Department of Public Health

University of Helsinki
Musculoskeletal disorders and psychosocial factors at work
Effects of a participatory ergonomics intervention in a
cluster randomized controlled trial
Eija Haukka
ACADEMIC DISSERTATION
To be presented by permission of the Faculty of Medicine of the University of Helsinki,
for public examination in the Auditorium of the Finnish Institute of Occupational Health,
Haartmaninkatu 1 A (3
rd
floor), Helsinki, on Friday 5
th
November 2010, at 12 noon
Publications of Public Health M206:2010
ISSN 0355-7979
ISBN 978-952-10-4867-8 (paperback)
ISBN 978-952-10-4868-5 (pdf)
http://ethesis.helsinki.fi
Helsinki University Print
Helsinki 2010
Author's address: Finnish Institute of Occupational Health
Topeliuksenkatu 41 a A
FI-00250 Helsinki, Finland
E-mail:eija.haukka@ttl.fi
Supervisors: Docent Pivi Leino-Arjas
Finnish Institute of Occupational Health
Helsinki, Finland
Professor (emerita) Hilkka Riihimki
Finnish Institute of Occupational Health
Helsinki, Finland
Reviewers: Professor Clas-Hkan Nygrd
University of Tampere
Tampere, Finland
Docent Simo Taimela
University of Helsinki
Helsinki, Finland
Opponent: Professor Allard van der Beek
EMGO Institute, VU University Medical Centre
Amsterdam, The Netherlands
CONTENTS
ABSTRACT.........................................................................................................................................7
LIST OF ORIGINAL PUBLICATIONS ...........................................................................................10
ABBREVIATIONS............................................................................................................................11
1. INTRODUCTION..........................................................................................................................12
2. REVIEW OF THE LITERATURE................................................................................................14
2.1. Burden of musculoskeletal disorders...........................................................................................14
2.2. Risk factors for musculoskeletal disorders..................................................................................16
2.2.1. Physical workload...............................................................................................................18
2.2.2. Psychosocial factors at work...............................................................................................21
2.2.3. Health-related lifestyle........................................................................................................22
2.2.4. Individual factors ................................................................................................................24
2.3. Ergonomics intervention studies .................................................................................................28
2.3.1. Study designs ......................................................................................................................28
2.3.2. Methodological issues.........................................................................................................30
2.3.3. Challenges encountered in workplace intervention studies................................................32
2.3.4. The participatory approach .................................................................................................33
2.4. Evidence of the effectiveness of ergonomics interventions in preventing
musculoskeletal disorders...................................................................................................................35
2.4.1. Overview.............................................................................................................................35
2.4.2. Behavioural and physical interventions..............................................................................36
2.4.3. Organizational interventions...............................................................................................37
2.4.4. Participatory ergonomics interventions ..............................................................................38
2.5. Kitchen work as a target of an ergonomics intervention.............................................................50
3. THEORETICAL FRAMEWORK OF THE STUDY....................................................................52
4. AIMS OF THE STUDY.................................................................................................................54
5. MATERIALS AND METHODS ...................................................................................................55
5.1. General description of the study..................................................................................................55
5.2. Study designs and subjects ..........................................................................................................56
5.3. Sample size, randomization and blinding....................................................................................57
5.4. Data collection, worker turnover and loss to follow-up..............................................................58
5.5. The participatory ergonomics intervention .................................................................................59
5.6. Measures......................................................................................................................................62
5.6.1. Musculoskeletal disorders and multiple-site musculoskeletal pain....................................62
5.6.2. Psychosocial factors at work and mental stress ..................................................................63
5.6.3. Perceived physical workload ..............................................................................................64
5.6.4. Covariates ...........................................................................................................................64
5.7. Statistical analyses.......................................................................................................................66
5.7.1. Effects of the participatory ergonomics intervention on musculoskeletal
disorders and psychosocial factors at work ..................................................................................66
5.7.2. Multiple-site musculoskeletal pain and reciprocal associations with psychosocial
factors at work ..............................................................................................................................67
6. RESULTS.......................................................................................................................................68
6.1. Effects of the participatory ergonomics intervention in the prevention of
musculoskeletal disorders (I)..............................................................................................................68
6.1.1. Baseline characteristics.......................................................................................................68
6.1.2. Changes in ergonomics.......................................................................................................70
6.1.3. Effects on health outcomes .................................................................................................71
6.1.4. Effects on perceived physical workload.............................................................................74
6.2. Effects of the participatory ergonomics intervention on psychosocial factors at work
(II).......................................................................................................................................................74
6.3. Occurrence of multiple-site musculoskeletal pain (III)...............................................................77
6.4. Reciprocal associations of multiple-site musculoskeletal pain with psychosocial
factors at work (IV) ............................................................................................................................79
6.4.1. Time-lagged associations....................................................................................................79
6.4.2. Trajectories of multiple-site musculoskeletal pain and psychosocial factors at
work ..............................................................................................................................................80
6.4.3. Associations of psychosocial factors at baseline with multiple-site
musculoskeletal pain trajectories ..................................................................................................82
6.4.4. Association of multiple-site musculoskeletal pain at baseline with psychosocial
work factor trajectories .................................................................................................................83
6.4.5. Psychosocial work factor trajectories in relation to multiple-site pain trajectories............84
7. DISCUSSION.................................................................................................................................85
7.1. Main findings and comparison to earlier studies.........................................................................85
7.1.1. Effects of the participatory ergonomics intervention on musculoskeletal
disorders and psychosocial factors at work in the CRT................................................................85
7.1.2. The occurrence of multiple-site musculoskeletal pain .......................................................90
7.1.3. Association of multiple-site musculoskeletal pain with psychosocial factors at
work and mental stress over time .................................................................................................92
7.2. Methodological aspects ...............................................................................................................96
7.2.1. Validity of the CRT ............................................................................................................96
7.2.2. Clinical importance and generalization of the CRT findings .............................................99
7.2.3. Methodological aspects in studies of multiple-site musculoskeletal pain........................101
7.2.4. Self-reported outcome measures.......................................................................................102
7.3. Study findings in relation to the theoretical framework of the study........................................104
8. CONCLUSIONS..........................................................................................................................106
9. IMPLICATIONS OF THE RESULTS.........................................................................................106
ACKNOWLEDGEMENTS .............................................................................................................108
REFERENCES.................................................................................................................................110
7
ABSTRACT
The primary aim of this study was to examine the efficacy of a participatory ergonomics
(PE) intervention in preventing musculoskeletal disorders (MSDs). Secondly, the effects
of the intervention on psychosocial factors at work as intermediate outcomes were
examined. Thirdly, the occurrence of multiple-site musculoskeletal pain (MSP), as well as
its associations with psychosocial factors at work and mental stress, was studied.
A cluster randomized controlled trial of 504 workers of 119 municipal kitchens in four
large cities in Finland was conducted during the years 2002-2005. The kitchens were
randomized to an intervention (n = 59) and control (n = 60) group. The intervention lasted
11 to 14 months. In eight 3-5 hour workshops, the workers were guided to identify
strenuous work tasks and to seek solutions for decreasing physical and mental workload,
as assisted by a researcher trained in ergonomics. The main outcome measures were the
occurrence of and trouble caused by musculoskeletal pain in seven anatomical sites (neck,
shoulders, forearms/hands, low back, hips, knees, and ankles/feet), localized
musculoskeletal fatigue after the working day, and sick leaves due to musculoskeletal
complaints. The following psychosocial factors were studied as intermediate outcomes:
mental stress, mental strenuousness of work, hurry, job satisfaction, job control, skill
discretion, co-worker relationships, and supervisor support. Data were collected by
questionnaire at baseline and thereafter every three months during the intervention and the
12-month post-intervention follow-up. A total of 402 changes related to ergonomics were
implemented. In the control group, 80 such changes were spontaneously implemented as a
part of normal activity. The intervention did not reduce perceived physical workload and
no systematic differences in any health outcomes were found between the intervention and
control groups during the intervention or during the 12-month post-intervention follow-up.
The effects on the psychosocial factors at work were adverse. At the end of the
intervention, the workers in the intervention group reported more job dissatisfaction,
greater mental stress and poor co-worker relationships compared to the control group. The
adverse overall effects were mainly due to a joint effect of the intervention and
unconnected organizational reforms taking place in the foodservice in two of the
participating cities. The results suggest that intervention as implemented in the present
8
trial is not useful in reducing perceived physical workload, improving unsatisfactory
psychosocial working conditions or preventing MSDs. If organizational reforms are
expected to occur, the introduction of other concomitant workplace interventions should
be avoided. There is a need for further high quality trials to elucidate the effectiveness of
PE interventions.
The co-occurrence of pain in seven anatomical sites was studied in a cross-sectional study
(n= 495 women) at baseline, before the intervention started. About 73% of the women
were experiencing pain in at least two, 36% in four or more, and 10% in six to seven sites.
Thus, MSP was more common than single-site pain (14%) or no pain at all (13%).The
seven pain symptoms occurred in 83 different combinations, in addition to those with no
pain and with pain in one site only. When the co-occurrence of pain was studied in three
larger anatomical areas (neck and low back, upper limbs, lower limbs), concurrent pain in
all three areas was the most common combination (36%). Concurrent pain in all three
areas increased successively with age from 22% in the youngest group ( 40 years) up to
49% in the oldest group (51 years).
Associations of psychosocial factors at work and mental stress with MSP were studied in a
two-year longitudinal study (n= 385 women). Questionnaire data which were collected at
three-month intervals in the intervention study were used. Since no effects of the
intervention were found on MSDs, the occurrence of MSP, or perceived physical
workload, the randomized design was relaxed. The 3-month prevalence of MSP (defined
as pain at 3 of seven sites) varied between 50% and 61% during the two-year follow-up.
In time-lagged generalized estimating equations, psychosocial factors at work predicted
MSP being reported three months later, and vice versa, MSP predicted adverse
psychosocial factors at work. Developmental patterns of MSP and psychosocial factors
were identified by trajectory analysis. Four trajectories of MSP prevalence emerged: Low,
Descending, Ascending, and High. A two trajectory model- Ascending or High vs. Low -
yielded the best fit for the psychosocial factors. Poor co-worker relationships, mental
stress and hurry at baseline predicted belonging to the High MSP trajectory. MSP at
baseline predicted belonging to the trajectories of Ascending low job control and mental
stress. Adverse changes in most psychosocial factors were associated with belonging to
the High and Ascending MSP trajectories. The reciprocality of these relationships implies
9
either the existence of two mutually dependent processes, or some shared common
underlying factor(s).
In the daily work of occupational and healthcare professionals, in epidemiological
research, and in connection with further interventions, the evaluation and prevention of
pain at multiple anatomical sites might be useful in addition to considering single-site
pain. Since MSP and psychosocial factors at work seem to be strongly linked together, the
assessment of both parameters in parallel is recommended.
10
LIST OF ORIGINAL PUBLICATIONS
This dissertation is based on the following original articles referred to in the text by their
Roman numerals:
I Haukka E, Leino-Arjas P, Viikari-Juntura E, Takala E-P, Malmivaara A, Hopsu L,
Mutanen P, Ketola R, Virtanen T, Pehkonen I, Holtari-Leino M, Nyknen J, Stenholm
S, Nykyri E, Riihimki H. A randomized controlled trial on whether a participatory
ergonomics intervention could prevent musculoskeletal disorders. Occup Environ
Med 2008;65:849-56. Originally published online 16 Apr 2008.
doi:10.1136/oem.2007.034579.
II Haukka E, Pehkonen I, Leino-Arjas P, Viikari-Juntura E, Takala E-P, Malmivaara A,
Hopsu L, Mutanen P, Ketola R, Virtanen T, Holtari-Leino M, Nyknen J, Stenholm S,
Ojajrvi A, Riihimki H. Effect of a participatory ergonomics intervention on
psychosocial factors at work in a randomized controlled trial. Occup Environ Med
2010;67:170-177. Originally published online 7 September 2009.
doi:10.1136/oem.2008.043786.
III Haukka E, Leino-Arjas P, Solovieva S, Ranta R, Viikari-Juntura E, Riihimki H. Co-
occurrence of musculoskeletal pain among female kitchen workers. Int Arch Occup
Environ Health 2006;80:141-8.
IV Haukka E, Leino-Arjas P, Ojajrvi A, Takala E-P, Viikari-Juntura E, Riihimki H.
Mental stress and psychosocial factors at work in relation to multiple-site
musculoskeletal pain: a longitudinal study of kitchen workers. Eur J Pain (accepted,
9 September 2010).
The papers are reprinted with the permission of the original publishers.
11
ABBREVIATIONS
BL Baseline
BMI Body mass index
CI Confidence interval
CRT Cluster randomized controlled trial
CTS Carpal tunnel syndrome
CWP Chronic widespread pain
FIOH Finnish Institute of Occupational Health
I Intervention phase
ICC Intraclass correlation
LBP Low back pain
MSD Musculoskeletal disorder
MSP Multiple-site musculoskeletal pain
OA Osteoarthritis
OR Odds ratio
Org - No organizational reform
Org + Organizational reform
PE Participatory ergonomics
PI Post-intervention follow-up phase
PIA Post-intervention assessment (at the end of the intervention)
PI
12
The 12-month post-intervention follow-up
PR Prevalence rate
RCT Randomized controlled trial
RR Relative risk
12
1. INTRODUCTION
Musculoskeletal disorders (MSDs) are an important and costly occupational health
problem with consequences for workers, employers and society. About 40 million workers
are affected by work-related MSDs. Almost a quarter of the European workforce report
that they have experienced muscular pain in their neck, shoulders and upper limbs, and
about one in every three suffers from low back pain (LBP). Within the European Union,
MSDs are the single most common cause of sickness absence from work, early retirement,
and disability payments. It is estimated that the direct annual costs of the MSDs account
for 2% of the European gross domestic product (Bevan et al. 2009).
There is a consensus that the etiology of MSDs is multifactorial. There has been a special
emphasis on the effect of physical workload (awkward working postures, forceful
movements, repetitive work, etc.) on MSDs. However, recently the role of psychosocial
factors at work has become a focus of epidemiologic research. Several psychosocial
factors have been found to associate with the occurrence of MSDs, e.g. low job
satisfaction, high job demands, low job control, and low workplace social support.
However, most of the evidence is based on cross-sectional studies. Cross-sectional studies
are valuable if one wishes to elucidate associations between variables, but they do not
usually provide information about the time sequence of the 'explanatory' and 'outcome'
variables. Thus, the need for conducting more longitudinal studies has been emphasized.
In recent years, several studies have observed that multiple-site musculoskeletal pain
(MSP) is even more frequently reported than any single-site pain. It has been suggested
that the basic expectation should be that one regional pain accompanies another,
irrespective of pain site (Croft et al. 2007; Croft 2009). Recently, Swedish researhers in
their population based (n=2329) five-year follow-up study concluded, that having
concurrent low back and neck-shoulder disorders was associated with a higher risk for
sickness absence [odds ratio, (OR) 1.7)] as well as a risk for long-term sickness absence
(OR 2.5) compared to having only low back or only neck-shoulder disorders (Nyman et al.
2007). The majority of studies to date have concentrated on pain and its risk factors at a
13
specific site, most often in the low back, and the risk factors of MSP are not well known.
The need to study the co-occurrence of pain over time and to provide information on why
pain tends to occur at multiple sites has recently been emphasized.
Preventive ergonomics interventions are widely applied in order to deal with MSDs.
However, evidence on which of the interventions are the most effective is lacking. The
participatory ergonomics (PE) approach is commonly recommended when the goal is to
reduce MSDs, and its potential as an effective means to improve unsatisfactory
psychosocial working environment has been emphasized. However, few high quality
studies have been reported and evidence on the effectiveness of PE is scanty. It has been
stated that a robust knowledge on the effectiveness of interventions must be based on
several comparable high quality studies in which the effect of ergonomics interventions on
the prevention or reduction of the MSDs has been rigorously evaluated (Bongers 2009).
With respect to LBP prevention, The European Guidelines Working Group recently
concluded that there is an urgent need for good quality randomized controlled trials (RCT)
to examine the effectiveness of physical, psychosocial and organizational ergonomics
interventions (Burton et al. 2006).
The studies described in this dissertation are part of a larger research programme entitled
'Effectiveness of an ergonomics intervention - a randomized controlled trial' (ERGO-
study) conducted by the Finnish Institute of Occupational Health (FIOH) during the years
2002-2005 in collaboration with municipal kitchens in four large cities in Finland. This
thesis reports the effects of the PE intervention on MSDs and on psychosocial factors at
work and mental stress as intermediate outcomes of the intervention (studies I-II). In
addition, the occurrence of MSP and associations of MSP with psychosocial factors at
work over time are presented (studies III-IV).
14
2. REVIEW OF THE LITERATURE
2.1. Burden of musculoskeletal disorders
MSDs constitute a major public health problem in the industrialized countries. MSDs
cause individual suffering, trouble in daily living, and considerable economic and societal
consequences due to short- and long-term work disability and productivity losses (Norlund
and Waddel 2000; Baldwin 2004; Buckle 2005). Low back pain, neck pain and shoulder
disorders are common among working populations and are the main reasons for work-
related consultations in general practice (Weevers et al. 2005; Taimela et al. 2007).
In general population studies in different countries, the lifetime prevalence of back
disorders has varied between 30% and 84% and that of neck disorders has been about 70%
(Riihimki 2005a). According to representative samples of the Finnish population, chronic
low back and neck disorders have decreased in both genders during the last two decades of
the 20th century (Kaila-Kangas 2007). Still, the lifetime prevalence of back pain was as
high as 76% among women and 77% among men, and that of neck pain 68% among
women and 54% among men. The respective one-month prevalence figures of back pain
were 33% and 28%, of neck pain 37% and 24%, of shoulder pain 23% and 18%, of hip
pain 12% and 8%, and of knee pain 21% and 18%. A recent survey based on self-reported
data from European workers claimed that about 33% of Finnish workers reported work-
related pain in their neck, arms and shoulders. In the descending order of pain reporting,
the Finnish workers ranked as the eighth of all EU member states (Bevan et al. 2009).
According to Finnish statistics in 2007, MSDs were the most common reason for receiving
sickness absence benefits accounting for a total of 35% of all absence periods. There were
almost 126 000 incident sickness allowance periods due to MSDs in 2007. The resulting
benefit expenditures were about 273 million euros making up 36% of all the benefits.
With respect to disability pensions, 24% were due to MSDs. Both the numbers of sickness
benefit days and disability pensions were more common in women (Finnish Centre for
Pensions 2008; The Social Insurance Institution 2008). The direct and indirect costs due to
musculoskeletal pain in Finnish primary health care were about 1 billion euros and when
15
combined with costs due to permanent work disability, the overall losses were estimated to
amount to 3% of the gross domestic product (Mntyselk et al. 2001; Mntyselk et al.
2002). In all 27 EU Member States, MSDs are the main cause of absence from work, and
in some countries 40% of the costs of workers' compensation are attributable to MSDs
(Podniece 2008).
Multiple-site musculoskeletal pain
Concomitant pain in several sites seems to be common. About three quarters of people
report more than one site of pain, with a median count of three out of a possible count of
ten (Croft 2009). The definition for MSP is not well established. The American College of
Rheumatology (ACR) has developed a definition for chronic widespread pain (CWP).
This definition requires the presence of pain that has been present for at least three months
in the axial skeleton, above and below the waist, and on the left and right side of body
(Wolfe et al. 1990). However, pain in multiple body sites that does not meet these
classification criteria may also deserve attention. For example, MSP might have a stronger
impact on mental health, quality of life and healthcare utilization than single-site pain,
irrespective of whether or not the pain experienced by individuals satisfies the criteria for
CWP (Davies et al. 1998; Carnes et al. 2007).
Several studies have reported that MSP is even more frequent than single-site pain both in
the general population (Picavet and Schouten 2003; Walker-Bone et al. 2004; Carnes et al.
2007; Kamaleri et al. 2008c) and in occupational groups (Molano et al. 2001; Yeung et al.
2002; Alexopoulos et al. 2004; Ijzelenberg and Burdorf 2004; Solidaki et al. 2010). In the
Dutch population, single-site pain during the past 12-months was reported by 25% of the
sample, while 29% reported pain in 2-3 sites and 21% in four or more sites (Picavet and
Schouten 2003). Among the Norwegian population, the respective proportions of subjects
with pain during the past seven days were 17%, 27% and 27% (Kamaleri et al. 2008c).
