” for financial support during my stay
in Boston.
Last but not least a huge thank you to my lovely wife and daughter for being
supportive and keeping up with me in general in this, at times, stressful period
of our lives.
Reference List
(1) Mehlum IS, Kristensen P, Veiersted KB, Waersted M, Punnett L. Does
the threshold for reporting musculoskeletal pain or the probability of attributing work-relatedness vary by socioeconomic position or sex? J Occup Environ Med 2013 Aug;55(8):901-9.
(2) Wijnhoven HA, de Vet HC, Picavet HS. Prevalence of musculoskeletal disorders is systematically higher in women than in men. Clin J Pain 2006 Oct;22(8):717-24.
(3) Bingefors K, Isacson D. Epidemiology, co-morbidity, and impact on health-related quality of life of self-reported headache and musculoskeletal pain - a gender perspective. European Journal of Pain 2004 Oct;8(5):435-50.
(4) Dahlberg R, Karlqvist L, Bildt C, Nykvist K. Do work technique and musculoskeletal symptoms differ between men and women performing the same type of work tasks? Applied Ergonomics 2004 Nov;35(6):521-9.
(5) Hooftman WE, van Poppel MN, van der Beek AJ, Bongers PM, van MW. Gender differences in the relations between work-related physical and psychosocial risk factors and musculoskeletal complaints. Scand J Work Environ Health 2004 Aug;30(4):261-78.
(6) de Zwart BC, Frings-Dresen MH, Kilbom A. Gender differences in upper extremity musculoskeletal complaints in the working population. Int Arch Occup Environ Health 2001 Jan;74(1):21-30.
(7) Nordander C, Ohlsson K, Akesson I, Arvidsson I, Balogh I, Hansson GA, et al. Risk of musculoskeletal disorders among females and males in repetitive/constrained work. Ergonomics 2009 Oct 1;52(10):1226-39.
(8) Nordander C, Ohlsson K, Balogh I, Hansson GA, Axmon A, Persson R, et al. Gender differences in workers with identical repetitive industrial tasks: exposure and musculoskeletal disorders. Int Arch Occup Environ Health 2008 Aug;81(8):939-47.
(9) Gerr F, Marcus M, Ensor C, Kleinbaum D, Cohen S, Edwards A, et al. A prospective study of computer users: I. Study design and incidence of musculoskeletal symptoms and disorders. Am J Ind Med 2002 Apr 1;41(4):221-35.
(10) Côté JN. A critical review on physical factors and functional characteristics that may explain a sex/gender difference in work-related neck/shoulder disorders. Ergonomics 2011 Aug 17;55(2):173-82.
(11) Artazcoz L, Borrell C, Cortes I, Escriba-Aguir V, Cascant L. Occupational epidemiology and work related inequalities in health: a gender perspective for two complementary approaches to work and health research. J Epidemiol Community Health 2007 Dec;61 Suppl 2:ii39-ii45.
(12) Messing K, Mager SJ. Sex, gender and women's occupational health: the importance of considering mechanism. Environ Res 2006 Jun;101(2):149-62.
(13) Niedhammer I, Saurel-Cubizolles MJ, Piciotti M, Bonenfant S. How is sex considered in recent epidemiological publications on occupational risks? Occup Environ Med 2000 Aug;57(8):521-7.
(14) Hurley RW, Adams MC. Sex, gender, and pain: an overview of a complex field. Anesth Analg 2008 Jul;107(1):309-17.
(15) Ladwig KH, Marten-Mittag B, Formanek B, Dammann G. Gender differences of symptom reporting and medical health care utilization in the German population. Eur J Epidemiol 2000 Jun;16(6):511-8.
(16) Mondelli M, Aprile I, Ballerini M, Ginanneschi F, Reale F, Romano C, et al. Sex differences in carpal tunnel syndrome: comparison of surgical and non-surgical populations. European Journal of Neurology 2005 Dec 1;12(12):976-83.
(17) Myers CD, Riley JL, III, Robinson ME. Psychosocial contributions to sex-correlated differences in pain. Clin J Pain 2003 Jul;19(4):225-32.
(18) Riley JL, III, Robinson ME, Wise EA, Myers CD, Fillingim RB. Sex differences in the perception of noxious experimental stimuli: a meta-analysis. Pain 1998 Feb;74(2-3):181-7.
(19) van Wijk CM, Kolk AM. Sex differences in physical symptoms: the contribution of symptom perception theory. Soc Sci Med 1997 Jul;45(2):231-46.
(20) Locke SJ, Colt JS, Stewart PA, Armenti KR, Baris D, Blair A, et al. Identifying gender differences in reported occupational information from three US population-based case Çôcontrol studies. Occupational and Environmental Medicine 2014 Mar 28.
(21) Kennedy SM, Koehoorn M. Exposure assessment in epidemiology: does gender matter? Am J Ind Med 2003 Dec;44(6):576-83.
(22) Leijon O, Bernmark E, Karlqvist L, Harenstam A. Awkward work postures: association with occupational gender segregation. Am J Ind Med 2005 May;47(5):381-93.
(23) Lundberg U. Psychophysiology of work: stress, gender, endocrine response, and work-related upper extremity disorders. Am J Ind Med 2002 May;41(5):383-92.
(24) Eng A, 't Mannetje A, McLean D, Ellison-Loschmann L, Cheng S, Pearce N. Gender differences in occupational exposure patterns.
Occupational and Environmental Medicine 2011 Dec 1;68(12):888-94.
(25) Messing K, Dumais L, Courville J, Seifert AM, Boucher M. Evaluation of exposure data from men and women with the same job title. J Occup Med 1994 Aug;36(8):913-7.
(26) Meyland J, Heilskov-Hansen T, Alkjar T, Koblauch H, Mikkelsen S, Svendsen S, et al. Sex differences in muscular load among house painters performing identical work tasks. Eur J Appl Physiol 2014;1-11.
(27) Lindbeck L, Kjellberg K. Gender differences in lifting technique. Ergonomics 2001 Feb 1;44(2):202-14.
(28) Bird CE, Sharman Z. Gender-Based Analysis Is Essential to Improving Women's Health and Health Care. Women's health issues : official publication of the Jacobs Institute of Women's Health 24[2], e163-e164. 1-3-2014.
Ref Type: Abstract
(29) Gillespie RM, Herbert R, Punnett L. Chapter 41 - Work-Related Musculo-Skeletal Disorders. In: Goldman MB, Troisi R, Rexrode KM, editors. Women and Health (Second Edition). Academic Press; 2013. p. 613-28.
(30) Gallagher S, Heberger JR. Examining the Interaction of Force and Repetition on Musculoskeletal Disorder Risk: A Systematic Literature Review. Human Factors: The Journal of the Human Factors and Ergonomics Society 2013 Feb 1;55(1):108-24.
(31) Nordlund A, Palsson B, Ohlsson K, Skerfving S. Economic consequences of occupational disorders in women with repetitive industrial work. The European Journal of Public Health 2000 Jun 1;10(2):127-32.
(32) Treaster DE, Burr D. Gender differences in prevalence of upper extremity musculoskeletal disorders. Ergonomics 2004 Apr 15;47(5):495-526.
(33) Hooftman WE, van der Beek AJ, Bongers PM, van MW. Is there a gender difference in the effect of work-related physical and psychosocial risk factors on musculoskeletal symptoms and related sickness absence? Scand J Work Environ Health 2009 Mar;35(2):85-95.
(34) Arvidsson I, Arvidsson M, Axmon A, Hansson G, Johansson CR, Skerfving S. Musculoskeletal disorders among female and male air traffic controllers performing identical and demanding computer work. Ergonomics 2006 Sep 15;49(11):1052-67.
(35) da Costa BR, Vieira ER. Risk factors for work-related musculoskeletal disorders: a systematic review of recent longitudinal studies. Am J Ind Med 2010 Mar 1;53(3):285-323.
(36) Punnett L, Wegman DH. Work-related musculoskeletal disorders: the epidemiologic evidence and the debate. J Electromyogr Kinesiol 2004 Feb;14(1):13-23.
(37) Punnett L. Musculoskeletal disorders and occupational exposures: How should we judge the evidence concerning the causal association? Scandinavian Journal of Public Health 2014 Mar 1;42(13 suppl):49-58.
(38) NIOSH. Musculoskeletal Disorders and Workplace Factors - A Critical Review of Epidemiologic Evidence for Work-Related Musculoskeletal Disorders of the Neck, Upper Extremity, and Low Back. U.S. DEPARTMENT OF HEALTH AND HUMAN SERVICES; 1997 Jul. Report No.: 97-141.
(39) Roquelaure Y, Ha C, Rouillon C, Fouquet N, Leclerc A, Descatha A, et al. Risk factors for upper-extremity musculoskeletal disorders in the working population. Arthritis Rheum 2009 Oct 15;61(10):1425-34.
(40) Viester L, Verhagen EA, Oude Hengel KM, Koppes LL, van der Beek AJ, Bongers PM. The relation between body mass index and musculoskeletal symptoms in the working population. BMC Musculoskelet Disord 2013;14:238.
(41) Latko WA, Armstrong TJ, Franzblau A, Ulin SS, Werner RA, Albers JW. Cross-sectional study of the relationship between repetitive work and the prevalence of upper limb musculoskeletal disorders. Am J Ind Med 1999 Aug 1;36(2):248-59.
(42) Thomsen JF, Mikkelsen S, Andersen JH, Fallentin N, Loft IP, Frost P, et al. Risk factors for hand-wrist disorders in repetitive work. Occupational and Environmental Medicine 2007 Aug 1;64(8):527-33.
(43) Fillingim RB, King CD, Ribeiro-Dasilva MC, Rahim-Williams B, Riley JL, III. Sex, gender, and pain: a review of recent clinical and experimental findings. J Pain 2009 May;10(5):447-85.
(44) Hunter SK. Sex differences in human fatigability: mechanisms and insight to physiological responses. Acta Physiol (Oxf) 2014 Apr;210(4):768-89.
(45) Wüst RCI, Morse CI, de Haan A, Jones DA, Degens H. Sex differences in contractile properties and fatigue resistance of human skeletal muscle. Experimental Physiology 2008 Jul 1;93(7):843-50.
(46) Westh E, Kongsgaard M, Bojsen-Moller J, Aagaard P, Hansen M, Kjaer M, et al. Effect of habitual exercise on the structural and mechanical properties of human tendon, in vivo, in men and women. Scand J Med Sci Sports 2008 Feb;18(1):23-30.
(47) Magnusson SP, Hansen M, Langberg H, Miller B, Haraldsson B, Westh EK, et al. The adaptability of tendon to loading differs in men and women. Int J Exp Pathol 2007 Aug;88(4):237-40.
(48) Sullivan BE, Carroll CC, Jemiolo B, Trappe SW, Magnusson SP, Dossing S, et al. Effect of acute resistance exercise and sex on human patellar tendon structural and regulatory mRNA expression. J Appl Physiol (1985 ) 2009 Feb;106(2):468-75.
(49) Miller BF, Hansen M, Olesen JL, Schwarz P, Babraj JA, Smith K, et al. Tendon collagen synthesis at rest and after exercise in women. J Appl Physiol (1985 ) 2007 Feb;102(2):541-6.
(50) Fedorowich L, Emery K, Gervasi B, Côté JN. Gender differences in neck/shoulder muscular patterns in response to repetitive motion induced fatigue. Journal of Electromyography and Kinesiology 2013 Oct;23(5):1183-9.
(51) Johansen T, Samani A, Antle D, Côté JN, Madeleine P. Gender effects on the coordination of subdivisions of the trapezius muscle during a repetitive box-folding task. Eur J Appl Physiol 2013;113(1):175-82.
(52) Maughan RJ, Watson JS, Weir J. Strength and cross-sectional area of human skeletal muscle. The Journal of Physiology 1983 May 1;338(1):37-49.
(53) Miller AE, MacDougall JD, Tarnopolsky MA, Sale DG. Gender differences in strength and muscle fiber characteristics. Eur J Appl Physiol Occup Physiol 1993;66(3):254-62.
(54) Won EJ, Johnson PW, Punnett L, Dennerlein JT. Upper extremity biomechanics in computer tasks differ by gender. Journal of Electromyography and Kinesiology 2009 Jun;19(3):428-36.
(55) Nordander C, Ohlsson K, Akesson I, Arvidsson I, Balogh I, Hansson GA, et al. Exposure-response relationships in work-related musculoskeletal disorders in elbows and hands - A synthesis of group-level data on exposure and response obtained using uniform methods of data collection. Appl Ergon 2013 Mar;44(2):241-53.
(56) Roepstorff C, Thiele M, Hillig T, Pilegaard H, Richter EA, Wojtaszewski JF, et al. Higher skeletal muscle alpha2AMPK activation and lower energy charge and fat oxidation in men than in women during submaximal exercise. J Physiol 2006 Jul 1;574(Pt 1):125-38.
(57) Hägg G. Static work loads and occupational myalgia - a new explanation model. In: P.A.Anderson, D.J.Hobart, J.V.Danhoff, editors. Electromyographycal kinesiology. Elsevier Science Publishers; 1991. p. 141-4.
(58) Hauke A, Flintrop J, Brun E, Rugulies R. The impact of work-related psychosocial stressors on the onset of musculoskeletal disorders in specific body regions: A review and meta-analysis of 54 longitudinal studies. Work & Stress 2011 Jul;25(3):243-56.
(59) van den Heuvel SG, van der Beek AJ, Blatter BM, Hoogendoorn WE, Bongers PM. Psychosocial work characteristics in relation to neck and upper limb symptoms. Pain 2005 Mar;114:47-53.
(60) Bongers PM, Kremer AM, ter LJ. Are psychosocial factors, risk factors for symptoms and signs of the shoulder, elbow, or hand/wrist?: A review of the epidemiological literature. Am J Ind Med 2002 May;41(5):315-42.
(61) Atroshi I, Englund M, Turkiewicz A, Tagil M, Petersson IF. Incidence of physician-diagnosed carpal tunnel syndrome in the general population. Arch Intern Med 2011 May 23;171(10):943-4.
(62) Gerr F, Fethke NB, Merlino L, Anton D, Rosecrance J, Jones MP, et al. A Prospective Study of Musculoskeletal Outcomes Among Manufacturing Workers: I. Effects of Physical Risk Factors. Human Factors: The Journal of the Human Factors and Ergonomics Society 2014 Feb 1;56(1):112-30.
(63) Violante FS, Bonfiglioli R, Hagberg M, Rempel D. Carpal tunnel syndrome diagnosis in occupational epidemiological studies. Occupational and Environmental Medicine 2014 May 30.
(64) Harrington JM, Carter JT, Birrell L, Gompertz D. Surveillance case definitions for work related upper limb pain syndromes. Occup Environ Med 1998 Apr;55(4):264-71.
(65) Vasiliadis HS, Georgoulas P, Shrier I, Salanti G, Scholten RJ. Endoscopic release for carpal tunnel syndrome. Cochrane Database Syst Rev 2014;1:CD008265.
(66) Fernandes CH, Nakachima LR, Hirakawa CK, Gomes Dos Santos JB, Faloppa F. Carpal tunnel release using the Paine retinaculotome inserted through a palmar incision. Hand (N Y ) 2014 Mar;9(1):48-51.
(67) Fnais N, Gomes T, Mahoney J, Alissa S, Mamdani M. Temporal trend of carpal tunnel release surgery: a population-based time series analysis. PLoS One 2014;9(5):e97499.
(68) Baker NA, Livengood HM. Symptom Severity and Conservative Treatment for Carpal Tunnel Syndrome in Association With Eventual Carpal Tunnel Release. J Hand Surg Am 2014 Jun 5.
(69) Silverstein BA, Fine LJ, Armstrong TJ. Occupational factors and carpal tunnel syndrome. Am J Ind Med 1987;11(3):343-58.
(70) Silverstein BA, Fan ZJ, Bonauto DK, Bao S, Smith CK, Howard N, et al. The natural course of carpal tunnel syndrome in a working population. Scand J Work Environ Health 2010 Sep;36(5):384-93.
(71) Feuerstein M, Miller VL, Burrell LM, Berger R. Occupational upper extremity disorders in the federal workforce. Prevalence, health care expenditures, and patterns of work disability. J Occup Environ Med 1998 Jun;40(6):546-55.
(72) Falkiner S, Myers S. When exactly can carpal tunnel syndrome be considered work-related? ANZ J Surg 2002 Mar;72(3):204-9.
(73) Bonfiglioli R, Mattioli S, Fiorentini C, Graziosi F, Curti S, Violante F. Relationship between repetitive work and the prevalence of carpal tunnel syndrome in part-time and full-time female supermarket cashiers: a quasi-experimental study. Int Arch Occup Environ Health 2007;80(3):248-53.
(74) Descatha A, Dale AM, Franzblau A, Coomes J, Evanoff B. Comparison of research case definitions for carpal tunnel syndrome. Scand J Work Environ Health 2011 Jul;37(4):298-306.
(75) Hegmann KT, Thiese MS, Wood EM, Garg A, Kapellusch JM, Foster J, et al. Impacts of Differences in Epidemiological Case Definitions on Prevalence for Upper-Extremity Musculoskeletal Disorders. Human Factors: The Journal of the Human Factors and Ergonomics Society 2014 Feb 1;56(1):191-202.
(76) Palmer KT, Harris EC, Linaker C, Cooper C, Coggon D. Optimising case definitions of upper limb disorder for aetiological research and prevention: a review. Occupational and Environmental Medicine 2012 Jan 1;69(1):71-8.
(77) Rempel D, Evanoff B, Amadio PC, de KM, Franklin G, Franzblau A, et al. Consensus criteria for the classification of carpal tunnel syndrome in epidemiologic studies. Am J Public Health 1998 Oct;88(10):1447-51.
(78) Shiri R, Miranda H, Heli+Âvaara M, Viikari-Juntura E. Physical work load factors and carpal tunnel syndrome: a population-based study. Occupational and Environmental Medicine 2009 Jun 1;66(6):368-73.
(79) van Rijn RM, Huisstede BM, Koes BW, Burdorf A. Associations between work-related factors and the carpal tunnel syndrome--a systematic review. Scand J Work Environ Health 2009 Jan;35(1):19-36.
(80) Salaffi F, De AR, Grassi W. Prevalence of musculoskeletal conditions in an Italian population sample: results of a regional community-based study. I. The MAPPING study. Clin Exp Rheumatol 2005 Nov;23(6):819-28.
(81) Patil A, Rosecrance J, Douphrate D, Gilkey D. Prevalence of carpal tunnel syndrome among dairy workers. Am J Ind Med 2012 Feb;55(2):127-35.
(82) Kim JY, Kim JI, Son JE, Yun SK. Prevalence of carpal tunnel syndrome in meat and fish processing plants. J Occup Health 2004 May;46(3):230-4.
(83) Nordander C, Ohlsson K, Balogh I, Rylander L, Palsson B, Skerfving S. Fish processing work: the impact of two sex dependent exposure profiles on musculoskeletal health. Occup Environ Med 1999 Apr;56(4):256-64.
(84) Silverstein BA, Hughes RE. Upper extremity musculoskeletal disorders at a pulp and paper mill. Appl Ergon 1996 Jun;27(3):189-94.
(85) Leclerc A, Franchi P, Cristofari MF, Delemotte B, Mereau P, Teyssier-Cotte C, et al. Carpal tunnel syndrome and work organisation in repetitive work: a cross sectional study in France. Study Group on Repetitive Work. Occup Environ Med 1998 Mar;55(3):180-7.
(86) Wang LY, Pong YP, Wang HC, Su SH, Tsai CH, Leong CP. Cumulative trauma disorders in betel pepper leaf-cullers visiting a rehabilitation clinic: experience in Taitung. Chang Gung Med J 2005 Apr;28(4):237-46.
(87) Frost P, Andersen JH, Nielsen VK. Occurrence of carpal tunnel syndrome among slaughterhouse workers. Scand J Work Environ Health 1998 Aug;24(4):285-92.
(88) Roquelaure Y, Ha C, Leclerc A, Touranchet A, Sauteron M, Melchior M, et al. Epidemiologic surveillance of upper-extremity musculoskeletal disorders in the working population. Arthritis Rheum 2006 Oct 15;55(5):765-78.
(89) Roquelaure Y, Ha C, Pelier-Cady MC, Nicolas G, Descatha A, Leclerc A, et al. Work increases the incidence of carpal tunnel syndrome in the general population. Muscle Nerve 2008 Apr;37(4):477-82.
(90) Roquelaure Y, Ha C, Fouquet N, Descatha A, Leclerc A, Goldberg M, et al. Attributable risk of carpal tunnel syndrome in the general population: implications for intervention programs in the workplace. Scand J Work Environ Health 2009 Oct;35(5):342-8.
(91) Walker-Bone K, Palmer KT, Reading I, Coggon D, Cooper C. Prevalence and impact of musculoskeletal disorders of the upper limb in the general population. Arthritis & Rheumatism 2004 Aug 15;51(4):642-51.
(92) Barcenilla A, March LM, Chen JS, Sambrook PN. Carpal tunnel syndrome and its relationship to occupation: a meta-analysis. Rheumatology 2012 Feb 1;51(2):250-61.
(93) Palmer KT, Harris EC, Coggon D. Carpal tunnel syndrome and its relation to occupation: a systematic literature review. Occupational Medicine 2007 Jan 1;57(1):57-66.
(94) Bonfiglioli R, Mattioli S, Armstrong TJ, Graziosi F, Marinelli F, Farioli A, et al. Validation of the ACGIH TLV for hand activity level in the OCTOPUS cohort: a two-year longitudinal study of carpal tunnel syndrome. Scand J Work Environ Health 2013 Mar 1;39(2):155-63.
(95) Geoghegan JM, Clark DI, Bainbridge LC, Smith C, Hubbard R. Risk factors in carpal tunnel syndrome. The Journal of Hand Surgery: British & European Volume 2004 Aug;29(4):315-20.
(96) Atroshi I, Gummesson C, Johnsson R, Ornstein E, Ranstam J, Rosen I. Prevalence of carpal tunnel syndrome in a general population. JAMA 1999 Jul 14;282(2):153-8.
(97) Bland JDP, Rudolfer SM. Clinical surveillance of carpal tunnel syndrome in two areas of the United Kingdom, 1991 2001. Journal of Neurology, Neurosurgery & Psychiatry 2003 Dec 1;74(12):1674-9.
(98) Violante FS, Armstrong TJ, Fiorentini C, Graziosi F, Risi A, Venturi S, et al. Carpal tunnel syndrome and manual work: a longitudinal study. J Occup Environ Med 2007 Nov;49(11):1189-96.
(99) Mattioli S, Baldasseroni A, Bovenzi M, Curti S, Cooke R, Campo G, et al. Risk factors for operated carpal tunnel syndrome: a multicenter population-based case-control study. BMC Public Health 2009;9(1):343.
(100) Dale AM, Harris-Adamson C, Rempel D, Gerr F, Hegmann K, Silverstein B, et al. Prevalence and incidence of carpal tunnel syndrome in US working populations: pooled analysis of six prospective studies. Scand J Work Environ Health 2013 Sep 1;39(5):495-505.
(101) Ferry S, Hannaford P, Warskyj M, Lewis M, Croft P. Carpal tunnel syndrome: a nested case-control study of risk factors in women. Am J Epidemiol 2000 Mar 15;151(6):566-74.
(102) Nathan PA, Meadows KD, Doyle LS. Relationship of age and sex to sensory conduction of the median nerve at the carpal tunnel and association of slowed conduction with symptoms. Muscle Nerve 1988 Nov;11(11):1149-53.
(103) Nathan PA, Istvan JA, Meadows KD. A longitudinal study of predictors of research-defined carpal tunnel syndrome in industrial workers: findings at 17 years. The Journal of Hand Surgery: British & European Volume 2005 Dec;30(6):593-8.
(104) Roquelaure Y, Ha C, Pelier-Cady MC, Nicolas G, Descatha A, Leclerc A, et al. Work increases the incidence of carpal tunnel syndrome in the general population. Muscle Nerve 2008 Apr;37(4):477-82.
(105) Werner R. Evaluation of Work-Related Carpal Tunnel Syndrome. J Occup Rehabil 2006;16(2):201-16.
(106) Palmer KT. Carpal tunnel syndrome: The role of occupational factors. Best Practice & Research Clinical Rheumatology 2011 Feb;25(1):15-29.
(107) Bovenzi M, Zadini A, Franzinelli A, Borgogni F. Occupational musculoskeletal disorders in the neck and upper limbs of forestry workers exposed to hand-arm vibration. Ergonomics 1991 May;34(5):547-62.
(108) Bovenzi M. Hand-arm vibration syndrome and dose-response relation for vibration induced white finger among quarry drillers
and stonecarvers. Italian Study Group on Physical Hazards in the Stone Industry. Occup Environ Med 1994 Sep;51(9):603-11.
(109) Cannon LJ, Bernacki EJ, Walter SD. Personal and occupational factors associated with carpal tunnel syndrome. J Occup Med 1981 Apr;23(4):255-8.
(110) Chatterjee DS, Barwick DD, Petrie A. Exploratory electromyography in the study of vibration-induced white finger in rock drillers. Br J Ind Med 1982 Feb;39(1):89-97.
(111) Farkkila M, Pyykko I, Jantti V, Aatola S, Starck J, Korhonen O. Forestry workers exposed to vibration: a neurological study. Br J Ind Med 1988 Mar;45(3):188-92.
(112) Koskimies K, Farkkila M, Pyykko I, Jantti V, Aatola S, Starck J, et al. Carpal tunnel syndrome in vibration disease. Br J Ind Med 1990 Jun;47(6):411-6.
(113) Nathan PA, Meadows KD, Istvan JA. Predictors of carpal tunnel syndrome: An 11-year study of industrial workers. The Journal of Hand Surgery 2002 Jul;27(4):644-51.
(114) Nordstrom DL, Vierkant RA, Layde PM, Smith MJ. Comparison of self-reported and expert-observed physical activities at work in a general population. Am J Ind Med 1998 Jul 1;34(1):29-35.
(115) Wieslander G, Norback D, Gothe CJ, Juhlin L. Carpal tunnel syndrome (CTS) and exposure to vibration, repetitive wrist movements, and heavy manual work: a case-referent study. Br J Ind Med 1989 Jan;46(1):43-7.
(116) Chiang HC, Ko YC, Chen SS, Yu HS, Wu TN, Chang PY. Prevalence of shoulder and upper-limb disorders among workers in the fish-processing industry. Scand J Work Environ Health 1993 Apr;19(2):126-31.
(117) de Krom MC, Kester AD, Knipschild PG, Spaans F. Risk factors for carpal tunnel syndrome. Am J Epidemiol 1990 Dec;132(6):1102-10.
(118) Tanaka S, Wild DK, Cameron LL, Freund E. Association of occupational and non-occupational risk factors with the prevalence of self-reported carpal tunnel syndrome in a national survey of the working population. Am J Ind Med 1997 Nov;32(5):550-6.
(119) Thomsen JF, Hansson GA, Mikkelsen S, Lauritzen M. Carpal tunnel syndrome in repetitive work: a follow-up study. Am J Ind Med 2002 Oct;42(4):344-53.
(120) Yagev Y, Carel RS, Yagev R. Assessment of work-related risks factors for carpal tunnel syndrome. Isr Med Assoc J 2001 Aug;3(8):569-71.
(121) Roquelaure Y, Mechali S, Dano C, Fanello S, Benetti F, Bureau D, et al. Occupational and personal risk factors for carpal tunnel syndrome in
industrial workers. Scand J Work Environ Health 1997 Oct;23(5):364-9.
(122) Gell N, Werner R, Franzblau A, Ulin S, Armstrong T. A Longitudinal Study of Industrial and Clerical Workers: Incidence of Carpal Tunnel Syndrome and Assessment of Risk Factors. J Occup Rehabil 2005;15(1):47-55.
(123) Burt S, Crombie K, Jin Y, Wurzelbacher S, Ramsey J, Deddens J. Workplace and individual risk factors for carpal tunnel syndrome. Occup Environ Med 2011 Dec;68(12):928-33.
(124) Burt S, Deddens JA, Crombie K, Jin Y, Wurzelbacher S, Ramsey J. A prospective study of carpal tunnel syndrome: workplace and individual risk factors. Occup Environ Med 2013 Aug;70(8):568-74.
(125) Cosgrove JL, Chase PM, Mast NJ, Reeves R. Carpal tunnel syndrome in railroad workers. Am J Phys Med Rehabil 2002 Feb;81(2):101-7.
(126) Roquelaure Y, Mariel J, Dano C, Fanello S, Penneau-Fontbonne D. Prevalence, incidence and risk factors of carpal tunnel syndrome in a large footwear factory. Int J Occup Med Environ Health 2001;14(4):357-67.
(127) Andersen JH, Thomsen JF, Overgaard E, Lassen CF, Brandt LP, Vilstrup I, et al. Computer use and carpal tunnel syndrome: a 1-year follow-up study. JAMA 2003 Jun 11;289(22):2963-9.
(128) Hou WH, Hsu JH, Lin CH, Liang HW. Carpal tunnel syndrome in male visual display terminal (VDT) workers. Am J Ind Med 2007 Jan;50(1):1-7.
(129) Mediouni Z, de RA, Dumontier C, Becour B, Garrabe H, Roquelaure Y, et al. Is carpal tunnel syndrome related to computer exposure at work? A review and meta-analysis. J Occup Environ Med 2014 Feb;56(2):204-8.
(130) Stevens JC, Witt JC, Smith BE, Weaver AL. The frequency of carpal tunnel syndrome in computer users at a medical facility. Neurology 2001 Jun 12;56(11):1568-70.
(131) Thomsen J, Gerr F, Atroshi I. Carpal tunnel syndrome and the use of computer mouse and keyboard: A systematic review. BMC Musculoskeletal Disorders 2008;9(1):134.
(132) Ali KM, Sathiyasekaran BW. Computer professionals and Carpal Tunnel Syndrome (CTS). Int J Occup Saf Ergon 2006;12(3):319-25.
(133) Werner RA, Franzblau A, Gell N, Hartigan AG, Ebersole M, Armstrong TJ. Incidence of carpal tunnel syndrome among automobile assembly workers and assessment of risk factors. J Occup Environ Med 2005 Oct;47(10):1044-50.
(134) van den Heuvel SG, van der Beek AJ, Blatter BM, Hoogendoorn WE, Bongers PM. Psychosocial work characteristics in relation to neck and upper limb symptoms. Pain 2005 Mar;114:47-53.
(135) Harris-Adamson C, Eisen EA, Dale AM, Evanoff B, Hegmann KT, Thiese MS, et al. Personal and workplace psychosocial risk factors for carpal tunnel syndrome: a pooled study cohort. Occupational and Environmental Medicine 2013 Aug 1;70(8):529-37.
(136) Abbas MF, Faris RH, Harber PI, Mishriky AM, El-Shahaly HA, Waheeb YH, et al. Worksite and personal factors associated with carpal tunnel syndrome in an Egyptian electronics assembly factory. Int J Occup Environ Health 2001 Jan;7(1):31-6.
(137) Liss GM, Jesin E, Kusiak RA, White P. Musculoskeletal problems among Ontario dental hygienists. Am J Ind Med 1995 Oct;28(4):521-40.
(138) Mondelli M, Giannini F, Giacchi M. Carpal tunnel syndrome incidence in a general population. Neurology 2002 Jan 22;58(2):289-94.
(139) Seta JA, Sundin DS, Pedersen DH. National occupational exposure survey field guidelines. Vol. I. U.S. Department of Health and Human Services, Centers for Disease Control, National Institute for Occupational Safety and Health; 1988. Report No.: DHHS (NIOSH) Publication No. 88-106.
(140) Landau K, Imhof-Gildein B, M++cke S. On the analysis of sector-related and gender-related stresses at the workplace. An analysis of the AET data bank. International Journal of Industrial Ergonomics 1996 Feb;17(2):175-86.
(141) McDiarmid M, Oliver M, Ruser J, Gucer P. Male and Female Rate Differences in Carpal Tunnel Syndrome Injuries: Personal Attributes or Job Tasks? Environmental Research 2000 May;83(1):23-32.
(142) Nieuwenhuijsen MJ. Exposure assessment in occupational and environmental epidemiology. New York: Oxford university press Inc.; 2010.
(143) Balogh I, Oerbaek P, Ohlsson K, Nordander C, Unge J, Winkel J, et al. Self-assessed and directly measured occupational physical activitiesnfluence of musculoskeletal complaints, age and gender. Applied
Ergonomics 2004 Jan;35(1):49-56.
(144) David GC. Ergonomic methods for assessing exposure to risk factors for work-related musculoskeletal disorders. Occupational Medicine 2005 May 1;55(3):190-9.
(145) Spielholz P, Silverstein B, Morgan M, Checkoway H, Kaufman J. Comparison of self-report, video observation and direct measurement methods for upper extremity musculoskeletal disorder physical risk factors. Ergonomics 2001 May 1;44(6):588-613.
(146) van der Beek AJ, Frings-Dresen MH. Assessment of mechanical exposure in ergonomic epidemiology. Occup Environ Med 1998 May;55(5):291-9.
(147) Hansson GA, Balogh I, Bystrom JU, Ohlsson K, Nordander C, Asterland P, et al. Questionnaire versus direct technical measurements in assessing postures and movements of the head, upper back, arms and hands. Scand J Work Environ Health 2001 Feb;27(1):30-40.
(148) Viikari-Juntura E, Rauas S, Martikainen R, Kuosma E, Riihimaki H, Takala EP, et al. Validity of self-reported physical work load in epidemiologic studies on musculoskeletal disorders. Scand J Work Environ Health 1996 Aug;22(4):251-9.
(149) Dale AM, Rohn AE, Patton A, Standeven J, Evanoff B. Variability and misclassification of worker estimated hand force. Appl Ergon 2011 Nov;42(6):846-51.
(150) Palmer KT, Haward B, Griffin MJ, Bendall H, Coggon D. Validity of self reported occupational exposures to hand transmitted and whole body vibration. Occupational and Environmental Medicine 2000 Apr 1;57(4):237-41.
(151) Hoozemans MJM, Knelange EB, Frings-Dresen MHW, Veeger HEJ, Kuijer PPFM. Are pushing and pulling work-related risk factors for upper extremity symptoms? A systematic review of observational studies. Occupational and Environmental Medicine 2014 Jul 17.
(152) Wells R, Mathiassen SE, Medbo L, Winkel J. Time A key issue for musculoskeletal health and manufacturing. Applied Ergonomics 2007 Nov;38(6):733-44.
(153) Takala EP, Pehkonen I, Forsman M, Hansson GA, Mathiassen SE, Neumann WP, et al. Systematic evaluation of observational methods assessing biomechanical exposures at work. Scand J Work Environ Health 2010 Jan;36(1):3-24.
(154) Liv P, Mathiassen SE, Svendsen SW. Theoretical and Empirical Efficiency of Sampling Strategies for Estimating Upper Arm Elevation. Annals of Occupational Hygiene 2011 May 1;55(4):436-49.
(155) van der Beek AJ, Kuiper JI, Dawson M, Burdorf A, Bongers PM, Frings-Dresen MH. Sources of variance in exposure to nonneutral trunk postures in varying work situations. Scand J Work Environ Health 1995 Jun;21(3):215-22.
(156) Balogh I, Ohlsson K, Nordander C, Skerfving S, Hansson GA. Precision of measurements of physical workload during standardized manual handling part III: goniometry of the wrists. J Electromyogr Kinesiol 2009 Oct;19(5):1005-12.
(157) Hansson GA, Asterland P, Holmer NG, Skerfving S. Validity and reliability of triaxial accelerometers for inclinometry in posture analysis. Med Biol Eng Comput 2001 Jul;39(4):405-13.
(158) Hansson GA, Balogh I, Ohlsson K, Skerfving S. Measurements of wrist and forearm positions and movements: effect of, and compensation for, goniometer crosstalk. J Electromyogr Kinesiol 2004 Jun;14(3):355-67.
(159) Hansson GA, Arvidsson I, Ohlsson K, Nordander C, Mathiassen SE, Skerfving S, et al. Precision of measurements of physical workload during standardised manual handling. Part II: Inclinometry of head, upper back, neck and upper arms. J Electromyogr Kinesiol 2006 Apr;16(2):125-36.
(160) Balogh I, Hansson GA, Ohlsson K, Stromberg U, Skerfving S. Interindividual variation of physical load in a work task. Scand J Work Environ Health 1999 Feb;25(1):57-66.
(161) Coury HJCG, Porcatti IA, Alem MER, Oishi J. Influence of gender on work-related musculoskeletal disorders in repetitive tasks. International Journal of Industrial Ergonomics 2002 Jan;29(1):33-9.
(162) Mergler D, Brabant C, Vezina N, Messing K. The weaker sex? Men in women's working conditions report similar health symptoms. J Occup Med 1987 May;29(5):417-21.
(163) Hooftman WE, van der Beek AJ, van de Wal BG, Knol DL, Bongers PM, Burdorf A, et al. Equal task, equal exposure? Are men and women with the same tasks equally exposed to awkward working postures? Ergonomics 2009 Sep 1;52(9):1079-86.
(164) Rijs KJ, van der Pas S, Geuskens GA, Cozijnsen R, Koppes LL, van der Beek AJ, et al. Development and validation of a physical and psychosocial job-exposure matrix in older and retired workers. Ann Occup Hyg 2014 Mar;58(2):152-70.
(165) Rubak T, Svendsen S, Andersen J, Haahr JP, Kryger A, Jensen L, et al. An expert-based job exposure matrix for large scale epidemiologic studies of primary hip and knee osteoarthritis: The Lower Body JEM. BMC Musculoskeletal Disorders 2014;15(1):204.
(166) Evanoff B, Zeringue A, Franzblau A, Dale AM. Using Job-Title-Based Physical Exposures From O*NET in an Epidemiological Study of Carpal Tunnel Syndrome. Human Factors: The Journal of the Human Factors and Ergonomics Society 2014 Feb 1;56(1):166-77.
(167) Kauppinen T, Heikkila P, Plato N, Woldbaek T, Lenvik K, Hansen J, et al. Construction of job-exposure matrices for the Nordic Occupational Cancer Study (NOCCA). Acta Oncol 2009;48(5):791-800.
(168) Benke G, Sim M, Fritschi L, Aldred G. Beyond the Job Exposure Matrix (JEM): the Task Exposure Matrix (TEM). Annals of Occupational Hygiene 2000 Sep 1;44(6):475-82.
(169) Mathiassen SE, Nordander C, Svendsen SW, Wellman HM, Dempsey PG. Task-based estimation of mechanical job exposure in occupational groups. Scand J Work Environ Health 2005 Apr;31(2):138-51.
(170) Svendsen SW, Mathiassen SE, Bonde JP. Task based exposure assessment in ergonomic epidemiology: a study of upper arm elevation in the jobs of machinists, car mechanics, and house painters. Occup Environ Med 2005 Jan;62(1):18-27.
(171) Perotto A. Anatomical guide for the electromyographer: the limbs and trunk. Springfield: Charles C Thomas; 2005.
(172) Rudroff T. Kinesiological fine wire EMG. 1st ed. Scottsdale: Noraxon USA.Inc.; 2008.
(173) Jonsson B. Quantitative electromyographic evaluation of muscular load during work. Scand J Rehabil Med Suppl 1978;6:69-74.
(174) Arvidsson I, Balogh I, Hansson GA, Ohlsson K, Akesson I, Nordander C. Rationalization in meat cutting - consequences on physical workload. Appl Ergon 2012 Nov;43(6):1026-32.
(175) Hansson GA, Balogh I, Ohlsson K, Rylander L, Skerfving S. Goniometer measurement and computer analysis of wrist angles and movements applied to occupational repetitive work. J Electromyogr Kinesiol 1996 Mar;6(1):23-35.
(176) Hansson GA, Asterland P, Kellerman M. Modular data logger system for physical workload measurements. Ergonomics 2003 Mar 15;46(4):407-15.
(177) Pedersen CB, Gotzsche H, Moller JO, Mortensen PB. The Danish Civil Registration System. A cohort of eight million persons. Dan Med Bull 2006 Nov;53(4):441-9.
(178) Carstensen B. Lexis macro for SAS. 1-11-2003. Ref Type: Online Source
(179) Carstensen B. Age period cohort models for the Lexis diagram. Statist Med 2007 Jul 10;26(15):3018-45.
(180) Hansson GA, Balogh I, Ohlsson K, Granqvist L, Nordander C, Arvidsson I, et al. Physical workload in various types of work: Part I. Wrist and forearm. International Journal of Industrial Ergonomics 2009 Jan;39(1):221-33.
(181) Hansson GA, Balogh I, Ohlsson K, Granqvist L, Nordander C, Arvidsson I, et al. Physical workload in various types of work: Part II. Neck, shoulder and upper arm. International Journal of Industrial Ergonomics 2010 May;40(3):267-81.
(182) Lidegaard M, Jensen RB, Andersen CH, Zebis MK, Colado JC, Wang Y, et al. Effect of brief daily resistance training on occupational
neck/shoulder muscle activity in office workers with chronic pain: randomized controlled trial. Biomed Res Int 2013;2013:262386.
(183) 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(1):31-41.
(184) Staudenmann D, Roeleveld K, Stegeman DF, van Dieen JH. Methodological aspects of SEMG recordings for force estimation--a tutorial and review. J Electromyogr Kinesiol 2010 Jun;20(3):375-87.
(185) Unge J, Hansson GA, Ohlsson K, Nordander C, Axmon A, Winkel J, et al. Validity of self-assessed reports of occurrence and duration of occupational tasks. Ergonomics 2005 Jan;48(1):12-24.
(186) Wells R, Norman R, Neumann P, Andrews D, Frank J, Shannon H, et al. Assessment of physical work load in epidemiologic studies: common measurement metrics for exposure assessment. Ergonomics 1997 Jan;40(1):51-61.
(187) Kapellusch JM, Garg A, Bao SS, Silverstein BA, Burt SE, Dale AM, et al. Pooling job physical exposure data from multiple independent studies in a consortium study of carpal tunnel syndrome. Ergonomics 2013 May 22;56(6):1021-37.
(188) Luckhaupt SE, Dahlhamer JM, Ward BW, Sweeney MH, Sestito JP, Calvert GM. Prevalence and work-relatedness of carpal tunnel syndrome in the working population, United States, 2010 national health interview survey. Am J Ind Med 2013 Jun 1;56(6):615-24.
(189) Trask C, Mathiassen SE, Wahlstrom J, Heiden M, Rezagholi M. Data collection costs in industrial environments for three occupational posture exposure assessment methods. BMC Med Res Methodol 2012;12:89.
(190) Trask C, Mathiassen SE, Jackson J, Wahlstrom J. Data processing costs for three posture assessment methods. BMC Med Res Methodol 2013;13:124.
(191) Trask C, Mathiassen SE, Wahlstrom J, Forsman M. Cost-efficient assessment of biomechanical exposure in occupational groups, exemplified by posture observation and inclinometry. Scand J Work Environ Health 2014 May 1;40(3):252-65.
(192) Skotte J, Korshoj M, Kristiansen J, Hanisch C, Holtermann A. Detection of physical activity types using triaxial accelerometers. J Phys Act Health 2014 Jan;11(1):76-84.
(193) Lagersted-Olsen J, Korshoj M, Skotte J, Carneiro IG, Sogaard K, Holtermann A. Comparison of objectively measured and self-reported time spent sitting. Int J Sports Med 2014 Jun;35(6):534-40.
(194) Korshoj M, Skotte JH, Christiansen CS, Mortensen P, Kristiansen J, Hanisch C, et al. Validity of the Acti4 software using ActiGraph
GT3X+accelerometer for recording of arm and upper body inclination in simulated work tasks. Ergonomics 2014;57(2):247-53.
(195) ACGIH. ACGIH - TLVs and BEIs. 2014. American Conference of Governmental Industrial Hygienist.
Ref Type: Online Source
(196) Coggon D, Ntani G, Harris E, Linaker C, Van der Star R, Cooper C, et al. Differences in risk factors for neurophysiologically confirmed carpal tunnel syndrome and illness with similar symptoms but normal median nerve function: a case-control study. BMC Musculoskeletal Disorders 2013;14(1):240.
(197) Ntani G, Palmer K, Linaker C, Harris E, Van der Star R, Cooper C, et al. Symptoms, signs and nerve conduction velocities in patients with suspected carpal tunnel syndrome. BMC Musculoskeletal Disorders 2013;14(1):242.
(198) Silverstein B, Fan ZJ, Smith CK, Bao S, Howard N, Spielholz P, et al. Gender adjustment or stratification in discerning upper extremity musculoskeletal disorder risk? Scand J Work Environ Health 2009 Mar;35(2):113-26.
(199) Messing K, Punnett L, Bond M, Alexanderson K, Pyle J, Zahm S, et al. Be the fairest of them all: Challenges and recommendations for the treatment of gender in occupational health research. Am J Ind Med 2003 Jun 1;43(6):618-29.
(200) Messing K, Stock SR, Tissot F. Should studies of risk factors for musculoskeletal disorders be stratified by gender? Lessons from the 1998 Quebec Health and Social Survey. Scand J Work Environ Health 2009 Mar;35(2):96-112.
(201) Giersiepen K, Eberle A, Pohlabeln H. Gender differences in carpal tunnel syndrome? occupational and non-occupational risk factors in a population-based case-control study. Annals of Epidemiology 2000 Oct;10(7):481.
(202) Leclerc A, Chastang JF, Niedhammer I, Landre MF, Roquelaure Y. Incidence of shoulder pain in repetitive work. Occup Environ Med 2004 Jan;61(1):39-44.
(203) Harenstam A. Exploring gender, work and living conditions and health - suggestions for contextual and comprehensive approaches. Scand J Work Environ Health 2009 Mar;35(2):127-33.
(204) Collins LM, Lanza ST. Latent Class Analysis with Covariates. Latent Class and Latent Transition Analysis. John Wiley & Sons, Inc.; 2009. p. 149-77.
(205) Strazdins L, Bammer G. Women, work and musculoskeletal health. Soc Sci Med 2004 Mar;58(6):997-1005.
(206) Yang JF, Cho CY. Comparison of posture and muscle control pattern between male and female computer users with musculoskeletal symptoms. Applied Ergonomics 2012 Jul;43(4):785-91.