Carnes et al. (2007) found that over three quarters of a British sample who reported pain
suffered from pain stemming from multiple sites.
Having pain at one site has been shown to increase the risk of developing pain at other
sites and the risk seems to increase with the number of pain sites at baseline (Croft et al.
2007; Kamaleri et al. 2009). MSP was associated with decreased level of functional ability
16
among employees of the City of Helsinki (Saastamoinen et al. 2006) and a sample of the
Norwegian general population (Kamaleri et al. 2008c). Among a representative sample of
Finnish adults, the odds ratio of poor perceived physical work ability and the respondent's
own prognosis of poor future work ability increased from two for single-site pain up to
eight for pain at four sites (Miranda et al. 2010). MSP has been shown to predict both
short- and long-term sickness absence (Natvig et al. 2002; Morken et al. 2003; Nyman et
al. 2007) and permanent work disability (Kamaleri et al. 2008b). MSP commonly co-
occurs with other syndromes (Aggarwal et al. 2006; Hagen et al. 2006) and is associated
with poor mental vitality (Walker-Bone et al. 2004; Hagen et al. 2006). MSP is a
persistent phenomenon and it predicts the risk of a current problem for becoming chronic
(Papageorgiou et al. 2002; Andersson 2004; Kamaleri et al. 2009).
MSP is common already of school age and in adolescence (Jones et al. 2003; Adamson et
al. 2007; Larsson and Sund 2007; Auvinen et al. 2009; Paananen et al. 2010). Thus, in
schoolchildren the one-month prevalence of MSP has varied between 8% (Adamson et al.
2007) and 15% (Jones et al. 2003), depending on the criteria used for MSP and the age of
the studied group. The average number of pain sites appears to be settled by age 20 and
little variation seems to occur thereafter (Croft 2009; Kamaleri et al. 2009).
Based on this knowledge, MSP seems to be more disabling and more severe compared to
local pain and it is a special challenge for occupational and health care professionals.
Nonetheless, the majority of studies examining the occurrence and prognostic factors of
musculoskeletal pain have focused on a specific anatomical site (Mallen et al. 2007) and
the risk factors of MSP are inadequately understood.
2.2. Risk factors for musculoskeletal disorders
MSDs are the single largest category of work-related illness, representing a third or more
of all registered occupational diseases in the Nordic countries, the United States and
Japan. It has been estimated that about 40% of all upper limb disorders in the total US
employed population were attributable to occupational exposures (Punnett and Wegman
2004). In global terms, 37% of LBP appear to be caused by occupation, and work-related
17
LBP has been estimated to be responsible for 818,000 disability-adjusted life years being
lost annually (Punnett et al. 2005). Accordingly, risk factors related to work have been
extensively studied. The World Health Organization has defined 'Work-related
musculoskeletal diseases' as diseases that can be partly caused by adverse working
conditions, or which may be aggravated, accelerated or exacerbated by work place
exposures, or such that may impair working capacity (WHO 1985).
Epidemiologic evidence has accumulated that both physical and psychosocial factors at
work and individual factors play a role although the mechanisms involved in the
development of MSDs are not yet completely understood. These factors interact, they may
reinforce each other and their effects may be mediated by cultural or societal factors. All
of them vary over time from one situation to the next, complicating attempts to clarify
their relationships with the development of MSDs (National Research Council 2001;
Bongers et al. 2006; Marras et al. 2009). Psychosocial risk factors for work-related MSDs
can be categorised into those that are specific to the workplace (low social support at
work, job satisfaction, low skill discretion, low job control etc.) and those that are
individual psychosocial or psychological characteristics, such as depression, anxiety and
mental stress (Sauter and Swanson 1996; Bernard 1997; National Research Council 2006).
Non-workplace factors, which contribute to work-related MSDs possibly influencing
individual responses to workplace exposures, are considered as individual factors. Among
these, gender, age, education, marital status, weight, height, overweight/obesity, smoking,
exercise or sports, other hobbies, drug use, personality, and various comorbidities
(Bernard 1997; Burdorf and Sorock 1997; National Research Council 2001; Buckle and
Devereux 2002; Cole and Rivilis 2006).
The following section will contain an overview of the risk factors of MSDs these being
mainly based on existing review articles. The electronic databases Pubmed and Scopus
were searched.
18
2.2.1. Physical workload
Low back disorders
The reviews on the relationship of physical workload with low back disorders by
Hoogendoorn et al. (1999) and Hansson and Westerholm (2001) were based on cohort and
case-control studies. The former covered studies from the period 1949-1997 and the latter
with publications from 1971- 2000. A recent review (Bakker et al. 2009) considered
cohort studies only from years 1997-2007. The reviews indicate that the evidence is strong
for manual materials handling, bending and twisting postures at work, and whole body
vibration as being risk factors for LBP. With respect to patient handling, the evidence is
moderate. Furthermore, the evidence is strong on standing, walking, and sitting at work
for not increasing the risk of LBP. The summary of the review of Lis et al. (2007) claimed
that prolonged merely sitting did not increase the risk of LBP, whereas sitting more than
half of a workday in combination with whole body vibration and/or awkward postures did
increase the risk. However, these conclusions were based mostly on cross-sectional
studies. A review by Chen et al. (2009), based on 10 prospective cohort and 5 case-control
studies, confirmed that sedentary life style (at work and leisure time) was not a risk factor
for LBP.
Neck and upper limb disorders
The Bone and Joint Decade Task Force has recently summarized the determinants of neck
pain among the working (Ct et al. 2008) and the general populations (Hogg-Johnson et
al. 2008). Both reviews examined the literature from 1980 to 2006. These reviews are the
first in which a sufficient number of cohort studies were available to allow reliable
conclusions to be drawn. The best evidence synthesis among the working population was
based on 19 cohort studies and 1 RCT. The following risk factors of neck pain were
identified: prolonged sitting, repetitive and precision work, prolonged neck flexion, elbow
and shoulder posture while working at a computer, inadequate mouse and keyboard
position, and use of telephone shoulder rests. Preliminary evidence was found that
awkward work postures, poor physical work environment, and exposure to glare might be
associated with neck pain.
19
Some reviews have concluded that repetitive work and static work with arms abducted or
elevated at more than 60 degrees are associated with shoulder disorders (Bernard 1997;
Hansson and Westerholm 2001; Walker-Bone and Cooper 2005). Potential risk factors
include also heavy workload and vibration (van der Windt et al. 2000). In a survey with a
20-year follow-up among the Finnish general population, work exposure to repetitive
movements at baseline increased the risk of incident shoulder disorder with an OR of 2.3
and exposure to vibration with an OR of 2.5. These adverse effects were seen even among
those older than 75 years at follow-up (Miranda et al. 2008a). Hagberg (2002) reviewed
experimental and epidemiological studies and concluded that the scientific evidence is
weak for vibration per se as a risk factor for MSDs, although job tasks with vibrating tools
are associated with MSDs. In a two-year prospective study among newly employed
workers (Harkness et al. 2003), lifting heavy weights with one or two hands, carrying on
one shoulder, lifting at or above shoulder level and pushing and pulling were predictive
factors of new onset shoulder pain.
Two recent reviews have summarised the literature on risk factors related to the carpal
tunnel syndrome (CTS). The occurrence of CTS was associated with high levels of hand-
arm vibration, prolonged work with a flexed or extended wrist, high requirements of hand
force (> 4 kg), high repetitiveness (cycle time < 10 seconds, or performing the same
movements for > 50% of cycle time), and their combination (van Rijn et al. 2009).
Previously Palmer at al. (2007a) had found evidence that regular and prolonged use of
hand-held vibratory tools doubled the risk of CTS. There was also substantial evidence
found for similar or even higher risks from prolonged repetitive flexion or extension of the
wrist, especially in combination with a forceful grip. Repetitive movements, use of hand
force, non-neutral wrist postures, and their combinations, all seem to be associated with
epicondylitis and hand/wrist tendinitis (Bernard 1997; Hansson and Westerholm 2001).
However, most studies on occupational risks of tenosynovitis and epicondylitis have been
crticized since they have been based on job titles rather than specific physical activities at
work (Palmer et al. 2007b).
20
Lower limb disorders
According to D'Souza et al. (2005), few reports exist on occupational factors and disorders
of the lower limbs other than osteoarthritis (OA), but it appears that standing and heavy
physical workload may be associated with ankle and foot pain and plantar fascitis. A
history of occupational activities involving prolonged standing may be associated with
chronic plantar heel pain (Irving et al. 2006). Heavy physical work demands have
increased the prevalence of knee complaints in repeated surveys among male employees in
the Netherlands (deZwart et al. 1997). However, in a prospective study of forestry
workers, no work-related factors predicted incident knee pain (Miranda et al. 2002).
Moderate to strong evidence exists that high physical workload is a risk factor of hip and
knee OA (Bierma-Zeinstra and Koes 2007). Recently Jensen (2008b; a) made a best
evidence synthesis evaluating the associations between hip and knee OA with more
detailes about physical work demands. With respect to hip OA, moderate to strong
evidence was found for heavy lifting, but the burdens have to be at least 10-20 kg and the
duration of exposure at least 10-20 years to clearly increase the risk of hip OA. For
example, for farmers the risk seems to double after about 10 years of farming. With regard
to knee OA, evidence was moderate for a relationship of kneeling and heavy lifting and
moderate for the combination of kneeling/squatting and heavy lifting. Insufficient or
limited evidence was found for climbing stairs or ladders as causes of either hip or knee
OA. There were limitations in some of the studies, e.g. small sample size, poor description
of exposure, and the use of differing diagnostic criteria.
Multiple-site pain
Knowledge on risk factors for MSP is based on single studies since no systematic reviews
are available. In dentists, a high perceived physical workload was associated with the co-
occurrence of musculoskeletal pain at different sites (Alexopoulos et al. 2004). In a study
of nurses, office workers, and postal clerks (Solidaki et al. 2010) the association was
examined with the number of pain sites and a physical demands score which was based on
heavy lifting, working with hands above the shoulder level, repeated bending of the elbow,
repeated hand-arm movements, and kneeling, squatting or climbing stairs. In the
multivariate analysis, the score had a strong graded relationship with the number of pain
sites. Lifting of loads (>15 lbs) with one hand or with two hands (> 24 lbs), pulling (> 56
21
lbs), prolonged squatting, and prolonged working with hands at or above shoulder level
were associated with the onset of widespread pain in a two-year prospective study among
newly employed workers (Harkness et al. 2004). In a three-year follow-up study among a
sample of adults in Manchester, Great Britain, pushing/pulling heavy weights, repetitive
wrist movements, and kneeling were associated with the onset of CWP (McBeth et al.
2003). Repetitive movements of the arm or wrists predicted [relative risks (RR) 4.1 and
3.4] future episodes of forearm pain that commonly co-occurred with other regional pain
(Macfarlane et al. 2000).
2.2.2. Psychosocial factors at work
Several reviews on associations of psychosocial factors at work with MSDs have been
published during the past decade. Many psychosocial factors at work have been found to
be associated with the occurrence of MSDs, such as rapid work pace, monotonous work,
low job satisfaction, low workplace social support, high job demands, low job control,
work stress, non-work-related stress, and high and low skill discretion (Davis and Heaney
2000; Hoogendoorn et al. 2000; Arins et al. 2001; Linton 2001; Bongers et al. 2002;
Bongers et al. 2006). However, the conclusions of the reviews tend to be inconsistent with
most of the evidence being based on cross-sectional studies. The review by Hartvigsen et
al. (2004) only considered prospective studies and concluded that there was moderate
evidence of no association between LBP with perception of work, organizational aspects,
and social support at work; with respect to stress at work, the evidence was insufficient.
Macfarlane et al. (2009) evaluated the evidence on associations between psychosocial
factors at work and MSDs based on published reviews. With regard to back pain, the
conclusions were the most consistent for an association with high job demands, low job
satisfaction and low work support. For neck/shoulder pain, consistency in conclusions was
found both regarding high work demands (four reviews out of six reported an association)
and for low job demands (two out of three reviews found an association). For lower limb
disorders, there appears to be only a single review regarding knee pain. Subsequently, a
review of studies among the working population (Ct et al. 2008) concluded that high job
demands, low social support at work and job insecurity are risk factors for neck pain.
Preliminary evidence was found for poor job satisfaction, stress at work and frequently
22
experiencing technical problems with a computer. van Rijn et al. (2009) found no
association between any psychosocial risk factors and CTS.
Multiple-site pain
In the review of Mallen et al. (2007), low social support was one of the 11 generic
prognostic factors associated with at least two regional pain complaints. In a population
based study, dissatisfaction with support from colleagues or supervisors was observed to
increase the risk (RR 4.7) of future episodes of forearm pain that commonly co-occurred
with other regional pain (Macfarlane et al. 2000). Poor social relationships at work and
poor job control predicted the increase of pain and clinical findings in the neck and upper
limbs, low back, and lower limbs in a 10-year follow-up of industrial workers (Leino and
Hnninen 1995). Adverse work-related psychosocial factors and high levels of individual
psychological distress predicted the onset of musculoskeletal pain at a one-year follow-up
with similar effects across the anatomical sites (Nahit et al. 2003). In a two-year
prospective study among newly employed workers, those who reported low job
satisfaction, low social support and monotonous work had an increased risk of new-onset
widespread pain (Harkness et al. 2004).
Leino and Magni (1993) conducted a 10-year follow-up study among employees in the
engineering industry and found that symptoms related to mental stress predicted an
increase in a sum score of musculoskeletal pain at several sites, and that the pain sum
score predicted an increase in stress symptoms.
2.2.3. Health-related lifestyle
Shiri at al. recently published two meta-analyses on the relationships of overweight,
obesity and smoking with LBP (Shiri et al. 2010a; b). Overweight and obesity increased
the prevalence of LBP. The associations were strongest for seeking care for LBP (pooled
OR 1.6) and chronic LBP (OR 1.4). Obesity also increased the incidence of LBP (OR 1.5).
The findings pointed to stronger associations in women than in men. Both current and
former smokers were observed to have a higher prevalence and incidence of LBP than
never smokers, with the associations being strongest for disabling (OR 2.1) and chronic
(OR 1.9) LBP. Male smokers were at a higher risk of LBP than female smokers. In their
23
previous systematic review, Shiri at al. (2007) concluded that overweight, long smoking
history, high physical activity and a high serum C-reactive protein level were associated
with lumbar radicular pain and sciatica. In her review Kauppila (2009) found that smoking
and high serum cholesterol levels were most consistently associated with disc
degeneration and LBP. Miranda et al. (2008b) studied the one-year incidence of LBP
according to age group in a Finnish industrial population. Health behaviour, defined as a
sum score of overweight, smoking and lack of physical activity, increased the risk of LBP
only among those 50 years or older (RR up to 2.8), whereas physical workload predicted
LBP among those younger than that age.
Two recent systematic reviews published by Hogg-Johnson et al. (2008) and Ct et al.
(2008) evaluated the determinants of neck pain. The former found evidence that exposure
to passive smoking in adolescents increased the risk of neck pain. The latter review
summarized that smoking increased the risk of neck pain among the working population.
A review byViikari-Juntura et al. (2008) reported some associations between body weight
(weight, BMI) and shoulder disorders. Smoking was associated with shoulder disorders
but only in studies examining occupational populations. Obesity is reported to be a risk
factor for CTS (Walker-Bone et al. 2003; Hooper 2006) and hip and knee OA (Hooper
2006; Wearing et al. 2006; Bierma-Zeinstra and Koes 2007). In the systematic review of
Irwing et al. (2006) increased BMI was associated with chronic plantar heel pain in a non-
athletic population. Obesity may also have a profound effect on soft-tissue structures, such
as tendons, fasciae, and the cartilage, in addition to being a risk factor for bone and joints
disorders (Wearing et al. 2006).
There is moderate to strong evidence that high-intensity sporting activities are risk factors
for hip OA (Bierma-Zeinstra and Koes 2007), but leisure time sports or exercise were not
found to be risk factors of LBP (Hoogendoorn et al. 1999; Bakker et al. 2009). A Dutch
research group (Hamberg-van Reenen et al. 2007) carried out the first review of the
relationship between physical capacity and future LBP and neck/shoulder pain based on
longitudinal studies. They found strong evidence that there is no relationship between
trunk muscle endurance and the risk of LBP and inconclusive evidence for a relationship
between trunk muscle strength and mobility of the lumbar spine and the risk of LBP.
Evidence was inconclusive also for the associations between physical capacity measures
24
with the risk of neck/shoulder pain. Instead, Ct et al. (2008) claimed that low to
moderate physical capacity of neck and shoulder musculature was associated with an
increased risk of neck pain.
Multiple-site pain
Current smoking was associated with an increased risk of chronic pain in multiple
locations and with CWP in both genders among the general rural population (Andersson et
al. 1998). Smokers, overweight subjects and those with low physical activity have
reported a greater number of pain sites (Walker-Bone et al. 2004; Kamaleri et al. 2008a).
Jones at al. (2003) stated that high levels of sports activity predicted the onset of
widespread pain among schoolchildren (RR 1.9). Finnish researchers have studied MSP
among adolescents (Paananen et al. 2010). They found that a high physical activity level,
long sitting time, short sleeping time, and smoking were associated with MSP in both boys
and girls though the associations were stronger in girls. In addition, MSP was associated
with overweight in girls.
2.2.4. Individual factors
Women tend to report more MSDs than men. The gender difference seems to be more
distinct for neck and upper limb disorders (Hooftman et al. 2004; Strazdins and Bammer
2004). Strazdins and Bammer (2004) examined why employed women are much more
likely than men to experience upper body disorders. The gender differences were
explained by risk factors at work (repetitive work, poor ergonomic equipment) and at
home (less opportunity to relax and exercise outside of work). Parenthood exacerbated the
gender difference with mothers reporting having the least time to relax or exercise. The
gender segregation of women into sedentary, repetitive and routine work and the
persisting gender imbalance in domestic work were concluded to be interlinking factors
that accounted for the gender differences in MSDs.
A Danish research group has recently published two large population based studies of
twins aged 20-71 years (Hartvigsen et al. 2009; Leboeuf-Yde et al. 2009). A cross-
sectional survey examined 34 902 twins (Leboeuf-Yde et al. 2009) and it showed that
women were more likely to report MSDs and to have had pain for longer periods than
25
men. Neck pain and LBP was common already at the age of 20, with the prevalence
increasing slowly with age until reaching a peak around middle age. Similar patterns were
noted for radiating pain. Pain was reported to be more long lasting in the older groups. In
the analyses of 15 328 twins, genetic susceptibility explained about 38% of lumbar, 32%
of thoracic, and 39% of neck pain (Hartvigsen et al. 2009).
Other twin studies have suggested that both disc degeneration and back pain contain a
genetic component. A variety of definitions of back pain problems have been used in
genetic studies that could influence responses and heritability estimates since these have
varied widely in the scientific literature, explaining from 0% to 57% of the variance in
back pain reporting (Battie et al. 2007). A recent narrative review by Kalichman and
Hunter (2008) reviewed knowledge of heritability of intervertabral disc degeneration. The
review covered the literature from 1950 to 2007. Familial predisposition to disc
degeneration was observed to be high. Heritability estimates ranged from 34% to 61% in
different spine locations. Fejer et al. (2006) examined a large population based group of
twins (n= 33 794) and noted that genes play a significant role in neck pain, particularly in
women. The overall genetic effect of lifetime neck pain was observed to be 44%. There
was a statistically significant difference in heritability between males and females (34%
vs. 52%). However, the genetic effect became gradually less important with increasing
age. In other words, environmental factors dominate almost completely in the older age
groups. Hartvigsen et al. (2005) reported that genetic factors do not play an important role
in the liability to neck pain in individuals 70 years of age or older. Previous
musculoskeletal pain is a risk factor for neck pain and ethnicity is possibly associated with
neck pain (Ct et al. 2008).
Multiple-site pain
MSP generally increases with age (Bingefors and Isacson 2004). Women have been
shown to report a greater number of pain sites than men (Picavet and Schouten 2003;
Walker-Bone et al. 2004; Kamaleri et al. 2008a) and this trend has been seen already in
adolescents. Paananen et al. (2010) reported that 40% of girls and 23% of boys had
experienced MSP during the past six months. Auvinen et al. (2009) found that the
prevalence of MSP during the past six months increased from 15% in girls and 9% in boys
at the age of 16 years up to 27% and 15%, respectively, at the age of 18 years. Kamaleri et
26
al. (2009) observed a relative stable pattern of reporting the number of pain sites across
adulthood. About 46% of the participants reported the same number of pain sites 14 years
later, plus or minus one site. The average number of pain sites appears to be set by the age
of 20. Little variation in the stability was seen by age or birth cohort, except for a small
decline occurring between 60 and 74 years (Croft 2009; Kamaleri et al. 2009).
Genetic factors may be associated with reporting pain at multiple sites (Zubieta et al.
2003). In the Danish twin study (Hartvigsen et al. 2009), patterns of pain occurrence in
three spinal regions (cervical, thoracic, lumbar) were examined. It was found that the
heritability of having pain in all three areas was 35%, i.e. the highest of all combinations.
Mikkelsson et al. (2001) examined the pattern of familial aggregation of widespread pain
in 11-year-old Finnish twins. In girls 56% and in boys 35% of the variation in liability was
accounted for by familial factors.
In summary, there is a considerable body of epidemiologic literature which has examined
associations of physical workload with MSDs, particularly with back disorders. In
addition, associations of psychosocial and individual factors with MSDs have been
studied. However, prospective studies are in the minority and evidence is largely based on
cross-sectional or case-control studies. This is particularly true of neck and upper limb
disorders. There are few studies on lower limb disorders these being mainly concerned
with hip and knee OA. Most of the existing evidence on associations of psychosocial
factors at work with MSDs is based on cross-sectional studies. Although MSP is more
common than single-site pain, the studies have typically examined separately at pain in
different single anatomical sites and thus the risk factors for MSP remain largely
unknown. However, it seems that single-site pain and MSP may share many predictive
factors. A summary of findings on associations of physical and psychosocial risk factors
with the various MSDs based on the literature is shown in Table 1.
Available evidence of risk factors suggests that workplace interventions aiming at
reducing physical workload and improving psychosocial working conditions could have
potential for the prevention of MSDs. In addition, interventions targeted to improve health
behaviour (avoiding overweight/obesity and smoking) may be beneficial. There is varying
evidence about the role of leisure time physical activity, and the role of physical capacity
27
as a protective factor for MSDs is unclear. More and more research addressing genetic
factors in relation to MSDs has been conducted over the past decades, and evidence has
accumulated that there are major genetic influences involved in the development of
MSDs. Nonetheless, little is known about the roles of individual genes or the pathways
and mechanisms through which they influence pain problems (Batti et al. 2007).
Table 1. Associations of work-related physical and psychosocial risk factors with MSDs, as
based on literature. Osteoarthritis (OA), multiple-site pain (MSP)
*Both physical and psychosocial risk factors for MSP were based on single prospective studies.
Based on reviews: low back disorders (cohort and case-control studies), neck disorders (cohort
studies, 1 RCT), upper limb disorders (mostly cross-sectional studies), hip and knee OA (mostly
case-control and cohort studies).