(207) Hooftman WE, van der Beek AJ, Bongers PM, van MW. Gender differences in self-reported physical and psychosocial exposures in jobs with both female and male workers. J Occup Environ Med 2005 Mar;47(3):244-52.
(208) van der Beek AJ, Mathiassen SE, Burdorf A. Efficient assessment of exposure to manual lifting using company data. Appl Ergon 2013 May;44(3):360-5.
(209) Kauppinen T, Vincent R, Liukkonen T, Grzebyk M, Kauppinen A, Welling I, et al. Occupational exposure to inhalable wood dust in the member states of the European Union. Ann Occup Hyg 2006 Aug;50(6):549-61.
(210) Thygesen LC, Ersboll AK. Danish population-based registers for public health and health-related welfare research: introduction to the supplement. Scand J Public Health 2011 Jul;39(7 Suppl):8-10.
(211) Thygesen LC, Daasnes C, Thaulow I, Bronnum-Hansen H. Introduction to Danish (nationwide) registers on health and social issues: structure, access, legislation, and archiving. Scand J Public Health 2011 Jul;39(7 Suppl):12-6.
(212) Petersson F, Baadsgaard M, Thygesen LC. Danish registers on personal labour market affiliation. Scand J Public Health 2011 Jul;39(7 Suppl):95-8.
(213) Baadsgaard M, Quitzau J. Danish registers on personal income and transfer payments. Scand J Public Health 2011 Jul;39(7 Suppl):103-5.
(214) Keir PJ, Bach JM, Hudes M, Rempel DM. Guidelines for Wrist Posture Based on Carpal Tunnel Pressure Thresholds. Human Factors: The Journal of the Human Factors and Ergonomics Society 2007 Feb 1;49(1):88-99.
(215) McGorry RW, Fallentin N, Andersen JH, Keir PJ, Hansen TB, Pransky G, et al. Effect of grip type, wrist motion, and resistance level on pressures within the carpal tunnel of normal wrists. J Orthop Res 2014 Apr 1;32(4):524-30.
(216) Werner R, Armstrong TJ, Bir C, Aylard MK. Intracarpal canal pressures: the role of finger, hand, wrist and forearm position. Clinical Biomechanics 1997 Jan;12(1):44-51.
(217) You D, Smith AH, Rempel D. Meta-Analysis: Association Between Wrist Posture and Carpal Tunnel Syndrome Among Workers. Safety and Health at Work 2014 Mar;5(1):27-31.
(218) Klum M, Wolf MB, Hahn P, Lecl+¿re FM, Bruckner T, Unglaub F. Normative Data on Wrist Function. The Journal of Hand Surgery 2012 Oct;37(10):2050-60.
(219) Messing K. Physical Exposures in Work Commonly Done by Women. Can J Appl Physiol 2004 Oct 1;29(5):639-56.
(220) Wergeland EL, Veiersted B, Ingre M, Olsson B, Akerstedt T, Bjornskau T, et al. A shorter workday as a means of reducing the occurrence of musculoskeletal disorders. Scand J Work Environ Health 2003 Feb;29(1):27-34.
(221) Azur MJ, Stuart EA, Frangakis C, Leaf PJ. Multiple imputation by chained equations: what is it and how does it work? Int J Methods Psychiatr Res 2011 Mar 1;20(1):40-9.
Appendix I: Borg CR-10 scale
Appendix II: Log-book for biomechanical measurements
Appendix III: Task- and job exposure matrices
Appendix IV: SHARM questionnaire (in Danish)
Left upper arm elevation
99thpercentile (°) Men 125 (15) 102 (6) 115 (16) 118 (15) 88 (26) 96 (18) 120 (23) 120 (14) 86 (24) Women 110 (15) 120 (27) 113 (20) 120 (17) 94 (18) 97 (16) 111 (16) 116 (15) 80 (15)
>90° (% time) Men 7.6 (5.4) 2.5 (1) 7.9 (7) 10.3 (12) 1.5 (2) 2.4 (3.1) 4.8 (3.5) 6.6 (5.4) 1.7 (3.2) Women 3.8 (2.8) 8 (5.9) 5.7 (4.7) 13.4 (13) 2.6 (3.8) 1.7 (1.4) 6.8 (12.3) 5.6 (3.9) 1.1 (2.2)
Within-minute variation (°)
Men 67 (17) 50 (3) 57 (12) 64 (18) 41 (11) 41 (12) 55 (12) - 33 (15) Women 57 (14) 70 (20) 55 (14) 69 (18) 47 (13) 41 (4) 52 (10) - 26 (9)
Between-minute variation (°)
Men 30 (2) 23 (3) 26 (4) 27 (5) 19 (7) 21 (11) 28 (7) - 20 (8) Women 25 (3) 24 (10) 25 (6) 26 (6) 21 (6) 20 (4) 27 (5) - 19 (5)
Median velocity (°/s) Men 42 (12) 44 (10) 36 (15) 45 (21) 41 (19) 32 (13) 31 (12) 35 (11) 18 (19) Women 42 (12) 55 (19) 31 (11) 43 (15) 42 (15) 28 (11) 35 (11) 35 (11) 10 (8)
Exposure values for the postures and movements per task/gender, left wrist. Data are shown for the 7 tasks that constitute the work. Additionally data are shown for total work and pauses. For flexion/extension and ulnar/radial deviation positive angles denote flexion and ulnar deviation and negative angles extension and radial deviation. MPF = mean power frequency.
Full
leveling
Sanding (by
hand)
Painting (brush)
Painting (roll)
Covering, carrying
and cleaning
Driving Other Total work
Pause
Flexion/extension
Percentile (°) 10th Men -47 (8) -54 (10) -50 (7) -50 (14) -42 (12) -48 (10) -50 (10) -49 (8) -45 (15) Women -52 (15) -60 (17) -53 (13) -52 (11) -47 (13) -48 (11) -48 (9) -51 (10) -46 (8)
50th Men -20 (7) -21 (7) -20 (8) -17 (9) -14 (8) -22 (12) -21 (9) -18 (8) -15 (10) Women -21 (11) -27 (14) -23 (10) -25 (11) -19 (9) -18 (10) -21 (11) -21 (10) -18 (9)
90th Men 2 (9) 2 (11) 3 (12) 5 (11) 9 (9) 10 (13) 5 (12) 6 (11) 13 (16) Women 6 (9) 0 (14) 6 (13) 3 (14) 7 (9) 6 (9) 7 (11) 8 (11) 12 (14)
95-5th Men 64 (7) 76 (11) 70 (10) 70 (11) 74 (16) 73 (12) 70 (12) 74 (6) 73 (18) Women 77 (9) 76 (19) 77 (15) 72 (15) 70 (16) 72 (16) 75 (12) 77 (11) 74 (13)
Median velocity (°/s) Men 11 (4) 10 (3) 10 (4) 12 (8) 10 (3) 9 (4) 9 (4) 9 (4) 6 (5) Women 11 (4) 15 (4) 8 (5) 12 (6) 12 (5) 8 (3) 10 (3) 10 (4) 4 (2)
Repetitiveness (MPF; Hz) Men .26 (.03) .23 (.05) .20 (.04) .24 (.08) .23 (.05) .25 (.05) .22 (.03) .22 (.04) .19 (.05) Women .22 (.03) .24 (.06) .20 (.05) .24 (.06) .27 (.07) .26 (.07) .24 (.03) .22 (.04) .18 (.04)
Ulnar/radial deviation
Percentile (°) 10th Men -12 (9) -22 (5) -21 (10) -22 (9) -16 (7) -16 (7) -17 (9) -21 (9) -21 (11) Women -16 (7) -21 (10) -28 (11) -22 (10) -20 (11) -19 (8) -23 (9) -22 (9) -21 (9)
50th Men 0 (7) -7 (5) -5 (8) -7 (8) -4 (7) -4 (8) 3 (8) -5 (8) -5 (9) Women -2 (5) -5 (9) -7 (8) -5 (8) -3 (6) -5 (7) -4 (9) -4 (7) -5 (9)
90th Men 18 (5) 6 (5) 12 (10) 9 (10) 11 (8) 10 (9) 12 (8) 11 (9) 9 (9) Women 13 (4) 9 (9) 9 (8) 10 (8) 11 (6) 8 (6) 12 (7) 11 (7) 10 (9)
95-5th Men 38 (6) 37 (3) 43 (12) 39 (7) 36 (9) 32 (4) 36 (5) 42 (9) 39 (10) Women 39 (5) 42 (8) 48 (11) 42 (6) 42 (13) 35 (5) 43 (5) 43 (6) 39 (9)
Median velocity (°/s) Men 7 (2) 6 (1) 5 (3) 7 (4) 6 (2) 5 (2) 5 (2) 5 (2) 3 (3) Women 6 (2) 10 (4) 5 (4) 8 (4) 7 (3) 5 (2) 6 (2) 6 (3) 2 (2)
Repetitiveness (MPF; Hz) Men .24 (.02) .23 (.03) .21 (.05) .25 (.07) .23 (.05) .28 (.07) .23 (.04) .23 (.05) .20 (.06) Women .24 (.05) .27 (.06) .19 (.05) .25 (.06) .25 (.07) .27 (.05) .23 (.04) .23 (.05) .19 (.03)
Number of recordings Men 5 5 14 13 12 8 10 25 23 Women 7 8 17 15 16 8 15 25 25
Mean recording duration in minutes
Men 88 102 141 128 49 55 118 280 45
Women 158 51 149 102 55 41 77 317 61
Paper I: Sex differences in muscular load among house painters per-
forming identical work tasks.
(Eur J Appl Physiol 2014;1-11)
Paper II: Sex differences in task distribution and task exposures among
Danish house painters: An observational study combining
questionnaire data with biomechanical measurements.
(Submitted to PLOS ONE)
Paper III: Exposure-response relationship between postures and move-
ments of the wrist and carpal tunnel syndrome among male
and female house painters: a retrospective cohort study.
(un-submitted)
1 3
Eur J Appl PhysiolDOI 10.1007/s00421-014-2918-6
OrIgIn Al ArtIclE
Sex differences in muscular load among house painters performing identical work tasks
Jacob Meyland · Thomas Heilskov-Hansen · Tine Alkjær · Henrik Koblauch · Sigurd Mikkelsen · Susanne Wulff Svendsen · Jane Frølund Thomsen · Gert-Åke Hansson · Erik B. Simonsen
received: 22 December 2013 / Accepted: 16 May 2014 © Springer-Verlag Berlin Heidelberg 2014
newton, based on ramp calibrations, were assessed. Sex dif-ferences were tested using a mixed model approach.Results Women worked at about 50 % higher relative mus-cular loads than men in the supraspinatus and forearm mus-cles at all percentiles and in all tasks. Women exerted about 30 % less force in the trapezius muscle at the 50th percentile.Conclusions Female house painters had a higher relative muscular load than their male colleagues without exerting more force. the effects of a higher relative muscular load accumulated over years of work may in part explain why musculoskeletal complaints in the upper body occur more frequently among women than men.
Keywords Electromyography · EMg · Surface EMg · Intramuscular EMg · Force · Upper extremity · Musculoskeletal complaints · Occupational medicine
AbbreviationsAPDF Amplitude probability distribution functionEMg ElectromyographyEMg max Maximum EMg amplitudecI confidence intervalMSD Musculoskeletal disordersMVc Maximal voluntary contractionP10 10 % percentileP50 50 % percentileP90 90 % percentileSD Standard deviation
Introduction
Musculoskeletal disorders (MSD) affect a large number of people and have important socioeconomic consequences in terms of costs in health care, sickness absence and early
Abstract Purpose the present study aimed to estimate possible differ-ences in upper body muscular load between male and female house painters performing identical work tasks. Sex-related differences in muscular load may help explain why women, in general, have more musculoskeletal complaints than men.Methods In a laboratory setting, 16 male and 16 female house painters performed nine standardised work tasks com-mon to house painters. Unilateral electromyography (EMg) recordings were obtained from the supraspinatus muscle by intramuscular electrodes and from the trapezius, extensor and flexor carpi radialis muscles by surface electrodes. rela-tive muscular loads in %EMgmax as well as exerted force in
communicated by Peter Krustrup.
J. Meyland (*) · t . Heilskov-Hansen · S. Mikkelsen · J. F. thomsen Department of Occupational and Environmental Medicine, Bispebjerg University Hospital, Bispebjerg Bakke 23, 2400 copenhagen, Denmarke-mail: [email protected]
t . Alkjær · H. Koblauch · E. B. Simonsen Department of neuroscience and Pharmacology, University of copenhagen, copenhagen, Denmark
S. W. Svendsen Danish ramazzini centre, Department of Occupational Medicine, regional Hospital West Jutland–University research clinic, Herning, Denmark
g. Hansson Occupational and Environmental Medicine, lund University, lund, Sweden
g. Hansson Occupational and Environmental Medicine, University and regional laboratories region Scania, lund, Sweden
Eur J Appl Physiol
1 3
retirement (norlund et al. 2000). therefore, it is of great importance to understand which factors contribute to the development of MSD.
Several studies have shown that women report musculo-skeletal complaints in the upper parts of the body more fre-quently than men (Bingefors and Isacson 2004; Dahlberg et al. 2004; de Zwart et al. 2001; Hooftman et al. 2004; Wijnhoven et al. 2006), but the causes of this difference are unclear. It could be explained in three different ways: (1) women are exposed to higher relative physical loads than men (Hooftman et al. 2004; Kennedy and Koehoorn 2003; leijon et al. 2005; lundberg 2002; nordander et al. 2008); (2) women are more vulnerable to physical loads than men due to anatomical, physical or other biological differences (de Zwart et al. 2001); (3) women report complaints at a lower threshold than men (Hurley and Adams 2008).
these hypotheses are under investigation in an ongoing epidemiological study focusing on Danish male and female house painters [the Shoulder Hand Arm (SHArM) study]. In Denmark, approximately 1/3 of house painters are women, and male and female house painters are assumed to perform virtually identical work tasks. the purpose of the present part of the SHArM-study was to examine whether there were differences in relative muscular load and exerted force between male and female house painters when they performed identical house painting work tasks under con-trolled experimental conditions.
Methods
Subjects
the study included 32 right-handed house painters: 16 men [mean (SD): weight 82.4 (12.9) kg, height 182 (6.1) cm, age 25.5 (3.9) years] and 16 women [mean (SD): weight 64.9 (7.1) kg, height 169 (6.2) cm, age 28.3 (6.2) years]. Sub-jects with previous reports of discomfort in the right shoul-der or the right or left arm and/or hand were excluded from the study. the subjects were recruited through the Painters’ Union in Denmark. Each subject participated voluntarily in the study and signed an informed consent according to the Helsinki Declaration. the study was approved by the local ethics committee (J. no.: H-3-2011-157).
Electromyography
the descending part of the trapezius muscle was selected as representative of elevation of the shoulder girdle, the supraspinatus muscle as an important abductor at the shoul-der joint and clinically relevant as it is often reported to suffer from degenerative alterations of its tendon (Svendsen et al. 2004b). the flexor and extensor radialis muscles were
chosen because of the painters’ frequent use of dorsiflexion and palmar flexion of the wrist joint.
EMg of the trapezius muscle (descending part) and mm. extensor and flexor carpi radialis on the right side was studied using bipolar silver/silver chloride surface electrodes (Multi Bio Sensors, tX, USA) with a fixed inter-electrode distance of 20 mm. According to Per-otto (2005), the electrodes were placed on the trapezius between the neck and the shoulder, on the extensor carpi radialis approximately 5 cm distal to the lateral epicon-dyle and on the flexor carpi radialis approximately 10 cm distal to the medial epicondyle. Prior to mounting the electrodes, the skin was shaved and cleaned with alcohol. the supraspinatus muscle was recorded by two intramus-cular wire electrodes (500 µm diameter, 1,500 µm un-insulated tip, stainless steel) [Spes Medica, Battipaglia (SA), Italy]. the site of insertion was found according to rudroff (2008). With the subject in prone position, the two wire electrodes were inserted into the supraspinatus muscle, penetrating the trapezius muscle with hypoder-mic needles. An inter-electrode distance of 20 mm was sought. A reference electrode was placed on the acromion (left side). All electrodes were connected to lightweight pre-amplifiers containing a 10–1,000 Hz band-pass filter and a 16 bit analog/digital converter. the sampling rate was 2,048 Hz. the digitized signals were lead through wires to a small box (120 g) fixed to the subject’s right upper arm (MQ16, Marq-Medical, Farum, Denmark) and transmitted wirelessly to a Pc using Bluetooth technol-ogy. the noise level of the MQ16 system was specified to be less than 3 μV. In the MQ16 system, the signals were pre-amplified and AD converted close to the electrodes to reduce movement artefacts.
the EMg amplitude of each muscle was calibrated to an isometric external force by simultaneously sampling EMg and the signal from a force transducer. A gradually increasing ramp contraction (Jonsson 1978) was performed over 15 s going from absolute relaxation increasing steadily until maximum effort was achieved. An additional period of 5 s was given to make sure that maximal voluntary con-traction (MVc) was reached, 30 % MVc was then calcu-lated and the relation between force and EMg was deter-mined by a linear relationship up to 30 % MVc. this was obtained by a least square fit forced through zero (Fig. 1) (Jonsson 1982; Mathiassen 1996).
In all exercises, a force transducer (Hottinger Bald-win Messtechnik, Darmstadt gmbH, germany), anchored to the floor, was connected in series with a nylon strap applied to the force exerting limb. For the forearm mus-cles, the subject was seated with the right forearm fixed in a supinated position for flexion and pronated for exten-sion, respectively. the hand was fixed by the nylon strap applied to the distal end of metacarpals 2–5. For the
Eur J Appl Physiol
1 3
supraspinatus muscle, the subject was standing with the right arm abducted 45° pressing the loop of the strap firmly outwards/upwards by the back of the hand. For the trape-zius muscle, the subject was standing with two nylon straps placed bilaterally on the acromion. the subject was asked to elevate both shoulders while keeping heel contact with the floor. Only the nylon strap on the subject’s right side was connected to the force transducer.
Additionally, MVc was measured for the supraspina-tus and trapezius muscles with the arm abducted 90° in the frontal plane. the strap was placed just proximal to the elbow. the two forearm muscles were also tested during power grip by use of a hand grip dynamometer (JAMAr, Sammons Preston, Bolingbrook Il, USA). In these two tests, the subjects were given three trials to reach the high-est possible force.
For the supraspinatus and trapezius muscles, the mean maximal EMg amplitude was, for both men and women, higher during arm abduction at 90°, than during arm abduc-tion at 45° and shoulder elevation, respectively (table 1). the maximum EMg amplitude during arm abduction at 90° was, for all subjects, used for normalisation when cal-culating the relative muscular load (see below) for these muscles.
A subjective measure of the muscular load was obtained immediately after completion of each task by asking the participants to rate perceived exertion using the Borg cr10 scale (Borg 1990; Borg et al. 1987).
nine work tasks, chosen in collaboration with the Paint-er’s Union in Denmark were investigated. they covered everyday work tasks in the field of house painting and were performed in the order presented: 1: Sanding (by hand), 2: painting (brush), 3: mounting glass-felt, 4: painting wall (roll), 5: painting ceiling (roll), 6: full levelling wall, 7: full levelling ceiling, 8: sanding wall (“giraffe” dry-wall-sander) and 9: sanding ceiling (“giraffe” dry-wall-sander) (Fig. 2). the number of square meters in each task was pre-determined. the subject had time to prepare the necessary tools for the task. the investigator verbally started the task and the subject signalled when the task was completed. Data was missing from two female subjects in task 9, one due to a technical failure and one being too short to per-form the task.
Data processing
Using Matlab r2008b (the MathWorks Inc., USA), EMg data were filtered by a fourth-order Butterworth filter with a band pass of 10–500 Hz for surface electrodes and 30–1,000 Hz for intramuscular measurements. By visual
Fig. 1 EMg–force calibration for a typical male subject, here show-ing the flexor carpi radialis muscle. the recording was performed during a steadily increasing force from 0 % to about 30 % of MVc for 15 s (EMG electromyography, MVC maximal voluntary contrac-tion)
Table 1 Mean MVc and EMg max of men (N = 16) and women (N = 16)
the P values refer to differences (t test) between men and women
MVc (n) P EMg max (mV) P
Men Women Men Women
Mean SD Mean SD Mean SD Mean SD
M. supraspinatus
Abd45 109 21 66 15 <0.001 1.40 0.63 1.10 0.56 0.17
Abd90 268 39 176 35 <0.001 1.44 0.66 1.14 0.59 0.19
M. trapezius
Elevation 467 120 299 137 <0.001 0.72 0.51 0.66 0.45 0.71
Abd90 268 39 176 35 <0.001 0.99 0.59 1.10 0.60 0.60
M. ext. carpi radialis
Extension 203 56 125 34 <0.001 1.31 0.56 1.00 0.58 0.14
Power grip 500 92 333 42 <0.001 1.00 0.59 0.80 0.46 0.28
M. flex. carpi radialis
Flexion 230 51 135 42 <0.001 1.36 0.56 1.10 0.40 0.14
Power grip 500 92 333 42 <0.001 0.58 0.18 0.51 0.20 0.27
Eur J Appl Physiol
1 3
inspection of the signals, short periods with high interfer-ence or noise were rejected from further analysis. less than 8 % of the total recording time was excluded. root mean
square amplitudes were calculated for epochs of 125 ms throughout the sampling period. Amplitude probability dis-tribution functions (APDF) (Jonsson 1982) were obtained
Fig. 2 the work tasks investigated, from the top left: Sanding (by hand); painting (brush); mounting glass felt; painting wall (roll); painting ceiling (roll); full levelling, wall; full levelling, ceiling; sand-
ing wall (“giraffe” dry-wall-sander); and sanding ceiling (“giraffe” dry-wall-sander)
Eur J Appl Physiol
1 3
for each subject, muscle and task. this was performed by sorting an entire measurement period in ascending order, thereby achieving a vector of EMg values wherein the relative position of an EMg value within the vector des-ignated the proportion of the measurement period less or equal to that particular EMg value. Percentiles was used to identify values obtained from the APDF, e.g. the 10 % percentile (p10) designates the EMg value for which 10 % of the measurement period was less or equal. three percen-tiles were selected for statistical analysis: the 10 % percen-tile (p10) the “static load,” the 50 % percentile (p50) the median load and the 90 % percentile (p90) the peak load (Jensen et al. 1993; Jonsson 1978) (Fig. 3). relative mus-cular load and exerted force were obtained by normalis-ing the EMg amplitudes to EMgmax and by expressing the EMg amplitudes in newton via the EMg-to-force calibra-tions, respectively.
Statistics
normality of EMg responses was examined using QQ plots and showed a need for logarithmic transformation. Differences were back-transformed giving results as ratios. Each muscle and percentile level was analysed using a two factor mixed model examining dependences of task, sex and task–sex interaction.
the ratings for perceived physical exertion were ana-lysed for differences between men and women within each task, using an un-paired double-sided t test with unequal variance. correlations between %EMgmax and Borg cr10 ratings were tested using Pearson’s correla-tion coefficient.
level of significance was set to 5 %. Statistical analy-ses were conducted with SAS 9.2 (SAS Institute Inc., cary, nc, USA).