Based on two reviews: one comprised of cohort studies and one

RCT, the other was an overview of the reviews (most studies were cross-sectional).

Low
back
Neck Upper
limbs
Hip
OA
Knee
OA
MSP*
Physical workloaa
f


Manual materials handling
Bending and twisting/awkward postures
Whole body vibration
Patient handling/nursing
Prolonged sedentary work position
Repetitive work
Precision work
Prolonged neck Ilexion
Upper extremity posture
Mouse/ keyboard position
Static work/static work with arms abducted or/
elevated more than 60 degrees

Heavy workload
Vibration
Repetitive movements
Non-neutral wrist postures
Use oI hand Iorce
Combination oI repetitive movements, use oI
hand Iorce, non-neutral wrist postures

LiIting
Kneeling
Combination oI kneeling/squatting, heavy liIting
Pushing/pulling
Prolonged squatting
Prolonged working with hands at or above
shoulder level


Psychosocial factors at work

High job demands
Low job demands (monotonous job, insuIIicient
use oI skills)

Low job satisIaction
Low social work support
Low job control
Poor social relationships at work
Stress at work
Adverse work-related psychosocial Iactors
Job insecurity
28
2.3. Ergonomics intervention studies
In order to control and deal with the burden of MSDs, a variety of preventive ergonomics
interventions have been conducted. Studies with rigorous designs and methods are needed
in order to determine which of these are the most (cost) effective in reducing MSDs.
Causal inference is often challenging in intervention research. If changes are observed, can
these be attributed to the intervention? The ideal type of study in establishing causality in
terms of internal validity is a RCT. The results of RCTs are considered as the 'gold
standard' and the basis of 'evidence-based practice' in intervention research. Quasi-
experimental studies are frequently used, when random assignment is not possible or
feasible (Rothman and Greenland 1998; Grimshaw et al. 2000; Franceshi and Plummer
2005).
2.3.1. Study designs
Randomized controlled trials
In RCTs, individuals, and in cluster randomized trials (CRTs), groups of people or intact
social units (e.g. workplaces, schools, medical practices, cities), are randomly allocated to
an intervention or a control group. In CRTs, outcomes are measured on the individuals
within clusters. CRTs are used to evaluate group interventions and individual interventions
where group level effects are relevant. CRTs are also considered when the intervention is
given to individuals but might affect others within that cluster. In the evaluation of
complex interventions in which several factors may act interdependently at various levels
of an organization and have an effect on many processes and outcomes, cluster designs
may be useful (Donner and Klar 2000; Atienza and King 2002; Medical Research Council
2002).
The aim of randomization is to avoid confounding and selection bias by creating
comparable groups with respect to any known or unknown potential confounding factors.
A control group is needed to differentiate between change and effect. The prospective
design with at least one measurement before and at least one after the intervention makes
it possible to study changes over time. The effect is estimated by comparing what
happened in the intervention group with what happened to those without the intervention.
29
Blinding and use of placebo treatment are other methodological advantages offered by
RCTs. The purpose of blinding is to eliminate bias resulting from the expectations of
participants or researchers with respect to outcomes. In single-blind trials, it is either the
participants or the researchers who are unaware of the results of the randomized
assignment. In double-blind trials, neither the participants nor the researchers or those
administering the interventions and assessing the outcomes (care providers, outcome
assessors) know who belongs to the control and the intervention group. Placebo treatment
is important to distinguish the effect of a specific intervention from the effect of a sham
intervention. Unfortunately, blinding and use of the placebo are commonly impossible in
ergonomics and other workplace interventions (Shannon et al. 1999; Jni et al. 2001;
Buring 2002; Green 2002; Kristensen 2005).
When compared with an individually randomized trial, the CRT is more complex to
design and it requires more participants to obtain equivalent statistical power. The power
of a CRT depends more on the number of groups randomized rather than on their size. The
number of groups randomized should be large enough in order to yield similar
distributions of baseline characteristics among the groups. Since participants within a
cluster are more likely to have similar outcomes, the outcomes are not completely
independent. This clustering effect and variation within and between clusters must be
taken into account in the statistical analyses (Rothman and Greenland 1998; Campbell et
al. 2000; Donner and Klar 2000; Atienza and King 2002; Green 2002; Medical Research
Council 2002; Campbell et al. 2004). Reporting of CRTs requires additional consideration,
and an extended CONSORT statement including a checklist of items that should be
included in the trial report, has been published (Campbell et al. 2004).
Quasi-experimental designs
In controlled before-after studies (pre-post / post-only non-equivalent control group
design), the groups are assigned to intervention and comparison groups without random
assignment. It is important that there should be minimal differences between the
intervention and control group at baseline regarding the outcome measures, confounding
factors and other variables thought to influence the estimation of the effect of the
intervention. The major problem with such a design is that the groups might not be the
similar before the intervention. Uncontrolled before-after studies (the pre-post/ post-only
30
one-group designs) with no random assignment and no comparison group are the weakest
designs. In these studies, it is difficult or impossible to attribute changes in the outcome to
the intervention (Shannon et al. 1999; Grimshaw et al. 2000).
Study design quality has been observed to be inversely associated with the reported
magnitude of effects; it seems that the more rigorous the study design, the weaker the
effects which can be found (Volinn 1999; Grimshaw et al. 2000; Neumann et al. 2010).
The effects based on uncontrolled before-after studies have been larger than those based
on controlled before-after studies. In randomized studies, the effects have been smaller
than in studies using quasi-experimental designs.
2.3.2. Methodological issues
Internal and external validity in randomized trials
In RCTs, internal validity is threatened by several sources of bias. Selection bias may
occur if there is an error in the allocation of individuals or groups to the intervention and
control groups. The researchers may have conscious or attitudinal preferences with respect
to the allocation, the consequences of which are avoided if the allocation sequence can be
generated by using a computer algorithm. The allocation sequence must be concealed
from the researchers enrolling the participants. Dropouts may lead to increasingly different
groups (attrition bias) if those who drop out are systematically different from those who
remain in the study. Unequal additional interventions distinct from the intervention under
evaluation (performance bias), and an influence by a knowledge of the subjects'
assignment on the assessment of outcome (detection bias) are other threats to internal
validity in randomized designs (Jni et al. 2001).
There are further potential influences that can diminish the effect of an intervention. The
subjects in the control group may also receive the intervention (contamination), e.g. when
participants move from one group to another during the intervention. Non-compliance
occurs when subjects in the intervention group do not follow the intervention protocol, i.e.
they do not 'take the pill' that they are supposed to take. The 'Hawthorne effect' refers to
the possibility that the behaviour of the subjects in the intervention group may alter simply
31
because they consider that they are under observation (Shannon et al. 1999; Jni et al.
2001; Buring 2002; Green 2002; Kristensen 2005).
External validity refers to the generalisability of the results. Causal inference based on a
study has external validity if it can be transferred from the study's unique settings to other
situations: other subjects, populations or workplaces. Information pertinent to the
assessment of external validity includes a detailed description of participants (age, gender,
occupation, work experience etc). The manner of recruitment (volunteers vs. all workers)
is relevant, since results obtained from volunteers are often not generalizable to the wider
population. The number of dropouts and the differences between them and non-dropouts
should be reported. Details of the intervention process and content need to be explicitly
described to enable the assessment of external validity. Contextual factors should be
considered as they can have an influence on the effectiveness of an intervention. As an
example, an ergonomics intervention might have a larger impact on a workplace where
ergonomics has not been considered compared to the situation in a workplace where
attention has already been paid to ergonomics (Goldenhar and Schulte 1996; Shannon et
al. 1999; Jni et al. 2001).
Study quality
In the literature, various proposals for improvement of study quality will be put forward.
Intervention studies should preferably have a theoretical basis that includes a model of the
possible causal processes involved. If there is no knowledge or theory on the links
between the intervention and the desired outcome, it is difficult to understand how an
intervention works (Goldenhar and Schulte 1996; Shannon et al. 1999; Kristensen 2005).
Attention should be paid to the selection of samples, adequate sample size, and duration,
exposure to and intensity of intervention (Goldenhar and Schulte 1996; Buring 2002;
Green 2002; Silverstein and Clark 2004; Kristensen 2005). In order to detect changes in
outcomes, a sufficiently long enough follow-up is needed due to latency periods in the
development of MSDs. No absolute limits can be set, but a follow-up time of one year, or
at least six months, has been proposed (Westgaard and Winkel 1997; Driessen et al. 2010).
Since the latencies are for the most part still unknown, it is recommended that multiple
follow-ups should be attempted if feasible, and both short- and long-term effects should be
32
measured (Karsh 2006). Valid and reliable measurements are needed to ascertain whether
the intervention was carried out as intended, whether the intervention led to the intended
changes in exposure, and whether the changes in exposure led to the anticipated effect on
health outcomes (Kristensen 2005). With all interventions, but especially when using a
multiple- component intervention, measurement and reporting on the effects of the
intervention on intermediate outcomes needs to be emphasized (Karsh et al. 2001).
Measurements of potential confounders or effect modifiers, appropriate and adequate
statistical methods and accounting for losses to follow-up are needed. Analyses should
follow an intention- to- treat principle, i.e. regardless of whether or not the intervention is
completed or received, the subjects or clusters are kept in their original groups. Evidence
of compliance with the intervention, potential contamination between the groups, and co-
interventions also need to be considered (Buring 2002; Green 2002; Silverstein and Clark
2004).
Process evaluation
A careful evaluation of the intervention process is important, based on a systematic
documentation of how the intervention was carried out. This can help in the interpretation
of the study findings and explain how and why changes were or were not achieved. The
evaluation provides feedback for improving the intervention in the future. The results can
be used when the intervention is repeated in another setting and help others to avoid
pitfalls (Goldenhar et al. 2001). It is important to detect if the intervention, even one
deemed effective, was not fully implemented such that not all particpants were exposed to
it (programme failure) or whether it worked poorly although all participants were involved
(theory failure) (Kristensen 2005).
2.3.3. Challenges encountered in workplace intervention studies
Executing controlled intervention studies in workplaces is difficult. It is also time-
consuming and costly (Schulte et al. 1996; Volinn 1999; Karsh et al. 2001). Worksites
with a large number of workers doing the same tasks largely belong to the past and studies
are confronted with unplanned changes beyond the researchers' control. Workforce or
manager turnover must be taken into account. Rapid business or market plan changes, and
changes in production processes that influence work assignments and exposure levels may
33
occur and these are largely unforeseen when the intervention is being planned (Cole et al.
2003; Silverstein and Clark 2004; Heaney and Fujishiro 2006).
It is essential that workplaces approve of the intervention study and perceive that they will
gain benefits from participating in it. Both management and workers should comply with
the concept of random assignment between an intervention and a control group. This is not
always the case. Workplaces participating in an intervention and believing that it will be
effective might have difficulties to accept the use of a control group. If the intervention
being tested is believed to be effective, it might be difficult for the workplaces to justify
having a control group due to financial aspects (Braunholtz et al. 2001; Karsh et al. 2001;
Kristensen 2005). Those outside the trial should not receive inferior care simply because
of not being in the trial. Thus, close collaboration, adequate communication, and mutual
commitment between researchers, workers and management are all necessities. The
successful conduct of an intervention requires interdisciplinary teams, in which
understanding and respect between disciplines can promote the ability to meaningfully
work together (Schulte et al. 1996; Kristensen 2005).
2.3.4. The participatory approach
The participatory method is commonly used in the design and analysis of work systems
(Brown 2005) and its advantages have been discussed especially in relation to PE (Kogi
2006). Along with an increased utilization of participatory methods, it has been recognised
that ergonomics experts alone cannot achieve successful and widespread implementation
of ergonomics in the workplace (Haines and Wilson 1998; Kogi 2006). PE is a complex
and ambiguous concept. No general agreement about the term exists (Haines and Wilson
1998). Wilson and Haines (1997) have proposed that PE refers to 'the involvement of
people in planning and controlling a significant amount of their own work activities, with
sufficient knowledge and power to influence both processes and outcomes in order to
achieve desirable goals'. Imada (1991) defines PE as 'a macroergonomic approach to the
implementation of technology in organizations that requires end users to be highly
involved in developing and implementing the technology' and Nagamashi (1995) as 'the
workers active involvement in implementing ergonomic knowledge and procedures in
their workplace supported by their supervisors and managers, in order to improve
34
working conditions and product quality' whereas Kuorinka (1997) considers PE as
'practical ergonomics with participation of the necessary actors in problem solving'.
A characteristic of most PE interventions is the formation of some type of team or
committee consisting of workers or their representatives, managers, health and safety
personnel, ergonomists, and sometimes research experts. This team usually receives
training in ergonomics from an expert to become familiar with ergonomic principles,
workplace specific risk factors and how this knowledge can instigate improvements
(Theberge et al. 2006; IWH 2008; van Eerd et al. 2008). The expert may also give
training in interpersonal, communication and team-building skills. The expert may
function as a resource person, i.e. an individual familiar with the available skills and
knowledge to help in problem-solving and who can contribute to the design of the
intervention. The role of the expert is varied and complex but in all participatory
approaches he/she acts as an agent or facilitator of change (Brown 2005).
Furthermore, a successful PE program requires that the responsibilities of team members
in problem-solving, and developing and implementing changes, are well defined, that the
group makes decisions through group consultation and that the management is involved in
any decision requiring resources and implementation. Support, time and financial
resources from the top levels of management are key factors in the success (IWH 2008;
van Eerd et al. 2008).
The PE approach has several advantages. When the workers are involved in the process as
the best experts of their own work, a feeling of solution ownership is generated, leading to
increased job satisfaction and commitment to the changes being implemented.
Participation represents a learning experience for the individuals involved, as well as
giving them more self confidence, competence and independence. It has been suggested
that worker participation can play a role in reducing stress at work.
There can be disadvantages encountered in PE, e.g. the workers' unwillingness to
participate, or the organizational structure may limit the degree of worker participation.
True commitment from the top management may be difficult to obtain. The process may
encourage workers to develop unrealistic expectations or it may provoke envy and
35
dissatisfaction among those not involved in the improvement process (Haines and Wilson
1998; Brown 2005).
2.4. Evidence of the effectiveness of ergonomics interventions in
preventing musculoskeletal disorders
2.4.1. Overview
When the current intervention study was planned, the scientific literature was scrutinized,
coming the conclusion that no high-quality studies had been reported on the effectiveness
of ergonomics interventions and that sound evidence was lacking (van Poppel et al. 1997;
Westgaard and Winkel 1997; Volinn 1999; Lincoln et al. 2000; Linton and van Tulder
2001). In recent years, a few RCTs have been reported and the number of systematic
reviews in this area has increased. Most of the studies have focused on behavioural (e.g.
physical training of the worker, advice or education about working methods and
techniques or the use of personal protective equipment) and physical interventions (e.g.
redesign of physical work environment, working aids or tools, lifting and transfer aids for
manual materials handling), whereas organizational intervention studies are still sparse.
Organizational interventions occur at company, job or task level and focus on work
processes, practices, and policies of workplaces (i.e. job rotation, job enlargement,
modification of the production system). Such interventions often have an impact on both
physical and psychosocial load of the workers (Goldenhar and Schulte 1996; Westgaard
and Winkel 1997; Zwerling et al. 1997). The PE approach is recommended to reduce
MSDs (Cole et al. 2005; Hignett et al. 2005; van der Molen et al. 2005a; Rivilis et al.
2006; Rivilis et al. 2008) and its potential as an effective means to improve unsatisfactory
psychosocial work factors has been emphasized (Haims and Carayon 1998; Heaney and
Fujishiro 2006; St-Vincent et al. 2006). It has also been claimed to be the most efficient
way to improve ergonomics in the workplace (Hagberg et al. 1995). However, evidence on
its effectiveness continues to be scanty. In multi-component interventions, different
intervention types are combined (Feuerstein et al. 2000; Karsh et al. 2001; Gatty et al.
2003). They might have a greater chance of success than single interventions in preventing
MSDs (Karsh et al. 2001; Gatty et al. 2003; Silverstein and Clark 2004; Tveito et al. 2004;
36
Cole et al. 2006), but it is not possible to recommend which components should be
included and how they should be balanced (Burton et al. 2006).
Despite the increasing number of studies, they seem mostly to suffer from low
methodological quality. All the reviews have concluded that good quality RCTs are
urgently needed to provide convincing evidence on the effectiveness of the interventions.
Burton et al. (2006) highlighted the need for conducting especially CRTs targeted to study
the effectiveness of physical, psychosocial and organizational ergonomics interventions.
This literature review is mainly based on recent systematic reviews. In addition, some
original randomized or non-randomized studies will be described. Interventions using the
PE approach are reviewed separately. Tertiary prevention interventions were excluded
because they are beyond the scope of this thesis. The reviews were favoured in which the
authors had used some of the several guidelines developed for the reporting and a quality
assessment of studies, e.g. the Delphi list (Verhagen et al. 1998), the GATE checklist
(EPIQ 2004), or the criteria recommended by the Cochrane back review group (van Tulder
et al. 2003).
2.4.2. Behavioural and physical interventions
Low back pain
Back schools are not recommended for the prevention of LBP (Linton and van Tulder
2001, Burton et al. 2006, Bigos et al. 2009). The use of back belts, lumbar supports
(Jellema et al. 2001; Linton and van Tulder 2001; van Poppel et al. 2004; Ammendolia et
al. 2005; Bigos et al. 2009), and shoe inserts or orthoses have also been shown to be
ineffective (Burton et al. 2006, Bigos et al. 2009). Training in manual materials handling
or the use of lifting equipments is not effective if it is used as the only intervention (Bos et
al. 2006; Martimo et al. 2008; Bigos et al. 2009). Training and education combined with
the use of mechanical or other aids may be effective, which can be partly explained by a
decrease in the frequency of manual lifting (Bos et al. 2006).
When technical lifting devices were part of a more comprehensive intervention, a
reduction in physical work demands and low back disorders was found, particularly when
the PE approach was used (van der Molen et al. 2005a). A recent systematic review by
37
Driessen et al. (2010) included only RCTs and surveyed studies from the time period 1988
to 2008. These investigators concluded that physical ergonomics interventions were not
effective in the prevention of LBP. Thus, there seems to be insufficient evidence to
recommended physical (or organizational) ergonomics interventions alone for the
prevention of LBP (Linton and van Tulder 2001; Burton et al. 2006).
Pain in the neck and upper limbs
Intervention studies concerning neck and upper limb disorders are considerably less
common than those on LBP. The systematic review of Brewer et al. (2006) examined
articles published from 1980 to 2005. They found moderate evidence for a positive effect
of alternative pointing devices and no effect of workstation adjustment, rest breaks, and
exercise during the breaks among computer users. The systematic review by Driessen et
al. (2010) showed that physical ergonomics interventions were not more effective on short
and long term neck pain prevalence or incidence, but possibly could reduce the intensity
of neck pain in the long run. However, the number of studies was limited and targeted
mostly office workers, and populations, interventions and outcomes were heterogeneous.
Boocock et al. (2007) reported moderate evidence that mouse and keyboard design led to
positive effects in neck and upper limb conditions among video display unit (VDU)
workers, but among manufacturing workers, the evidence was insufficient. Norwegian
researchers have published protocols of Cochrane systematic reviews on workplace
interventions for neck pain (and LBP), and it can be predicted that the results of these
reviews will add to knowledge in this area in the near future (Aas et al. 2009a; Aas et al.
2009b).
Lower limb pain
No literature on the effectiveness of work-related interventions regarding lower limb pain
was found.
2.4.3. Organizational interventions
A narrative review by Westgaard and Winkel (1997) and a systematic review conducted
ten years later by Boocock et al. (2007) showed consistently that there was insufficient
evidence for demonstrating any benefits of organizational interventions. In the latter
review, only two relevant studies were identified. These involved no randomization, were
38
of low quality, and showed no improvements in outcome measures. Driessen et al. (2010)
stated that RCTs on organizational ergonomics interventions to prevent and reduce neck
pain and LBP were lacking. Many other authors have recently concluded that evidence on
the effectiveness of organizational interventions is scanty (Murphy and Sauter 2004;
Bongers et al. 2006; Burton et al. 2006; Heaney and Fujishiro 2006; St-Vincent et al.
2006; Podniece 2008). Few studies exist on the effectiveness of work-rest schedules,
breaks, task rotation and task enrichment, but the results are contradictory and no
conclusive evidence is available (Bongers et al. 2006; Podnieze 2008). Two reviews have
reported on the effects of workplace reorganization on psychosocial factors and health,
with reference to the 'demand-control-support' model (Bambra et al. 2007; Egan et al.
2007). Limited and inconsistent evidence was found to support health benefits when
employee control improved, demands decreased or support increased. Two studies of
participatory interventions occurring alongside redundancies reported worsening of
employee health (Egan et al. 2007). Task-restructuring interventions that increased
demand and decreased control had an adverse effect on health, and an increase in
workplace support did not appear to mediate this relationship (Bambra et al. 2007). van
der Klink et al. (2001) reported a small and non-significant effect size for the few
organizational interventions that were focused on work-related stress.
One CRT (Tsutsumi et al. 2009) and one non-randomized controlled before-after study
(Kobayashi et al. 2008) have been published after the above mentioned reviews. They
showed positive effects on improving mental health and job performance by using the
participatory approach. These studies are reviewed below.
2.4.4. Participatory ergonomics interventions
Only one systematic review on the effectiveness of the PE interventions was found. Rivilis
et al. (2008) used a best evidence synthesis where 12 studies were evaluated to be of
sufficient methodological quality to be included. Nine 'medium' and three 'higher' quality
studies showed partial to moderate evidence that the PE interventions had had a
favourable impact on musculoskeletal symptoms and reduced workers' compensation
claims and sickness absence. However, random allocation or adjustments for differences
between groups at baseline were used only in three studies. Other methodological flaws
39
were insufficient documentation of participation in the PE process and poor reporting of
consideration of potential confounders. In the studies graded as 'low' quality, also deficient
reporting of co-interventions and of risk factor measurement at baseline and at follow-up,
lack of comparison groups, and inappropriate statistical analysis were observed.
Seven RCTs were detected. Two studies (Straker et al. 2004; van der Molen et al. 2005b)
were not included because they did not have musculoskeletal health or psychosocial
factors at work as outcomes. Finally, three of the five RCTs were cluster randomized
studies (one being still conducted), and two were individually randomized. Only two
studies were organizational interventions specifically on mental health and psychosocial
job conditions. The contents of the PE interventions are briefly described below. A more
detailed summary of the reviewed original PE intervention studies and their results are
presented in Table 2.
Non-randomized controlled before-after studies
Kobayashi et al. (2008) examined the effects of a participatory organizational intervention
on job stressors and mental health among workers in a manufacturing company. A work
environment improvement team supported each participating department throughout the
six-month intervention. A checklist called the Mental Health Action Checklist for a Better
Workplace Environment (MHACL) was designed for use in a group-based workshop that
was arranged in the departments. In the workshops, workers were informed about stress
and mental health at work and concepts related to improving work environments. Based
on the MHACL, each group listed action items that might be useful for better worker
mental health in their department. The team encouraged the implementation of the plans in
each department after the workshops. Among women, skill underutilization, supervisor
and co-worker support, psychological distress and job satisfaction developed significantly
more favorably in the intervention group than in the control group. No such effect was
observed among men, and no effect was observed on the number of sick leaves in either
gender.
The group of Laing studied the effectiveness of a PE intervention in reducing worker pain
severity through changes aimed at reducing physical demands (Laing et al. 2005) and
improving communication and the psychosocial work environment (Laing et al. 2007) in
40
an automotive parts manufacturing factory. An ergonomics change team (ECT) was
responsible for identifying the targets for ergonomic improvements and developing and
implementing solutions based on the PE Blueprint (identification, assessment of
ergonomic risk factors, solution building/testing, and implementation). The researchers
supported the activities of the ECT during an 11-month period (1 month training, 10
months intervention). The ECT was trained to take into account basic anatomy, physical
and psychosocial risk factors, and ergonomics principles and assessments tools. The
intervention had no significant effect on the workers' mechanical exposures, perceived
effort, perceptions of decision latitude, or pain severity levels. Communication dynamics
regarding ergonomics was significantly enhanced in the intervention group compared to
the control group.
The study of Rivilis et al. (2006) conducted in a depot of a large courier company shared
similarities with the study of Laing et al. An ECT was responsible for the 14-month
intervention that was based on the PE Blueprints. The ECT was trained in ergonomics,
workplace risk factors, and the tools needed to perform ergonomic assessments. Greater
participation in the intervention process was associated with increased levels of job
influence and communication. Improvements in communication were associated with
reduced pain intensity and improved work role function. The lower levels of pain post-
intervention were related to greater work role function.
Evanoff et al. (1999) studied the effect of a PE team on rates of injury, lost time and
MSDs among hospital orderlies. They also hypothesized that direct worker participation in
problem-solving would improve job satisfaction. The PE team received training in team
building, risk identification and control, and undertook supervised exercises in observation
and measurement. The PE team was responsible for identifying and prioritizing safety
problems and implementing solutions. The primary safety intervention implemented was
the development of standardized lifting techniques. All of the orderlies received training
on the standardized procedures according to a lifting manual written by the team. New
employees were required to complete the training before being allowed to work
independently. The rates of injuries and lost days decreased following the two- year
intervention. At the 15-month follow-up, there were significant reductions in the
proportion of employees who reported MSDs. Statistically significant improvements were
41
observed in job satisfaction, perceived job stressors and social support among the
orderlies. However, based on the presented results, it is not clear whether there were any
significant differences between the intervention and control group.
Randomized controlled trials
The aim of the CRT concucted by Tsutsumi et al. (2009) was to study the effect of a
participatory intervention for workplace improvement on mental health and job
performance among blue collar workers manufacturing electronic equipment. The
intervention process lasted approximately 15 months. The intervention consisted of a half-
day training workshop for facilitators of participative actions at their workplaces. The
facilitators received information on mental health and the participatory approach for stress
prevention. Supervisors were educated on the significance of positive mental health and
improvement in the work environment, and given examples of good practices. The
supervisors were trained to identify occupational stressors and to make suggestions for
ways to reduce conflicts and to eliminate the sources of stress. In a workshop, the
intervention group workers were provided with an introduction to the methodology for
improving the work environment, and a lecture on hazard identification. They had group
discussions on work improvements and presentations of action plans. During the 5- month
implementation phase, two follow-up workshops were arranged. Mental health remained
at the same level in the intervention group, but deteriorated in the control group. Job
performance increased statistically significantly in the intervention group compared to the
control group. However, after imputation of missing values, no statistically significant
differences between the groups were observed.
The effect of a one-year PE program on MSDs and individual coping, job demands, job
control and social support was studied in a CRT among operators in the aluminium
industry (Morken et al. 2002). Three intervention groups, 'shift group with supervisor',
'shift group without supervisor', and 'managers only', were formed. The PE program
consisted of 10 didactic sessions (1h 15 min) and discussion (45 min) chaired by a
physiotherapist acquainted with ccupational health services in the aluminum plants. Each
session had a different topic related to physical, psychosocial, work-organizational, and
individual risk factors of MSDs, basics of ergonomics, and coping with MSDs. Active
participation of the group members responsible for identifying, prioritizing and
42
implementing solutions was emphasized. The implemented changes focused on redesign
of workplace, working aids or tools, reducing repetitive-motion stress points and
modifying the work processes to promote job variation. No significant changes on MSDs
were found. Operators in the 'shift group without a supervisor' used coping strategies more
often and social support improved slightly. Job demands and control did not differ
between the groups.
Two individually randomized trials were targeted at the ergonomics of office work
involving VDU. Ketola et al. (2002) evaluated the effect of an intensive ergonomics
approach and education. In the intensive ergonomics group, two physiotherapists visited
each worker and introduced an ergonomic checklist for VDU work. The workers assessed
independently their workstations with the aid of the list. Potential improvements based on
the workers own views and the physiotherapist's observations were discussed. The
workers were encouraged to participate actively in the redesign in their workstations.
Advice to take care of their work postures and to add short pauses into work was given. In
the education group, in the 1-hour training session, the workers were instructed on the
principles of VDU work, they received the same checklist, and were encouraged to
evaluate their own workstation and implement the changes. The same advice considering
work postures and pauses during the work was given. The control group received only a
one-page leaflet. The intensive ergonomics and ergonomic education groups showed less
musculoskeletal discomfort than the control group after a two-month follow-up. No
significant differences in the strain level or in pain or any long-term effects on discomfort,
strain, or pain were found at the 10-month follow-up.
In Bohrs' study (2000), the control group had no education session. The traditional group
had a 1-hour education session consisting of a lecture and informational handouts about
office ergonomics. The participatory education group received a 2- hour education session
with similar content but incorporating discussion, problem-solving exercises, and
evaluation and modification of work areas according to the information received. In
follow-ups at three, six and 12- months, those who received education reported less
pain/discomfort or psychosocial stress than those in the control group, but there was no
indication that the differences were related to better work area configuration or improved
worker postures. The workers in the participatory group reported a better health status than
43
those in the control or the traditional education group. Later Bohr (2002) reported that
there was no evidence that the PE method was more effective than the traditional method
in encouraging the workers to optimize the positions of their work equipment or to
maintain good working postures.
A large CRT ('Stay@Work') aimed to investigate the effectiveness of a PE to prevent LBP
and neck pain among blue and white collar workers is underway in the Netherlands
(Driessen et al. 2008). The intervention group is participating in a six-step PE programme.
A working group was created including eight workers, a representative of the
management, and an occupational and health and safety coordinator. During a one-day
meeting, the working group followed the steps of the PE program and identified the most
important risk factors, the most fundasmental ergonomic measures on the basis of group
consensus, and prepared the implementation plan. If needed, a second meeting could be
arranged to evaluate the status of the implementation phase. The results are expected in
2010.
In summary, there is little evidence for the effectiveness of preventive ergonomics
interventions. The physical and organizational ergonomics interventions do not seem to be
more effective than interventions including no ergonomics. However, the amount of high-
quality studies is insufficient to allow any final conclusions to be drawn. Evidence of the
effectiveness of PE is lacking since most studies suffer from methodological flaws and
only a few RCTs exist.
T
a
b
l
e

2
.