Results
In all strength measurements, male subjects were signifi-cantly stronger than female subjects regarding absolute forces (P < 0.001). Men were on average 50–70 % stronger than women in the different measurements of MVc. no statistically significant differences between sexes were found for EMgmax (P = 0.14), although men had numeri-cally higher mean EMgmax than women in 7 out of 8 tests (t able 1).
t otal time spent on all nine work tasks varied between subjects from 27 to 47 min with a mean of 36 min. t otal work duration did not vary significantly (P = 0.18) between men [mean (SD): 34 min 43 s (5 min 25 s)] and women [mean (SD): 37 min 16 s (4 min 58 s)]. the Borg cr10 ratings of perceived exertion showed statistically
significant differences between sexes for tasks 5–9 (Fig. 4). In all tasks, women on average rated their physical exertion higher than men did.
relative muscular load
Figure 5 depicts the average relative muscular load for each muscle, percentile and task. For all muscles, percen-tiles and tasks (except for m. trapezius, p10, task 9) women had a higher relative muscular load. For the supraspinatus muscle, average muscular load for p10 was 2.07 %EMg-
max for the women and 1.48 %EMgmax for the men, p50 was 8.38 %EMgmax for the women and 5.60 %EMgmax for the men, and p90 was 21.88 %EMgmax for the women
Fig. 3 Example of an APDF curve obtained from the trapezius mus-cle of a typical subject during task 3. Dashed lines show p10, p50 and p90, respectively (APDF amplitude probability distribution functions)
Fig. 4 Borg cr10 scale for perceived physical exertion in nine tasks, females (grey bars) and males (shaded bars). Mean values and 95 % confidence intervals (indicated by error bars). Asterisks denote sig-nificant differences between sexes
Eur J Appl Physiol
1 3
and 17.85 %EMgmax for the men. For the trapezius mus-cle, average muscular load for p10 was 1.92 %EMgmax for the women and 1.59 %EMgmax for the men, p50 was 9.00 %EMgmax for the women and 6.80 %EMgmax for the men, and p90 was 21.78 %EMgmax for the women and 17.47 %EMgmax for the men. For the extensor carpi radialis
muscle, average muscular load for p10 was 4.62 %EMgmax for the women and 2.83 %EMgmax for the men, p50 was 9.86 %EMgmax for the women and 6.50 %EMgmax for the men, and p90 was 19.82 %EMgmax for the women and 14.16 %EMgmax for the men. For the flexor carpi radialis muscle, average muscular load for p10 was 2.51 %EMgmax
Fig. 5 relative muscular load in nine tasks, p10 (top), p50 (middle) and p90 (bottom), females (N = 16 grey bars) and males (N = 16 shaded bars). Mean values and 95 % confidence intervals (indicated by error bars) were calculated on logarithmic data, and retransformed to linear scales for presentation. t asks are: 1: Sanding (by hand), 2: painting (brush), 3: mounting glass felt, 4: painting wall (roll), 5:
painting ceiling (roll), 6: full levelling, wall, 7: full levelling, ceiling, 8: sanding wall (“giraffe” dry-wall- sander), and 9: sanding ceiling (“giraffe” dry-wall-sander) [p10, p50 and p90 are the 10th, 50th and 90th percentiles, respectively]. p10, p50 and p90 for the supraspina-tus and extensor and flexor carpi radialis muscles showed significant effect of sex adjusted for task
Eur J Appl Physiol
1 3
for the women and 1.74 %EMgmax for the men, p50 was 7.57 %EMgmax for the women and 5.36 %EMgmax for the men, and p90 was 15.97 %EMgmax for the women and 11.25 %EMgmax for the men.
the task–sex interaction was found to be insignifi-cant regarding relative muscular load. leaving only main effects (sex and task) in the model, p10, p50 and p90 for the supraspinatus and extensor and flexor carpi radia-lis muscles were found to have a significant effect of sex adjusted for task. the trapezius muscle showed an effect in the same direction although not statistically significant. All significant sex effects were due to women being exposed to a higher load than men (table 2).
Exerted force
Average exerted force for each muscle, percentile and task can be seen in Fig. 6. For the supraspinatus mus-cle, average exerted force for p10 was 1.68 n for the women and 1.54 n for the men, p50 was 6.70 n for the women and 7.60 n for the men, and p90 was 17.72 n for the women and 23.81 n for the men. For the trapezius muscle, average exerted force for p10 was 7.86 n for the women and 11.08 n for the men, p50 was 36.82 n for the women and 47.21 n for the men, and p90 was 89.14 n for the women and 121.35 n for the men. For the exten-sor carpi radialis muscle, average exerted force for p10, was 7.92 n for the women and 8.48 n for the men, p50 was 16.91 n for the women and 19.43 n for the men, and p90 was 33.99 n for the women and 42.39 n for the men. For the flexor carpi radialis muscle, average exerted force for p10, was 4.79 n for the women and 4.52 n for the men, p50 was 14.70 n for the women and 13.77 n for the men, and p90 was 30.50 n for the women and 29.29 n for the men.
the task–sex interaction was found to be significant in a single case of p10 of the trapezius muscle (P = 0.02), and post hoc analysis showed the difference to be found in task 9 (sanding ceiling with the “giraffe” dry-wall-sander), which is also visible in Fig. 6.
leaving only main effects (sex and task) in the model, only p50 of the trapezius muscle was found to be signifi-cantly affected by sex adjusted for task. Estimates of sex effects adjusted for task are shown in t able 3, which shows that in the significant case of p50 of the trapezius muscle, women exerted 30 % less force compared with men.
Discussion
In this study, we found that women experienced signifi-cantly higher relative muscular loads during house painting tasks than men did. three of the four muscles investigated, the supraspinatus and extensor and flexor carpi radialis muscles, showed that women experienced a significantly higher relative muscular load in the three EMg parameters: p10, p50 and p90, which furthermore corresponded to the subjective ratings using Borg cr10. Additionally, no sig-nificant task–sex interaction was found in terms of relative muscular load. Women were shown to exert a lower p50 of force in trapezius in all tasks as well as a lower p10 in task 9, while the other muscles investigated showed no signifi-cant differences between men and women.
Measures of exposure can be assessed by self-reports (Borg and Kaijser 2006; Mcgorry et al. 2010), observa-tions (Moore and garg 1995) or technical measurements by, e.g. EMg. Previously, EMg has been used extensively to quantify muscular load during work situations (Hans-son et al. 2000, 2009, 2010; Jensen et al. 1993, 1999; Jon-sson 1982; Veiersted et al. 1990). t raditionally, EMg is reported in %EMgmax giving a relative muscular load. It is, however, well known, and verified in the present study, that on average men are about 50 % stronger than women (Maughan et al. 1983; Miller et al. 1993) without EMgmax in men being correspondingly higher. therefore, a meas-ure of exerted force was considered relevant for the present study. A linear relationship is in many cases a reasonable description of the relationship between EMg amplitude and force (Staudenmann et al. 2010). However, the rela-tionship between EMg and force is not linear over the full
Table 2 Estimated sex effect on relative muscular loads, adjusted for task
load for women (N = 16) compared to that of men (N = 16), male level being 100 %a 0.001 < P < 0.010b 0.010 < P < 0.050
p10 (%) p50 (%) p90 (%)
Mean 95 % cI Mean 95 % cI Mean 95 % cI
M. supraspinatus 187a 131–266 156a 118–207 131a 109–158
M. trapezius 119 83–171 120 93–155 127 99–162
M. ext. carpi radialis 180a 120–270 164a 114–237 149b 107–207
M. flex. carpi radialis 159b 112–225 140b 102–193 136b 101–183
Eur J Appl Physiol
1 3
range of force, and therefore, we limited the calibration range to about 30 % of MVc and estimated a linear EMg/force relation for each muscle in each subject. the esti-mated force should of course be interpreted with caution
when the load exceeded 30 % MVc, and the resulting force in newton should not be taken as a precise measure of the exerted force of the individual muscle, since this is influ-enced by, e.g. the length of the muscle, contributions from
Fig. 6 Exerted force in nine tasks, p10 (top), p50 (middle) and p90 (bottom), females (N = 16 grey bars) and males (N = 16 shaded bars). Mean values and 95 % confidence intervals (indicated by error bars) were calculated on logarithmic data and retransformed to linear scales for presentation. t asks are: 1: Sanding (by hand), 2: painting (brush), 3: mounting glass felt, 4: painting wall (roll), 5: painting ceil-
ing (roll), 6: full levelling, wall, 7: full levelling, ceiling, 8: sanding wall (“giraffe” dry-wall-sander), and 9: sanding ceiling (“giraffe” dry-wall-sander) [p10, p50 and p90 are the 10th, 50th and 90th per-centiles, respectively]. Asterisk denotes significant task–sex inter-action. P50 of the trapezius muscle showed significant effect of sex adjusted for task
Eur J Appl Physiol
1 3
synergistic muscles, and eccentric and concentric contrac-tions. the external force that is exerted, e.g. on a tool, is even harder to predict from the recorded EMg since this is affected by adjacent muscles and joint geometry. Using EMg normalisation from one contractile condition to another may introduce some disparities. However, a recent review paper (Burden 2010) emphasises that this method is well accepted and is not inferior to other methods, i.e. numerous MVc’s at different rOM’s, which are consid-ered much more labour intensive.
the role of the supraspinatus muscle is of special inter-est due to its functional importance for the shoulder joint and the frequent occurrence of shoulder complaints with clinical findings presumed to originate from, e.g. degen-erative alterations of the supraspinatus tendon, which are linked to work tasks with arms elevated above 90° (Svend-sen et al. 2004a, b; van rijn et al. 2010; Jensen et al. 1994). Most studies of occupational EMg in the upper extremity have not measured the supraspinatus muscle, probably due to the deep location of the muscle underneath the trape-zius muscle. Others have tried to assess muscular load of the supraspinatus muscle by the use of surface electrodes (Waite et al. 2010). the measurement of supraspinatus activity by use of surface electrodes has, however, shown a weak correlation with intramuscular measurements (Waite et al. 2010) stressing the need for examining the supraspi-natus muscle by the use of intramuscular electrodes during work. the forearm extensor and flexor muscles as well as the trapezius muscle can all be measured validly by surface electrodes.
this study investigated the work of house painters in a condensed way without pauses between tasks. In compari-son with results obtained in whole day recordings, the mus-cular load measured in the present study will probably be high because this study investigated condensed periods of work. However, this deviation is expected to be equal for men and women.
Several studies have found the Borg cr10 scale of perceived exertion to have a good correlation with physi-ological responses (Borg et al. 1987; capodaglio 2001; Pincivero et al. 2004; DiDomenico and nussbaum 2008). In accordance with previous studies (Borg et al. 1987; Pin-civero et al. 2001; t roiano et al. 2008), we found that the Borg cr10 had a linear relationship with relative muscular
load. these results indicate that the Borg cr10 scale may be a valid subjective tool to assess sex differences of a given work-related task with respect to relative muscular load.
In several tasks of the present study, it was observed that p10 of relative muscular load was lower, while p90 was higher compared to dentists (Akesson et al. 1997), hair-dressers (chen et al. 2010) and assembly plant employees (christensen 1986). thus, the tasks performed by house painters were in these cases of a more diverse and dynamic character.
the highest reported difference between men and women was found by chen et al. (2010) in a study of barbers and hairdressers. When normalised to EMgmax, women had an EMg activity that was 100–200 % higher than men. this could, however, be partly explained by large standard deviations as reported in that particular study.
the differences in relative muscular load found in the present study corroborated the findings of nordander et al. (2008) in a study of industrial workers, regarding the trape-zius and the forearm extensors. likewise, Won et al. (2009) found that during keyboard typing, women had a higher relative muscular load than men for the trapezius and fore-arm extensor and flexor muscles, while no differences were found in the exerted force.
the study of keyboard typing by Won et al. (2009) showed that men and women exerted the same force, while a study of postal workers (van der Beek et al. 2000) showed that men exerted greater forces, which was also found in a few cases in the present study. A reasonable explanation is that the more strength demanding the task is, the more apparent the sex difference will be because men are capable of exerting more force if required.
It is interesting that the relative muscular load in the supraspinatus, extensor carpi radialis and flexor carpi radi-alis muscles was between 30 and 90 % greater in women than in men, considering the fact that these muscles are subject to work-induced MSD (supraspinatus tendinopathy, medial and lateral epicondylitis) (nordander et al. 2013; van rijn et al. 2009, 2010). A recent study by nordander et al. (2013) linked work with a high relative muscular load to the presence of pain and indicated that accumulation of effects caused by a high relative muscular load may be linked to MSD.
Table 3 Estimated sex effect of exerted force, adjusted for task
Force exerted by women (N = 16) compared to that of men (N = 16), male level being 100 %a 0.010 < P < 0.050
p10 (%) p50 (%) p90 (%)
Mean 95 % cI Mean 95 % cI Mean 95 % cI
M. supraspinatus 111 78–158 95 70–129 80 62–103
M. trapezius 70 48–101 70a 49–99 74 50–108
M. ext. carpi radialis 104 69–158 95 67–136 86 62–121
M. flex. carpi radialis 116 79–171 103 71–148 99 70–141
Eur J Appl Physiol
1 3
the lower exerted force of women found in the trapezius muscle indicates that the work was possibly performed dif-ferently by men and women with respect to postures and movements.
there were no significant task–sex interactions with respect to relative muscular load, indicating that sex dif-ferences with respect to muscular load were similar across work tasks. However, by differentiating the tasks so that women avoid the most physically challenging tasks, the total daily exposure would be lowered in women. Another way of reducing the difference in relative muscular load could—in theory—be to increase maximal muscle strength of women by strength training. However, it is most unlikely that MVc in female house painters could be increased by 30–90 %, which would be necessary to match the MVc of men. nevertheless, previous studies have found strength training to be an effective method of reducing pain in other occupational groups (Andersen et al. 2010, 2012).
Conclusions
Female house painters worked at a higher relative muscular load than their male colleagues even though men in some cases exerted more force during work. It is possible that the effects of a higher relative muscular load among female house painters can be accumulated over years contribut-ing to the explanation of why upper body complaints occur more frequently among women. We encourage prospective studies to investigate effects of relative muscular load and exerted force on development of MSD.
Acknowledgments the present study was supported by grants from the Danish rheumatism Association and the Danish Working Environment research Fund. the authors wish to thank co-workers at the University Hospital in lund and Bispebjerg University Hospital, especially student Hatice Yilmaz for assistance during data collection and Jacob louis Marott for statistical guidance. Also the collaboration of the Painters’ Union in Denmark is highly appreciated.
Conflict of interest the authors declare that they have no conflict of interest.
References
Akesson I, Hansson gA, Balogh I, Moritz U, Skerfving S (1997) Quantifying work load in neck, shoulders and wrists in female dentists. Int Arch Occup Environ Health 69(6):461–474
Andersen ll, christensen KB, Holtermann A, Poulsen OM, Sjogaard g, Pedersen Mt, Hansen EA (2010) Effect of physical exercise interventions on musculoskeletal pain in all body regions among office workers: a one-year randomized controlled trial. Man ther 15(1):100–104
Andersen ll, Jakobsen MD, Pedersen Mt, Mortensen OS, Sjogaard g, Zebis MK (2012) Effect of specific resistance training on fore-arm pain and work disability in industrial technicians: cluster ran-domised controlled trial. BMJ Open 2(1):e000412
Bingefors K, Isacson D (2004) Epidemiology, co-morbidity, and impact on health-related quality of life of self-reported headache and mus-culoskeletal pain—a gender perspective. Eur J Pain 8(5):435–450
Borg g (1990) Psychophysical scaling with applications in physi-cal work and the perception of exertion. Scand J Work Environ Health 16(Suppl 1):55–58
Borg E, Kaijser l (2006) A comparison between three rating scales for perceived exertion and two different work tests. Scand J Med Sci Sports 16(1):57–69
Borg g, Hassmen P, lagerstrom M (1987) Perceived exertion related to heart rate and blood lactate during arm and leg exercise. Eur J Appl Physiol Occup Physiol 56(6):679–685
Burden A (2010) How should we normalize electromyograms obtained from healthy participants? What we have learned from over 25 years of research. J Electromyogr Kinesiol 20:1023–1035
capodaglio EM (2001) comparison between the cr10 Borg’s scale and the VAS (visual analogue scale) during an arm-cranking exercise. J Occup rehabil 11(2):69–74
chen Hc, chang cM, liu YP, chen cY (2010) Ergonomic risk fac-tors for the wrists of hairdressers. Appl Ergon 41(1):98–105
christensen H (1986) Muscle activity and fatigue in the shoulder muscles of assembly-plant employees. Scand J Work Environ Health 12(6):582–587
Dahlberg r, Karlqvist l, Bildt c, nykvist K (2004) Do work tech-nique and musculoskeletal symptoms differ between men and women performing the same type of work tasks? Appl Ergon 35(6):521–529
de Zwart Bc, Frings-Dresen MH, Kilbom A (2001) gender differ-ences in upper extremity musculoskeletal complaints in the work-ing population. Int Arch Occup Environ Health 74(1):21–30
DiDomenico A, nussbaum MA (2008) Interactive effects of physical and mental workload on subjective workload assessment. Int J Ind Ergon 38(11–12):977–983
Hansson gA, Balogh I, Ohlsson K, Palsson B, rylander l, Skerfv-ing S (2000) Impact of physical exposure on neck and upper limb disorders in female workers. Appl Ergon 31(3):301–310
Hansson gA, Balogh I, Ohlsson K, granqvist l, nordander c, Arvidsson I, Akesson I, Unge J, rittner r, Stromberg U, Skerfv-ing S (2009) Physical workload in various types of work: Part I. Wrist and forearm. Int J Ind Ergon 39(1):221–233
Hansson gA, Balogh I, Ohlsson K, granqvist l, nordander c, Arvidsson I, Akesson I, Unge J, rittner r, Stromberg U, Skerfv-ing S (2010) Physical workload in various types of work: Part II. neck, shoulder and upper arm. Int J Ind Ergon 40(3):267–281
Hooftman WE, van Poppel Mn, van der Beek AJ, Bongers PM, van Mechelen W (2004) gender differences in the relations between work-related physical and psychosocial risk factors and musculo-skeletal complaints. Scand J Work Environ Health 30(4):261–278
Hurley rW, Adams Mc (2008) Sex, gender, and pain: an overview of a complex field. Anesth Analg 107(1):309–317
Jensen Br, Schibye B, Sogaard K, Simonsen EB, Sjogaard g (1993) Shoulder muscle load and muscle fatigue among industrial sewing-machine operators. Eur J Appl Physiol Occup Physiol 67(5):467–475
Jensen Br, Jorgensen K, Sjogaard g (1994) the effect of prolonged isometric contractions on muscle fluid balance. Eur J Appl Phys-iol Occup Physiol 69(5):439–444
Jensen c, Finsen l, Hansen K, christensen H (1999) Upper trapezius muscle activity patterns during repetitive manual material han-dling and work with a computer mouse. J Electromyogr Kinesiol 9(5):317–325
Jonsson B (1978) Quantitative electromyographic evaluation of mus-cular load during work. Scand J rehabil Med Suppl 6:69–74
Jonsson B (1982) Measurement and evaluation of local muscular strain in the shoulder during constrained work. J Hum Ergol 11(1):73–88
Eur J Appl Physiol
1 3
Kennedy SM, Koehoorn M (2003) Exposure assessment in epidemi-ology: does gender matter? Am J Ind Med 44(6):576–583
leijon O, Bernmark E, Karlqvist l, Harenstam A (2005) Awkward work postures: association with occupational gender segregation. Am J Ind Med 47(5):381–393
lundberg U (2002) Psychophysiology of work: stress, gender, endo-crine response, and work-related upper extremity disorders. Am J Ind Med 41(5):383–392
Mathiassen SE (1996) the statistical confidence of load estimates based on ramp calibration of upper trapezius EMg. J Electromy-ogr Kinesiol 6(1):59–65
Maughan rJ, Watson JS, Weir J (1983) Strength and cross-sectional area of human skeletal muscle. J Physiol 338:37–49
Mcgorry r W, lin JH, Dempsey Pg, casey JS (2010) Accuracy of the Borg cr10 scale for estimating grip forces associated with hand tool tasks. J Occup Environ Hyg 7(5):298–306
Miller AEJ, Macdougall JD, t arnopolsky MA, Sale Dg (1993) gen-der differences in strength and muscle-fiber characteristics. Eur J Appl Physiol Occup Physiol 66(3):254–262
Moore JS, garg A (1995) the strain index: a proposed method to ana-lyze jobs for risk of distal upper extremity disorders. Am Ind Hyg Assoc J 56(5):443–458
nordander c, Ohlsson K, Balogh I, Hansson gA, Axmon A, Persson r, Skerfving S (2008) gender differences in workers with identi-cal repetitive industrial tasks: exposure and musculoskeletal dis-orders. Int Arch Occup Environ Health 81(8):939–947
nordander c, Ohlsson K, Akesson I, Arvidsson I, Balogh I, Hans-son gA, Stromberg U, rittner r, Skerfving S (2013) Exposure-response relationships in work-related musculoskeletal disorders in elbows and hands—a synthesis of group-level data on expo-sure and response obtained using uniform methods of data collec-tion. Appl Ergon 44(2):241–253
norlund A, Palsson B, Ohlsson K, Skerfving S (2000) Economic con-sequences of occupational disorders in women with repetitive industrial work. Eur J Public Health 10(2):127–132
Perotto A (2005) Anatomical guide for the electromyographer: the limbs and trunk. charles c thomas, Springfield
Pincivero DM, coelho AJ, campy rM, Salfetnikov Y, Bright A (2001) the effects of voluntary contraction intensity and gender on perceived exertion during isokinetic quadriceps exercise. Eur J Appl Physiol 84(3):221–226
Pincivero DM, coelho AJ, campy rM (2004) gender differences in perceived exertion during fatiguing knee extensions. Med Sci Sports Exerc 36(1):109–117
rudroff t (2008) Kinesiological fine wire EMg, 1st edn. noraxon U.S.A., Inc., Scottsdale
Staudenmann D, roeleveld K, Stegeman DF, van Dieen JH (2010) Methodological aspects of SEMg recordings for force esti-mation—a tutorial and review. J Electromyogr Kinesiol 20(3):375–387
Svendsen SW, Bonde JP, Mathiassen SE, Stengaard-Pedersen K, Frich lH (2004a) Work related shoulder disorders: quantitative exposure-response relations with reference to arm posture. Occup Environ Med 61(10):844–853
Svendsen SW, gelineck J, Mathiassen SE, Bonde JP, Frich lH, Stengaard-Pedersen K, Egund n (2004b) Work above shoul-der level and degenerative alterations of the rotator cuff ten-dons: a magnetic resonance imaging study. Arthritis rheum 50(10):3314–3322
t roiano A, naddeo F, Sosso E, camarota g, Merletti r, Mesin l (2008) Assessment of force and fatigue in isometric contractions of the upper trapezius muscle by surface EMg signal and per-ceived exertion scale. gait Posture 28(2):179–186
van der Beek AJ, Kluver BD, Frings-Dresen MH, Hoozemans MJ (2000) gender differences in exerted forces and physiologi-cal load during pushing and pulling of wheeled cages by postal workers. Ergonomics 43(2):269–281
van rijn rM, Huisstede BM, Koes BW, Burdorf A (2009) Asso-ciations between work-related factors and specific disorders at the elbow: a systematic literature review. rheumatology 48(5):528–536
van rijn rM, Huisstede BM, Koes BW, Burdorf A (2010) Associa-tions between work-related factors and specific disorders of the shoulder—a systematic review of the literature. Scand J Work Environ Health 36(3):189–201
Veiersted KB, Westgaard rH, Andersen P (1990) Pattern of muscle-activity during stereotyped work and its relation to muscle pain. Int Arch Occup Environ Health 62(1):31–41
Waite Dl, Brookham rl, Dickerson cr (2010) On the suitability of using surface electrode placements to estimate muscle activity of the rotator cuff as recorded by intramuscular electrodes. J Elec-tromyogr Kinesiol 20(5):903–911
Wijnhoven HA, de Vet Hc, Picavet HS (2006) Prevalence of mus-culoskeletal disorders is systematically higher in women than in men. clin J Pain 22(8):717–724
Won EJ, Johnson PW, Punnett l, Dennerlein Jt (2009) Upper extrem-ity biomechanics in computer tasks differ by gender. J Electromy-ogr Kinesiol 19(3):428–436
1
Sex differences in task distribution and task exposures among Danish
house painters: An observational study combining questionnaire data with
biomechanical measurements.