E
f
f
e
c
t
s

o
f

p
a
r
t
i
c
i
p
a
t
o
r
y

e
r
g
o
n
o
m
i
c
s

i
n
t
e
r
v
e
n
t
i
o
n
s

o
n

m
u
s
c
u
l
o
s
k
e
l
e
t
a
l

d
i
s
o
r
d
e
r
s

(
M
S
D
s
)

a
n
d

p
s
y
c
h
o
s
o
c
i
a
l

f
a
c
t
o
r
s

a
t

w
o
r
k
F
i
r
s
t

a
u
t
h
o
r
Y
e
a
r
I
n
d
u
s
t
r
y
/
s
e
c
t
o
r
S
t
u
d
y

d
e
s
i
g
n
P
a
r
t
i
c
i
p
a
n
t
s
F
o
l
l
o
w
-
u
p
I
n
t
e
r
v
e
n
t
i
o
n
/

A
i
m
E
f
f
e
c
t
s

o
n

M
S
D
s
/

p
s
y
c
h
o
s
o
c
i
a
l
f
a
c
t
o
r
s

a
t

w
o
r
k
T
o
t
a
l
I
n
t
e
r
v
e
n
t
i
o
n
C
o
n
t
r
o
l
N
o
n
-
r
a
n
d
o
m
i
z
e
d

c
o
n
t
r
o
l
l
e
d

t
r
i
a
l
s
*
K
o
b
a
y
a
s
h
i
,

2
0
0
8
;
J
a
p
a
n
,
m
a
n
u
f
a
c
t
u
r
i
n
g
e
n
t
e
r
p
r
i
s
e
B
e
f
o
r
e
-
a
f
t
e
r
w
i
t
h

c
o
n
t
r
o
l
g
r
o
u
p
1
4
3
4

w
o
r
k
e
r
s

f
r
o
m

4
5
d
e
p
a
r
t
m
e
n
t
s
.
P
r
e
-
p
o
s
t

i
n
t
e
r
v
e
n
t
i
o
n

m
a
t
c
h
e
d
q
u
e
s
t
i
o
n
n
a
i
r
e
n
=
1
0
7
1
n
=

3
2
1
w
o
r
k
e
r
s
f
r
o
m

9
d
e
p
a
r
t
m
e
n
t
s
n
=
7
5
0
w
o
r
k
e
r
s
f
r
o
m

3
6
d
e
p
a
r
t
m
e
n
t
s
P
r
e
-
p
o
s
t
q
u
e
s
t
i
o
n
n
a
i
r
e
s
a
d
m
i
n
i
s
t
e
r
e
d
1
2

m
o
n
t
h
s
a
p
a
r
t

S
i
c
k

l
e
a
v
e
d
a
t
a

w
e
r
e
o
b
t
a
i
n
e
d

f
r
o
m
t
h
e

c
o
m
p
a
n
y
r
e
g
i
s
t
r
i
e
s
E
f
f
e
c
t
s

o
f

a

p
a
r
t
i
c
i
p
a
t
o
r
y
o
r
g
a
n
i
z
a
t
i
o
n
a
l
i
n
t
e
r
v
e
n
t
i
o
n

o
n

t
h
e
r
e
d
u
c
t
i
o
n

o
f

j
o
b

s
t
r
e
s
s
o
r
s
a
n
d

t
h
e

e
n
h
a
n
c
e
m
e
n
t

o
f
t
h
e

m
e
n
t
a
l

h
e
a
l
t
h

w
e
r
e
s
t
u
d
i
e
d

A
m
o
n
g

w
o
m
e
n
,

s
k
i
l
l
u
n
d
e
r
u
t
i
l
i
z
a
t
i
o
n

(
p
=
0
.
0
4
8
)
,
s
u
p
e
r
v
i
s
o
r

(
p
=
0
.
0
0
0
)

a
n
d

c
o
-
w
o
r
k
e
r
s
u
p
p
o
r
t

(
p
=
0
.
0
3
6
)
,

p
s
y
c
h
o
l
o
g
i
c
a
l
d
i
s
t
r
e
s
s

(
v
i
g
o
r

p
=
0
.
0
2
6
,

d
e
p
r
e
s
s
i
o
n
p
=

0
.
0
1
0
)
,

j
o
b

s
a
t
i
s
f
a
c
t
i
o
n

(
0
.
0
4
3
)
,
a
n
d

h
e
a
l
t
h

r
i
s
k
s

(
H
R
)

a
s
s
o
c
i
a
t
e
d
w
i
t
h

j
o
b

s
t
r
e
s
s
o
r
s

(
t
o
t
a
l

H
R
p
=
0
.
0
0
0
,

H
R

a
s
s
o
c
i
a
t
e
d

w
i
t
h

j
o
b
s
t
r
a
i
n

p
=

0
.
0
3
0
,

a
n
d

H
R

a
s
s
o
c
i
a
t
e
d
w
i
t
h

w
o
r
k
s
i
t
e

s
u
p
p
o
r
t

p
=
0
.
0
0
0
)
c
h
a
n
g
e
d

s
i
g
n
i
f
i
c
a
n
t
l
y

i
n

t
h
e
i
n
t
e
r
v
e
n
t
i
o
n

g
r
o
u
p
.

I
m
p
r
o
v
e
m
e
n
t
s

i
n

t
h
e

o
u
t
c
o
m
e
s
w
e
r
e

m
o
r
e

p
r
o
m
i
n
e
n
t

a
m
o
n
g
d
e
p
a
r
t
m
e
n
t
s

w
i
t
h

a

5
0
%

o
r

h
i
g
h
e
r
p
a
r
t
i
c
i
p
a
t
i
o
n

r
a
t
e

i
n

t
h
e

p
l
a
n
n
i
n
g
w
o
r
k
s
h
o
p
s

(
p
=
0
.
0
0
0
-
0
.
0
3
0
)

a
n
d
a
m
o
n
g

d
e
p
a
r
t
m
e
n
t
s

w
i
t
h

5
0
%

o
r
h
i
g
h
e
r

r
a
t
e

o
f

i
m
p
l
e
m
e
n
t
e
d

v
s
.
p
l
a
n
n
e
d

a
c
t
i
o
n
s

(
p
=
0
.
0
0
1
-
0
.
0
3
9
)
.
E
f
f
e
c
t

s
i
z
e
s
:

n
o
t

g
i
v
e
n

N
o

f
a
v
o
u
r
a
b
l
e

e
f
f
e
c
t

a
m
o
n
g

m
e
n

N
o

e
f
f
e
c
t

o
n

s
i
c
k

l
e
a
v
e

a
m
o
n
g
e
i
t
h
e
r

g
e
n
d
e
r
*
L
a
i
n
g
,

2
0
0
5
,
2
0
0
7
;

C
a
n
a
d
a
,

a
n
m
u
l
t
i
n
a
t
i
o
n
a
l
a
u
t
o
m
o
t
i
v
e

p
a
r
t
s
m
a
n
u
f
a
c
t
u
r
i
n
g
c
o
m
p
a
n
y
B
e
f
o
r
e
-
a
f
t
e
r
w
i
t
h

c
o
n
t
r
o
l
g
r
o
u
p
T
h
e

c
o
m
p
l
e
t
e

p
r
e
-
p
o
s
t
m
a
t
c
h
e
d

q
u
e
s
t
i
o
n
n
a
i
r
e
n

=

8
3
n

=

4
4
A

s
i
s
t
e
r

p
l
a
n
t
i
n

t
h
e
c
o
m
p
a
n
y

a
s

a
c
o
n
t
r
o
l

g
r
o
u
p
n

=

3
9
P
r
e
-
p
o
s
t
q
u
e
s
t
i
o
n
n
a
i
r
e
s
a
d
m
i
n
i
s
t
e
r
e
d
1
0

m
o
n
t
h
s
a
p
a
r
t
E
f
f
e
c
t
i
v
e
n
e
s
s

o
f

a

P
E
i
n
t
e
r
v
e
n
t
i
o
n
1
)

i
n

r
e
d
u
c
i
n
g

w
o
r
k
e
r
p
a
i
n

s
e
v
e
r
i
t
y

t
h
r
o
u
g
h
c
h
a
n
g
e
s

a
i
m
e
d

a
t
r
e
d
u
c
i
n
g

p
h
y
s
i
c
a
l
1
)

M
i
n
o
r

i
n
f
l
u
e
n
c
e

o
n

m
e
c
h
a
n
i
c
a
l
e
x
p
o
s
u
r
e
s

N
o

s
y
s
t
e
m
a
t
i
c

c
h
a
n
g
e
s

i
n
p
e
r
c
e
i
v
e
d

e
f
f
o
r
t

o
r

p
a
i
n

s
e
v
e
r
i
t
y
l
e
v
e
l
s
4
4
4
5
T
a
b
l
e

2

(
c
o
n
t
i
n
u
e
d
)
F
i
r
s
t

a
u
t
h
o
r
Y
e
a
r
I
n
d
u
s
t
r
y
/
s
e
c
t
o
r
S
t
u
d
y

d
e
s
i
g
n
P
a
r
t
i
c
i
p
a
n
t
s
F
o
l
l
o
w
-
u
p
I
n
t
e
r
v
e
n
t
i
o
n
/

A
i
m
E
f
f
e
c
t
s

o
n

M
S
D
s
/
p
s
y
c
h
o
s
o
c
i
a
l
f
a
c
t
o
r
s

a
t

w
o
r
k
T
o
t
a
l
I
n
t
e
r
v
e
n
t
i
o
n
C
o
n
t
r
o
l
d
e
m
a
n
d
s

(
2
0
0
5
)
2
)

i
n

i
m
p
r
o
v
i
n
g
c
o
m
m
u
n
i
c
a
t
i
o
n
(
e
r
g
o
n
o
m
i
c
s
-
r
e
l
a
t
e
d

a
n
d
o
v
e
r
a
l
l
)

a
n
d

p
s
y
c
h
o
s
o
c
i
a
l
w
o
r
k

e
n
v
i
r
o
n
m
e
n
t
(
d
e
c
i
s
i
o
n

l
a
t
i
t
u
d
e
,
i
n
f
l
u
e
n
c
e
)

(
2
0
0
7
)
2
)

C
o
m
m
u
n
i
c
a
t
i
o
n

d
y
n
a
m
i
c
s
r
e
g
a
r
d
i
n
g

e
r
g
o
n
o
m
i
c
s

w
a
s
e
n
h
a
n
c
e
d

i
n

t
h
e

i
n
t
e
r
v
e
n
t
i
o
n

g
r
o
u
p
(
p

=

0
.
0
0
4
-
0
.
0
4
5

f
o
r

a
l
l

5
q
u
e
s
t
i
o
n
s
)
.

N
o

d
i
f
f
e
r
e
n
c
e
s

b
e
t
w
e
e
n
t
h
e

g
r
o
u
p
s

w
e
r
e

o
b
s
e
r
v
e
d

f
o
r

t
h
e

2
q
u
e
s
t
i
o
n
s

o
n

o
v
e
r
a
l
l

w
o
r
k
p
l
a
c
e
c
o
m
m
u
n
i
c
a
t
i
o
n

d
y
n
a
m
i
c
s

(
b
e
t
w
e
e
n
l
a
b
o
u
r

a
n
d

m
a
n
a
g
e
m
e
n
t

a
n
d
b
e
t
w
e
e
n

o
n
e

a
n
d

h
i
s
/
h
e
r

c
o
-
w
o
r
k
e
r
s
)
.
E
f
f
e
c
t

s
i
z
e
s
:

n
o
t

g
i
v
e
n

N
o

s
t
a
t
i
s
t
i
c
a
l
l
y

s
i
g
n
i
f
i
c
a
n
t
d
i
f
f
e
r
e
n
c
e
s

i
n

d
e
c
i
s
i
o
n

l
a
t
i
t
u
d
e

o
r
i
n
f
l
u
e
n
c
e
*
R
i
v
i
l
i
s
,

2
0
0
6
;
C
a
n
a
d
a
,
a

c
o
u
r
i
e
r

c
o
m
p
a
n
y
B
e
f
o
r
e
-
a
f
t
e
r
w
i
t
h

c
o
n
t
r
o
l
g
r
o
u
p
T
h
e

c
o
m
p
l
e
t
e

p
r
e
-
p
o
s
t
m
a
t
c
h
e
d

q
u
e
s
t
i
o
n
n
a
i
r
e
n

=

1
2
2
o
n
e

d
e
p
o
t
n

=

7
1
a

n
e
a
r
b
y

d
e
p
o
t
i
n

t
h
e

s
a
m
e
r
e
g
i
o
n

a
s

a
c
o
n
t
r
o
l

g
r
o
u
p
n

=

5
1
P
r
e
-
p
o
s
t
q
u
e
s
t
i
o
n
n
a
i
r
e
s
a
d
m
i
n
i
s
t
e
r
e
d
1
4

m
o
n
t
h
s
a
p
a
r
t

A
t

b
a
s
e
l
i
n
e
,
i
n
t
e
r
v
e
n
t
i
o
n
g
r
o
u
p

w
o
r
k
e
r
s
w
e
r
e

a
n
a
v
e
r
a
g
e

5
.
2
y
e
a
r
s

o
l
d
e
r
(
p
=

0
.
0
2
8
9
)
a
n
d
e
x
p
e
r
i
e
n
c
e
d
m
o
r
e

p
a
i
n
(
p
=
0
.
0
2
1
)

t
h
a
n

E
v
a
l
u
a
t
i
o
n

o
f

a

P
E
i
n
t
e
r
v
e
n
t
i
o
n

a
i
m
e
d

a
t
i
m
p
r
o
v
i
n
g
m
u
s
c
u
l
o
s
k
e
l
e
t
a
l

h
e
a
l
t
h
;
A

h
y
p
o
t
h
e
s
i
s

t
h
a
t

w
i
t
h
g
r
e
a
t
e
r

p
r
o
c
e
s
s
i
m
p
l
e
m
e
n
t
a
t
i
o
n

i
n
t
e
n
s
i
t
y
,
g
r
e
a
t
e
r

c
h
a
n
g
e

i
n
e
x
p
o
s
u
r
e

t
o

M
S
D

r
i
s
k
f
a
c
t
o
r
s

w
o
u
l
d

b
e
o
b
s
e
r
v
e
d

a
n
d

r
e
s
u
l
t

i
n
i
m
p
r
o
v
e
m
e
n
t
s

i
n

h
e
a
l
t
h
o
u
t
c
o
m
e
s

G
r
e
a
t
e
r

p
a
r
t
i
c
i
p
a
t
i
o
n

i
n

t
h
e
p
r
o
c
e
s
s

w
a
s

a
s
s
o
c
i
a
t
e
d

w
i
t
h
i
n
c
r
e
a
s
e
d

l
e
v
e
l
s

o
f

j
o
b

i
n
f
l
u
e
n
c
e
(
p
=

0
.
0
0
6
)

a
n
d

c
o
m
m
u
n
i
c
a
t
i
o
n
(
p
=

0
.
0
9
4
)

I
m
p
r
o
v
e
m
e
n
t
s

i
n

c
o
m
m
u
n
i
c
a
t
i
o
n
l
e
v
e
l
s

w
e
r
e

a
s
s
o
c
i
a
t
e
d

w
i
t
h

r
e
d
u
c
e
d
p
a
i
n

i
n
t
e
n
s
i
t
y

(
p
=

0
.
0
0
8
)

a
n
d
i
m
p
r
o
v
e
d

w
o
r
k

r
o
l
e

f
u
n
c
t
i
o
n
(
W
R
F
,

h
o
w

m
u
c
h

o
f

t
i
m
e

a
t

w
o
r
k

i
s
l
i
m
i
t
e
d

b
y

m
u
s
c
u
l
o
s
k
e
l
e
t
a
l

p
a
i
n
,
p
=
0
.
0
2
5
)

L
o
w
e
r

l
e
v
e
l
s

o
f

p
a
i
n

p
o
s
t
-
i
n
t
e
r
v
e
n
t
i
o
n

w
e
r
e

r
e
l
a
t
e
d

t
o

g
r
e
a
t
e
r
W
R
F

(
p
=

0
.
0
5
0
)
4
6
T
a
b
l
e

2

(
c
o
n
t
i
n
u
e
d
)
F
i
r
s
t

a
u
t
h
o
r
Y
e
a
r
I
n
d
u
s
t
r
y
/
s
e
c
t
o
r
S
t
u
d
y

d
e
s
i
g
n
P
a
r
t
i
c
i
p
a
n
t
s
F
o
l
l
o
w
-
u
p
I
n
t
e
r
v
e
n
t
i
o
n
/

A
i
m
E
f
f
e
c
t
s

o
n

M
S
D
s
/
p
s
y
c
h
o
s
o
c
i
a
l
f
a
c
t
o
r
s

a
t

w
o
r
k
T
o
t
a
l
I
n
t
e
r
v
e
n
t
i
o
n
C
o
n
t
r
o
l
t
h
e

w
o
r
k
e
r
s

i
n
t
h
e

c
o
n
t
r
o
l
g
r
o
u
p
.

A
g
e
a
n
d

b
a
s
e
l
i
n
e
p
a
i
n

l
e
v
e
l
w
e
r
e

t
a
k
e
n
i
n
t
o

t
h
e
a
n
a
l
y
s
e
s

M
e
a
s
u
r
e
s
/
o
u
t
c
o
m
e
s
;
p
e
r
c
e
i
v
e
d

p
h
y
s
i
c
a
l
d
e
m
a
n
d
s
,

w
o
r
k
o
r
g
a
n
i
z
a
t
i
o
n
a
l

r
i
s
k
f
a
c
t
o
r
s

(
p
e
r
c
e
p
t
i
o
n
s

o
f
j
o
b

i
n
f
l
u
e
n
c
e
,
c
o
m
m
u
n
i
c
a
t
i
o
n
d
y
n
a
m
i
c
s
)
,

s
e
l
f
-
r
e
p
o
r
t
e
d
p
a
i
n
/
d
i
s
c
o
m
f
o
r
t

a
n
d
w
o
r
k

f
u
n
c
t
i
o
n

o
u
t
c
o
m
e
s
E
f
f
e
c
t

s
i
z
e
s
:

n
o
t

g
i
v
e
n
*
E
v
a
n
o
f
f
,

1
9
9
9
;
U
S
A
,
h
o
s
p
i
t
a
l

o
r
d
e
r
l
i
e
s
P
r
o
s
p
e
c
t
i
v
e
i
n
t
e
r
v
e
n
t
i
o
n
t
r
i
a
l
(
b
e
f
o
r
e
-
d
u
r
i
n
g
-
a
f
t
e
r
)
M
e
a
n

e
m
p
l
o
y
m
e
n
t

d
u
r
i
n
g
s
t
u
d
y

p
e
r
i
o
d

1
0
0
-
1
1
0

o
r
d
e
r
l
i
e
s
6
7

o
r
d
e
r
l
i
e
s
p
r
e
-
i
n
t
e
r
v
e
n
t
i
o
n
a
n
d

8
8

p
o
s
t
-
i
n
t
e
r
v
e
n
t
i
o
n
c
o
m
p
l
e
t
e
d
q
u
e
s
t
i
o
n
n
a
i
r
e
O
t
h
e
r

h
o
s
p
i
t
a
l
w
o
r
k
e
r
s

u
s
e
d
a
s

a
c
o
n
c
u
r
r
e
n
t
c
o
n
t
r
o
l

1

m
o
n
t
h

7

m
o
n
t
h
s

1
5

m
o
n
t
h
s

2
4

m
o
n
t
h
s

T
h
e

a
i
m

w
a
s

t
o

s
t
u
d
y

i
f
t
h
e

i
m
p
l
e
m
e
n
t
a
t
i
o
n

o
f

a
P
E

t
e
a
m

w
o
u
l
d

r
e
s
u
l
t

i
n
l
o
w
e
r

r
a
t
e
s

o
f

i
n
j
u
r
y
,

l
o
s
t
t
i
m
e

a
n
d

m
u
s
c
u
l
o
s
k
e
l
e
t
a
l
s
y
m
p
t
o
m
s

(
m
e
a
n

b
o
d
y
c
o
m
f
o
r
t

r
a
t
e
,

1
=
u
n
c
o
m
-
f
o
r
t
a
b
l
e
,

5
=

c
o
m
f
o
r
t
a
b
l
e
)

I
t

w
a
s

h
y
p
o
t
h
e
s
i
z
e
d

t
h
a
t
d
i
r
e
c
t

w
o
r
k
e
r

p
a
r
t
i
c
i
p
a
-
t
i
o
n

i
n

p
r
o
b
l
e
m

s
o
l
v
i
n
g
w
o
u
l
d

i
m
p
r
o
v
e

j
o
b
s
a
t
i
s
f
a
c
t
i
o
n

(
t
h
r
e
e
s
u
m
m
a
r
y

s
c
a
l
e
s
;

t
h
e

J
o
b
S
a
t
i
s
f
a
c
t
i
o
n

S
c
a
l
e

(
3
i
t
e
m
s
)
,

t
h
e

W
o
r
k

A
p
g
a
r
(
s
o
c
i
a
l

s
u
p
p
o
r
t
,
i
n
f
l
u
e
n
c
e
,

7

i
t
e
m
s
)
,

a
n
d
t
h
e

P
s
y
c
h
o
s
o
c
i
a
l
S
t
r
e
s
s
o
r
s

s
c
a
l
e

(
3

i
t
e
m
s
)

I
n

t
h
e

i
n
t
e
r
v
e
n
t
i
o
n

g
r
o
u
p
,
d
e
c
r
e
a
s
e
d

r
i
s
k
s

o
f

w
o
r
k

i
n
j
u
r
y
(
R
R

=
0
.
5
0
,

9
5
%

C
I

0
.
3
5
-
0
.
7
2
)
,

l
o
s
t
t
i
m
e

i
n
j
u
r
y

(
R
R
=
0
.
2
6
,

9
5
%

C
I
0
.
1
4
-
0
.
4
8
)

a
n
d

i
n
j
u
r
y

w
i
t
h

3

o
r
m
o
r
e

d
a
y
s

o
f

t
i
m
e

l
o
s
s

(
R
R
=

0
.
1
9
,
9
5
%

C
I

0
.
0
7
-
0
.
5
3
)
.

T
o
t
a
l

l
o
s
t

d
a
y
s
d
e
c
l
i
n
e
d

f
r
o
m

1
3
6

t
o

2
3

a
n
n
u
a
l
l
y
p
e
r

1
0
0

f
u
l
l
-
t
i
m
e

w
o
r
k
e
r
e
q
u
i
v
a
l
e
n
t
s

(
F
T
E
)
.

I
n

t
h
e

i
n
t
e
r
v
e
n
t
i
o
n

g
r
o
u
p
,

m
e
a
n
d
i
s
c
o
m
f
o
r
t

r
a
t
i
n
g

c
h
a
n
g
e
d

f
r
o
m

3
.
5
t
o

3
.
9
,

p
<
0
.
0
5

n
e
c
k
;

3
.
0

t
o

3
.
4
,

p
<
0
.
0
1

l
o
w
e
r

b
a
c
k
;

3
.
7

t
o

4
.
0

p
<

0
.
0
1
f
o
r
e
a
r
m
;

3
.
3

t
o

3
.
8
,

p
<
0
.
0
5

k
n
e
e
.
B
a
s
e
d

o
n

t
h
e

p
r
e
s
e
n
t
e
d

r
e
s
u
l
t
s
,

i
t

i
s
n
o
t

c
l
e
a
r

w
h
e
t
h
e
r

t
h
e
r
e

w
e
r
e
s
t
a
t
i
s
t
i
c
a
l
l
y

s
i
g
n
i
f
i
c
a
n
t

e
f
f
e
c
t
s
b
e
t
w
e
e
n

t
h
e

g
r
o
u
p
s
.

I
n

t
h
e

i
n
t
e
r
v
e
n
t
i
o
n

g
r
o
u
p
,
i
m
p
r
o
v
e
m
e
n
t
s

i
n

j
o
b

s
a
t
i
s
f
a
c
t
i
o
n
(
f
r
o
m

7
.
7

t
o

8
.
8
,

p
<
0
.
0
1
)
,

i
n

w
o
r
k
A
p
g
a
r

(
f
r
o
m

1
4
.
6

t
o

1
5
.
8
,

p
<
0
.
0
5
)
,
a
n
d

i
n

p
e
r
c
e
i
v
e
d

p
s
y
c
h
o
s
o
c
i
a
l
4
7
T
a
b
l
e

2

(
c
o
n
t
i
n
u
e
d
)
F
i
r
s
t

a
u
t
h
o
r
Y
e
a
r
I
n
d
u
s
t
r
y
/
s
e
c
t
o
r
S
t
u
d
y

d
e
s
i
g
n
P
a
r
t
i
c
i
p
a
n
t
s
F
o
l
l
o
w
-
u
p
I
n
t
e
r
v
e
n
t
i
o
n
/

A
i
m
E
f
f
e
c
t
s

o
n

M
S
D
s
/
p
s
y
c
h
o
s
o
c
i
a
l
f
a
c
t
o
r
s

a
t

w
o
r
k
T
o
t
a
l
I
n
t
e
r
v
e
n
t
i
o
n
C
o
n
t
r
o
l
R
a
n
d
o
m
i
z
e
d

c
o
n
t
r
o
l
l
e
d

t
r
i
a
l
s
s
t
r
e
s
s
o
r
s

(
f
r
o
m

8
.
2

t
o

7
.
2
,

p

<
0
.
0
1
)
.
B
a
s
e
d

o
n

t
h
e

p
r
e
s
e
n
t
e
d

r
e
s
u
l
t
s
,

i
t

i
s
n
o
t

c
l
e
a
r

w
h
e
t
h
e
r

t
h
e
r
e

w
e
r
e
s
t
a
t
i
s
t
i
c
a
l
l
y

s
i
g
n
i
f
i
c
a
n
t

e
f
f
e
c
t
s
b
e
t
w
e
e
n

t
h
e

g
r
o
u
p
s
.
*
D
r
i
e
s
s
e
n
,

2
0
0
8
;
T
h
e

N
e
t
h
e
r
l
a
n
d
s
,
b
l
u
e

a
n
d

w
h
i
t
e
c
o
l
l
a
r

w
o
r
k
e
r
s

o
f
f
o
u
r

D
u
t
c
h
c
o
m
p
a
n
i
e
s

S
t
u
d
y
(
S
t
a
y
@
W
o
r
k
)
u
n
d
e
r

w
a
y
A

c
l
u
s
t
e
r
-
r
a
n
d
o
m
i
z
e
d
c
o
n
t
r
o
l
l
e
d
t
r
i
a
l

I
n
c
l
u
s
i
o
n

c
r
i
t
e
r
i
a
:

1
)
a
g
e
d

1
8
-
6
5

y


2
)

n
o

c
u
m
u
l
a
t
i
v
e

s
i
c
k

l
e
a
v
e
p
e
r
i
o
d

l
o
n
g
e
r

t
h
a
n

4

w
e
e
k
s

d
u
e

t
o

L
B
P

o
r

n
e
c
k

p
a
i
n

i
n

t
h
e

p
a
s
t

3
m
o
n
t
h
s

b
e
f
o
r
e

t
h
e

s
t
a
r
t

o
f

t
h
e

i
n
t
e
r
v
e
n
t
i
o
n

3
)

n
o
t

p
r
e
g
n
a
n
t

5
7
5
9

w
o
r
k
e
r
s

w
o
r
k
i
n
g

a
t

3
6

d
e
p
a
r
t
m
e
n
t
s

w
e
r
e

e
x
p
e
c
t
e
d

t
o
p
a
r
t
i
c
i
p
a
t
e

a
t

b
a
s
e
l
i
n
e

T
h
e

d
e
p
a
r
t
m
e
n
t
s

c
o
n
s
i
s
t
i
n
g

o
f

a
b
o
u
t

1
5
0

w
o
r
k
e
r
s

w
e
r
e

p
r
e
-
s
t
r
a
t
i
f
i
e
d

a
n
d

r
a
n
d
o
m
i
z
e
d

B
a
s
e
d

o
n

p
o
w
e
r

a
n
a
l
y
s
i
s

(
I
C
C

0
.
7
3
,

p
o
w
e
r

0
.
8
0
,

0
.
0
5
,

2
5
%
d
e
c
r
e
a
s
e

b
e
t
w
e
e
n

t
h
e

c
o
m
p
a
r
e
d

g
r
o
u
p
s
)

1
6
6
2

w
o
r
k
e
r
s

w
o
u
l
d

b
e
n
e
e
d
e
d
(
A
b
o
u
t


2
4
-
3
6

d
e
p
a
r
t
m
e
n
t
s

w
i
t
h


8
3
1

w
o
r
k
e
r
s

/

g
r
o
u
p
)

E
x
p
e
c
t
i
n
g

a

d
r
o
p
-
o
u
t

r
a
t
e

o
f

2
0
%
,

a
n

i
n
i
t
i
a
l

s
a
m
p
l
e

o
f

2
0
7
6
w
o
r
k
e
r
s

w
o
u
l
d

b
e

n
e
e
d
e
d
.

6

m
o
n
t
h
s

1
2

m
o
n
t
h
s

D
a
t
a

o
n

t
h
e
p
r
i
m
a
r
y
o
u
t
c
o
m
e
s
,
i
n
t
e
n
s
i
t
y

o
f
p
a
i
n
,

s
i
c
k
l
e
a
v
e
,

w
o
r
k
p
r
o
d
u
c
t
i
v
i
t
y
,
a
n
d

h
e
a
l
t
h
c
a
r
e

c
o
s
t
s

a
r
e
c
o
l
l
e
c
t
e
d
e
v
e
r
y

3
m
o
n
t
h
s
T
h
e

s
t
u
d
y

w
i
l
l
i
n
v
e
s
t
i
g
a
t
e

t
h
e

(
c
o
s
t
)
e
f
f
e
c
t
i
v
e
n
e
s
s

o
f


P
E

t
o
p
r
e
v
e
n
t

L
B
P

a
n
d

n
e
c
k
p
a
i
n
R
e
s
u
l
t
s

a
r
e

e
x
p
e
c
t
e
d


i
n

2
0
1
0

P
r
i
m
a
r
y

o
u
t
c
o
m
e
:

a
n

e
p
i
s
o
d
e

o
f
L
B
P

a
n
d

n
e
c
k

p
a
i
n

S
e
c
o
n
d
a
r
y

o
u
t
c
o
m
e
s
:

a
c
t
u
a
l

u
s
e

o
f
e
r
g
o
n
o
m
i
c

m
e
a
s
u
r
e
s
,

p
h
y
s
i
c
a
l
w
o
r
k
l
o
a
d
,

p
s
y
c
h
o
s
o
c
i
a
l

w
o
r
k
l
o
a
d
,
i
n
t
e
n
s
i
t
y

o
f

p
a
i
n
,

g
e
n
e
r
a
l

h
e
a
l
t
h
s
t
a
t
u
s
,

s
i
c
k

l
e
a
v
e

a
n
d

w
o
r
k
p
r
o
d
u
c
t
i
v
i
t
y

C
o
s
t
-
e
f
f
e
c
t
i
v
e
n
e
s
s

a
n
a
l
y
s
i
s

w
i
l
l

b
e
p
e
r
f
o
r
m
e
d

f
r
o
m

t
h
e

s
o
c
i
e
t
a
l

a
n
d
c
o
m
p
a
n
y

p
e
r
s
p
e
c
t
i
v
e
*
T
s
u
t
s
u
m
i
,

2
0
0
9
;
J
a
p
a
n
,

b
l
u
e
-
c
o
l
l
a
r
w
o
r
k
e
r
s
m
a
n
u
f
a
c
t
u
r
i
n
g
e
l
e
c
t
r
o
n
i
c
e
q
u
i
p
m
e
n
t
A

c
l
u
s
t
e
r
-
r
a
n
d
o
m
i
z
e
d
c
o
n
t
r
o
l
l
e
d
t
r
i
a
l
n

=

1
1

a
s
s
e
m
b
l
y

l
i
n
e
s
n

=

9
7

w
o
r
k
e
r
s
n
=

6

l
i
n
e
s
n

=

4
7

w
o
r
k
e
r
s
n
=

5

l
i
n
e
s
n
=

5
0

w
o
r
k
e
r
s
A
p
p
r
o
x
i
m
a
t
e
-
l
y

1
2

m
o
n
t
h
s
T
o

e
x
p
l
o
r
e

t
h
e

e
f
f
e
c
t

o
f

a
p
a
r
t
i
c
i
p
a
t
o
r
y

i
n
t
e
r
v
e
n
t
i
o
n
f
o
r

w
o
r
k
p
l
a
c
e

i
m
p
r
o
v
e
-
m
e
n
t

o
n

m
e
n
t
a
l

h
e
a
l
t
h
a
n
d

j
o
b

p
e
r
f
o
r
m
a
n
c
e

G
e
n
e
r
a
l

H
e
a
l
t
h
Q
u
e
s
t
i
o
n
n
a
i
r
e

(
G
H
Q
)
a
n
d

H
e
a
l
t
h

a
n
d

W
o
r
k
P
e
r
f
o
r
m
a
n
c
e
Q
u
e
s
t
i
o
n
n
a
i
r
e

(
H
P
Q
)

G
H
Q

s
c
o
r
e
s

r
e
m
a
i
n
e
d

a
t

t
h
e

s
a
m
e
l
e
v
e
l

i
n

t
h
e

i
n
t
e
r
v
e
n
t
i
o
n

g
r
o
u
p
,

b
u
t
d
e
t
e
r
i
o
r
a
t
e
d

i
n

t
h
e

c
o
n
t
r
o
l

g
r
o
u
p

H
P
Q

s
c
o
r
e
s

i
n
c
r
e
a
s
e
d

i
n

t
h
e
i
n
t
e
r
v
e
n
t
i
o
n

g
r
o
u
p
,

b
u
t

d
e
c
r
e
a
s
e
d

i
n
t
h
e

c
o
n
t
r
o
l

g
r
o
u
p

(
p

=

0
.
0
4
8
)
.

A
f
t
e
r
i
m
p
u
t
a
t
i
o
n

o
f

m
i
s
s
i
n
g

v
a
l
u
e
s
,

t
h
e
r
e
w
a
s

n
o

s
t
a
t
i
s
t
i
c
a
l
l
y

s
i
g
n
i
f
i
c
a
n
t
d
i
f
f
e
r
e
n
c
e

b
e
t
w
e
e
n

t
h
e

g
r
o
u
p
s
.

E
f
f
e
c
t

s
i
z
e
:

n
o
t

g
i
v
e
n
4
8
T
a
b
l
e

2

(
c
o
n
t
i
n
u
e
d
)
F
i
r
s
t

a
u
t
h
o
r
Y
e
a
r
I
n
d
u
s
t
r
y
/
s
e
c
t
o
r
S
t
u
d
y

d
e
s
i
g
n
P
a
r
t
i
c
i
p
a
n
t
s
F
o
l
l
o
w
-
u
p
I
n
t
e
r
v
e
n
t
i
o
n
/

A
i
m
E
f
f
e
c
t
s

o
n

M
S
D
s
/
p
s
y
c
h
o
s
o
c
i
a
l
f
a
c
t
o
r
s

a
t

w
o
r
k
T
o
t
a
l
I
n
t
e
r
v
e
n
t
i
o
n
C
o
n
t
r
o
l
*
M
o
r
k
e
n
,

2
0
0
2
;
N
o
r
w
a
y
,
a
l
u
m
i
n
i
u
m

i
n
d
u
s
t
r
y
A

c
l
u
s
t
e
r
-
r
a
n
d
o
m
i
z
e
d
c
o
n
t
r
o
l
l
e
d
t
r
i
a
l
A
l
l

w
o
r
k
e
r
s

f
r
o
m

8

p
l
a
n
t
s
c
o
m
p
l
e
t
e
d

t
h
e

q
u
e
s
t
i
o
n
n
a
i
r
e
.
B
a
s
e
l
i
n
e

n
=
5
6
5
4
,

f
o
l
l
o
w
-
u
p
n
=
5
1
4
3
,

m
a
t
c
h
e
d

n
=

3
3
2
1
,
n
=

9
5

g
r
o
u
p
s
n
=

2
1
8
1

o
p
e
r
a
t
o
r
s
W
o
r
k
e
r
s

f
r
o
m

t
h
e

p
r
o
d
u
c
t
i
o
n
l
i
n
e

w
e
r
e

r
a
n
d
o
m
i
z
e
d
n
=

4
0

g
r
o
u
p
s
,
n
=

4
1
4
o
p
e
r
a
t
o
r
s
3

I
n
t
e
r
v
e
n
t
i
o
n
g
r
o
u
p
s

(
I
)
:
I
1
)

'
s
h
i
f
t

g
r
o
u
p
w
i
t
h

s
u
p
e
r
v
i
s
o
r
'
(
n
=
2
0

g
r
o
u
p
s

o
f
o
p
e
r
a
t
o
r
s

&
s
u
p
e
r
v
i
s
o
r
s
,

n
=

1
3
2
o
p
e
r
a
t
o
r
s
)
I
2
)

'
s
h
i
f
t

g
r
o
u
p
w
i
t
h
o
u
t
s
u
p
e
r
v
i
s
o
r
s
'
(
n
=
1
8

g
r
o
u
p
s

o
f
o
p
e
r
a
t
o
r
s

o
n
l
y
,
n
=

1
3
5
)
I
3
)

'
m
a
n
a
g
e
r
s
o
n
l
y
'
,

(
n
=

2
g
r
o
u
p
s

o
f
s
u
p
e
r
v
i
s
o
r
s

&
m
a
n
a
g
e
r
s

a
n
d
1
0

g
r
o
u
p
s

o
f
o
p
e
r
a
t
o
r
s
n
=

1
4
7
)
n

=

5
5

g
r
o
u
p
s
n
=

1
7
6
7
o
p
e
r
a
t
o
r
s
C
o
n
t
r
o
l

g
r
o
u
p
A

(
C
A
)
n
=

5
5

g
r
o
u
p
s
n
=

4
2
3
o
p
e
r
a
t
o
r
s
C
o
n
t
r
o
l

g
r
o
u
p
B

(
C
B
)
F
r
o
m

o
t
h
e
r
p
a
r
t
s

o
f

t
h
e
s
a
m
e

p
l
a
n
t
,

n
=
1
3
4
4
o
p
e
r
a
t
o
r
s
,
n
o
n

r
a
n
d
o
m
i
-
z
e
d
1
6

m
o
n
t
h
s
E
x
a
m
i
n
a
t
i
o
n

o
f

t
h
e

e
f
f
e
c
t
s
o
f

a

1
-
y
e
a
r

P
E

p
r
o
g
r
a
m

o
n
M
S
D
s
,

c
o
p
i
n
g

s
t
r
a
t
e
g
i
e
s
,
j
o
b

d
e
m
a
n
d
s
,

c
o
n
t
r
o
l

a
n
d
s
o
c
i
a
l

s
u
p
p
o
r
t

M
S
D
s

i
n
c
r
e
a
s
e
d

i
n

t
h
e

C
B
(
s
h
o
u
l
d
e
r
s

p
=
0
.
0
1
4
;

h
a
n
d
s
p
<
0
.
0
0
1
;

e
l
b
o
w
s

p
<
0
.
0
0
1
)
.

N
o
s
t
a
t
i
s
t
i
c
a
l
l
y

s
i
g
n
i
f
i
c
a
n
t

c
h
a
n
g
e
s
i
n

t
h
e

o
t
h
e
r

g
r
o
u
p
s
.

T
h
e

c
h
a
n
g
e
d
i
d

n
o
t

d
i
f
f
e
r

s
t
a
t
i
s
t
i
c
a
l
l
y
s
i
g
n
i
f
i
c
a
n
t
l
y

b
e
t
w
e
e
n

t
h
e
i
n
t
e
r
v
e
n
t
i
o
n

a
n
d

c
o
n
t
r
o
l

g
r
o
u
p
s
.

T
h
e

i
n
t
e
r
v
e
n
t
i
o
n

g
r
o
u
p
s

u
s
e
d
m
o
r
e

c
o
p
i
n
g

s
t
r
a
t
e
g
i
e
s

t
h
a
n
c
o
n
t
r
o
l

g
r
o
u
p
s

(
p
=
0
.
0
4
3
)
.

I
2
(
'
o
p
e
r
a
t
o
r
s

w
i
t
h
o
u
t

a
s
u
p
e
r
v
i
s
o
r
s
'
)

i
n
c
r
e
a
s
e
d

t
h
e

m
o
s
t
(
m
e
a
n

c
h
a
n
g
e
=
0
.
0
4
1
,

9
5
%

C
I
0
.
0
0
5
-
0
.
0
7
7
)
.

I
2

a
n
d

C
B

d
i
f
f
e
r
e
d
(
p
=
0
.
0
1
7
)
.

S
o
c
i
a
l

s
u
p
p
o
r
t

i
n

t
h
e
I
2

i
m
p
r
o
v
e
d

s
l
i
g
h
t
l
y

f
r
o
m

p
r
e

t
o
p
o
s
t
.

A
l
l

o
t
h
e
r

g
r
o
u
p
s

t
e
n
d
e
d

t
o
d
e
c
l
i
n
e

(
p
=
0
.
1
0
)
.

J
o
b

d
e
m
a
n
d
s
a
n
d

c
o
n
t
r
o
l

d
i
d

n
o
t

d
i
f
f
e
r
s
t
a
t
i
s
t
i
c
a
l
l
y

s
i
g
n
i
f
i
c
a
n
t
l
y
.

E
f
f
e
c
t

s
i
z
e
:

n
o
t

g
i
v
e
n
4
9
T
a
b
l
e

2

(
c
o
n
t
i
n
u
e
d
)
F
i
r
s
t

a
u
t
h
o
r
Y
e
a
r
I
n
d
u
s
t
r
y
/
s
e
c
t
o
r
S
t
u
d
y

d
e
s
i
g
n
P
a
r
t
i
c
i
p
a
n
t
s
F
o
l
l
o
w
-
u
p
I
n
t
e
r
v
e
n
t
i
o
n
/

A
i
m
E
f
f
e
c
t
s

o
n

M
S
D
s
/
p
s
y
c
h
o
s
o
c
i
a
l
f
a
c
t
o
r
s

a
t

w
o
r
k
T
o
t
a
l
I
n
t
e
r
v
e
n
t
i
o
n
C
o
n
t
r
o
l
K
e
t
o
l
a
,

2
0
0
2
;
F
i
n
l
a
n
d
,
m
u
n
i
c
i
p
a
l

o
f
f
i
c
e
w
o
r
k
e
r
s

u
s
i
n
g
v
i
d
e
o

d
i
s
p
l
a
y

u
n
i
t
s
Ar
a
n
d
o
m
i
z
e
d
c
o
n
t
r
o
l
l
e
d
t
r
i
a
l
n

=

1
2
4
I
n
t
e
r
v
e
n
t
i
o
n
g
r
o
u
p
s
:
1
)

i
n
t
e
n
s
i
v
e
e
r
g
o
n
o
m
i
c
s
g
r
o
u
p
,

n
=

3
7
2
)

e
d
u
c
a
t
i
o
n
g
r
o
u
p
,

n
=

3
5
n
=

3
9

2

m
o
n
t
h
s

1
0

m
o
n
t
h
s
E
v
a
l
u
a
t
i
o
n

o
f

t
h
e

e
f
f
e
c
t

o
f
a
n

i
n
t
e
n
s
i
v
e

e
r
g
o
n
o
m
i
c
a
p
p
r
o
a
c
h

a
n
d

e
d
u
c
a
t
i
o
n

o
n
w
o
r
k
s
t
a
t
i
o
n

c
h
a
n
g
e
s

a
n
d
M
S
D
s

(
d
i
s
c
o
m
f
o
r
t
,

s
t
r
a
i
n
p
a
i
n
)
.

A
t

2

m
o
n
t
h
s

f
o
l
l
o
w
-
u
p
,

i
n

t
h
e
i
n
t
e
n
s
i
v
e

v
s
.

c
o
n
t
r
o
l

g
r
o
u
p
,
s
i
g
n
i
f
i
c
a
n
t

(
p
=
0
.
0
0
1
-
0
.
0
2
5
)
r
e
d
u
c
t
i
o
n

i
n

d
i
s
c
o
m
f
o
r
t

i
n

t
h
e
n
e
c
k

(
m
e
a
n

2
.
7

v
s
.

3
.
3
)
,

a
r
e
a
b
e
t
w
e
e
n

n
e
c
k

a
n
d

s
h
o
u
l
d
e
r
(
r
i
g
h
t

s
i
d
e
)

(
2
.
5

v
s
.

3
.
1
)
,
s
h
o
u
l
d
e
r
s

(
2
.
2

v
s
.

2
.
8

r
i
g
h
t
,

1
.
9
v
s
.

2
.
4

l
e
f
t
)
,

r
i
g
h
t

f
o
r
e
a
r
m

(
2
.
1
v
s
.

2
.
5
)
,

l
e
f
t

f
i
n
g
e
r
s

(
1
.
8

v
s
.
2
.
3
)
,

a
n
d

u
p
p
e
r

b
a
c
k

(
2
.
2

v
s
.
2
.
9
)
.

I
n

t
h
e

e
d
u
c
a
t
i
o
n

g
r
o
u
p
,
s
i
m
i
l
a
r

c
h
a
n
g
e
s

(
p
=

0
.
0
0
2
-
0
.
0
0
9
)

e
x
c
e
p
t

s
h
o
u
l
d
e
r
s
.

A
t

1
0
m
o
n
t
h
s

f
o
l
l
o
w
-
u
p
,

n
o
s
t
a
t
i
s
t
i
c
a
l
l
y

s
i
g
n
i
f
i
c
a
n
t

c
h
a
n
g
e
s
w
e
r
e

d
e
t
e
c
t
e
d

i
n

b
o
t
h

g
r
o
u
p
s
c
o
m
p
a
r
e
d

t
o

c
o
n
t
r
o
l

g
r
o
u
p

i
n
d
i
s
c
o
m
f
o
r
t
,

s
t
r
a
i
n

o
r

p
a
i
n
.

E
f
f
e
c
t

s
i
z
e
:

n
o
t

g
i
v
e
n
*
B
o
h
r
,

2
0
0
0
;

S
t
.
L
o
u
i
s

M
i
s
s
o
u
r
i
,
U
S
A
,
r
e
s
e
r
v
a
t
i
o
n
c
e
n
t
e
r
/
o
f
f
i
c
e

(
u
s
i
n
g
m
o
s
t
l
y

c
o
m
p
u
t
e
r
)
Ar
a
n
d
o
m
i
z
e
d
c
o
n
t
r
o
l
l
e
d
t
r
i
a
l
S
u
b
j
e
c
t
s

w
e
r
e

s
e
l
e
c
t
e
d

a
t
r
a
n
d
o
m

f
r
o
m

a

l
i
s
t

o
f
v
o
l
u
n
t
e
e
r
s

w
h
o

w
e
r
e
e
m
p
l
o
y
e
d

a
s

a
g
e
n
t
s

a
t

t
h
e
c
e
n
t
r
a
l
i
z
e
d

r
e
s
e
r
v
a
t
i
o
n
,

u
s
e
d
c
o
m
p
u
t
e
r
s

a
t

l
e
a
s
t

5

h

p
e
r
w
o
r
k

d
a
y
n
=

1
5
4

a
t

b
a
s
e
l
i
n
e
n
=
1
2
4

a
t

t
h
e

1
2

m
o
n
t
h
s
2

i
n
t
e
r
v
e
n
t
i
o
n
g
r
o
u
p
s
:
1
)

t
r
a
d
i
t
i
o
n
a
l
e
d
u
c
a
t
i
o
n
,
n
=
3
9
2
)

p
a
r
t
i
c
i
p
a
t
o
-
r
y

e
d
u
c
a
t
i
o
n
n
=

3
8
n
=

4
7

3

m
o
n
t
h
s

6

m
o
n
t
h
s

1
2

m
o
n
t
h
s
I
n
v
e
s
t
i
g
a
t
i
o
n

o
f

t
h
e

e
f
f
i
c
a
c
y
o
f

w
o
r
k
e
r

e
d
u
c
a
t
i
o
n
p
r
o
g
r
a
m
s

i
n

p
r
e
v
e
n
t
i
n
g
M
S
D
s

P
a
i
n
/
d
i
s
c
o
m
f
o
r
t

i
n

t
h
e
;
u
p
p
e
r

b
o
d
y

(
s
c
o
r
e
s

i
n
c
l
u
d
e
d
s
y
m
p
t
o
m
s

f
o
r

n
e
c
k
,

u
p
p
e
r
b
a
c
k
,

s
h
o
u
l
d
e
r
/
u
p
p
e
r

a
r
m
,
f
o
r
e
a
r
m
,

a
n
d

w
r
i
s
t
/

h
a
n
d
)
a
n
d

l
o
w
e
r

b
o
d
y

(
l
o
w
e
r

b
a
c
k
,
h
i
p
/
b
u
t
t
o
c
k
s
,

k
n
e
e
s
,

a
n
d
f
e
e
t
/
a
n
k
l
e
)

P
s
y
c
h
o
s
o
c
i
a
l

w
o
r
k

s
t
r
e
s
s

(
6
i
t
e
m
s
)

U
p
p
e
r

b
o
d
y

s
c
o
r
e
:

c
o
n
t
r
o
l
g
r
o
u
p

r
e
p
o
r
t
e
d

h
i
g
h
e
r

f
r
e
q
u
e
n
c
y
o
f

p
a
i
n
/
d
i
s
c
o
m
f
o
r
t

t
h
r
o
u
g
h
o
u
t
t
h
e

s
t
u
d
y

t
h
a
n

e
i
t
h
e
r
i
n
t
e
r
v
e
n
t
i
o
n

g
r
o
u
p
s
[
F

(
2
,
1
5
1
)

=

4
.
8
6
,

p
<
0
.
0
1
)
]

N
o
s
t
a
t
i
s
t
i
c
a
l
l
y

s
i
g
n
i
f
i
c
a
n
t
d
i
f
f
e
r
e
n
c
e

b
e
t
w
e
e
n

t
r
a
d
i
t
i
o
n
a
l
a
n
d

p
a
r
t
i
c
i
p
a
t
o
r
y

e
d
u
c
a
t
i
o
n
g
r
o
u
p
.