Thomas Heilskov-Hansen1 MHSc, Susanne Wulff Svendsen2 PhD, Jane Frølund Thomsen1 PhD, Sigurd
Mikkelsen1 DMSc, Gert-Åke Hansson3 PhD
-
Correspondence to: Thomas Heilskov-Hansen, Department of Occupational and Environmental Medicine,
Bispebjerg University Hospital, Bispebjerg Bakke 23, DK-2400 Copenhagen NW, Denmark.
E-mail: [email protected]. Phone: +45 35 31 37 88
Character count: 20.655
Word count: 3.760
Number of tables: 3
Number of figures: 2
2
-
-
,
-
-
- -
- - -
-
3
Introduction
Musculoskeletal disorders (MSDs) account for a large part of the population's use of health care, sickness
absence, and health related pensioning before the normal retirement age [1–3]. Women constitute about
half of the working population in many industrialized countries, but female populations are
underrepresented in occupational epidemiological studies of MSDs [4–6]. It is well known that women
report more musculoskeletal complaints in the upper extremities than their male co-workers [7–14], and to
some extent this may reflect a higher prevalence of work-related MSDs. Three main hypotheses have been
proposed to explain this difference between the two sexes. First, women may have a lower threshold for
reporting complaints [7,15–19]. Second, occupational biomechanical exposures may be higher among
women even within the same job because of sex-segregated tasks [10,20–22], different postures and
movements while performing the same task, or higher use of force relative to their maximum [23]. Third,
women may be more vulnerable to specific exposures [15,24]. Increased knowledge on reasons for sex
differences could potentially open up perspectives for the prevention of MSDs.
To evaluate the vulnerability hypothesis, studies should focus on MSDs that are as
objectively assessed as possible and include men and women who are as equally exposed as possible so
that the reporting and exposure hypotheses cannot explain any differences in the occurrence of MSDs. In
Denmark, the house painters’ trade seems well suited to investigate the vulnerability hypothesis. About 1/3
of the house painters are women, and house painters have exposures that are risk factors for MSDs of the
upper extremities [25]. Throughout the period 1998 to 2007, female house painters were 2 to 3 times as
likely as male house painters to report claims of MSDs of the upper extremities to the Danish National
Board of Industrial Injuries (Rolf Petersen, personal communication). The question is to which degree men
and women in the house painters’ trade actually have equal biomechanical exposures to the upper
extremities.
To evaluate exposure differences between men and women within the same job, task-based
exposure assessment may be a practicable approach [26]. Using this approach, the job exposure of an
4
individual is estimated by weighing task exposures (i.e., the specific exposures to a specific body part which
result from performing a specific task) according to the individual’s task distribution (i.e., the occurrence
and duration of the different tasks within the job; task proportions designate the duration of each task
relative to the whole working time) [27]. Task exposures may be aggregated in a task exposure matrix
(TEM) in the same way as job exposures may be aggregated in a job exposure matrix [26]. Task-based
exposure assessment is particularly likely to be successful if there are large exposure differences between
the tasks, and it has previously been shown that there are significant differences in biomechanical
exposures between the tasks of male house painters [28].
The aim of this study was to determine to which degree there are sex differences in task
distribution and task exposures with respect to postures and movements of the wrist, shoulder, and head
among Danish house painters, and to establish sex-specific TEMs. The study was conducted as a part of the
SHARM (Shoulder-Hand-ARM) study. The study aims were fulfilled.
5
Methods
This study combined questionnaire information on task distribution with task exposures measured by
goniometry and inclinometry.
Questionnaire
In 2011, we sent a questionnaire about musculoskeletal health and work to all members of the Painters’
Union in Denmark, who were born in 1940 or later (N=9364, 3128 (33.4%) women and 6236 (66.6%) men).
A maximum of two reminders were sent. One question concerned the task distribution during a typical
week. The question listed the 11 most common tasks according to representatives of the Painters’ Union,
and a task labelled “other”: 1) removing wallpaper.; 2) levelling; 3) sanding (hand); 4) sanding (giraffe
drywall sander); 5) painting (brush); 6) painting (roll); 7) spraying; 8) hanging wallpaper.; 9) covering,
carrying materials and equipment, or cleaning; 10) pause; 11) driving; 12) other. The question was phrased:
“This question regards your work after 1990. The question is a bit difficult. Try your best to make the hours
add up to your total working hours in a typical week. Please, state the average number of hours you have
spent on each task. Start by writing 0 for tasks that you have usually not performed. Then distribute your
working hours so that the numbers add up to your total working hours in a typical week”. The participants
were encouraged to draft and add up the hours before filling in the questionnaire. We converted the self-
reported task distribution to task proportions for each individual and tested for sex differences in the
average percentage of time spent on each task. Due to large sex differences in age, the questionnaire data
was analysed divided into 5 age-groups.
Exposure measurements
All painters’ workshops (N=267) in the Capital Region of Denmark were identified in the Danish Central
Business Register. In a random order, we contacted the workshops by postal mail, followed by a phone-call,
and asked them if 1-4 of their house painters (preferably the same number of men and women) would be
willing to participate in exposure measurements during one whole working day per person. This procedure
was continued until 25 men and 25 women were included. Originally, we intended to ask the companies to
6
provide a list of all employed house painters, so that we could ask a random sample to participate [28], but
in many cases there were only a few employees or few employees expressed willingness to participate, so
we gave up this procedure.
After contact to 53 companies, 22 companies had agreed to participate, and the
predetermined number of participants had been achieved. For each participant, one whole day
measurement was performed on an ordinary working day from Monday to Thursday (Friday was avoided
because it normally had fewer working hours). The measurements were performed in the period from May
2011 to March 2012. Only right handed persons without current upper extremity complaints were
included. Investigators met with the participants at their worksite or at the workshop. Preparation of the
measurements was carried out by one of two investigators and lasted approximately one hour including
questions on background characteristics. The time spent on preparation of the measurements was paid for
by the employer. After preparation of the measurements, the investigator left and then returned at the end
of the day to remove the equipment. During the measurements, the participants filled in the clock time for
changes between tasks in a logbook. The tasks were predefined and corresponded to the ones in the
questionnaire. Each participant was given a synchronized clock radio so that the time could be read from a
digital display. Thus, the measurements could subsequently be divided according to tasks.
Due to less than 5 sex-specific measurements per task, the tasks 1. removing wallpaper; 4.
sanding (giraffe); 7. spraying; and 8. hanging wallpaper were added to task 12. other. In order to avoid
attenuation of differences between task exposures due to imprecise indication of time for task-change, two
minutes were excluded from the measurements at the beginning and end of each task before we calculated
task exposures for the TEMs. For the overall job exposure “total work”, the entire recording was used. If a
participant performed a given task more than once, the recordings were pooled.
Biaxial goniometers (SG75, Biometrics Ltd, Newport, UK) were used for measuring wrist
postures and movements. The goniometers were placed on the dorsal side of each wrist with the distal part
over the third metacarpal bone and the proximal part in the midline between ulna and radius [29–31].
7
Initially, recordings were made in the anatomical reference position in order to define the neutral posture,
i.e., 0° of flexion/extension and radial/ulnar deviation. This was done with the participants sitting down,
resting their lower arms and hands on a table with their palms facing down.
Triaxial inclinometers (Logger Teknologi HB, Åkarp, Sweden) were used for measuring
inclination of the head and elevation of the upper arms [32]. The inclinometers were placed on the
forehead and on the lateral side of both upper arms just beneath the protrusion of the middle deltoid
muscle [33]. To define a neutral posture, reference positions were initially recorded. For the head, the
participants were standing, looking at an object 2 to 4 meters away at eye-level. For the arms, the
participants were sitting sideways on a chair leaning against the backrest with their arm hanging vertically,
holding a 2 kg dumbbell in their hand [33].
The goniometry and inclinometry data was recorded by two person-worn data-loggers
(Logger Teknologi HB, Åkarp, Sweden) with a sampling frequency of 20 Hz [34]. Data from both sides were
analysed. For each participant, the exposure measures were derived for “total work” and for the different
tasks. For the wrist, the 10th, 50th, and 90th percentiles for the flexion/extension movement were used to
represent the extended, median, and flexed wrist-posture, respectively. The corresponding percentiles for
radial/ulnar deviation were used to represent radial, median, and ulnar deviation, respectively. The 5th-95th
interpercentile range described the range of motion. Movement velocity was represented by its median,
and the mean power frequency (MPF) was used as a measure of repetitiveness; for a strictly cyclic
movement, MPF is identical to the inverse value of the cycle time [29]. Non-neutral postures of the wrist
were defined as the % time with angles exceeding 45° flexion/extension or 20° ulnar/radial deviation.
. For head posture the 1st, 50th, and 90th percentiles were used to represent the backward,
median, and forward inclination, respectively (49). Upper arm elevation was characterized by the 99th
percentile (the angle that is exceeded for 1% of the time) and the % time spent with an elevation above
90°. As measures of variation, the 5th-95th interpercentile range was calculated for each minute. The mean
8
value of these one minute samples was defined as the “within-minute variation” and the standard
deviation as the “between-minute variation” [35].
Statistical analyses
We used Cochran-Mantel-Haenszel statistics to test for statistically significant (p<0.05) sex differences in
task proportions within each age-group. To evaluate sex differences in task exposures, we used an unpaired
double-sided t-test with post-hoc Bonferroni correction. Differences between task exposures for the right
and left side were tested using a paired double-sided t-test. Statistical analyses were made using SAS
statistical software (v9.2 Cary, NC, USA).
Ethics Statement
The study was accepted by the Regional Scientific Ethics Committee, Capital region of Denmark (j.no.: H-C-
FSP-2010-036). Participants in the exposure measurements gave informed written consent. Permission to
store the personal information about the participants was given by the Danish Data Protection Agency
(j.no.: 2010-41-5325). Data was anonymised before performing the analyses.
9
Results
Questionnaire
The proportion who responded was 53% (n=4957), 59% among women and 50% among men. The mean
age of the non-responders was 31 years among men and 41 years among women. Table 1 shows
characteristics of the questionnaire respondents. The age distribution differed considerably between men
and women with a higher percentage of older men.
Figure 1 displays mean task proportions for men and women, respectively, according to age-
group. Within tasks, mean task proportions differed by up to 5% between age-groups. Age-group patterns
were similar for men and women.
Figure 2 presents sex differences in task proportions, according to age-group. Some
statistically significant differences between men and women were found, but none exceeding ±4%. Men
had higher task proportions for levelling, sanding (giraffe), and spraying, whereas women had higher task
proportions for painting with brush and roll.
Table 1. Characteristics of questionnaire respondents and participants in exposure measurements
Questionnaire respondents Participants in exposure measurements
Men (n=3124)
respondents
Women (n=1833)
respondents
Men (n=25)
Measured
Women (n=25)
measured Mean SD % Mean SD % Mean SD % Mean
()
SD %
Seniority (years) 27.1 14.3 - 13.8 10.3 - 22.8 14.4 - 8.6 7.2) -
Working hours per week 36.8 2.8 - 36.4 3.2 - 37.0 0 - 37.0 0 -
≤30 - - 2 - - 3 - - 0 - - 0
>30 - ≤35 - - 1 - - 6 - - 0 - - 0
>35 - - 98 - - 91 - - 100 - - 100
Age (years) 49.7 14.8 - 35.2 11.4 - 45.1 13.1 - 31.9 10.5 -
17-25 - - 9 - - 24 - - 4 - - 28
26-35 - - 12 - - 32 - - 20 - - 44
36-45 - - 17 - - 25 - - 28 - - 16
46-55 - - 16 - - 13 - - 24 - - 8
56-70 - - 46 - - 6 - - 24 - - 4
Height (cm) 178.7 7.3 - 167.4 6.3 - 180.4 6.0 - 166.2 4.9 -
Weight (kg) 84.2 14.7 - 70.4 14.3 - 85.6 14.2 - 65.8 12.7 -
Body mass index (kg/m2) 26.4 4.3 - 25.1 4.9 - 23.7 3.9 - 19.8 3.9 -
Left hand dominance - - 9 - - 10 - - 0 - - 0
Right hand dominance - - 79 - - 83 - - 100 - - 100
No dominant hand - - 12 - - 8 - - 0 - - 0
10
Figure 1. Mean task distribution for men and women, respectively, by age-group.
Figure 2. Sex differences in mean task proportions, by age-group. Women are reference-group.
* Indicates a statistically significant (p<0.05) sex difference in the specific age-group.
Exposure measurements
Table 2 shows the TEM for postures and movements of the right wrist for each sex. “Total work” represents
the overall job exposure. No statistically significant sex differences were observed. For both sexes, there
were clear differences in exposures between tasks. For example, the median velocity for flexion/extension
during painting (brush) was approximately 4 °/s less than for sanding (hand) for both men and women. The
11
median velocity was approximately 50% higher for flexion/extension of the wrist than for ulnar/radial
deviation in all tasks due to the higher range of motion for flexion/extension than for ulnar/radial deviation.
MPF was approximately the same for flexion/extension as for deviations since this measure is sensitive to
the frequency, but not to the amplitude of the movements. Some measures seem to reflect the same task
properties to a great extent. For example the 50th percentiles for flexion/extension and non-neutral
postures showed the same difference between men and women within tasks.
Table 3 shows the TEM for postures and movements of the head and right upper arm for
each sex; again, job exposures are presented as well. There were no statistically significant sex differences.
Between-minute variation was higher for “total work” than for any of the tasks that constituted the work.
This shows that, unlike the rest of the exposure measures, job exposures in terms of between-minute
variation cannot even approximately be derived by a straight forward time weighting of task exposures.
For both sexes, median velocity was distinctively lower for the wrist and upper arms in the
tasks “driving” and “pause” than in any other task (tables 2 and 3).
Job exposures differed statistically significantly (p<0.05) between left and right sides. For
flexion/extension and ulnar/radial deviation, both men and women had a higher median velocity and MPF
on the right side; the same was present for median velocity of shoulder elevation. Sex-specific TEMs were
also established for the left side, i.e., the non-dominant side since all participants were right-handed. The
TEMs for the left side are available from the supporting information file.
12
Tab
le 2
. Tas
k ex
po
sure
mat
rix
for
po
stu
res
and
mo
vem
ents
of
the
righ
t w
rist
fo
r ea
ch s
ex. D
ata
are
dis
pla
yed
fo
r th
e 7
tas
ks t
hat
co
nst
itu
te t
he
wo
rk. A
dd
itio
nal
ly, d
ata
are
sho
wn
fo
r to
tal w
ork
an
d p
ause
. Fo
r fl
exio
n/e
xten
sio
n a
nd
uln
ar/r
adia
l dev
iati
on
, po
siti
ve a
ngl
es d
eno
te f
lexi
on
an
d u
lnar
dev
iati
on
, res
pec
tive
ly, a
nd
neg
ativ
e an
gles
ext
ensi
on
an
d r
adia
l-d
evia
tio
n, r
esp
ecti
vely
, [SD
=sta
nd
ard
dev
iati
on
; M
PF=
mea
n p
ow
er f
req
uen
cy]
Leve
llin
g Sa
nd
ing
(han
d)
Pai
nti
ng
(bru
sh)
Pai
nti
ng
(ro
ll)
Co
veri
ng
Dri
vin
g O
ther
To
tal w
ork
P
ause
Mea
n
SD
Mea
n
SD
Mea
n
SD
Mea
n
SD
Mea
n
SD
Mea
n
SD
Mea
n
SD
Mea
n
SD
Mea
n
SD
Fle
xio
n/e
xten
sio
n
Per
cen
tile
(°)
1
0th
M
en
-57
16
-47
6
-50
12
-43
11
-42
14
-46
14
-45
18
-48
12
-44
17
Wo
men
-4
6 8
-5
1 9
-5
2 1
0 -4
7 9
-5
2 1
2 -4
7 1
3 -4
8 8
-5
0 8
-4
8 8
5
0th
M
en
-26
10
-15
3
-23
10
-17
11
-17
18
-20
13
-20
14
-20
11
-15
12
W
om
en
-18
8
-26
13
-25
7
-23
8
-23
13
-13
9
-21
11
-22
8
-22
10
90
th
Men
0
1
0 1
0 2
3
9
8
1
3 1
0 1
5 5
1
3 5
1
2 7
1
1 1
2 1
5
Wo
men
6
8
0
1
1 5
9
6
1
1 5
9
1
1 8
7
1
0 7
8
9
1
4
Ran
ge o
f m
oti
on
5
th-
95
th
Men
7
5 1
0 7
4 9
6
9 1
0 6
5 1
2 6
6 2
0 6
7 1
2 6
5 1
1 7
2 7
7
0 1
7
Wo
men
6
7 4
6
5 1
1 7
4 1
4 7
0 1
5 7
2 1
3 7
6 1
4 7
4 9
7
5 1
0 7
4 1
1
Med
ian
vel
oci
ty (
°/s)
M
en
18
.1
8
19
.3
5
15
.5
7
17
.6
6
13
.3
8
10
.0
5
14
.0
8
14
.5
5
5.5
4
Wo
men
2
1.2
7
1
9.1
4
1
5.3
5
1
6.3
5
1
4.2
8
7
.9
3
14
.5
6
14
.6
4
4.8
4
Rep
etit
iven
ess
(MP
F;
Hz)
M
en
.28
.05
.29
.04
.26
.06
.30
.08
.29
.08
.28
.06
.29
.03
.27
.04
.29
.06
W
om
en
.33
.05
.28
.06
.25
.04
.28
.06
.26
.06
.24
.05
.28
.06
.27
.04
.20
.03
Uln
ar/r
adia
l de
viat
ion
P
erce
nti
le (
°)
10
th
Men
-8
5
-1
9 5
-1
5 1
0 -1
8 9
-9
8
-9
6
-1
3 1
1 -1
5 9
-1
5 9
W
om
en
-13
8
-21
16
-21
11
-22
9
-20
12
-17
9
-20
11
-19
11
-18
10
50
th
Men
9
3
-4
5
1
1
1 -2
7
6
9
6
6
4
1
0 1
9
0
9
Wo
men
3
6
-4
1
6 -5
1
0 -5
9
-4
8
-2
8
-4
1
1 -2
1
0 -3
1
0
9
0th
M
en
27
5
11
4
18
12
17
10
19
10
21
7
19
9
18
9
14
9
Wo
men
1
9 6
1
4 1
2 1
2 1
0 1
2 8
1
3 8
1
3 7
1
1 1
0 1
4 9
1
2 1
0
Ran
ge o
f m
oti
on
5
th-
95
th
Men
4
4 1
1 4
0 3
4
2 8
4
4 9
3
6 1
0 3
8 9
4
2 7
4
2 8
3
7 8
Wo
men
4
1 8
4
4 7
4
4 5
4
3 6
4
2 9
3
8 8
4
1 6
4
3 6
3
9 7
Med
ian
vel
oci
ty (
°/s)
M
en
10
.9
6
11
.5
3
10
.8
6
12
.9
7
8.1
4
5
.7
3
8.8
4
9
.0
3
3.6
3
Wo
men
1
2.3
5
1
5.3
6
9
.9
3
13
.1
6
8.9
7
4
.8
1
8.9
4
9
.2
3
3.2
2
Rep
etit
iven
ess
(MP
F;
Hz)
M
en
.29
.06
.33
.04
.28
.06
.30
.08
.31
.06
.29
.08
.27
.03
.28
.05
.21
.06
W
om
en
.33
.06
.30
.06
.27
.06
.31
.07
.28
.07
.25
.06
.28
.06
.28
.06
.20
.05
Co
mb
ine
d w
rist
po
stu
res
N
on
-neu
tral
po
stu
res
(% t
ime)
M
en
43
16
20
2
33
19
28
14
24
22
29
18
31
18
30
15
22
14
W
om
en
26
9
39
21
36
13
32
13
34
16
25
9
28
17
32
12
31
15
Nu
mb
er
of
reco
rdin
gs
Men
5
-
5
- 1
4 -
13
- 1
2 -
8
- 1
0 -
24
- 2
3 -
W
om
en
7
- 8
-
17
- 1
5 -
16
- 8
-
15
- 2
5 -
25
-
M
ean
re
cord
ing
du
rati
on
(m
inu
tes)
M
en
82
- 7
6 -
16
2 -
13
8 -
46
- 4
5 -
94
- 2
80
- 4
8 -
Wo
men
1
27
- 5
1 -
14
9 -
10
2 -
55
- 4
1 -
78
- 3
09
- 5
9 -
12
13
Ta
ble
3. T
ask
exp
osu
re m
atri
x fo
r p
ost
ure
s an
d m
ove
men
ts o
f th
e h
ead
an
d r
igh
t u
pp
er a
rm f
or
each
sex
. Dat
a ar
e d
isp
laye
d f
or
the
7 t
asks
th
at c
on
stit
ute
th
e w
ork
. Ad
dit
ion
ally
, dat
a ar
e sh
ow
n f
or
tota
l w
ork
an
d p
ause
. Fo
r fl
exio
n/e
xten
sio
n, p
osi
tive
an
gles
den
ote
fle
xio
n a
nd
ne
gati
ve a
ngl
es e
xten
sio
n. [
SD=s
tan
dar
d d
evia
tio
n]
Leve
llin
g Sa
nd
ing
(han
d)
Pai
nti
ng
(bru
sh)
Pai
nti
ng
(ro
ll)
Co
veri
ng
Dri
vin
g O
ther
To
tal w
ork
P
ause
Mea
n
SD
Mea
n
SD
Mea
n
SD
Mea
n
SD
Mea
n
SD
Mea
n
SD
Mea
n
SD
Mea
n
SD
Mea
n
SD
He
ad in
clin
atio
n
Per
cen
tile
(°)
1
st
Men
-4
6 1
2 -3
9 1
4 -4
5 1
3 -5
3 1
2 -2
6 2
2 -1
6 1
0 -3
8 1
3 -4
5 1
4 -2
2 1
4 W
om
en
-40
6
-40
17
-45
14
-53
15
-23
15
-18
4
-40
12
-47
12
-21
14
50
th
Men
1
8 1
4 2
4 8
1
6 1
6 8
1
3 2
5 1
6 1
5 1
4 1
9 1
5 1
7 1
4 1
6 1
3
Wo
men
1
6 5
1
8 1
1 1
5 1
1 7
2
2 2
7 1
2 1
3 9
1
7 1
1 1
7 8
1
3 1
2
9
0th
M
en
64
11
54
6
54
15
50
12
52
17
44
14
51
14
52
13
43
15
W
om
en
55
8
55
11
53
14
54
12
54
12
40
11
51
11
53
10
36
14
Rig
ht
up
pe
r ar
m e
leva
tio
n
9
9th
per
cen
tile
(°)
M
en
13
6 1
1 1
23
12
12
3 1
2 1
21
23
89
27
90
17
12
1 2
5 1
27
13
90
27
W
om
en
13
1 1
4 1
24
30
12
7 1
8 1
26
21
95
22
94
23
12
6 1
5 1
28
15
84
23
>9
0°
(% t
ime)
Men
1
5 8
6
2
1
2 9
8
7
2
2
4
8
6
5
9
4
2
3
Wo
men
9
4
1
1 6
1
2 8
1
1 7
3
4
3
6
1
0 1
1 9
5
1
1
Wit
hin
-min
ute
va
riat
ion
(°)
a M
en
75
18
62
2
68
15
65
17
42
10
39
11
59
20
62
10
32
11
W
om
en
73
16
75
21
70
20
68
16
47
11
46
16
62
13
62
11
30
15
B
etw
een
-min
ute
va
riat
ion
(°)
a M
en
29
5
29
6
28
6
26
9
18
8
18
6
28
6
31
5
20
7
W
om
en
31
5
26
12
29
7
27
7
19
8
19
3
30
5
32
6
22
7
M
edia
n v
elo
city
(°/
s)
Men
5
1.6
1
6 5
9.9
1
3 4
7.0
1
7 5
2.5
1
9 4
7.0
2
4 3
3.6
1
2 3
8.5
1
8 4
2.9
1
2 1
7.0
1
9
Wo
men
5
9.9
2
0 6
6.4
2
5 4
3.2
1
4 5
0.8
1
3 4
8.4
2
5 2
9.4
1
1 4
3.9
1
8 4
3.7
1
3 1
0.1
9
N
um
be
r o
f re
cord
ings
M
en
5
- 6
-
17
- 1
4 -
12
- 9
-
10
- 2
5 -
24
- W
om
en
7
- 8
-
17
- 1
5 -
16
- 8
-
15
- 2
5 -
25
-
M
ean
re
cord
ing
du
rati
on
(m
inu
tes)
M
en
88
- 9
5 -
15
8 -
13
0 -
52
- 5
1 -
12
4 -
31
3 -
52
- W
om
en
15
8 -
51
- 1
49
- 1
02
- 5
5 -
41
- 7
8 -
31
8 -
61
- a Th
e m
easu
res
of
vari
atio
n w
ere
calc
ula
ted
fro
m t
he
5th
-95
th in
terp
erce
nti
le r
ange
fo
r ea
ch m
inu
te
13
14
Discussion
This study showed only minor sex differences in task distribution and task-specific postures and movements
among Danish house painters. There was a considerable difference in age and seniority between male and
female respondents. Thus, reported task distributions for men could reflect a longer period back in time,
where task distributions may have differed from nowadays. However, we controlled for effects of age by
stratification, and the age-dependent patterns within each task were quite similar for men and women. The
proportion who returned the questionnaire was relatively low, and responders and non-responders
differed from each other with respect to sex and age distribution, but we think that the age-specific
comparisons counteracted the risk of non-response bias with respect to assessing sex differences in task
distributions.