L
o
w
e
r

b
o
d
y

s
c
o
r
e
:

n
o
s
t
a
t
i
s
t
i
c
a
l
l
y

s
i
g
n
i
f
i
c
a
n
t
d
i
f
f
e
r
e
n
c
e

a
c
r
o
s
s

t
h
e

g
r
o
u
p
s
.
50
2.5. Kitchen work as a target of an ergonomics intervention
The present study was targeted to kitchen work. Kitchen workers are mainly women and
they suffer a high prevalence of MSDs (Huang et al. 1988; Ono et al. 1997; Ono et al.
1998; Perki-Mkel et al. 2006; Nagasu et al. 2007; Shiue et al. 2008). The work is
physically demanding (Shibata et al. 1991; Ono et al. 1997; Ono et al. 1998) and it often
involves ergonomic problems (Huang et al. 1988; Shibata et al. 1991). The HORECA
sector (hotels, restaurants, catering) employs about eight million people in the European
Union. In 2000, almost 55% of the EU-15 workers in the sector reported that their job
involved painful or tiring positions compared to 45% across all sectors. The corresponding
figures for carrying or moving heavy loads were 43% compared to 36%, and for repetitive
hand or arm movements 64% compared to 56% overall. Pain in lower limbs is more
common in the HORECA sector than elsewhere (European Agency for Safety and Health
at Work 2008).
In a large Japanese cross-sectional survey among cooks (n= 5835) working in school
lunch services, the prevalence of LBP and its associations with various risk factors were
recently examined (Nagasu et al. 2007). About 83% of the workers were women (mean
age 48 years). Female gender was associated with LBP [prevalence ratio, (PR) 1.3].
Several other factors were significantly associated with the prevalence of LBP. In women,
these included no regular physical activity, the number of hours of sleep 7, and the
number of prepared lunches per cook. In both genders, current and past smoking, lack of
breaks in the morning session, physical kitchen environment (noisy surroundings, poor
state of drainage, presence of slippery surfaces, bumps and obstacles on the floor), and the
height of cooking equipment (e.g. counter-tables, kitchen sinks, cauldrons) were
associated with LBP. In addition, job dissatisfaction, stress at work, financial constraints,
health-related stress, and worries about the future were similarly associated. The PR
values varied from 1.1 to 1.7, being strongest for stress at work (PR 1.7), and current (PR
1.6) and past (PR 1.4) smoking. The annual incidence of MSDs among Chinese restaurant
cooks (n= 52 261) was observed to be higher (ORs 1.3-1.4) compared to the age and sex
matched general population. There was a trend for increasing MSD incidence with age in
both the cooks and the reference group (Shiue et al. 2008).
51
In one considers municipal occupations in Finland, then the kitchen workers are among
the top five occupations with the highest sickness absence and disability pension rates
(Forma 2004; Vahtera et al. 2008) as well as reporting high physical work load and fast
work pace, unsatisfactory working climate, poor perceived health, low correspondence
between knowhow and work, need for education, and fear of temporary dismissals or lay-
offs (Forma 2004). In the planning phase of the present intervention study, kitchen
workers were considered to represent good subjects for ergonomic improvements. Based
on observations in kitchens during the development phase of the study, many generic risk
factors for MSDs were observed. Awkward postures, repetitive and forceful movements,
and manual material handling typically occurred often in the daily routine. The work
imposed both static and dynamic load on the musculoskeletal system. The workers
performed several parallel tasks, they worked under time pressure and high mental and
physical workloads. In the study, seven main work tasks were distinguished: preparation
(e.g. washing, paring, and cutting of the groceries), cooking and baking, distribution and
serving of the food (e.g. dishing out and dosage of the food), dishwashing (e.g. sorting,
pre-washing and washing of the dirty dishes/ sorting and setting of the clean dishes),
cleaning and maintenance of room and equipment (e.g. cleaning tables and floors,
handling and transit of refuses), and receiving and storing of raw material. Packing food
into thermal transport cases was an extra task in kitchens which prepared the food to be
transported to other kitchens where the food was delivered to the clients.
Little previous research had focused on this challenging occupational group. Kitchens
were also the correct sized units suitable for a cluster randomized design. In particular,
municipal kitchens were thought to be sustaining workplaces that would allow a long
enough follow-up without an immediate threat to their continuing existence.
52
3. THEORETICAL FRAMEWORK OF THE STUDY
There are several theoretical models describing the multifactorial etiology of MSDs
(Huang et al. 2002; Karsh 2006). These are needed to target interventions that might
prevent or reduce MSDs. The theories also guide in the selection of variables that should
be controlled for in a study. The theoretical basis of this thesis (Figure 1) is based on the
ecological model of Sauter and Swanson (1996) that was also the framework of the entire
ERGO-study. This model, originally designed for office and VDU work, incorporates
biomechanical, psychosocial and cognitive factors, the last of these three components
being the one distinguishing it from the other models. According to this model, MSDs are
related to work technology that includes workplace characteristics, and the nature of work
processes and tools. The work technology is linked directly to physical work demands, as
defined by the physical connection between a tool and a worker, and to the work
organization. The pathway from the work organization to the physical demands proposes
that the physical demands of work are affected by organizational demands; for example,
increased repetition may be caused by increased specialization of work tasks. A direct path
is seen also between the work organization and psychosocial strain which can then affect
the biomechanical strain. The model suggests that the relationships between
biomechanical strain and the development of MSDs are mediated by complex cognitive
processes that involve the detection and sensation of symptoms and their attribution to the
musculoskeletal system. The work organization, psychological strain, and individual
factors have an influence on the connection between biomechanical strain and MSDs, and
on the manner by which workers detect and respond to physical sensations. Finally, the
model shows the reciprocal links between MSDs and the work organization and
psychological strain.
Study hypotheses based on the theoretical framework
It was hypothesized that by implementing ergonomics changes in tools, equipment and
technology aimed at optimisation of biomechanical and mental load and affecting factors
related to work organization, it would be possible to prevent MSDs and to improve the
psychosocial work environment among kitchen workers (I-II). At baseline, before the
intervention started, kitchen workers were assumed to have a high occurrence, not only of
53
single-site pain, but also of pain at multiple anatomical sites (III). In study IV, there were
two hypotheses assuming opposite time sequences: that psychosocial factors at work and
mental stress would predict the occurrence of MSP, and that MSP would predict the
occurrence of psychosocial factors at work and mental stress.
Figure 1. Framework of the study based on the conceptual model of work-related
musculoskeletal disorders, adapted from Sauter and Swanson 1996. Boxes with dash lines refer
to the measures used in the current study.

3+<6,&$/'(0$1'6

722/6
(48,30(17
7(&+12/2*<

:25.25*$1,=$7,21

%,20(&+$1,&$/675$,1
- perce|ved p|ys|ca| Wor||oad
086&8/26.(/(7$/
287&20(6
- s|r|e-s|le pa|r
- ru|l|p|e-s|le pa|r
- lrouo|e caused oy pa|r
- ruscu|os|e|ela| lal|ue
- s|c| |eaves

'(7(&7,21
6(16$7,21
/$%(/,1*
$775,%87,21
36<&+2/2*,&$/675$,1
- rerla| slress
- rerla| slreruousress ol Wor|
- |urry
- joo sal|slacl|or
- joo corlro|
- s|||| d|screl|or
- co-Wor|er re|al|ors||ps
- superv|sor supporl
,1',9,'8$/)$&7256
- ae, erder
- |ea|l|-re|aled ||le-sly|e
overWe||l / ooes|ly
sro||r, p|ys|ca| exerc|se
54
4. AIMS OF THE STUDY
The primary aim was to study the efficacy of a workplace PE intervention in preventing
MSDs among municipal kitchen workers. Second, the effects of the PE intervention on
psychosocial working conditions as intermediate outcomes were studied. Furthermore, the
occurrence of MSP and the associations between MSP and psychosocial factors at work
over time were examined.
The specific aims were:
1. To study the efficacy of a PE intervention in preventing MSDs (I).
2. To investigate the effects of a PE intervention on psychosocial factors at work as
intermediate outcomes of the intervention (II).
3. To examine the co-occurrence of musculoskeletal pain in seven distinct body sites
and their combinations in female kitchen workers (III).
4. To identify the developmental patterns of MSP and of psychosocial factors at work
during a two-year follow-up period. To study the interrelationships between
psychosocial factors at work and MSP in female kitchen workers (IV).
55
5. MATERIALS AND METHODS
5.1. General description of the study
The sub-studies comprising this dissertation are part of a larger research programme
carried out by the FIOH during the years 2002-2005 in collaboration with municipal
kitchens in four large cities in Finland. The Academy of Finland (Health Promotion
Research Programme), the Finnish Work Environment Fund, the Ministry of Labour, and
the Local Government Pensions Institution financially supported the study. The Ethics
Committee of the FIOH approved the study proposal.
At the beginning of the study, meetings were arranged in each city where the management
and kitchen workers were informed about the project and encouraged to participate. First,
meetings with representatives of foodservice management were held in order to obtain
their approval and commitment. Next, information sessions for the workers were held. In
these sessions, the aims of the study and the study protocol were described. A research
agreement was signed with each city and a written informed consent was obtained from
each kitchen and each worker who agreed to participate in the study.
Flow of the study
The flowchart of the study, with the eligibility criteria for participation, is presented in
Figure 2. Altogether 122 kitchens (60% of those eligible) were randomized into
intervention or control groups. Three kitchens dropped out immediately after the
randomization and thus 119 kitchens of schools, nurseries, and nursing homes, with a total
of 504 workers participated in the study. For feasibility reasons, the intervention (n= 59)
and control (n= 60) kitchens were divided into 16 series each including eight kitchens
(four intervention and four control kitchens) on average. The series entered the active
study phase sequentially in time. Within each series, the kitchens of both study arms
proceeded parallely. The first series started in the spring of 2002 and the last ones in the
autumn of 2003. By the end of 2004, the intervention phase was completed. Each series
had a 12-month post- intervention follow-up. The follow-up phase was completed by the
end of 2005.
56
Figure 2. Flowchart of enrollment and study design.
*
Beginning of the first series.

Beginning of the last series.

5.2. Study designs and subjects
Detailed descriptions of the study designs and study material analyzed in the various sub-
studies are given in Figure 3.
One of the sub-studies was cross-sectional (III) and focused on the co-occurrence of pain
in seven body sites at baseline. Here the subjects were composed of 523 workers from all
the initial 122 kitchens. Only 19 of the workers were men, and therefore the analyses were
restricted to women only (n=504).
Study I describes the effects of a PE intervention on MSDs and study II investigates the
effects on psychosocial factors at work. The subjects at baseline were 504 workers in 119
kitchens. Cross-sectional open sample data were collected at nine time points. The number
202 eligible kitchens
_ 3 Iull-time workers
working Ior 6 hours/day 80 kitchens reIused to participate
collective or administrative decision
_ 1/3 oI the workers individually reIused
reconstruction oI kitchen
Baseline questionnaire
3 kitchens dropped out
reconstruction oI kitchen (1)
too Iew workers participating (1)
122 kitchens randomized
(523 workers, study III)
study was considered too cumbersome (1)
119 kitchens participated
(504 workers, studies I, II, IV)
Kitchen series 1- Kitchen series 16
Intervention
kitchens:
i i
Control
kitchens:
Intervention
kitchens:
Control
kitchens:
..............
intervention
phase
11-14
months
normal
activity
intervention
phase
11-14
months
normal
activity
March 2002*
September 2003
f
57
of workers participating in different cross-sections, by study arm (59 intervention kitchens
and 60 control kitchens), are seen in Figure 3.
In study IV, the intervention and control groups were pooled together in a two- year
prospective study to investigate developmental patterns of MSP and to examine
associations between MSP and psychosocial factors at work over time. The final study
sample consisted of women with observed values of MSP in at least four time points of
nine (n=385).
Figure 3. Study designs and study materials in the original studies.
*BL; baseline, I; intervention phase, PI; post-intervention follow-up phase.

PIA, post-intervention
assessment. Five series finished the intervention at 9 months (I
9
, n = 157) and 11 series at 12
months (I
12
). The PIA was thus made either at 9 or 12 months after the start of the implementation
phase of the intervention (combined n = 452).
5.3. Sample size, randomization and blinding
Since the participants within any cluster are more likely to be similar with each other than
with the participants in another cluster, and also to have more similar outcomes, a larger
sample size is required in a CRT to obtain adequate statistical power compared to an
Study III Studies I II Study IV Study III Studies I-II Study IV
Cross-sectional Cluster randomized trial 2-year longitudinal
Randomization
3 kitchens dropped out
122 kitchens
523 workers
122 kitchens
Women Men Excluded
I t ti C t l
119 kitchens participated
Intervention Control
119 kitchens
n 504 n 19
Intervention Control
group n 59 group n 60
k l
group n 59 group n 60
263 workers 241 workers
InIormation on all
No workers Total
BL* 263 241 504
I
3
241 228 469
I 216 196 412
n 504
InIormation on all
7 anatomical sites
I
6
216 196 412
I
9
105 124 229
I
12
159 136 295
PIA
f
232 220 452
PI 142 116 258
Excluded Women Men
n 487 n 17
n 495
PI
3
142 116 258
PI
6
209 176 385
PI
9
110 122 232
PI
12
216 206 422
InIormation
_ 4 time points/ 9
n 385
58
individually randomized trial. Due to the cluster randomized design, within and between
cluster correlations were considered. Based on power calculations and assumptions
[intraclass correlation (ICC) 0.50, average of three subjects per kitchen, power of 0.80 to
detect a difference of 15% in rates between the groups with 0.05], the number of
kitchens needed was estimated as 80 per group. Although the final number of kitchens was
smaller, the sample size was sufficient because the ICC in the calculations proved to have
been overestimated. At baseline, the empirical ICCs for the musculoskeletal and
psychosocial work factor outcomes, as well as for perceived physical workload were
below 0.30. The ICCs of workers in the same city area varied from 0.0 to 0.06.
Randomization was based on stratification by area (city district) and type of kitchen
(school, nursery, home for senior citizens, other institution), and an assignment algorithm
(Alternate Ranks Design, ARD) (Bonate 2000). Kitchens in each stratum were ranked in a
descending order by the number of staff. The largest kitchen was randomized first, and
thereafter, the next two were systematically allocated in pairs to the other study arm and so
on. Randomization was carried out by an individual not otherwise involved in the study
during the field phase.
It is impossible to conduct a PE intervention in a blinded manner. However, during
baseline data collection, neither workers participating in the study nor the researchers
knew to which group the workers would be allocated since the randomization was done
after the baseline data collection (Figure 2). The researchers had no access to the
questionnaire data during the data collection, and data analysis was started only after the
follow-up data collection was completed.
5.4. Data collection, worker turnover and loss to follow-up
Data were gathered from questionnaires which a researcher distributed to all participating
kitchens in both study arms. The baseline information (BL) was collected before
randomization and thereafter every three months during the intervention (I3, I6, I9, I12) and
the 12-month post- intervention follow-up period (PI3, PI6, PI9, PI12). Due to summer
holidays at schools, at least one questionnaire was omitted in each kitchen series. Five
series finished the intervention at 9 months (I
9
, n = 157) and 11 series at 12 months (I
12
, n
59
= 295). The post-intervention assessment (PIA) was thus made at either 9 or 12 months
after the start of the implementation phase (combined n = 452). The next questionnaire
(PI3) was always three months after the PIA. Extensive questionnaire data were collected
at the BL and after the intervention phase (PIA) and a shorter questionnaire was
distributed every three months during the intervention and the 12-month post-intervention
follow-up period.
Response rates in different surveys varied between 92% and 99%. The proportion of
workers employed in the same kitchen throughout the intervention phase was 86% in the
intervention group and 84% in the control group. The respective values for the 12-month
follow-up were 70% and 71%.
5.5. The participatory ergonomics intervention
The framework of the intervention was based on a model developed at the FIOH
(Leppnen 2001). The intervention process is described in Figure 4. Active group work
was emphasized, and workers were regarded as actors identifying problems, planning and
evaluating changes, and implementing them in collaboration with the management and
technical staff. The researchers provided guidance and support, and facilitated the progress
of the process. The intervention consisted of a 2-month pre-implementation and 9-12-
month implementation phase promoted by eight workshops (28 hours in total). Thus the
duration of the intervention was 11-14 months. In two cities, a local steering group was
established for improving the exchange of information between the research group and
food service management.
The field study was carried out by four teams of two researchers trained in ergonomics
and in good practice in kitchen work. One researcher (ergonomist) was responsible for
supporting the intervention process. The main task of the other researcher was to assess
the state of the ergonomics, to document the implemented changes and to distribute the
questionnaires. Regular meetings were held to ascertain the similarity of working methods
between the research teams. The project coordinator supervised the workshops and
provided feedback to the researchers.
60
The foodservice managers and technical staff were invited to participate in the workshops.
The kitchens were not provided with extra funding in association with the CRT and the
ergonomics changes were implemented within their annual budgets. The municipal
authorities agreed to prioritize the intervention kitchens in case of special needs arising
during the intervention phase.
The control kitchens continued their normal activity. During 3-monthly visits, the
researchers distributed questionnaires and collected information of all spontaneously
implemented ergonomics changes and undertook short interviews. There were no other
contacts with the researchers in these kitchens.
Pre-implementation phase
In the 2-month pre-implementation phase, all staff of the intervention kitchens were
gathered together for two 5-hour workshops. In the first session, the workers were taught
the basics of ergonomics and guided to analyze their work tasks and processes with the
intention of identifying strenuous tasks and risk factors of MSDs. The kitchen staff had
one month's time to continue the work analysis in their own kitchens and develop ideas to
decrease their physical and mental workload. The ergonomist visited each kitchen once
and gave them one supportive phone call. In the second workshop, each kitchen decided
on their primary targets and planned how these would be best implemented.
Implementation phase
During the 9-12 month implementation phase, all staff of the intervention kitchens
convened six times for a 3-hour workshop. Each workshop included a specific theme
related to ergonomics and the progress of the intervention in each kitchen was thoroughly
discussed. The ergonomist visited the kitchens if requested and provided support to the
process. The workshops rotated from one kitchen to another to give the workers an
opportunity to learn from each other's solutions and practices. The personnel of each
kitchen kept a detailed diary about the ergonomic changes made. Based on the diaries, the
researchers carefully recorded all ergonomic changes and evaluated the significance of the
changes with regard to the load on the musculoskeletal system or occupational safety. The
changes were classified according to the work tasks and by target.
61
Figure 4. Intervention process. WS, workshop.
Unconnected organizational reforms
An organizational reform of foodservices occurred simultaneously with the intervention
study in two of the participating cities. In one city, the planning and implementation of the
reform were executed concurrently with the intervention and in one other city, the reform
was planned during the intervention and the new organization started during the 12-month
post-intervention follow-up. In the new model, the major change was that cooking was
centralized to large production kitchens, from which meals were delivered to other
kitchens to be distributed to the clients. In addition, discussions about the possible
outsourcing of the foodservices or organizing them as a public utility were underway in
the two cities. Altogether 31 intervention and 31control kitchens were involved in these
organizational reforms. In study II, that examined the effects of the ergonomics
intervention on psychosocial factors at work, independent and joint effects of the
intervention and organizational reforms were studied in a secondary analysis.
WS 5 WS 4 WS 3
I t ti
Work analysis WS 1
Implementation
visits by need
Intervention
kitchens:
Intervention
by workers continue
1 visit
1 phone-call
Analysis oI
work tasks and
processes
visits by need
workshop rotation
no extra Iunding
WS 2
Primary targets
WS 8 WS 7 WS 6
Control kitchens:
Normal activity
3-monthly visits: questionnaires, inIormation oI spontaneous ergonomic changes
2 months
Pre-implementation phase
9 -12 months
Implementation phase
In total 11-14 months
Figure 4. Intervention process. WS, workshop.
Unconnected organizational reforms
An organizational reform of foodservices occurred simultaneously with the intervention
study in two of the participating cities. In one city, the planning and implementation of the
reform were executed concurrently with the intervention and in one other city, the reform
was planned during the intervention and the new organization started during the 12-month
post-intervention follow-up. In the new model, the major change was that cooking was
centralized to large production kitchens, from which meals were delivered to other
kitchens to be distributed to the clients. In addition, discussions about the possible
outsourcing of the foodservices or organizing them as a public utility were underway in
the two cities. Altogether 31 intervention and 31control kitchens were involved in these
organizational reforms. In study II, that examined the effects of the ergonomics
intervention on psychosocial factors at work, independent and joint effects of the
intervention and organizational reforms were studied in a secondary analysis.
62
5.6. Measures
5.6.1. Musculoskeletal disorders and multiple-site musculoskeletal pain
Musculoskeletal outcomes were measured by modified questions from the validated
Nordic Musculoskeletal Questionnaire (Kuorinka et al. 1987).
Musculoskeletal pain, trouble caused by pain and sick leave due to any pain during the
past 3 months
The questionnaires contained the following question on musculoskeletal pain in the neck,
shoulders, forearms/hands, low back, hips, knees and ankles/feet: "Have you had 'x' pain
during the past 3 months (no/yes) ?" Trouble and sick leave due to pain were asked
separately for each of the seven anatomical sites by the questions: "Please, assess how
much trouble 'x' pain has caused during the past 3 months (1 = 'not at all', 7 = 'very much'),
and "Have you been on the sick leave due to 'x' pain during the past 3 months (no/yes)?
Localized fatigue after the working day during the past 7 days
Musculoskeletal fatigue was asked by a question: "How much bodily fatigue you have felt
after the working day during the past 7 days (1 = 'not at all', 6 = 'very much').
Multiple-site pain
A sum index (0 = 'no pain', 7 = 'pain in seven sites') was calculated to describe the
occurrence of pain in multiple body sites during the past 3 months. The sum dichotomized
to 0-2 pain sites (no MSP) and 3 pain sites (MSP) was used as an outcome measure. A
sensitivity analysis was conducted by using 4 pain sites as the cut-off point. To obtain a
view of concurrent pain in the body, the occurrence of all combinations of the seven
anatomical sites was examined. In addition, the sites were combined to three larger
anatomical areas: the axial (neck and low back), the upper limbs (shoulders,
forearms/hands), and lower limbs (hips, knees, ankles/feet).
63
5.6.2. Psychosocial factors at work and mental stress
Psychosocial factors at work were assessed using questions adapted from a validated
questionnaire (Elo et al. 1992). All items had initially five categories that were
dichotomized to represent the presence or absence of the outcome as shown below.
Mental stress during the past month. "Stress refers to a situation in which a person feels
tense, restless, nervous or anxious, or is unable to sleep at night because his/her mind is
troubled all the time. Have you felt like this during the past month?" (1= not at all, 2= only
a little, 3= to some extent 4= rather much, 5= very much); dichotomized as no (1-3)/ yes
(4-5).
Mental strenuousness of work. "Is your work mentally strenuous?" 1= not at all, 2= rather
light, 3= somewhat strenuous, 4= rather strenuous, 5= very strenuous. Dichotomized as no
(1-3)/ yes (4-5).
Hurry at work. "Do you have to hurry to get your work done?" (1= never, 2= rather
seldom, 3= now and then, 4= rather often, 5= constantly); dichotomized as no (1-3)/ yes
(4-5).
Job dissatisfaction. "How satisfied are you with your present work?" 1= very satisfied, 2=
rather satisfied, 3= neither satisfied nor dissatisfied, 4= rather dissatisfied, 5= very
dissatisfied. Dichotomized as no (1-3)/ yes (4-5).
Poor co-worker relationships. "How do workmates get along at your workplace?" (1=
very well, 2= rather well, 3=neither well nor badly, 4= there are some problems, 5=
badly); dichotomized as no (1-3)/ yes (4-5).
Low job control. "At work, can you influence matters concerning yourself?"
Low skill discretion. "Can you use your knowledge and skills in your work?"
Low supervisor support. "Does your supervisor provide support and help when needed"?
These questions had similar classes (1= very much, 2= rather much, 3= to some extent, 4=
very little, 5= not at all) and were dichotomized as no (1-3) / yes (4-5).
64
5.6.3. Perceived physical workload
The question regarding the perceived physical workload was constructed specifically for
the kitchen work since no suitable published methods for the purpose were available. The
perceived physical workload (1= 'not at all', 7= 'very strenuous') of seven different work
tasks (preparation, cooking and baking, distribution and serving of food, packing food to
be delivered to clients, dishwashing, cleaning and maintenance of room and equipment,
receiving and storing of raw material) was inquired. The mean was used in the analyses.
The use of seven categories was based on the previous findings suggesting that the
minimum number of categories should be between five to seven. In addition, up to seven,
the reliability of the measure increases, but thereafter addition of more categories causes a
decline in the reliability (Streiner and Norman 1995).
5.6.4. Covariates
Job title (foodservice manager, chef, cook, kitchen aid, other), years employed in kitchen
work and employment (permanent, full-time) were characteristics of work history. Age,
gender, height, weight, body mass index (BMI, kg/m
2
), overweight (no overweight; BMI
< 25/ overweight; BMI 25 kg/m
2
), current regular smoking (no/yes), and physical
exercise ( once a week/ > once a week) were other items included. The question
regarding physical exercise was worded as follows: "During the past 12 months, how
many times a week have you exercised at least 20 min per session, to the extent to cause
perspiration?" (not at all, less than once a week, once a week, 2-3 times a week, 4-5 times
a week, 6-7 times a week). A sum index of musculoskeletal pain (0 = 'no pain', 7 = 'pain in
seven sites'), the mean of the perceived physical workload, organizational reforms
(no/yes), study arm (intervention/control group), and city were also considered as
covariates.
Type of kitchen (school, nursery, nursing home, geriatric service centre, other) was used
as a cluster level characteristic.
The variables and their roles as outcomes, determinants or covariates, and statistical
analyses in the original studies are summarized in Table 3.
6
5
T
a
b
l
e