There is a potential for recall bias with a tendency for overestimating time proportions spent
on highly exposed tasks; in particular, women might tend to overestimate time proportions spent on force
demanding tasks because they use more force relative to their maximum than men do [23,24]. Even if this
was the case, the slight differences in task proportions, which we identified, seemed to be in a direction of
men doing more of the tasks that require high force [23].
During the whole day exposure measurements, a certain task could be performed by few
persons and the total time spent on the task could be very short. Based on recent recommendations [35],
we decided on a minimum of 5 measurements for each sex per task. In case of an insufficient number of
measurements for a specific task, the task was added to the task “other”, which diminished the number of
tasks from 12 to 8. An alternative procedure could have been to supplement the whole day measurements
by task-specific measurements until the pre-specified number of measurements per task was met, but this
would have been time consuming beyond our resources. As a tentative rule of thumb, a minimum of 120
minutes of measurement per task has been recommended when constructing a TEM [36]. For all 8 tasks in
our TEM, the total duration of task exposure measurements was well above this limit.
15
Self-reported logbook data on task occurrence and duration may be less precise than direct
observation [37]. In an attempt to exclude potential overflow between tasks due to imprecise timing, we
decided to exclude two minutes in the beginning and end of each task measurement. This step caused only
minor changes, which indicates a precise overall reporting. We did not apply a minimum duration of task
recording per period, but due to the 2 minutes cut-off, recordings lasting less than 4 minutes were
discarded.
Regarding non-neutral wrist postures, our limit for radial deviation (20°) was set higher than
reported normative data on range of motion [38]. This was done because flexion/extension and ulnar/radial
deviation movements are coupled [39,40] so that during normal activity, radial deviation angles exceed the
range of motion of radial deviation in a constrained neutral flexion/extension angle. This is caused by an
oblique orientation of the mechanical axes in relation to the anatomical axes of the wrist [41]. Our choice of
±45° for flexion/extension is consistent with the symmetrical properties of these movements and well
within reported range of motions [38]. However, the contributions of flexion/extension to non-neutral
postures were almost entirely due to excessive extension. This corresponds well with the 50th percentile for
flexion/extension which showed that all tasks were performed with a generally extended wrist; in this
posture the fingers can exert more power in the grip. Extended wrist postures have been shown to produce
particularly high carpal tunnel pressures, which has been linked to the development of carpal tunnel
syndrome and tendon related disorders [42–44]. Especially, the combination of non-neutral postures and a
high use of force has been reported as a risk factor for developing MSDs in the upper extremity [45].
Contrary to this, a recent study showed a protective effect of wrist extension >43° during heavy grip, but
this study mainly concerned de Quervain’s disease and the study had very few cases [46].
Exposure measures for work with elevated arms are commonly derived from the right upper
tail of the amplitude distribution function (ADF) of the elevation angle. A high percentile, e.g. the 99th, can
be selected, and the corresponding angle, xx°, derived from the ADF. The interpretation of this measure is
that for 1% of the time, the elevation angle exceeded xx°. Alternatively, a high elevation threshold, e.g. 90°
16
can be chosen, and the % time above this angle, yy%, can be derived; the interpretation is that for yy% of
the time, the elevation exceeded 90°. Percentiles are commonly used to express neutral and extreme
postures without any assumptions about the pathophysiological mechanisms [47–52]. A disadvantage with
percentiles is that time weighting of task exposure is an approximation [47], which may lead to bias for
short measurements [36]. Thresholds are more intuitive and are generally considered to be easily
assessable from low-cost observations and thus preferable for epidemiological studies and work place visits
e.g. by occupational physicians. However, observations may be as resource demanding as measurements,
and may introduce bias [53]. The choice of thresholds may be based on hypotheses on pathogenesis [27]
and safe lower levels of exposure. Exposures below the threshold are disregarded. For upper arm elevation
>90°, the SD in our material exceeded the group mean for some of the tasks, e.g. driving (table 3). This is a
drawback because it shows that the confidence intervals for these task exposures are wide.
Upper extremity MSDs have been demonstrated to be more prevalent in the dominant arm
[13,54]. This distinction is also evident within the house painters’ trade [25]. Our tables presented exposure
measurement data from the right side, which was the dominant side for all participants, by design. We
think that these task exposures can be assigned to the dominant side for left-handed persons because
house painters are not constrained to use the right hand due to the design of tools or characteristics of
worksites.
Small, but statistically significant sex differences were found for some tasks when comparing
the self-reported task distributions. Several studies have addressed the problem of sex-segregated tasks in
relation to sex differences in MSDs within a job [9,24,55–59]. In our study, task-specific postures and
movements did not show statistically significant sex differences, and it could therefore be considered to
use a common TEM for men and women, even in studies of sex differences in MSDs. However, the sex-
specific TEMs enable inclusion of the derived sex differences, when estimating the individual exposure in
future epidemiological studies. Moreover, the TEMs can be extended with sex-specific data regarding use
of force (23).
17
In conclusion, sex-specific TEMs were constructed for the head and the dominant and non-
dominant wrist and upper arm. Only minor sex differences were found in self-reported task distributions
and objectively measured task-specific postures and movements of the upper extremities. Thus, the house
painters’ trade seems well suited to investigate sex differences in vulnerability to exposures that may cause
upper extremity MSDs.
Acknowledgements
The authors wish to thank Rolf Petersen for compiling the data on workers compensation claims, which
initiated the study. We would also like to thank the Painters’ Union in Denmark and its members for
cooperating.
18
Reference List
1. Thiehoff R (2002) Economic significance of work disability caused by musculoskeletal disorders. Orthopade 31: 949-956.
2. Nordlund A, Palsson B, Ohlsson K, Skerfving S (2000) Economic consequences of occupational disorders in women with repetitive industrial work. The European Journal of Public Health 10: 127-132.
3. Wilson d'Almeida K, Godard C, Leclerc A, Lahon G+ (2008) Sickness absence for upper limb disorders in a French company. Occupational Medicine 58: 506-508.
4. Artazcoz L, Borrell C, Cortes I, Escriba-Aguir V, Cascant L (2007) Occupational epidemiology and work related inequalities in health: a gender perspective for two complementary approaches to work and health research. J Epidemiol Community Health 61 Suppl 2: ii39-ii45.
5. Messing K, Mager SJ (2006) Sex, gender and women's occupational health: the importance of considering mechanism. Environ Res 101: 149-162.
6. Niedhammer I, Saurel-Cubizolles MJ, Piciotti M, Bonenfant S (2000) How is sex considered in recent epidemiological publications on occupational risks? Occup Environ Med 57: 521-527.
7. Bingefors K, Isacson D (2004) Epidemiology, co-morbidity, and impact on health-related quality of life of self-reported headache and musculoskeletal pain - a gender perspective. European Journal of Pain 8: 435-450. doi: DOI: 10.1016/j.ejpain.2004.01.005.
8. Dahlberg R, Karlqvist L, Bildt C, Nykvist K (2004) Do work technique and musculoskeletal symptoms differ between men and women performing the same type of work tasks? Applied Ergonomics 35: 521-529. doi: DOI: 10.1016/j.apergo.2004.06.008.
9. de Zwart BC, Frings-Dresen MH, Kilbom A (2001) Gender differences in upper extremity musculoskeletal complaints in the working population. Int Arch Occup Environ Health 74: 21-30.
10. Hooftman WE, van Poppel MN, van der Beek AJ, Bongers PM, van MW (2004) Gender differences in the relations between work-related physical and psychosocial risk factors and musculoskeletal complaints. Scand J Work Environ Health 30: 261-278.
11. Nordander C, Ohlsson K, Akesson I, Arvidsson I, Balogh I, Hansson GA, Stromberg U, Rittner R, Skerfving S (2013) Exposure-response relationships in work-related musculoskeletal disorders in elbows and hands - A synthesis of group-level data on exposure and response obtained using uniform methods of data collection. Appl Ergon 44: 241-253. S0003-6870(12)00116-0 [pii];10.1016/j.apergo.2012.07.009 [doi].
12. Nordander C, Ohlsson K, Akesson I, Arvidsson I, Balogh I, Hansson GA, tramberg U, Rittner R, Skerfving S (2009) Risk of musculoskeletal disorders among females and males in repetitive/constrained work. Ergonomics 52: 1226-1239. doi: 10.1080/00140130903056071.
19
13. Shiri R, Varonen H, Heliovaara M, Viikari-Juntura E (2007) Hand dominance in upper extremity musculoskeletal disorders. J Rheumatol 34: 1076-1082. 07/13/034 [pii].
14. Wijnhoven HA, de Vet HC, Picavet HS (2006) Explaining sex differences in chronic musculoskeletal pain in a general population. Pain 124: 158-166.
15. Hurley RW, Adams MC (2008) Sex, gender, and pain: an overview of a complex field. Anesth Analg 107: 309-317.
16. Ladwig KH, Marten-Mittag B, Formanek B, Dammann G (2000) Gender differences of symptom reporting and medical health care utilization in the German population. Eur J Epidemiol 16: 511-518.
17. Myers CD, Riley JL, III, Robinson ME (2003) Psychosocial contributions to sex-correlated differences in pain. Clin J Pain 19: 225-232.
18. Riley JL, III, Robinson ME, Wise EA, Myers CD, Fillingim RB (1998) Sex differences in the perception of noxious experimental stimuli: a meta-analysis. Pain 74: 181-187.
19. van Wijk CM, Kolk AM (1997) Sex differences in physical symptoms: the contribution of symptom perception theory. Soc Sci Med 45: 231-246. S0277953696003401 [pii].
20. Kennedy SM, Koehoorn M (2003) Exposure assessment in epidemiology: does gender matter? Am J Ind Med 44: 576-583.
21. Leijon O, Bernmark E, Karlqvist L, Harenstam A (2005) Awkward work postures: association with occupational gender segregation. Am J Ind Med 47: 381-393.
22. Lundberg U (2002) Psychophysiology of work: stress, gender, endocrine response, and work-related upper extremity disorders. Am J Ind Med 41: 383-392.
23. Meyland J.K, Heilskov-Hansen T, Alkjaer T, Koblauch H, Mikkelsen S, Svendsen SW, Thomsen JF, Hansson GA, Simonsen EB (2014) Sex differences in muscular load among house painters performing identical work tasks. Eur J Appl Physiol .
24. Nordander C, Ohlsson K, Balogh I, Hansson GA, Axmon A, Persson R, Skerfving S (2008) Gender differences in workers with identical repetitive industrial tasks: exposure and musculoskeletal disorders. Int Arch Occup Environ Health 81: 939-947. 10.1007/s00420-007-0286-9 [doi].
25. Svendsen SW, Bonde JP, Mathiassen SE, Stengaard-Pedersen K, Frich LH (2004) Work related shoulder disorders: quantitative exposure-response relations with reference to arm posture. Occup Environ Med 61: 844-853.
26. Benke G, Sim M, Fritschi L, Aldred G (2000) Beyond the Job Exposure Matrix (JEM): the Task Exposure Matrix (TEM). Annals of Occupational Hygiene 44: 475-482.
27. Winkel J, Mathiassen SE (1994) Assessment of physical work load in epidemiologic studies: concepts, issues and operational considerations. Ergonomics 37: 979-988. doi: 10.1080/00140139408963711.
20
28. Svendsen SW, Mathiassen SE, Bonde JP (2005) Task based exposure assessment in ergonomic epidemiology: a study of upper arm elevation in the jobs of machinists, car mechanics, and house painters. Occup Environ Med 62: 18-27.
29. Hansson GA, Balogh I, Ohlsson K, Rylander L, Skerfving S (1996) Goniometer measurement and computer analysis of wrist angles and movements applied to occupational repetitive work. J Electromyogr Kinesiol 6: 23-35.
30. Balogh I, Ohlsson K, Nordander C, Skerfving S, Hansson GA (2009) Precision of measurements of physical workload during standardized manual handling part III: goniometry of the wrists. J Electromyogr Kinesiol 19: 1005-1012. S1050-6411(08)00113-2 [pii];10.1016/j.jelekin.2008.07.003 [doi].
31. Hansson GA, Balogh I, Ohlsson K, Skerfving S (2004) Measurements of wrist and forearm positions and movements: effect of, and compensation for, goniometer crosstalk. J Electromyogr Kinesiol 14: 355-367.
32. Hansson GA, Asterland P, Holmer NG, Skerfving S (2001) Validity and reliability of triaxial accelerometers for inclinometry in posture analysis. Med Biol Eng Comput 39: 405-413.
33. Hansson GA, Arvidsson I, Ohlsson K, Nordander C, Mathiassen SE, Skerfving S, Balogh I (2006) Precision of measurements of physical workload during standardised manual handling. Part II: Inclinometry of head, upper back, neck and upper arms. J Electromyogr Kinesiol 16: 125-136.
34. Hansson GA, Asterland P, Kellerman M (2003) Modular data logger system for physical workload measurements. Ergonomics 46: 407-415. 8W8F4P4QCUCUHULF [pii];10.1080/0014013021000034920 [doi].
35. Arvidsson I, Balogh I, Hansson GA, Ohlsson K, Akesson I, Nordander C (2012) Rationalization in meat cutting - consequences on physical workload. Appl Ergon 43: 1026-1032. S0003-6870(12)00029-4 [pii];10.1016/j.apergo.2012.03.001 [doi].
36. Mathiassen SE, Wahlstrom J, Forsman M (2012) Bias and imprecision in posture percentile variables estimated from short exposure samples. BMC Med Res Methodol 12: 36. 1471-2288-12-36 [pii];10.1186/1471-2288-12-36 [doi].
37. Unge J, Hansson GA, Ohlsson K, Nordander C, Axmon A, Winkel J, Skerfving S (2005) Validity of self-assessed reports of occurrence and duration of occupational tasks. Ergonomics 48: 12-24.
38. Klum M, Wolf MB, Hahn P, Lecl+¿re FM, Bruckner T, Unglaub F (2012) Normative Data on Wrist Function. The Journal of Hand Surgery 37: 2050-2060.
39. Garg R, Kraszewski P, Stoecklein HH, Syrkin G, Hillstrom HJ, Backus S, Lenhoff ML, Wolff L, Crisco JJ, Wolfe SW (Wrist Kinematic Coupling and Performance During Functional Tasks: Effects of Constrained Motion. The Journal of Hand Surgery .
40. Li ZM, Kuxhaus L, Fisk JA, Christophel TH (2005) Coupling between wrist flexion extension and
radial ulnar deviation. Clinical Biomechanics 20: 177-183.
21
41. Crisco JJ, Heard WMR, Rich RR, Paller DJ, Wolfe SW (2011) The Mechanical Axes of the Wrist Are Oriented Obliquely to the Anatomical Axes. The Journal of Bone & Joint Surgery 93: 169-177. doi: 10.2106/JBJS.I.01222.
42. Keir PJ, Bach JM, Hudes M, Rempel DM (2007) Guidelines for Wrist Posture Based on Carpal Tunnel Pressure Thresholds. Human Factors: The Journal of the Human Factors and Ergonomics Society 49: 88-99.
43. McGorry RW, Fallentin N, Andersen JH, Keir PJ, Hansen TB, Pransky G, Lin JH (2014) Effect of grip type, wrist motion, and resistance level on pressures within the carpal tunnel of normal wrists. J Orthop Res 32: 524-530.
44. Werner R, Armstrong TJ, Bir C, Aylard MK (1997) Intracarpal canal pressures: the role of finger, hand, wrist and forearm position. Clinical Biomechanics 12: 44-51.
45. Punnett L, Wegman DH (2004) Work-related musculoskeletal disorders: the epidemiologic evidence and the debate. J Electromyogr Kinesiol 14: 13-23. 10.1016/j.jelekin.2003.09.015 [doi];S1050641103001251 [pii].
46. Harris-Adamson C, You D, Eisen EA, Goldberg R, Rempel D (2014) The Impact of Posture on Wrist Tendinosis Among Blue-Collar Workers: The San Francisco Study. Human Factors: The Journal of the Human Factors and Ergonomics Society 56: 143-150.
47. Fethke NB, Gant LC, Gerr F (2011) Comparison of biomechanical loading during use of conventional stud welding equipment and an alternate system. Applied Ergonomics 42: 725-734.
48. Hansson GA, Balogh I, Ohlsson K, Granqvist L, Nordander C, Arvidsson I, Akesson I, Unge J, Rittner R, Stromberg U, Skerfving S (2009) Physical workload in various types of work: Part I. Wrist and forearm. International Journal of Industrial Ergonomics 39: 221-233. doi: DOI: 10.1016/j.ergon.2008.04.003.
49. Hansson GA, Balogh I, Ohlsson K, Granqvist L, Nordander C, Arvidsson I, Akesson I, Unge J, Rittner R, Stromberg U, Skerfving S (2010) Physical workload in various types of work: Part II. Neck, shoulder and upper arm. International Journal of Industrial Ergonomics 40: 267-281.
50. Balogh I, Ohlsson K, Hansson GA, Engstrom T, Skerfving S (2006) Increasing the degree of automation in a production system: Consequences for the physical workload. International Journal of Industrial Ergonomics 36: 353-365.
51. Delisle A, Lariviere C, Plamondon A, Imbeau D (2006) Comparison of three computer office workstations offering forearm support: impact on upper limb posture and muscle activation. Ergonomics 49: 139-160. doi: 10.1080/10610270500450739.
52. Arvidsson I, Hansson GA, Mathiassen SE, Skerfving S (2006) Changes in physical workload with implementation of mouse-based information technology in air traffic control. International Journal of Industrial Ergonomics 36: 613-622.
53. Trask C, Mathiassen SE, Wahlstrom J, Forsman M (2014) Cost-efficient assessment of biomechanical exposure in occupational groups, exemplified by posture observation and inclinometry. Scand J Work Environ Health 40: 252-265. 3416 [pii];10.5271/sjweh.3416 [doi].
22
54. Arvidsson I, Arvidsson M, Axmon A, Hansson G, Johansson CR, Skerfving S (2006) Musculoskeletal disorders among female and male air traffic controllers performing identical and demanding computer work. Ergonomics 49: 1052-1067. doi: 10.1080/00140130600733816.
55. Locke SJ, Colt JS, Stewart PA, Armenti KR, Baris D, Blair A, Cerhan JR, Chow WH, Cozen W, Davis F, De Roos AJ, Hartge P, Karagas MR, Johnson A, Purdue MP, Rothman N, Schwartz K, Schwenn M, Severson R, Silverman DT, Friesen MC (2014) Identifying gender differences in reported
occupational information from three US population-based case Çôcontrol studies. Occupational and Environmental Medicine .
56. Messing K, Punnett L, Bond M, Alexanderson K, Pyle J, Zahm S, Wegman D, Stock SR, de Grosbois S (2003) Be the fairest of them all: Challenges and recommendations for the treatment of gender in occupational health research. Am J Ind Med 43: 618-629.
57. Eng A, 't Mannetje A, McLean D, Ellison-Loschmann L, Cheng S, Pearce N (2011) Gender differences in occupational exposure patterns. Occupational and Environmental Medicine 68: 888-894.
58. Hooftman WE, van der Beek AJ, Bongers PM, van MW (2005) Gender differences in self-reported physical and psychosocial exposures in jobs with both female and male workers. J Occup Environ Med 47: 244-252. 00043764-200503000-00006 [pii].
59. Messing K, Dumais L, Courville J, Seifert AM, Boucher M (1994) Evaluation of exposure data from men and women with the same job title. J Occup Med 36: 913-917.