3
.
S
t
u
d
y

d
e
s
i
g
n
s
,

s
u
b
j
e
c
t
s
,

d
e
f
i
n
i
t
i
o
n

o
f

v
a
r
i
a
b
l
e
s

a
n
d

t
h
e
i
r

r
o
l
e
s

a
s

o
u
t
c
o
m
e
s
,

d
e
t
e
r
m
i
n
a
n
t
s

o
r

c
o
v
a
r
i
a
t
e
s
,

a
n
d

s
t
a
t
i
s
t
i
c
a
l

a
n
a
l
y
s
e
s

i
n

t
h
e

o
r
i
g
i
n
a
l
s
t
u
d
i
e
s

S
t
u
d
y

D
e
s
i
g
n

a
n
d

S
u
b
j
e
c
t
s

M
a
i
n

o
u
t
c
o
m
e
s

I
n
t
e
r
m
e
d
i
a
t
e

o
u
t
c
o
m
e
s
/

D
e
t
e
r
m
i
n
a
n
t
s

C
o
v
a
r
i
a
t
e
s

S
t
a
t
i
s
t
i
c
a
l

a
n
a
l
y
s
e
s

I

C
l
u
s
t
e
r

r
a
n
d
o
m
i
z
e
d


c
o
n
t
r
o
l
l
e
d

t
r
i
a
l

1
1
9

k
i
t
c
h
e
n
s

5
0
4

w
o
r
k
e
r
s


M
u
s
c
u
l
o
s
k
e
l
e
t
a
l

p
a
i
n


T
r
o
u
b
l
e

c
a
u
s
e
d

b
y

p
a
i
n











S
i
c
k

l
e
a
v
e

d
u
e

t
o

a
n
y

m
u
s
c
u
l
o
s
k
e
l
e
t
a
l


p
a
i
n





L
o
c
a
l

m
u
s
c
u
l
o
s
k
e
l
e
t
a
l

I
a
t
i
g
u
e

a
I
t
e
r

w
o
r
k
d
a
y


d
u
r
i
n
g

t
h
e

p
a
s
t

7

d
a
y
s

s
e
v
e
n

a
n
a
t
o
m
i
c
a
l

s
i
t
e
s
:

n
e
c
k
,

s
h
o
u
l
d
e
r
s
,

I
o
r
e
a
r
m
s
/
h
a
n
d
s
,

l
o
w

b
a
c
k
,

h
i
p
s
,

k
n
e
e
s
,

a
n
k
l
e
s
/
I
e
e
t

P
e
r
c
e
i
v
e
d

p
h
y
s
i
c
a
l

w
o
r
k
l
o
a
d

M
e
n
t
a
l

s
t
r
e
s
s

d
u
r
i
n
g

t
h
e

p
a
s
t

m
o
n
t
h

M
e
n
t
a
l

s
t
r
e
n
u
o
u
s
n
e
s
s

o
I

w
o
r
k

J
o
b

s
a
t
i
s
I
a
c
t
i
o
n

B
a
s
e
l
i
n
e

l
e
v
e
l

o
I

t
h
e

o
u
t
c
o
m
e


J
o
b

s
a
t
i
s
I
a
c
t
i
o
n

A
g
e

G
e
n
d
e
r

S
m
o
k
i
n
g

B
o
d
y

m
a
s
s

i
n
d
e
x

P
h
y
s
i
c
a
l

e
x
e
r
c
i
s
e

P
e
r
m
a
n
e
n
t

v
s
.

I
i
x
e
d
-
t
e
r
m


e
m
p
l
o
y
m
e
n
t

P
r
e
v
a
l
e
n
c
e

r
a
t
e
s

o
r

m
e
a
n

s
c
o
r
e
s

o
I

c
r
o
s
s
-
s
e
c
t
i
o
n
a
l

o
p
e
n

p
o
p
u
l
a
t
i
o
n

d
a
t
a

T
h
e

g
e
n
e
r
a
l
i
z
e
d

e
s
t
i
m
a
t
i
n
g

e
q
u
a
t
i
o
n
s
m
o
d
e
l
,

G
E
E

I
I

C
l
u
s
t
e
r

r
a
n
d
o
m
i
z
e
d

c
o
n
t
r
o
l
l
e
d

t
r
i
a
l

1
1
9

k
i
t
c
h
e
n
s

5
0
4

w
o
r
k
e
r
s


M
e
n
t
a
l

s
t
r
e
s
s

d
u
r
i
n
g

t
h
e

p
a
s
t

m
o
n
t
h

M
e
n
t
a
l

s
t
r
e
n
u
o
u
s
n
e
s
s

o
I

w
o
r
k

H
u
r
r
y

J
o
b

d
i
s
s
a
t
i
s
I
a
c
t
i
o
n

L
o
w

j
o
b

c
o
n
t
r
o
l

L
o
w

s
k
i
l
l

d
i
s
c
r
e
t
i
o
n

P
o
o
r

c
o
-
w
o
r
k
e
r

r
e
l
a
t
i
o
n
s
h
i
p
s

L
o
w

s
u
p
e
r
v
i
s
o
r

s
u
p
p
o
r
t


B
a
s
e
l
i
n
e

l
e
v
e
l

o
I

t
h
e

o
u
t
c
o
m
e


A
g
e

M
u
s
c
u
l
o
s
k
e
l
e
t
a
l

p
a
i
n

P
e
r
c
e
i
v
e
d

p
h
y
s
i
c
a
l

w
o
r
k
l
o
a
d

C
i
t
y


P
r
e
v
a
l
e
n
c
e

r
a
t
e
s

o
I

c
r
o
s
s
-
s
e
c
t
i
o
n
a
l

o
p
e
n

p
o
p
u
l
a
t
i
o
n

d
a
t
a

T
h
e

g
e
n
e
r
a
l
i
z
e
d

e
s
t
i
m
a
t
i
n
g

e
q
u
a
t
i
o
n
s
m
o
d
e
l
,

G
E
E

I
I
I

C
r
o
s
s
-
s
e
c
t
i
o
n
a
l

5
2
3

w
o
r
k
e
r
s


1
9

m
e
n

w
e
r
e

e
x
c
l
u
d
e
d

5
0
4

w
o
m
e
n

w
i
t
h

i
n
I
o
r
m
a
t
i
o
n

o
n

a
l
l

7

a
n
a
t
o
m
i
c
a
l

s
i
t
e
s

n

4
9
5

M
u
s
c
u
l
o
s
k
e
l
e
t
a
l

p
a
i
n

d
u
r
i
n
g

t
h
e

p
a
s
t

3

m
o
n
t
h
s

i
n

7

a
n
a
t
o
m
i
c
a
l

s
i
t
e
s
:

n
e
c
k
,

s
h
o
u
l
d
e
r
s
,

I
o
r
e
a
r
m
s
/
h
a
n
d
s
,

l
o
w

b
a
c
k
,

h
i
p
s
,

k
n
e
e
s
,

a
n
k
l
e
s
/
I
e
e
t

T
h
r
e
e

c
o
m
b
i
n
e
d

a
r
e
a
s
:

a
x
i
a
l


(
n
e
c
k

a
n
d

l
o
w

b
a
c
k
)
,

u
p
p
e
r

l
i
m
b
s
,

l
o
w
e
r

l
i
m
b
s


O
c
c
u
p
a
t
i
o
n
a
l

t
i
t
l
e

A
g
e

W
o
r
k

y
e
a
r
s

P
r
e
v
a
l
e
n
c
e

r
a
t
e
s

M
a
n
n
-
W
h
i
t
n
e
y

U
-
t
e
s
t

C
o
x

p
r
o
p
o
r
t
i
o
n
a
l

h
a
z
a
r
d
s

r
e
g
r
e
s
s
i
o
n

I
V

L
o
n
g
i
t
u
d
i
n
a
l

5
0
4

w
o
r
k
e
r
s

1
7

m
e
n

w
e
r
e

e
x
c
l
u
d
e
d

4
8
7

w
o
m
e
n

a
n
a
l
y
s
i
s

a
m
o
n
g

s
u
b
j
e
c
t
s

w
i
t
h

o
b
s
e
r
v
a
t
i
o
n
s

_

4

t
i
m
e

p
o
i
n
t
s
/

9

n

3
8
5













































R
e
c
i
p
r
o
c
a
l
l
y

M
u
l
t
i
p
l
e
-
s
i
t
e

m
u
s
c
u
l
o
s
k
e
l
e
t
a
l

p
a
i
n
,


_

3

p
a
i
n

s
i
t
e
s





























































M
e
n
t
a
l

s
t
r
e
s
s

d
u
r
i
n
g

t
h
e

p
a
s
t

m
o
n
t
h

H
u
r
r
y

L
o
w

j
o
b

c
o
n
t
r
o
l

L
o
w

s
k
i
l
l

d
i
s
c
r
e
t
i
o
n

P
o
o
r

c
o
-
w
o
r
k
e
r

r
e
l
a
t
i
o
n
s
h
i
p
s

L
o
w

s
u
p
e
r
v
i
s
o
r

s
u
p
p
o
r
t

B
a
s
e
l
i
n
e

l
e
v
e
l

o
I

t
h
e

o
u
t
c
o
m
e


A
g
e

S
m
o
k
i
n
g

B
o
d
y

m
a
s
s

i
n
d
e
x

P
h
y
s
i
c
a
l

e
x
e
r
c
i
s
e

P
e
r
c
e
i
v
e
d

p
h
y
s
i
c
a
l

w
o
r
k
l
o
a
d

O
r
g
a
n
i
z
a
t
i
o
n
a
l

r
e
I
o
r
m
s

S
t
u
d
y

a
r
m


S
e
m
i
p
a
r
a
m
e
t
r
i
c

g
r
o
u
p
-
b
a
s
e
d

m
e
t
h
o
d

(
t
r
a
j
e
c
t
o
r
y

a
n
a
l
y
s
i
s
)

L
o
g
i
s
t
i
c

r
e
g
r
e
s
s
i
o
n

m
o
d
e
l
s

T
i
m
e
-
l
a
g
g
e
d

g
e
n
e
r
a
l
i
z
e
d

e
s
t
i
m
a
t
i
n
g

e
q
u
a
t
i
o
n
s

m
o
d
e
l
,

G
E
E

D
u
r
i
n
g

t
h
e

p
a
s
t

3

m
o
n
t
h
s

66
5.7. Statistical analyses
Summary of statistical analyses used in the original studies are presented in table 3. All
analyses were performed using the SAS version 9.1 (SAS Institute Inc., Cary, NC, USA)
and SPSS version 12.0.1 (SPSS Inc., Chicago, IL, USA).
5.7.1. Effects of the participatory ergonomics intervention on musculoskeletal
disorders and psychosocial factors at work
In studies I and II, the data were analyzed according to the intention-to-treat principle and
the results were based on prevalence rates or mean scores of cross-sectional data of an
open sample (Ukoumunne and Thompson 2001; Atienza and King 2002). In analyses by
generalized estimating equations (GEE), the clustering of the data (workers, kitchens) was
taken into account.
Assessment of the effects of the intervention was based on the comparison of the
musculoskeletal outcomes between the groups (study arms) at each cross-section. The
effects of covariates (Table 3) on the results were tested at the PIA and at the 12-month
post-intervention follow-up (PI
12)
). Effects on musculoskeletal fatigue and trouble due to
pain were studied using mixed regression models, in which the kitchen series and
individual kitchens were interpreted as random variables and the study arm, city, and
potential covariates as fixed variables. In logistic regression models, which were used to
analyze the effects on pain and sick leaves, the series and individual kitchens were used as
subject effects. The repeated measures design was used in the re-analyses among the
subgroup of subjects that remained in employed in the same kitchens throughout the study
(cohort analysis,
n = 307).
Between-group differences in psychosocial factors at work were studied at BL, at PIA and
at PI
12
. The variable assessing the independent and joint effects of the intervention and the
unconnected organizational reforms on psychosocial factors at work was coded as follows:
1=no intervention, no organizational reform; 2= intervention, no organizational reform; 3=
no intervention, organizational reform; 4=intervention, organizational reform. No
67
intervention/ no organizational reform (1) was used as the reference group. Logistic
regression models were first adjusted only for the baseline level of the outcome variable,
and the second models also for other covariates (Table 8 and 9).
In all the models, ORs with their 95% confidence intervals (CI) were estimated as the
measure of effect. Two-sided tests of statistical significance (p < 0.05) were used.
5.7.2. Multiple-site musculoskeletal pain and reciprocal associations with
psychosocial factors at work
The occurrence of MSP at baseline was described in the cross-sectional study (III) among
female workers, who had information on all seven assessed anatomical sites (n= 495,
Figure 3). Differences in pain prevalence between age groups were assessed with the
Mann-Whitney U- test. PRs for pain in one anatomical site relative to another were
calculated using Cox proportional hazards regression. Adjustment was made for
occupational title, age, and work years.
In study IV, the intervention and control groups were pooled to be analyzed prospectively
over two-years (Figure 3). Since the number of men was so low, the analyses were carried
out among women only. The aim was to study associations of psychosocial factors at work
and mental stress with development of MSP over time, as well as associations in the
reversed time order. To assess the average time trends of the prevalence of MSP, a mixed-
effects logistic regression model was fitted to the data with linear time effects.
Associations between MSP and psychosocial factors at work were first studied by a time-
lagged generalized estimation equations model (Liang and Zeger 1986). Repeated
measurements of psychosocial factors were studied in relation to MSP three months later,
and vice versa. Second, a semi-parametric group-based approach (trajectory analysis)
(Nagin 1999; 2005; Jones and Nagin 2007) was used to identify developmental patterns of
MSP and psychosocial factors at work. This method identifies groups (trajectories) of
individuals who tend to have a similar profile over time. The Bayesian information
criterion (BIC) is used as the basis for selecting the optimal model, number of trajectories,
and their shape (intercept, linear trend, cubic or quadratic). Individuals are assigned to the
68
trajectory to which they have the highest probability of belonging. Ideally, the posterior
membership probability should be near 1. To minimize the risk of misclassification, a
model that includes trajectories with mean posterior assignment probabilities below 0.70 is
not recommended (Nagin 2005). Measurements should be available in at least 50% of the
assessed time points to reliably assess group membership probabilities, and in at least
three time points to detect trajectories of quadratic shape (Kokko 2004). Based on these
requirements, women with observed values of MSP in at least four time points of nine
during the two-year follow-up constituted the final data set (n=385, Figure 3). After
definition of the trajectories, associations of the baseline psychosocial factors at work with
MSP trajectories were analyzed and vice versa. Finally, psychosocial work factor
trajectories were studied in relation to MSP trajectories.
Logistic regression models with odds ratios (OR) and 95% confidence intervals (CI) were
used to assess the associations. All models were adjusted for age, BMI, smoking, physical
exercise, and perceived physical workload at baseline, and for the study arm (intervention
/control) and organizational reforms (no/yes). In the reciprocal time-lagged analyses, the
baseline value of the outcome was included among the independent variables.
6. RESULTS
6.1. Effects of the participatory ergonomics intervention in the
prevention of musculoskeletal disorders (I)
6.1.1. Baseline characteristics
At baseline, there were no statistically significant differences in the 3-month prevalence
rates of pain between the intervention and the control group. In both groups, pain was
most common in the neck, low back, and forearms/hands, followed by pain in the
shoulders, ankles/feet, knees and hips. No significant differences between the groups in
the occurrence of MSP, sick leaves, or adverse psychosocial working conditions were
observed. Compared to the control group, somewhat fewer permanent workers and more
69
smokers were assigned to the intervention group. Overall, randomization seemed to be
successful and the comparability between the groups was good (Table 4).
Table 4. Baseline information given at the cluster and individual levels
Baseline

Variable

Intervention group


Control group

Cluster level
Number oI kitchens 59 60
Type oI kitchens (No.)
School 43 42
Nursery 10 11
Nursing home or geriatric service centre 5 6
Other 1 1

Inaiviaual level
Number oI workers 263 241
Occupational title ()
Foodservice manager 2 2
CheI 19 19
Cook 20 21
Kitchen aid 59 57
Other 0.8 0.4
Employed in kitchen work (years), range (median) 0-40 (17) 0-43 (20)
Employment ()
Permanent 81 91
Full-time 95 97
Women () 96 98
Age (years), range (median) 19-63 (46) 19-62 (47)
Height (cm), range (median) 141-196 (165) 148-186 (165)
Weight (kg), range (median) 49-147 (69) 46-132 (67)
Body mass index (kg/m
2
), range (median) 18-51 (24.9) 16-46 (24.6)
Smokers () 29 20
Physical exercise ()
_ Once a week 37 41
_ 4 times/week 21 23
Any musculoskeletal pain during the past 3 months () 86 87
Multiple-site musculoskeletal pain (_ 3 pain sites) () 53 55
Sick leave due to any musculoskeletal pain during the past
3 months ()
15 18
Perceived physical workload, mean (standard deviation) 3.7 (1.1) 3.7 (1.1)
Mental stress during the past month () 8 10
Mental strenuousness oI work () 15 12
Hurry () 34 30
Job dissatisIaction () 3 2
Low job control () 15 13
Low skill discretion () 4 3
Poor co-worker relationships () 11 7
Low supervisor support () 11 13
70
6.1.2. Changes in ergonomics
Participation rates in the workshops during the intervention were excellent, on average
73%. Altogether 402 ergonomics changes, evaluated by the researchers as being beneficial
with regard to load on the musculoskeletal system or occupational safety, were
implemented in the intervention group during the intervention phase. In the control group,
80 changes were implemented spontaneously within normal activity during the same
period. In the intervention group, over 100 of the intended changes were not completed
(Table 5). Both in the intervention and control group, the changes mostly concerned the
ergonomics of dishwashing, cooking and baking, and the distribution and serving of food.
In the intervention group, the largest proportion of changes was targeted at work
organization and methods (41%), whereas machines, equipment, and tools received the
most attention (52%) in the control group (Table 6).
Examples of implemented changes and of good practices in kitchen work have been
collected to the web pages of the FIOH and they are freely accessible
(www.ttl.fi/keittiovinkit). Pehkonen et al. (2009a) have described some of the
implemented changes in the context of the description of a video-based observation
method aimed to assess musculoskeletal load in kitchen work.
Table 5. Number of implemented changes during the intervention and the 12-month post-
intervention follow-up in the intervention and control group, and uncompleted changes in the
intervention group. The number of changes per kitchen is given in brackets

Number oI changes
Intervention group
(n59)
Control group
(n 60)
During intervention phase 402 (6.8) 80 (1.3)
During Iollow-up phase 101 (1.7) 69 (1.2)
Uncompleted 113 (1.9) -


71
Table 6. Distribution of implemented changes in the intervention and control group
(402 vs. 80) subdivided according to work tasks and targets
*
E.g. removal of doorsteps.

E.g. new package sizes.

E.g. removing objects causing risk of injury.
6.1.3. Effects on health outcomes
No systematic differences in any health outcome were observed between the intervention
and control group during the intervention or during the 12-month post-intervention follow-
up. Prevalence rates of musculoskeletal pain and sick leave during the past 3 months were
similar in the intervention and control groups (Figure 5). A few statistically significant
differences were found in the cross-sections, but mainly to a non-hypothesized direction.
Furthermore, no differences in MSP ( 3 pain sites) were found between the groups.
During the study period, the 3-month prevalence of MSP varied between 51% and 61% in
the intervention group, and between 47% and 59% in the control group (Figure 6).
Intervention
group (n59)
Control group
(n 60)

Work task ()
Preparation 8 10
Cooking and baking 18 39
Distribution and serving oI Iood 18 17
Packing Iood to be delivered Ior clients 9 6
Dishwashing 23 17
Cleaning and maintenance oI room and equipment 15 8
Receiving and storing oI raw material 8 2
OIIice work 1 0

Target ()
Machines, equipment, tools 27 52
Layout, Iurniture
*
13 11
Work organization, methods and practices 41 16
Materials
f
6 15
Physical work environment, saIety

10 4
Other 3 2
72
Figure 5. Musculoskeletal pain in seven anatomical sites and sick leave due to any pain during
the past 3 months. Prevalence rates at baseline (BL), and every 3 months during the
intervention (I) and the 12-month post-intervention (PI) follow-up. Comparison of the
intervention and control group.
*p < 0.05.

Some of the kitchens finished the intervention at 9 months (I

9
) and some at 12 (I
12
)
months. Thus, the PIA, post-intervention assessment was a combination of these assessments.
Neck
0
20
40
60
80
BL I3 I6 I9 I12 PIAf PI3 PI6 PI9 PI12
Follow-up survey
P
r
e
v
a
l
e
n
c
e

o
f

p
a
i
n


(

)*
Hips
0
20
40
60
80
BL i3 i6 i9 i12 PIAf pi3 pi6 pi9 pi12
Follow-up survey
P
r
e
v
a
l
e
n
c
e

o
f

p
a
i
n


(

)
*
*
Shoulders
0
20
40
60
80
BL i3 i6 i9 i12 PIAf pi3 pi6 pi9 pi12
Follow-up survey
P
r
e
v
a
l
e
n
c
e

o
f


p
a
i
n


(

)
Knees
0
20
40
60
80
BL i3 i6 i9 i12 PIAf pi3 pi6 pi9 pi12
Follow-up survey
P
r
e
v
a
l
e
n
c
e

o
f

p
a
i
n


(

)
*

Forearms or hands
0
20
40
60
80
BL i3 i6 i9 i12 PIAf pi3 pi6 pi9 pi12
Follow-up survey
P
r
e
v
a
l
e
n
c
e

o
f

p
a
i
n


(

)
* *
*
Ankles or feet
0
20
40
60
80
BL i3 i6 i9 i12 PIAf pi3 pi6 pi9 pi12
Follow-up survey
P
r
e
v
a
l
e
n
c
e

o
f

p
a
i
n


(

)
Lowback
0
20
40
60
80
BL i3 i6 i9 i12 PIAf pi3 pi6 pi9 pi12
Follow-up survey
P
r
e
v
a
l
e
n
c
e

o
f



p
a
i
n

(

)
*
*
Sick leave
0
20
40
60
80
BLA i3 i6 i9 i12 PIA* pi3 pi6 pi9 pi12
Follow-up survey
P
r
e
v
a
l
e
n
c
e

o
f

s
i
c
k

l
e
a
v
e

(

)
*

BL I3 I6 I9 I12 PIA
f
PI3 PI6 PI9PI12
BL I3 I6 I9 I12 PIA
f
PI3 PI6 PI9PI12
BL I3 I6 I9 I12 PIA
f
PI3 PI6 PI9 PI12 BL I3 I6 I9 I12 PIA
f
PI3 PI6 PI9PI12 BL I3 I6 I9 I12 PIA
f
PI3 PI6 PI9PI12
BL I3 I6 I9 I12 PIA
f
PI3 PI6 PI9PI12
BL I3 I6 I9 I12 PIA
f
PI3 PI6 PI9PI12
BL I3 I6 I9 I12 PIA
f
PI3 PI6 PI9 PI12
BL I3 I6 I9 I12 PIA
f
PI3 PI6 PI9PI12
73
Figure 6. Multiple-site musculoskeletal pain ( 3 pain sites) during the past 3 months.
Prevalence rates at baseline (BL), and every 3 months during the intervention (I) and the 12-
month post-intervention (PI) follow-up. Comparison of the intervention and control group.
*p < 0.05.