1
Exposure-response relationship between postures and movements of the
wrist and carpal tunnel syndrome among male and female house painters:
a retrospective cohort study
Thomas Heilskov-Hansen1 MHSc, Sigurd Mikkelsen1 DMSc, Susanne Wulff Svendsen2 PhD, Lau Caspar
Thygesen3PhD, Gert-Åke Hansson4 PhD, Jane Frølund Thomsen1 PhD
-
Correspondence to: Thomas Heilskov-Hansen, Department of Occupational and Environmental Medicine,
Bispebjerg University Hospital, Bispebjerg Bakke 23, DK-2400 Copenhagen NW, Denmark.
E-mail: [email protected]. Phone: +45 35 31 37 88
Key words: Carpal Tunnel Syndrome, Cumulative Trauma Disorders, Epidemiology, Sex Characteristics,
Workload.
2
ABSTRACT
Objectives. The aim of this study was to investigate exposure-response relationships between measured
postures and movements of the wrist and the incidence rate of carpal tunnel syndrome (CTS) diagnoses
and CTS-surgery, and any modifications of these relationships by sex.
Methods. The cohort comprised all 9364 persons, who were members of the Painters’ Union in Denmark
by March 1st 2011, and who were born in 1940 or later. Exposure measures from sex-specific task exposure
matrices were weighted by self-reported task distributions to obtain individual estimates of exposure
intensity for the wrist in terms of median velocity of wrist flexion/extension, mean power frequency (MPF)
as a measure of repetitiveness ,and non-neutral postures. Exposure duration was assessed from seniority
and yearly working proportions. From Danish health registers, first time CTS-diagnoses and first time events
of CTS-surgery were collected as outcomes. The cohort was followed from January 1st 1994 to January 1st
2011. Log linear Poisson regression was used.
Results. The incidence rate ratios (IRRs) for both CTS outcomes increased significantly with increasing
wrist velocity [IRR=1.29 (95% CI 1.07-1.56) per °/s for CTS-diagnoses, IRR=1.32 (95% CI 1.06-1.65) per °/s for
CTS-surgery] and MPF [IRR=1.39 (95% CI 1.13-1.71) per 0.01 Hz for CTS-diagnoses, IRR=1.34 (95% CI 1.04-
1.71) per 0.01 Hz for CTS-surgery]. Non-neutral postures were not related to the outcomes and the same
was true of the proportion of time worked during the previous 1, 2, and 5 years. The IRRs for women were
higher than those for men [IRR=4.64 (95% CI 3.21-6.71) for CTS-diagnoses, IRR=6.03 (95% CI 3.89-9.34) for
CTS-surgery]. There were no significant interactions between exposure intensity and sex or between
exposure duration and sex. These results, based on log linear Poisson regression, imply that the absolute
CTS incidence rates of women increased at a steeper rate with increasing wrist velocity and repetition than
for men, indicating that women may be more vulnerable than men to these effects.
Conclusions. Exposure-response relationships were found between median velocity of flexion/extension of
the wrist and MPF and the incidence of both CTS-diagnoses and CTS-surgery. The relationships were not
modified by sex, but the results implied that the absolute incidence rates for women increased at a steeper
3
rate with increasing exposures than the incidence rates for men did. Thus women may be more vulnerable
to these effects than men.
4
INTRODUCTION
Carpal tunnel syndrome (CTS) is one of the most common wrist disorders associated with occupational
exposures (1;2). The exact pathogenesis of CTS is unknown, but clinical symptoms of recurrent or persistent
numbness, tingling, and burning pain in one or more of the first three digits usually presents as a result of
an increased pressure on the median nerve by the surrounding tissue in the carpal tunnel (3). Several risk
factors for CTS are well known including age, sex, body mass index (BMI), pregnancy, and medical
conditions such as: rheumatoid arthritis, diabetes, hypothyroidism and trauma (1). Forceful and repetitive
movements and awkward postures of the wrists as well as exposure to hand-arm vibrations have been
reported as occupational risk factors for CTS (1;4-8). Treatment options include rest, splinting, local
corticosteroid injections, and – if symptoms are refractory of if nerve conduction studies show severe
affection of the median nerve - open or endoscopic carpal tunnel decompression (9-11).
Many studies have investigated sex differences in occurrence of work-related musculoskeletal disorders
(WMSDs) in the upper extremity (12-14). When investigating exposure response-relationships most studies
have assumed that men and women within the same profession have the same exposures. This assumption
has been criticised because there could be large differences in exposures due to different sex-specific task
exposures and sex-segregated task distributions, even within the same profession (15-17). If so, results on
sex differences in the occurrence of WMSD’s could be biased by inaccurate exposure assessment. In a
previous study of Danish house painters (18), we found only minor sex-differences in task distributions and
task-specific postures and movements of the wrists. This led us to conclude that Danish house painters
would be well suited to examine if women are more vulnerable than men are at the same level of
exposure.
The aim of this study was to investigate exposure-response relationships between three different measures
of postures and movements of the wrist and the incidence rate (IR) of CTS-diagnoses and CTS-surgery, and
any modifications of these relationships by sex.
5
METHODS
Population
The study cohort comprised all members of the Painters’ Union in Denmark, who were born in 1940 or later
and were still a member on March 1st 2011 [N=9364, 6236 (66.6%) men and 3128 (33.4%) women]. All
members of the cohort received a questionnaire, which was sent by postal mail with up to two reminders.
Since 1968 every person in Denmark with permanent residency is assigned a ten digit personal
identification number (PIN) from the Danish Civil Registration System (19). The PIN enables linkage of
information at the individual level in all Danish national registers.
Outcomes
The Danish National Patient Register (DNPR) contains personal information on practically all contacts with
the Danish public health care system (and private hospitals since 2003) (20). Since 1994 diagnoses have
been reported using the International Classification of Diseases, version 10 (ICD-10) and since 1996 surgery
has been reported using the Danish version of Nomesco Classification of Surgical Procedures (NCSP-D).
Some patients are diagnosed and/or treated by a medical specialist within primary care. In Denmark, this
group of health care professionals are contracted by the Danish National Health Service and they receive
payment by reporting a service code, indicating the form of treatment they have provided, to the Danish
National Health Service Register (NHSR) (21). The service code 3146 (“nerve compression”) has been used
since 2002. We treated this code as both a diagnosis and a surgery code. From the DNPR and the NHSR, we
extracted information on CTS-diagnoses (ICD-10 code G56.0 or service code 3146) and CTS-surgery (NCSP-D
codes KACC51 and KACC61 or service code 3146) and the date of diagnosis or operation.
Exposure intensity
The individual physical exposure assessment was based on self-record of task distribution and objective
measurements of task specific movements and postures of the wrist. Details have been described in a
separate paper (18). Task distributions were constructed by having the questionnaire respondents divide
their work hours in a typical week between the 12 most common painting tasks (e.g. painting with a brush,
6
painting with a roller, spraying, , levelling, sanding, covering, driving, other tasks). The typical week should
represent their work as a painter since 1990. Sex-specific task exposure matrices (TEMs) of postures and
movements of the upper extremity were constructed using goniometry (wrists) and inclinometry
(shoulders, neck and forehead) to measure 50 full working days equally distributed between male and
female house painters. During the measurements, a log-book clearly stating the time spent on each task
was filled out by the participants (18). Exposure measures from the TEMs were weighted by the self-
reported task distributions to obtain individual estimates of exposure intensity for the wrist in terms of
median velocity (°/s) of flexion/extension of the wrist (wrist velocity), mean power frequency (MPF) (Hz) as
a measure of repetitiveness, and non-neutral postures (% time) as a combined measure of extreme
postures of flexion/extension and radial/ulnar deviation of the wrist (18).
Exposure duration
Information on seniority as a painter was obtained from the questionnaire. The yearly working proportion
was assessed from record linkage with Danish national registers. Information on periods of absence from
work without being available for employment (i.e. maternity leave, sickness absence, and periods of
education) was obtained from the Danish Register for Evaluation of Marginalisation, which contains data on
all Danish public transfer payments on a weekly basis (22). Data on periods of unemployment was obtained
from the Integrated Database for Labour Market Affiliation for Persons, which contains data on yearly
unemployment as a proportion of the total number of possible working days per year. Based on this
information, we calculated a yearly working proportion for each member of the cohort (range 0 to 1, 0=did
not work that year, 1=worked all working days that year). The working proportion was calculated for each
calendar year after entry into the study (see below).
Other covariates
Information on sex and date of birth was obtained from the Civil Registration System (REF). From the
questionnaire, we obtained Information on height and weight, fractures near the wrist and co-morbidity in
terms of hypothyroidism, diabetes, rheumatoid arthritis, gout, and connective tissue disorders (0=none,
7
1=any of these diseases). Body mass index (BMI) was calculated as weight (kg)/height (m)2. A pregnancy
variable was defined as a period of 7 months prior to and 5 months after giving birth according to dates of
birth obtained from DNPR.
Statistical analyses
The incidence rate ratios (IRRs) of CTS-diagnoses and of CTS-surgery were examined by survival analyses
with exposure intensity and exposure duration variables, sex, age, BMI, fractures near the wrist,
comorbidity, and pregnancy as explanatory factors. Age, seniority, working proportion, and pregnancy were
considered as time-dependent variables. The cohort members were followed from January 1st 1994 or start
of membership of the Painters’ Union in Denmark, whichever came latest, until a first time CTS-event or
January 1st 2011, whichever came first (the cohort only consisted of persons who were alive at the end of
follow up). Ninety one persons emigrated during the study period, with a combined risk time of 112.3
person years (0.1% of total risk time). No exclusions were made. Risk time was calculated from entry to exit
date for each calendar year, broken down by time-dependent variables. Risk time calculations were made
with a publicly accessible macro (23) using the methodology of a Lexis diagram (24). Persons with a CTS-
event before start of follow up were excluded. A CTS-event was defined as a CTS-diagnosis in the analyses
of CTS-diagnoses and as CTS-surgery in the analyses of CTS-surgery. We used a log linear Poisson model for
the analyses.
We first examined questionnaire responders, in total and by sex. To assess the effects of non-response we
also analysed the total material, including variables recorded for all participants (working proportion,
seniority, sex, age, and pregnancy) as covariates, supplemented by a questionnaire response variable
(yes/no).
The effects of exposure duration were examined in separate parallel analyses of the effects of the work
proportion variable defined for time windows of 1, 2 and 5 years and of total seniority before the year of
occurrence of a CTS-event.
8
In the analyses of data for both sexes, we examined if the effects of the exposure variables were different
for men and women by including an interaction term between sex and the exposure variable. We also
examined if there was an interaction between exposure intensity and duration to see if the effects of these
two exposure modalities were mutually dependent, which would indicate accumulation of exposure effects
over time. Non-significant interaction terms were excluded from the final models.
Pregnancy had no significant effect in any of the models examined and was therefore not included in the
final models.
All models were examined twice, once for the outcome of diagnoses and once for operations. Sensitivity
analyses were performed using only outcome data from DNPR. Statistical analyses were made using SAS
statistical software (v9.3 Cary, NC, USA). The GENMOD procedure was used to fit the Poisson survival
analysis models.
Ethics statement
The study protocol was accepted by the Regional Scientific Ethics Committee, Capital region of Denmark
(j.no.: H-C-FSP-2010-036). Individuals who participated in the biomechanical measurements gave written
informed consent. The Danish Data Protection Agency gave permission to store any personal information
concerning the participants (j.no.: 2010-41-5325).
9
RESULTS
The questionnaire was returned by 4957 (53%, 51% among men and 59% among women). In the DNPR and
the NHSR, we identified 464 and 102 registrations of CTS, respectively, of which 192 and 70 qualified as first
time diagnoses. After exclusion of CTS-diagnoses before entry into the Painters’ Union, 230 first time CTS-
diagnoses were available for the analyses. Out of the 192 DNPR first time diagnoses, 111 had a first time
event of CTS-surgery. Together with the 70 NHSR-cases, who all had surgery, and after exclusion of those
who were operated before entry into the Painters’ Union, 162 first time events of CTS-surgery were
available for the analyses (Figure 1).
Figure 1. Flow chart of CTS-diagnoses (Danish National Patient Register 1994-2011 and Danish National Health Service Register 2002-2011) and CTS-surgery (Danish National Patient Register 1996-2011 and Danish National Health Service Register 2002-2011) in the cohort of house painters (N=9364). CTS=carpal tunnel syndrome.
10
Table 1. Characteristics of the study population according to register information for the total cohort and for questionnaire responders.
Age (years) Seniority (years) Work
proportion Diagnoses of carpal tunnel
syndrome* Surgery for carpal tunnel
syndrome*
Mean SD Mean SD Mean SD
Risk time
(years)
Incidence rate per 10.000 years
Study period
prevalence (%)
Risk time
(years)
Incidence rate per 10.000 years
Study period
prevalence (%)
Total (n=9364) 41.5 15.7 18.2 14.7 0.76 0.35 104308 22.05 2.5 104792 15.46 1.7
Men (n=6236) 45.5 16.0 21.6 15.6 0.76 0.36 76694 13.17 1.6 76969 8.57 1.1
Women (n=3128) 33.4 11.1 11.5 9.7 0.74 0.34 27614 46.72 4.1 27823 34.50 3.1
Questionnaire responders (n=4957)
44.3 15.3 22.2 14.5 0.77 0.35 60993 26.56 3.3 61332 18.91 2.3
Men (n=3124) 49.7 14.3 27.1 14.3 0.77 0.36 43310 16.39 2.3 43509 11.03 1.5
Women (n=1833) 35.2 11.4 13.8 10.3 0.77 0.32 17683 51.46 5.0 17823 38.15 3.7
*Records from the Danish National Patient Register and the Danish National Health Register during the study period 1994-2011.
Table 2. Study population characteristics based on questionnaire information.
Questionnaire responders
(N=4957) Men
(N=3124) Women
(N=1833)
Mean SD Mean SD Mean SD
Body mass index (kg/m2) 25.9 4.5 26.4 4.2 25.1 4.9
n (%) n (%) n (%)
Fracture near the wrist 574 11.8* 390 12.9* 184 10.2*
Co-morbidity 798 16.7* 616 20.0* 182 10.2*
Hypothyroidism 140 2.9* 75 2.5* 65 3.6*
Diabetes 219 4.6* 200 6.7* 19 1.1*
Rheumatoid arthritis 410 8.6* 313 10.5* 97 5.4*
Gout 149 3.1* 138 4.6* 11 0.6*
Connective tissue disorders
53 1.1*
37 1.2*
16 0.9*
*Based on variable specific numbers of responders.
Tables 1 and 2 present characteristics of the total cohort and of the questionnaire responders. In the total
cohort, the female/male prevalence ratios of CTS diagnoses and surgery were 2.6 and 2.8, respectively, and
the corresponding IRRs were 3.6 and 4.0 (Table 1). There were few missing questionnaire values [BMI
n=183 (3.7%), fracture near the wrist n=134 (2.7%), and co-morbidity n=170 (3.4%), Table 2]. On average,
the men were 12 years older than the women were and had worked 10 years more as a painter. The mean
working proportion was slightly higher (0.02) for men than for women, corresponding to nearly one
working week per year. The BMI distribution and the prevalence of fractures near the wrist were similar for
the two sexes. The prevalence of co-morbidity was twice as high among men as among women (Table 2).
11
Information on task distribution was missing for 537 participants (11%), leaving 4420 participants for the
analyses. The task distributions were similar for men and women (results not shown). Table 3 shows the
distributions of the three exposure intensity variables and their Pearson correlations. For women there was
a strong correlation between MPF and wrist velocity (0.80) and a moderate correlation between MPF and
non-neutral postures (-0.60). Other sex-specific correlations reached a maximum of 0.21.
Table 3. Distributions of exposure intensity variables and their correlations. Questionnaire responders who provided information on task distribution (n=4420).
Median velocity for flexion/ extension of the wrist (°/s) Mean
Standard deviation
5th percentile
95th percentile
Correlation with mean power
frequency (Hz)
Total 15.6 0.8 14.4 16.7 0.38*
Men (n=2738) 15.6 0.7 14.4 16.7 0.15*
Women (n=1682) 15.6 0.9 14.4 16.8 0.80*
Mean power frequency (Hz) Correlation with non-neutral wrist postures (% time)
Total 0.281 0.007 0.266 0.289 -0.36*
Men (n=2738) 0.285 0.003 0.280 0.289 -0.15*
Women (n=1682) 0.274 0.007 0.263 0.286 -0.60*
Non-neutral wrist postures (% time)
Correlation with median velocity
(°/s)
Total 31.1 1.7 28.6 33.9 0.13*
Men (n=2738) 30.8 1.8 28.3 33.8 0.21*
Women (n=1682) 31.5 1.4 29.1 33.9 0.004
* Statistically significantly different from zero (p<0.0001)
12
Table 4. Incidence rate ratios (IRR) of CTS-diagnoses and CTS-surgery. Models with median velocity of flexion/extension of the right wrist, working
proportion, sex, age, body mass index, fractures near the wrist, co-morbidity, and questionnaire response as explanatory factors, depending on the
model. CTS=carpal tunnel syndrome, CI=confidence interval.
Median velocity of
flexion/ extension
of the wrist
(per 1 °/s)
Work
proportion
previous
year
Sex
(women/
men)
Age
(per 10
years)
Body mass
index
(per 5 units)
Fracture
near the
wrist
(yes/no)
Co-
morbidity
(yes/no)
Question-
naire
response
(yes/no)
Model 1. Crude associations
CTS diagnoses
IRR 1.351 0.752 3.552 0.992 1.251 1.471 1.371 1.692
95 % CI 1.10-1.64 0.54-1.05 2.73-4.60 0.90-1.09 1.09-1.43 0.97-2.24 0.96-1.96 1.27-2.25
P-value 0.0035 0.049 <0.0001 0.89 0.0012 0.070 0.083 0.0003
CTS surgery
IRR 1.39 0.75 4.02 1.02 1.29 1.22 1.45 1.79
95 % CI 1.10-1.76 0.50-1.11 2.94-5,50 0.91-1.14 1.11-1.50 0.72-2.07 0.96-2.20 1.27-2.51
P-value 0.0051 0.15 <0.0001 0.79 0.0009 0.56 0.079 0.0009
Model 2. Mutually adjusted associations.
All questionnaire responders (n=4198)
CTS diagnoses
IRR 1.29 0.80 4.64 1.25 1.27 1.57 1.40 -
95 % CI 1.07-1.56 0.52-1.22 3.21-6.71 1.08-1.44 1.12-1.45 1.03-2.40 0.97-2.03 -
P-value 0.0085 0.29 <0.0001 0.0030 0.0002 0.035 0.075 -
CTS surgery
IRR 1.32 0.86 6.03 1.41 1.32 1.33 1.42 -
95 % CI 1.06-1.65 0.52-1.41 3.89-9.34 1.18-1.67 1.14-1.52 0.78-2.25 0.92-2.19 -
P-value 0.014 0.55 <0.0001 0.0001 0.0002 0.30 0.11 -
Model 2a. Mutually adjusted associations.
Male questionnaire responders (n=2596)
CTS diagnoses
IRR 1.22 0.72 - 1.13 1.25 2.12 1.21 -
95 % CI 0.86-1.72 0.39-1.33 - 0.91-1.41 1.01-1.55 1.21-3.70 0.72-2.04 -
P-value 0.27 0.29 - 0.26 0.0394 0.0085 0.48 -
CTS surgery
IRR 1.30 0.80 - 1.33 1.33 1.47 1.10 -
95 % CI 0.85-1.99 0.38-1.68 - 1.00-1.76 1.06-1.66 0.69-3.13 0.58-2.08 -
P-value 0.23 0.56 - 0.048 0.015 0.33 0.77 -
Model 2b. Mutually adjusted associations.
Female questionnaire responders (n=1602)
CTS diagnoses
IRR 1.32 0.80 - 1.34 1.30 1.13 1.60 -
95 % CI 1.05-1.64 0.44-1.46 - 1.11-1.62 1.11-1.53 0.58-2.17 0.96-2.69 -
P-value 0.016 0.47 - 0.0026 0.0015 0.73 0.073 -
CTS surgery
IRR 1.32 0.85 - 1.45 1.31 1.22 1.78 -
95 % CI 1.02-1.71 0.43-1.71 - 1.16-1.80 1.09-1.58 0.58-2.55 1.003-3.16 -
P-value 0.037 0.65 - 0.0010 0.0041 0.60 0.049 -
Model 3. Mutually adjusted associations.
Total cohort (n=9364)
CTS diagnoses
IRR - 0.84 4.63 1.25 - - - 1.44
95 % CI - 0.59-1.20 3.41-6.29 1.11-1.41 - - - 1.08-1.92
P-value - 0.34 <0.0001 0.0002 - - - 0.0013
CTS surgery
IRR - 0.86 5.71 1.34 - - - 1.47
95 % CI - 0.57-1.38 3.95-8.24 1.16-1.54 - - - 1.04-2.09
P-value - 0.87 <0.0001 <0.0001 - - - 0.028
Bold highlights statistically significant IRRs (p<0.05). 1. Questionnaire responders (n=4420 for wrist velocity, n=4773 for BMI, n=4825 for fractures near
the wrist, n=4787 for co-morbidity). 2. Total material (n=9364)
13
Table 4 shows the results of survival analyses from models with wrist velocity as the measure of exposure
intensity. We omitted seniority as a covariate (see below). Owing to missing values, the analyses of
questionnaire responders (models 2a and 2b) were based on 4198 participants (44.8% of the total cohort).
Crude IRRs were similar to the IRRs in the adjusted models, except for sex and age. However, when these
two covariates were mutually adjusted, their IRRs became similar to those of the other models (results not
shown); this pattern reflected the composition of the cohort with men being older and having less CTS than
women.
Increasing wrist velocity was associated with statistically significantly higher IRRs for both CTS-diagnoses
and CTS-surgery. The IRRs were quite stable across models, approximately 1.30 per 1 °/s, except in model
2a (men only) where it was a little lower (1.22) and not statistically significant. The working proportion in
the previous year was not significantly related to the outcomes. The IRRs for women versus men ranged
from 4.6 to 4.9 for CTS-diagnoses and from 6.0 to 6.1 for CTS-surgery. The IRRs increased significantly with
increasing age and BMI. For fractures and co-morbidity the IRR’s also increased. The estimates of BMI
effects were of similar size and statistically significant for men and women and for the two CTS outcomes.
Wrist-near fractures and comorbidity seemed to have different effects for the two sexes (models 2a and
2b). The IRRs of questionnaire responders were higher than those of non-responders (model 1 and model
3).
The IRRs of working proportion, sex, and age were similar for questionnaire responders (model 2) and for
the total material (model 3).
14
Table 5. Incidence rate ratios of CTS-diagnoses and CTS-surgery. Models with mean power frequency and non-neutral postures for the right wrist, working proportion, sex, age, BMI, fractures near the wrist, co-morbidity, and questionnaire response as explanatory factors, depending on the model. Same models as in Table 4, apart from the exposure intensity variables. IRR=incidence rate ratio, CI=confidence interval.
Mean power frequency (Hz) (per 0.010 Hz*)
Non-neutral wrist postures (% time)
Model 1. Crude associations (n=4420)
CTS diagnoses
IRR 1.38 .98
95 % CI 1.12-1.70 0.88-1.08
P-value 0.0026 0.67
CTS surgery
IRR 1.33 0.97
95 % CI 1.04-1.71 0.86-1.10
P-value 0.025 0.62
Model 2. Mutually adjusted associations. All questionnaire responders (n=4198)
CTS diagnoses
IRR 1.39 0.98
95 % CI 1.13-1.71 0.88-1.08
P-value 0.0022 0.66
CTS surgery
IRR 1.34 0.97
95 % CI 1.04-1.72 0.86-1.09
P-value 0.023 0.59
Model 2a. Mutually adjusted associations. Male questionnaire responders (n=2596)
CTS diagnoses
IRR 1.33 1.07
95 % CI 0.57-3.13 0.94-1.21
P-value 0.51 0.31
CTS surgery
IRR 1.30 1.02
95 % CI 0.46-3.66 0.87-1.20
P-value 0.62 0.80
Model 2b. Mutually adjusted associations. Female questionnaire responders (n=1602)
CTS diagnoses
IRR 1.40 0.86
95 % CI 1.13-1.74 0.74-1.002
P-value 0.0024 0.053
CTS surgery
IRR 1.35 0.91
95 % CI 1.04-1.75 0.77-1.08
P-value 0.025 0.28
Bold highlights statistically significant IRRs (p<0.05). * Almost equal to the interquartile range.
15
Table 5 shows results for the two other exposure intensity variables than wrist velocity, analysed in the
same models as in Table 4. We have omitted the results for the other covariates since they were very
similar to those presented in Table 4. The IRRs associated with an increase in MPF by 0.01 Hz ranged from
1.33 to 1.40. This effect was statistically significant in all models with the exception of model 2a (male
questionnaire responders). Non-neutral wrist postures were not related to the outcomes.
The models presented in Tables 4 and 5 were also examined using working proportions during the previous
two and five years instead of only the previous year. The results of these analyses were very similar to
those using the working proportion in the previous year only. None of the interaction terms between
exposure intensity or exposure duration variables with sex were significant. Neither were the interaction
terms between exposure intensity and exposure duration variables. Seniority was not included in the
models (Tables 4 and 5) since it had a very high correlation with age (0.83, 0.82 among men and 0.77
among women). When seniority was included instead of age, the IRRs for seniority were close to 1. When
age and seniority were included in the same model, the association was statistically significant for age but
insignificant for seniority in all models except model 3 for wrist velocity and CTS-diagnoses (p=0.04).The
estimates increased for age and decreased for seniority when both variables were included in the models.
Sensitivity analyses, limiting the outcomes to diagnoses and surgery listed in the DNPR showed slightly
increased IRRs (results not shown).
16
DISCUSSION
We found that the IRRs of both CTS-outcomes increased with increasing velocity of wrist flexion/extension
and with increasing MPF, but not with non-neutral wrist postures. The working proportion during 1, 2, and
5 years prior to a CTS-event had no significant effect, and the effects of exposure intensity were not
modified by these measures of exposure duration. These findings suggested that repetitive and high
velocity wrist movements are occupational risk factors for CTS, whereas non-neutral wrist postures are not,
and that effects of repetitiveness and velocity do not accumulate over time. Thus, the results indicated that
CTS may develop after only a short period with high velocity and repetitiveness. The relationships between
CTS -outcomes and the exposure intensity and duration variables were not modified by sex.
The relationships between CTS -outcomes and the exposure intensity and duration variables were not
modified by sex, indicating that these exposures affect men and women equally. However, this was on a
logarithmic IR-scale. On the back-transformed scale, the absolute increase of IRs will depend on the IR at
zero exposure (the intercept in the analyses). The independent effect of sex means that the IR of women at
zero exposure is higher than that of men, and as a consequence the back-transformed increase in IRs is
steeper for women than for men. Figure 2 illustrates the absolute IRs of the two CTS-outcomes based on
the effect estimates of wrist velocity and MPF from the crude models (model 1 in Tables 4 and 5). The
figure also shows the observed IRs of CTS-diagnoses for men and women by tertiles of the exposure. The
observed IRs of CTS-diagnoses for men were 1.70, 1.33, and 1.91 per 1000 person-years by increasing
tertiles of wrist velocity. The corresponding figures for women were 4.13, 4.81, and 6.51. The observed
differences in IRs between men and women were 2.43, 3.48, and 4.60 per 1000 person-years by increasing
tertiles. A similar pattern of results was found for CTS-surgery versus wrist velocity and for the two CTS-
outcomes and MPF (results not shown). The raw data, thus, do not seem to be at odds with assumptions of
the log-linear models, neither were these rejected by PROC GENMOD fit statistics. Nevertheless, one might
consider supplementary analyses in additive Poisson models to examine the additive linear outcome-
17
exposure pattern and if the difference of absolute incidence rates between men and women were
statistically significant.
Figure 2. Observed incidence rates of CTS-outcomes by tertiles of wrist velocity and MPF (grey and black columns) with estimated rates overlaid. Estimates are based on crude associations (Model 1 in tables 4 and 5)
CTS-diagnoses incidence rates by sex and tertiles of median wrist velocity
Median wrist velocity (deg./s)
8 10 12 14 16 18 20 22
CT
S-in
cide
nce
per
1000
per
son
year
s at
ris
k
0
5
10
15
20
25
FemaleMaleMaleFemale
CTS-operations incidence rates by sex and tertiles of median wrist velocity
Median wrist velocity (deg./s)
8 10 12 14 16 18 20 22
CT
S-in
cide
nce
per
1000
per
son
year
s at
ris
k
0
2
4
6
8
10
12
14
16
18
20
FemaleMaleMaleFemale
CTS-diagnoses incidence rates by sex and tertiles of mean power frequency
MPF (Hz)
0.24 0.26 0.28 0.30 0.32 0.34
CT
S-in
cide
nce
per
1000
per
son
year
s at
ris
k
0
5
10
15
20
25
30
35
FemaleMaleMaleFemale
CTS-operations incidence rates by sex and tertiles of mean power frequency
MPF (Hz)
0.24 0.26 0.28 0.30 0.32 0.34
CT
S-in
cide
nce
per
1000
per
son
year
s at
ris
k
0
5
10
15
20
25
FemaleMaleMaleFemale
18
We decided not to include seniority in the final models due to its high correlation with age, lack of
significant effects on CTS with and without adjustment for age, and change in effect direction after
adjustment for age. We considered seniority as a measure of cumulative exposure, but the results are
difficult to interpret in this context. Furthermore, seniority may be influenced by a secondary healthy
worker effect if persons with CTS left the painting trade before the cohort was established.
Our cohort consisted of painters who were members of the PUD when the study was initiated in 2011. This
means that a secondary healthy worker effect could have biased our results towards unity through non-
differential selection if independent of exposure, and even more if related to high exposures. However, we
did find a study period prevalence and incidence rates resembling those reported in the literature (25-32).
Shiri et al. (6) reported a prevalence of CTS-diagnoses of 2.1% in men and 5.3% in women. In comparison,
we found a prevalence of CTS-diagnoses of 2.3 % in men and 5.0 % in women among questionnaire
responders (Table 1). Annual IRs of CTS-diagnoses of 18.2 per 10.000 person years for men and 42.8 for
women were reported by Atroshi et al. (25). Similarly, we found IRs of 16.4 per 10.000 person years for
men and 51.5 for women (Table1).
The proportion that returned the questionnaire was only 53%. Furthermore, among questionnaire
responders, quite a few had missing values, especially for task distributions, leaving only 45% of the total
cohort for the full model survival analyses (model 2 in Tables 4 and 5). Questionnaire responders had a
higher IR of CTS-outcomes than non-responders; this effect became smaller after adjustment for age, sex,
and working proportion, but remained significant (model 3 in Table 4). However, the IRRs of CTS-outcomes
related to sex, age, and working proportion, recorded for the total cohort, were similar for questionnaire
responders and the total cohort (Table 4, models 2 and 3). Furthermore, the IRRs of the CTS-outcomes
associated with the exposure intensity variables changed only marginally after adjustment for these factors
(Tables 4 and 5, models 1 and 2). This was also the case when further adjustment was made for BMI, wrist-
near fractures and co-morbidity, neither did adjustment for the questionnaire based covariates change the
effect estimates of register-based covariates (data not shown). Considering the rather stable IRRs for the
19
exposure intensity variables across the different models, we think that it is unlikely that our results were
seriously affected by non-response bias.
We have combined information from several Danish national registers. The quality of the Civil Registration
System is very high and is continuously validated (19). Both DNPR and NHSR (20;21) function as tools for
determining the payment to the health care provider. Since 2003 it has been mandatory for private
hospitals to report all services, but a 2008 estimate from the Danish National Board of Health still showed
an underreporting of 5 % in surgeries reported to the DNPR (20). We consider this misclassification as non-
differential in relation to our study and the effect, therefore, may only be to attenuate our findings.
In the NHSR, the service code 3146 (“nerve compression”) is not uniquely used for CTS, but a personal
contact to all Danish practitioners in this field has assured us that the vast majority of these registrations
concern CTS. The service code was not introduced until 2002. Before then, more general codes, which we
did not include, were used to cover treatment for CTS. Therefore we performed sensitivity analyses, where
the case definitions were limited to those identified in the DNPR. This resulted in stronger effect estimates
for wrist velocity and MPF. Still, we chose to leave NHSR cases in the final model since we think most of
these were in fact CTS.
Individual exposure intensity measures were based on questionnaire information on distribution of
painting tasks during a typical working week for painting work after 1990. Obviously, this is a difficult
question, especially if there was a large individual variation in task distributions over years. According to
representatives of the Painters’ Union in Denmark, work as a painter had not changed much over the years,
although some changes had taken place, e.g. more sanding with a giraffe, more use of glass-felt, and less
wallpaper hanging. If status as a CTS-case influenced the recall of the typical task distribution, this may have
resulted in inflated effect estimates. If so, we think that CTS-cases would probably report a higher
proportion of tasks with high force requirements rather than high velocity and repetitiveness, among other
things because it would be difficult for the participants to rank typical painting tasks by velocity and
20
repetitiveness without knowledge of measurements. Analyses comparing the sex-specific task distributions
for all questionnaire responders with the subgroup of CTS cases revealed that male cases primarily did
more “spraying” and “other tasks” and less “wallpaper hanging” and had fewer pauses (results not shown).
Female cases reported more “full leveling” and “hanging wallpaper” and less “driving”. Some of the tasks
that are usually considered among the hardest (“giraffe drywall sander” and “removing wallpaper”) were
reported for a lower proportion of the time among both male and female cases. This indicated that cases
did not just report more of the tasks which were felt as physically demanding. Also, if recall bias was to
explain our results, we would expect that non-neutral wrist postures had also been related to the
outcomes. Knowing that the composition of a questionnaire can influence the responses (33;34), we tried
not to use phrases which implied a special interest in CTS, and the questionnaire contained questions on
many different personal and physical aspects.
Our results concerning the effects of wrist velocity and MPF are in accordance with findings in other
studies. Thus, a 2012 meta-analysis (1) concluded that repetitive wrist movements may cause CTS
We did not find any association between non-neutral risk postures and CTS. This is in contrast to a recent
meta-analysis (5) which showed a relative risk of 2.01 (CI 1.32-2.97) with high exposures to non-neutral
wrist postures. However this meta-analysis only included studies where non-neutral wrist postures were
assessed by self-report or an observer. Our results supported a meta-analysis from 2012 (1) and a
systematic review from 2009 (35), which, in concurrence with others (36-38), found insufficient evidence to
support that non-neutral wrist postures increase the risk of CTS. In our study we found no effect of
duration of work before the occurrence of CTS-events. This result is supported by the results of Shiri et al.
(6), who could only find associations between CTS and the most recent job.
The main limitations of this study is a low questionnaire response rate and the task distribution assessment,
Another limitation to the study is that no measure of muscular load was included. In a previous study (39)
we have shown that men and women exert the same amount of absolute force doing the same tasks.
Owing to their lower physical strength, women are therefore exposed to a higher relative muscular load
21
than men (40). It has been suggested (41) that a sex difference in relative muscular load accumulated over
years may play a role in the development of WMSDs.
The main strengths of the study is that the outcomes are diseases diagnosed and recorded independently
of the exposure, and that analyses is based on incidence rates of physician diagnosed cases.
The range of exposure intensity variables were not very wide compared to other studies and our results
may not apply to exposures outside this range. Our study was limited to Danish building painters and the
results should therefore be interpreted with caution when applied to other populations.
Our results are in accordance with the hypothesis that CTS may be caused by high velocity and
repetitiveness of wrist movements. This effect was related to the intensity but not to the duration of the
exposure. Women had a higher risk of developing CTS than men, independent of other risk factors. Their
absolute risk of CTS also increased more than those for men with increasing wrist velocity and
repetitiveness. The steeper exposure-response relation for women compared to men may be interpreted
as a higher vulnerability of women to adverse musculoskeletal effects of these exposures. However, the
results may partly depend on the multiplicative Poisson model used to analyse the data and further
analyses are required to reach a firm conclusion on this problem.
22
Reference List
(1) Barcenilla A, March LM, Chen JS, Sambrook PN. Carpal tunnel syndrome and its relationship to occupation: a meta-analysis. Rheumatology 2012 Feb 1;51(2):250-61.
(2) Gerr F, Fethke NB, Merlino L, Anton D, Rosecrance J, Jones MP, et al. A Prospective Study of Musculoskeletal Outcomes Among Manufacturing Workers: I. Effects of Physical Risk Factors. Human Factors: The Journal of the Human Factors and Ergonomics Society 2014 Feb 1;56(1):112-30.
(3) Violante FS, Bonfiglioli R, Hagberg M, Rempel D. Carpal tunnel syndrome diagnosis in occupational epidemiological studies. Occupational and Environmental Medicine 2014 May 30.
(4) Viikari-Juntura E, Silverstein B. Role of physical load factors in carpal tunnel syndrome. Scand J Work Environ Health 1999 Jun;25(3):163-85.
(5) You D, Smith AH, Rempel D. Meta-Analysis: Association Between Wrist Posture and Carpal Tunnel Syndrome Among Workers. Safety and Health at Work 2014 Mar;5(1):27-31.
(6) Shiri R, Miranda H, Heli+Âvaara M, Viikari-Juntura E. Physical work load factors and carpal tunnel syndrome: a population-based study. Occupational and Environmental Medicine 2009 Jun 1;66(6):368-73.
(7) Gallagher S, Heberger JR. Examining the Interaction of Force and Repetition on Musculoskeletal Disorder Risk: A Systematic Literature Review. Human Factors: The Journal of the Human Factors and Ergonomics Society 2013 Feb 1;55(1):108-24.
(8) Bonfiglioli R, Mattioli S, Fiorentini C, Graziosi F, Curti S, Violante F. Relationship between repetitive work and the prevalence of carpal tunnel syndrome in part-time and full-time female supermarket cashiers: a quasi-experimental study. Int Arch Occup Environ Health 2007;80(3):248-53.
(9) Ghasemi-Rad M, Nosair E, Vegh A, Mohammadi A, Akkad A, Lesha E, et al. A handy review of carpal tunnel syndrome: From anatomy to diagnosis and treatment. World J Radiol 2014 Jun 28;6(6):284-300.
(10) Roquelaure Y, Petit A, Fouquet B, Descatha A. [Work-related musculoskeletal disorders: priority to prevention and coordination of the interventions]. Rev Prat 2014 Mar;64(3):350-7.
(11) Lalonde DH. Evidence-based medicine: Carpal tunnel syndrome. Plast Reconstr Surg 2014 May;133(5):1234-40.
(12) Gillespie RM, Herbert R, Punnett L. Chapter 41 - Work-Related Musculo-Skeletal Disorders. In: Goldman MB, Troisi R, Rexrode KM, editors. Women and Health (Second Edition). Academic Press; 2013. p. 613-28.
(13) Hooftman WE, van der Beek AJ, Bongers PM, van MW. Is there a gender difference in the effect of work-related physical and psychosocial risk factors on musculoskeletal symptoms and related sickness absence? Scand J Work Environ Health 2009 Mar;35(2):85-95.
23
(14) Treaster DE, Burr D. Gender differences in prevalence of upper extremity musculoskeletal disorders. Ergonomics 2004 Apr 15;47(5):495-526.
(15) Eng A, 't Mannetje A, McLean D, Ellison-Loschmann L, Cheng S, Pearce N. Gender differences in occupational exposure patterns. Occupational and Environmental Medicine 2011 Dec 1;68(12):888-94.
(16) Kennedy SM, Koehoorn M. Exposure assessment in epidemiology: does gender matter? Am J Ind Med 2003 Dec;44(6):576-83.
(17) Leijon O, Bernmark E, Karlqvist L, Harenstam A. Awkward work postures: association with occupational gender segregation. Am J Ind Med 2005 May;47(5):381-93.
(18) Heilskov-Hansen T, Svendsen SW, Thomsen JF, Mikkelsen S, Hansson GA. Sex differences in task distribution and task exposure among Danish house painters: An observational study combining questionnaire data with biomechanical measuremnets. 2014.
Ref Type: Unpublished Work
(19) Pedersen CB, Gotzsche H, Moller JO, Mortensen PB. The Danish Civil Registration System. A cohort of eight million persons. Dan Med Bull 2006 Nov;53(4):441-9.
(20) Lynge E, Sandegaard JL, Rebolj M. The Danish National Patient Register. Scandinavian Journal of Public Health 2011 Jul 1;39(7 suppl):30-3.
(21) Sahl Andersen J, De Fine Olivarius N, Krasnik A. The Danish National Health Service Register. Scandinavian Journal of Public Health 2011 Jul 1;39(7 suppl):34-7.
(22) Hjollund NH, Larsen FB, Andersen JH. Register-based follow-up of social benefits and other transfer payments: Accuracy and degree of completeness in a Danish interdepartmental administrative database compared with a population-based survey. Scandinavian Journal of Public Health 2007 Oct 1;35(5):497-502.
(23) Carstensen B. Lexis macro for SAS. 1-11-2003. Ref Type: Online Source
(24) Carstensen B. Age period cohort models for the Lexis diagram. Statist Med 2007 Jul 10;26(15):3018-45.
(25) Atroshi I, Englund M, Turkiewicz A, Tagil M, Petersson IF. Incidence of physician-diagnosed carpal tunnel syndrome in the general population. Arch Intern Med 2011 May 23;171(10):943-4.
(26) Atroshi I, Gummesson C, Johnsson R, Ornstein E, Ranstam J, Rosen I. Prevalence of carpal tunnel syndrome in a general population. JAMA 1999 Jul 14;282(2):153-8.
(27) Descatha A, Dale AM, Franzblau A, Coomes J, Evanoff B. Comparison of research case definitions for carpal tunnel syndrome. Scand J Work Environ Health 2011 Jul;37(4):298-306.
(28) Luckhaupt SE, Dahlhamer JM, Ward BW, Sweeney MH, Sestito JP, Calvert GM. Prevalence and work-relatedness of carpal tunnel syndrome in the working population, United States, 2010 national health interview survey. Am J Ind Med 2013 Jun 1;56(6):615-24.
24
(29) Mondelli M, Giannini F, Giacchi M. Carpal tunnel syndrome incidence in a general population. Neurology 2002 Jan 22;58(2):289-94.
(30) Roquelaure Y, Ha C, Leclerc A, Touranchet A, Sauteron M, Melchior M, et al. Epidemiologic surveillance of upper-extremity musculoskeletal disorders in the working population. Arthritis Rheum 2006 Oct 15;55(5):765-78.
(31) Salaffi F, De AR, Grassi W. Prevalence of musculoskeletal conditions in an Italian population sample: results of a regional community-based study. I. The MAPPING study. Clin Exp Rheumatol 2005 Nov;23(6):819-28.
(32) Walker-Bone K, Palmer KT, Reading I, Coggon D, Cooper C. Prevalence and impact of musculoskeletal disorders of the upper limb in the general population. Arthritis & Rheumatism 2004 Aug 15;51(4):642-51.
(33) Balogh I, Oerbaek P, Ohlsson K, Nordander C, Unge J, Winkel J, et al. Self-assessed and directly
measured occupational physical activities nfluence of musculoskeletal complaints, age and gender. Applied Ergonomics 2004 Jan;35(1):49-56.
(34) Viikari-Juntura E, Rauas S, Martikainen R, Kuosma E, Riihimaki H, Takala EP, et al. Validity of self-reported physical work load in epidemiologic studies on musculoskeletal disorders. Scand J Work Environ Health 1996 Aug;22(4):251-9.
(35) van Rijn RM, Huisstede BM, Koes BW, Burdorf A. Associations between work-related factors and the carpal tunnel syndrome--a systematic review. Scand J Work Environ Health 2009 Jan;35(1):19-36.
(36) Thomsen JF, Mikkelsen S, Andersen JH, Fallentin N, Loft IP, Frost P, et al. Risk factors for hand-wrist disorders in repetitive work. Occupational and Environmental Medicine 2007 Aug 1;64(8):527-33.
(37) Roquelaure Y, Mariel J, Dano C, Fanello S, Penneau-Fontbonne D. Prevalence, incidence and risk factors of carpal tunnel syndrome in a large footwear factory. Int J Occup Med Environ Health 2001;14(4):357-67.
(38) NIOSH. Musculoskeletal Disorders and Workplace Factors - A Critical Review of Epidemiologic Evidence for Work-Related Musculoskeletal Disorders of the Neck, Upper Extremity, and Low Back. U.S. DEPARTMENT OF HEALTH AND HUMAN SERVICES; 1997 Jul. Report No.: 97-141.
(39) Meyland J, Heilskov-Hansen T, Alkjar T, Koblauch H, Mikkelsen S, Svendsen S, et al. Sex differences in muscular load among house painters performing identical work tasks. Eur J Appl Physiol 2014;1-11.
(40) Miller AE, MacDougall JD, Tarnopolsky MA, Sale DG. Gender differences in strength and muscle fiber characteristics. Eur J Appl Physiol Occup Physiol 1993;66(3):254-62.
(41) Nordander C, Ohlsson K, Akesson I, Arvidsson I, Balogh I, Hansson GA, et al. Exposure-response relationships in work-related musculoskeletal disorders in elbows and hands - A synthesis of group-level data on exposure and response obtained using uniform methods of data collection. Appl Ergon 2013 Mar;44(2):241-53.
PhD ThesisThomas Heilskov-Hansen
2014
Physical work exposure and sex differences in work-related musculoskeletal disorders
Physical work exposure and sex differences in w
ork-related musculoskeletal disorders
Thomas Heilskov-Hansen – PhD thesis
ISBN 978-87-996042-6-5
UDGIVNE RAPPORTER
1. Rapport om Reproduktionsskader ved arbejde med Frisørkemikalier, 2013
2. Rapport om Arbejdsbetinget Eksem, 2013
3. Rapport om Gravides arbejdsmiljø. Udsættelse for Methyl metakrylat (MMA) blandt tandteknikere og operationspersonale, 2013
4. Report: Risk of disease following occupational exposure to Polychlorinated Biphenyls, 2013
5. Workplace Bullying, Depression and Cortisol – Maria Gullander, PhD thesis, 2014
6. Is retirement bene�icial for mental and cardiovascular health?Kasper Olesen, PhD thesis, 2014
7. Physical work exposure and sex differences in work-related musculoskeletal disordersThomas Heilskov-Hansen, PhD thesis 2014
heilskov_omslag.indd 1 04/11/14 13.10