Some of the kitchens finished the intervention at 9 months (I

9
) and some at 12 (I
12
)
months. Thus, the PIA, post-intervention assessment was a combination of these assessments.
The mean scores of trouble caused by pain during the past 3 months and local
musculoskeletal fatigue after a working day during the past 7 days displayed no significant
differences between the intervention and control group at any time point (Figure 7).
Figure 7. Mean values of trouble caused by any musculoskeletal pain during the past 3 months
and local musculoskeletal fatigue after a working day during the past 7 days. Minimum, 25
percentile, median, arithmetic mean (+), 75 percentile, and maximum. A comparison between
the intervention and control group.
*
Some of the kitchens finished the intervention at 9 months (I
9
) and some at 12 (I
12
) months. Thus,
the PIA, post-intervention assessment was a combination of these assessments.
Multiple-site pain
0
20
40
60
80
BL 3 6 9 12 PIA 3 6 9 12
Follow-up survey
P
r
e
v
a
l
e
n
c
e

o
f

M
S
P

(

)


BL I
3
I
6
I
9
I
12
PIA
f
PI
3
PI
6
PI
9
PI
12
74
Inclusion of the baseline covariates in the analyses did not change the results. The results
were similar among those who remained in the same kitchen throughout the study (cohort
analysis, n=307). Similarly, no differences between age groups ( 45 years vs. > 45 years)
in the effects of the intervention were observed. The effects of the intervention on health
outcomes were similar in the intervention and control groups in the cities where there were
organizational reforms and in the cities where no reforms took place (unpublished results).
6.1.4. Effects on perceived physical workload
Perceived physical workload showed no effect caused by the intervention. Overall mean
values for the intervention group were 3.7 [standard deviation (SD) 1.1] at baseline, 3.8
(1.0) at the end of the intervention, and 3.9 (1.2) at the 12-month post-intervention follow-
up. The respective figures for the control group were 3.7 (1.1), 3.8 (1.0), and 3.7 (1.2).
When subdivided according to work tasks, no significant differences between the groups
were observed (see the original publication I, Table 3).
6.2. Effects of the participatory ergonomics intervention on psychosocial
factors at work (II)
No favourable effects of the intervention on psychosocial factors at work were found.
Instead, a deterioration in several measures was observed. The adverse effects were
mainly due to a joint effect of the intervention and unconnected major organizational
reforms of foodservices in two of the participating cities.
Effects on psychosocial factors at work
Crude prevalence rates for each psychosocial factor at work at different time-points are
presented in Figure 8 and the effects of the intervention on psychosocial factors at work in
Table 7. At PIA, after adjustment for the baseline level of the outcome, workers in the
intervention group were more dissatisfied with their work (OR 3.0, 95% CI 1.1.-8.5), they
reported more mental stress (2.3, 1.2-4.7) and felt co-worker relationships to be poorer
(2.3, 1.0-5.2) compared to the workers in the control group. The effect on job
dissatisfaction persisted at PI
12
(3.0, 1.2-7.8). In the fully adjusted model, poor co-worker
75
relationships and job dissatisfaction at PIA remained statistically significantly poorer in
the intervention group (see the original publication II, Table 3).
Figure 8. Crude prevalence rates of psychosocial factors at work according to organizational
reform (Org -, no organizational reform; Org +, organizational reform). The rates are based on
cross-sectional data of an open sample. Comparison between the intervention and control group
at baseline (BL), at the end of the intervention (post-intervention assessment, PIA), and at the
12-month post-intervention follow-up (PI
12
).
*Some of the kitchens finished the intervention at 9 months and some at 12 months. Thus, the PIA
was a combination of these assessments.





76
Table 7. Effects of the intervention on psychosocial factors at work. Odds ratios (OR) and 95%
confidence intervals (CI) at the end of the intervention (post-intervention assessment, PIA), and
at the 12-month post-intervention follow-up (PI
12
) for intervention group compared to control
group
*Adjusted for baseline level of the outcome.
Joint effects of the intervention and organizational reforms
No statistically significant independent effects of either the intervention or the
organizational reforms were detected at PIA or PI
12.
Instead, Figure 8 illustrates that there
was a joint effect between the intervention and the unconnected organizational reforms, in
relation to the psychosocial factors. At PIA, even after adjustment for all covariates,
mental stress was more common (OR above 3) among those with exposure to both the
intervention and organizational reforms, compared to those who were exposed to neither
(Table 8). Similarly, the mental strenuousness of work (OR over 2), job dissatisfaction
(almost 12), low job control (almost 3), and poor co-worker relationships (almost 6)
developed unfavourably in those workers subjected to double-exposure, i.e. both the
intervention and the organizational reforms. At PI
12,
this kind of joint effect was found for
mental strenuousness of work, hurry, low skill discretion, poor co-worker relationships,
and low supervisor support.
PIA PI
12

Outcome OR (95 CI)*
(n 448-449)
OR (95 CI)*
(n 418-421)
Mental stress during the past month 2.31 (1.15 to 4.66) 1.22 (0.63 to 2.36)
Mental strenuousness oI work 1.31 (0.74 to 2.36) 1.36 (0.75 to 2.46)
Hurry 0.85 (0.51 to 1.40) 1.21 (0.65 to 2.23)
Job dissatisIaction 3.02 (1.08 to 8.45) 3.03 (1.18 to 7.82)
Low job control 1.51 (0.84 to 2.69) 1.34 (0.74 to 2.41)
Low skill discretion 1.16 (0.54 to 2.49) 1.85 (0.88 to 3.88)
Poor co-worker relationships 2.29 (1.00 to 5.22) 2.03 (0.89 to 4.63)
Low supervisor support 0.99 (0.56 to 1.74) 1.77 (0.95 to 3.32)
*Adjusted for baseline level of the outcome.
Joint effects of the intervention and organizational reforms
No statistically significant independent effects of either the intervention or the
organizational reforms were detected at PIA or PI
12.
Instead, Figure 8 illustrates that there
was a joint effect between the intervention and the unconnected organizational reforms, in
relation to the psychosocial factors. At PIA, even after adjustment for all covariates,
mental stress was more common (OR above 3) among those with exposure to both the
intervention and organizational reforms, compared to those who were exposed to neither
(Table 8). Similarly, the mental strenuousness of work (OR over 2), job dissatisfaction
(almost 12), low job control (almost 3), and poor co-worker relationships (almost 6)
developed unfavourably in those workers subjected to double-exposure, i.e. both the
intervention and the organizational reforms. At PI
12,
this kind of joint effect was found for
mental strenuousness of work, hurry, low skill discretion, poor co-worker relationships,
and low supervisor support.
77
Table 8. Joint effects of the intervention and organizational reforms on psychosocial factors at
work. Those with no exposure to the intervention and no exposure to the organizational reforms
as a reference group. Odds ratios (OR) and 95% confidence intervals (CI) based on cross-
sectional analyses at the end of the intervention (PIA) and at the 12- month post-intervention
follow-up (PI
12
). Neither the intervention nor the organizational reforms had any statistically
significant independent effects on psychosocial factors at work at PIA or at PI
12
*Adjusted for baseline level of the outcome.

Adjusted for baseline level of the outcome, age,
musculoskeletal pain, and physical workload.

Too few observations.

6.3. Occurrence of multiple-site musculoskeletal pain (III)
In studies III-IV, the study samples consisted of women only. At baseline, neck pain was
the most common (3-month prevalence 71%), followed by pain in the low back (50%),
forearms or hands (49%), shoulders (34%), ankles or feet (30%), knees (29%), and hips
(19%). The occurrence of pain at multiple-sites concurrently was more common than
single-site pain. Only 13% of the women reported no pain and 14% reported pain in only
one site. Instead, about 73% reported pain in at least two, 36% in four or more, and 10% in
six to seven sites. The symptoms occurred in 83 different combinations, in addition to
those with no pain and pain in one site only. The PRs for pain in a single site relative to
another varied from 1.3 to 5.8, e.g. neck pain was associated with pain in other sites with
PRs varying from 1.3 to 1.6, and ankle or foot pain with ratios between 1.9 and 2.4 (Table
9). When assessed in three larger anatomical areas, i.e. the axial skeleton (neck and low
back), upper limbs, and lower limbs, concurrent pain in all three areas was the most
PIA PI
12

Outcome

Model 1*
OR (95 CI)
(n 448-449)
Model 2
f

OR (95 CI)
(n 389-391)
Model 1*
OR (95 CI)
(n 418-421)
Model 2
f

OR (95 CI)
(n 331-332)
Mental stress
during the past month
3.45

(1.32 to 9.01) 3.46 (1.22 to 9.79) 2.24 (0.87 to 5.75) 1.96 (0.72 to 5.34)
Mental strenuousness oI
work
1.38 (0.63 to 3.00) 2.24 (1.02 to 4.89) 2.33 (0.98 to 5.54) 3.09

(1.20 to 7.94)
Hurry 1.01 (0.53 to 1.96) 1.17 (0.61 to 2.24) 2.23 (0.92 to 5.39) 2.52 (1.00 to 6.37)
Job dissatisIaction 11.0 (1.48 to 87.1) 11.6 (1.37 to 99.1) 5.10 (1.36 to 19.1)

Low job control 2.57 (1.07 to 6.13) 2.84 (1.03 to 7.84) 1.96 (0.90 to 4.25) 2.20 (0.99 to 4.89)
Low skill discretion 1.80 (0.60 to 5.38) 1.61 (0.49 to 5.25) 2.71 (0.93 to 7.93) 6.31 (1.34 to 29.6)
Poor co-worker
relationships
3.34 (1.00 to 11.1) 5.71 (1.53 to 21.3) 3.75 (1.41 to 10.1) 2.88 (0.94 to 8.84)
Low supervisor support 0.81 (0.34 to 1.92) 1.43 (0.56 to 3.66) 2.51 (1.01 to 6.20) 3.18 (1.16 to 8.77)
*Adjusted for baseline level of the outcome.

Adjusted for baseline level of the outcome, age,
musculoskeletal pain, and physical workload.

Too few observations.

6.3. Occurrence of multiple-site musculoskeletal pain (III)
In studies III-IV, the study samples consisted of women only. At baseline, neck pain was
the most common (3-month prevalence 71%), followed by pain in the low back (50%),
forearms or hands (49%), shoulders (34%), ankles or feet (30%), knees (29%), and hips
(19%). The occurrence of pain at multiple-sites concurrently was more common than
single-site pain. Only 13% of the women reported no pain and 14% reported pain in only
one site. Instead, about 73% reported pain in at least two, 36% in four or more, and 10% in
six to seven sites. The symptoms occurred in 83 different combinations, in addition to
those with no pain and pain in one site only. The PRs for pain in a single site relative to
another varied from 1.3 to 5.8, e.g. neck pain was associated with pain in other sites with
PRs varying from 1.3 to 1.6, and ankle or foot pain with ratios between 1.9 and 2.4 (Table
9). When assessed in three larger anatomical areas, i.e. the axial skeleton (neck and low
back), upper limbs, and lower limbs, concurrent pain in all three areas was the most
78
common (36%) combination and this increased statistically significantly (p< 0.005) with
age from 22% among the youngest group to 49% in the oldest group (Figure 9). The
longer the women had been employed in kitchen work, the higher was the prevalence of
pain in all three areas. There were no differences in the number of anatomical areas with
pain between kitchen aids, cooks and other occupations.
Table 9. Prevalence ratios (PR) for pain in a single anatomical site relative to another site
among women in kitchen work (n=495). Cox regression analysis
*Treated as the dependent variable.

Treated as the independent variable.

Prevalence ratios are

statistically significant (p< 0.001), adjusted for 42 multiple tests.
Figure 9. Prevalence (%) of the number of painful anatomical areas (neck and low back, upper
limbs, lower limbs) during the past three months among women in kitchen work (n=495) by age
group. Areas classified as no, one, two, and three.
Site Neck* Shoulder* Forearm
or hand*
Low
back *
Hip* Knee*

Ankle
or Ioot*
Neck pain
f
- 5.8

3.2

1.9

3.1

2.1

2.4

Shoulder pain
f
1.6

-
2.1

1.4 2.1

1.7

2.0

Forearm or hand pain
f
1.6

3.1

- 1.6

3.0

1.5 2.0

Low back pain
f
1.4 1.5 1.6

- 2.8

1.8

2.1

Hip pain
f
1.3 1.8

1.7

1.6

- 1.6 1.9

Knee pain
f
1.3 1.6 1.3 1.5 1.8 - 2.1

Ankle or Ioot pain
f
1.3 1.9

1.6

1.6

2.2

2.1

-
*Treated as the dependent variable.

Treated as the independent variable.

Prevalence ratios are

statistically significant (p< 0.001), adjusted for 42 multiple tests.
Painful anatomical areas
0 20 40 60 80 100
51 y (n162)
41-50 y (n185)
40 y (n148)
Age
No One Two Three

79
6.4. Reciprocal associations of multiple-site musculoskeletal pain with
psychosocial factors at work (IV)
6.4.1. Time-lagged associations
Since no overall positive effects of the PE intervention on musculoskeletal symptoms
(Figure 5) or on the occurrence of MSP (Figure 6) were found, the intervention and
control groups were pooled so that they could be analyzed in the two-year longitudinal
study.
All psychosocial factors at work, except low skill discretion and poor co-worker
relationships, predicted the occurrence of MSP (defined as pain at 3 of seven sites) three
months later, odds ratios varying from 1.4 to 2.1 after adjustment for all covariates (Table
10a). MSP predicted low job control, low supervisor suppport, and mental stress (ORs 1.4-
2.0), respectively (Table 10b).
Table 10a. Psychosocial factors at work as determinants of multiple-site musculoskeletal pain
(MSP, 3 pain sites) among female kitchen workers (n=385). In a time-lagged generalized
estimating equation (GEE) model, psychosocial factors at work were assessed 3 months before
MSP. Odds ratios (OR) and their 95% confidence intervals (CI)
*
Adjusted for study arm, organizational reforms, and the following factors at baseline: age, body
mass index, smoking, physical exercise, perceived physical workload, and MSP.
Multiple-site pain
OR (95 CI)*
Low job control 1.8 (1.3-2.5)
Low skill discretion 1.4 (0.7-2.9)
Low supervisor support 2.1 (1.4-3.2)
Poor co-worker relationships 1.5 (0.8-2.6)
Hurry 1.4 (1.1-1.8)
Mental stress during the past month 1.6 (1.1-2.3)
*
Adjusted for study arm, organizational reforms, and the following factors at baseline: age, body
mass index, smoking, physical exercise, perceived physical workload, and MSP.
80
Table 10b. Multiple-site musculoskeletal pain (MSP, 3 pain sites) as a determinant of
psychosocial factors at work among female kitchen workers (n=385). In a time-lagged
generalized estimation equation (GEE) model, MSP was assessed 3 months before psychosocial
factors at work. Odds ratios and their 95% confidence intervals
*
*
Adjusted for study arm, organizational reforms and the following factors at baseline: age, body
mass index, smoking, physical exercise, perceived physical workload, and psychosocial factor at
work.
6.4.2. Trajectories of multiple-site musculoskeletal pain and psychosocial factors at
work
The 3-month prevalence of MSP varied from 50% to 61% during the two-year follow-up
period. There was no statistically significant overall trend in the prevalence. In the
trajectory analysis of MSP, the best fit was a four-group model including two trajectories
with the intercept (no change over time) and two with a linear shape (Figure 10). The
trajectory labelled as Low was composed of the workers whose prevalence of MSP was
constantly low (n=129, 33%), and the High trajectory of those with a constantly high
prevalence of MSP (n=146, 37%) over the two-year follow-up. The Descending trajectory
consisted of 66 workers (19%) and followed a declining pattern. The Ascending trajectory
was the smallest (n=44, 11%), and had a low initial prevalence of MSP that increased
considerably toward the end of the follow-up. The mean trajectory-group assignment
probabilities varied between 0.69-0.92, indicating that there was relatively little
classification uncertainty. The results of a sensitivity analysis using 4 pain sites as the
cut-off point for MSP were similar. The four-group model with the similar shapes of the
trajectories had the best fit. The group sizes were 34% for Low, 28% for Descending, 13%
for Ascending, and 25% for the High trajectory, and the group assignment probabilities
varied from 0.77- 0.91.
Low
job control
Low
skill
discretion
Low
supervisor
support
Poor
co-worker
relationships
Hurry Mental stress
during the
past month
MSP 1.4 (1.1-2.0)

1.3 (0.9-2.0) 2.0 (1.4-2.8) 1.4 (0.9-2.3) 0.8 (0.6-1.1) 2.0 (1.3-2.9)

*
Adjusted for study arm, organizational reforms and the following factors at baseline: age, body
mass index, smoking, physical exercise, perceived physical workload, and psychosocial factor at
work.
6.4.2. Trajectories of multiple-site musculoskeletal pain and psychosocial factors at
work
The 3-month prevalence of MSP varied from 50% to 61% during the two-year follow-up
period. There was no statistically significant overall trend in the prevalence. In the
trajectory analysis of MSP, the best fit was a four-group model including two trajectories
with the intercept (no change over time) and two with a linear shape (Figure 10). The
trajectory labelled as Low was composed of the workers whose prevalence of MSP was
constantly low (n=129, 33%), and the High trajectory of those with a constantly high
prevalence of MSP (n=146, 37%) over the two-year follow-up. The Descending trajectory
consisted of 66 workers (19%) and followed a declining pattern. The Ascending trajectory
was the smallest (n=44, 11%), and had a low initial prevalence of MSP that increased
considerably toward the end of the follow-up. The mean trajectory-group assignment
probabilities varied between 0.69-0.92, indicating that there was relatively little
classification uncertainty. The results of a sensitivity analysis using 4 pain sites as the
cut-off point for MSP were similar. The four-group model with the similar shapes of the
trajectories had the best fit. The group sizes were 34% for Low, 28% for Descending, 13%
for Ascending, and 25% for the High trajectory, and the group assignment probabilities
varied from 0.77- 0.91.
81
Observed Predicted ----
Figure 10. Multiple-site musculoskeletal pain trajectories (MSP, 3 pain sites) among female
kitchen workes (n=385).
1 = Low (low prevalence of MSP), 2 = Descending (decrease in prevalence of MSP),
3 = Ascending (increase in prevalence of MSP), 4 = High (high prevalence of MSP).
Trajectories of psychosocial factors at work
The two-group model had the best fit for each of the psychosocial factors at work (Figure
11). Hurry had trajectories with the intercept shape with 23% of the workers belonging to
the High trajectory. All other psychosocial variables had one trajectory with the intercept
shape and the other with a linear ascending trend. The proportions of subjects belonging to
the ascending trajectories were 23% (low job control), 7% (low skill discretion), 21% (low
supervisor support), 10% (poor co-worker relationships), and 20% (mental stress). The
mean assignment probabilities for all psychosocial work factor trajectories were over 0.70.

Observed Predicted ----
Figure 10. Multiple-site musculoskeletal pain trajectories (MSP, 3 pain sites) among female
kitchen workes (n=385).
1 = Low (low prevalence of MSP), 2 = Descending (decrease in prevalence of MSP),
3 = Ascending (increase in prevalence of MSP), 4 = High (high prevalence of MSP).
Trajectories of psychosocial factors at work
The two-group model had the best fit for each of the psychosocial factors at work (Figure
11). Hurry had trajectories with the intercept shape with 23% of the workers belonging to
the High trajectory. All other psychosocial variables had one trajectory with the intercept
shape and the other with a linear ascending trend. The proportions of subjects belonging to
the ascending trajectories were 23% (low job control), 7% (low skill discretion), 21% (low
supervisor support), 10% (poor co-worker relationships), and 20% (mental stress). The
mean assignment probabilities for all psychosocial work factor trajectories were over 0.70.
82
Figure 11. Trajectories of psychosocial factors at work and mental stress among female kitchen
workers (n=385). Trajectories labelled as Low and Ascending (job control, skill discretion,
supervisor support, co-worker relationships, and mental stress), or Low and High (hurry).
6.4.3. Associations of psychosocial factors at baseline with multiple-site
musculoskeletal pain trajectories
After adjustment for covariates, poor co-worker relationships (OR 3.9, 95% CI 1.2-12.4),
mental stress (3.1, 1.1-9.1) and hurry (2.1, 1.1-3.7) at baseline predicted belonging to the
High vs. the Low MSP trajectory (Table 11; Baseline). None of the psychosocial work







Observed Predicted ----
83
factors at baseline predicted belonging to the Ascending vs. the Low or to the Descending
vs. the High MSP trajectory.
6.4.4. Association of multiple-site musculoskeletal pain at baseline with psychosocial
work factor trajectories
With the Low trajectory as the reference, MSP at baseline predicted belonging to the
Ascending trajectory of low job control (OR 2.2, 1.3-3.9) and of mental stress (3.2, 1.7-
6.0) (Table 12).
Table 11. Multiple-site musculoskeletal pain (MSP, 3 pain sites) trajectories in relation to
psychosocial factors at work and mental stress among female kitchen workers (n=385). The
psychosocial factors have been assessed at baseline and as trajectories
*
(two groups, always
contrasting Ascending/High vs. No). Comparison between the MSP trajectories as follows:
Ascending vs. Low, High vs. Low, and Descending vs. High

. Logistic regression analysis. Odds

ratios (OR) and their 95% confidence intervals (CI)
*
Development of psychosocial factors at work and mental stress over a two-year follow-up period.

Number of subjects in multiple-site pain trajectories: Ascending (n= 44), Low (n=129), High
(n=146), Descending (n=66).

Adjusted for study arm, organizational reforms and the following

factors at baseline: age, body mass index, smoking, physical exercise, perceived physical
workload.

Adjusted for study arm and organizational reforms, and the following factors at
baseline: age, smoking, physical exercise, perceived physical workload.

No observations in 'BL
yes' category in the Ascending trajectory of mental stress.
Contrast oI multiple-site pain trajectories
Determinant Ascending vs. Low
|
High vs. Low

Descending vs. High

OR (95 CI) OR (95 CI) OR (95 CI)
Low job control


Baseline, yes vs. no 1.8 (0.5-6.2) 2.0 (0.8-4.9) 1.4 (0.6-3.4)
Trajectories, Ascending vs. No 3.1 (1.1-8.6) 5.0 (2.4-10.3) 0.3 (0.2-0.8)

Low skill discretion


Baseline, yes vs. no 0.9 (0.1-9.7) 1.1. (0.2-5.4) 0.4 (0.0-3.5)

Trajectories, Ascending vs. No 1.9 (0.4-9.5) 3.1 (1.0-10.3) 0.1 (0.0-0.9)

Low supervisor support
Baseline, yes vs. no 0.8 (0.2-2.8) 1.3 (0.6-3.0) 1.4 (0.6-3.3)
Trajectories, Ascending vs. No 2.7 (1.0-6.9) 2.3 (1.2-4.7) 1.1 (0.5-2.3)

Poor co-worker relationships
Baseline, yes vs. no 1.2 (0.2-7.4) 3.9 (1.2-12.4) 0.8 (0.3-2.2)
Trajectories, Ascending vs. No 5.0 (1.1-21.6) 3.5 (1.1-11.4) 1.0 (0.4-2.7)

Hurry
Baseline, yes vs. no 1.2 (0.5-2.9) 2.1 (1.1-3.7) 1.0 (0.5-2.0)
Trajectories, High vs. No 1.1 (0.4-3.0) 1.7 (0.9-3.4) 0.9 (0.4-1.9)

Mental stress during the past month
Baseline, yes vs. no

3.1 (1.1-9.1) 0.9 (0.3-2.4)

Trajectories, Ascending vs. No 5.5 (1.6-19.0) 8.6 (3.3-22.1) 0.6 (0.3-1.3)
84
Table 12. Multiple-site musculoskeletal pain (MSP, 3 pain sites) at baseline as determinant of
trajectories of psychosocial work factors and mental stress among female kitchen workers
(n=385). Comparison between trajectories as follows: Ascending vs. Low* and High vs. Low

.
Logistic regression analysis. Odds ratios (OR) and their 95% confidence intervals (CI)

Adjusted for study arm, organizational reforms, and the following factors at baseline: age, body
mass index, smoking, physical exercise, perceived physical workload.
6.4.5. Psychosocial work factor trajectories in relation to multiple-site pain
trajectories
Trajectories of psychosocial factors at work as determinants of MSP trajectories are
presented in Table 11 (Trajectories). With the Low MSP trajectory as the reference,
adverse changes in all psychosocial factors at work, except in low skill discretion and
hurry, were associated with belonging to the Ascending MSP trajectory, with the odds
ratios varying from 2.7 to 5.5 after adjustment for covariates. Similarly, adverse changes
in all psychosocial factors, with the exception of hurry, were associated with belonging to
the High MSP trajectory (ORs 2.3 to 8.6). With the High MSP trajectory as the reference,
adverse changes in low job control and low skill discretion were inversely associated with
belonging to the Descending MSP trajectory (ORs 0.1-0.3).
These associations were in many cases stronger than the effects of the baseline levels of
the psychosocial factors on the MSP trajectories. Only poor co-worker relationships and
hurry at baseline had higher risk estimates compared with their trajectories in predicting
membership of the High vs. Low MSP trajectories. Hurry at baseline had also a slightly
higher risk estimate compared with its trajectory in predicting membership of the
Ascending vs. High MSP trajectory.
Low
job control
*

Low
skill
discretion
`

Low
supervisor
support
*

Poor
co-worker
relationships
*

Hurry
1
Mental stress
during the
past month
*

MSP 2.2 (1.3-3.9) 2.5 (0.9-7.2) 1.5 (0.9-2.7) 1.8 (0.8-4.1) 1.5 (0.9-2.5) 3.2 (1.7-6.0)

Adjusted for study arm, organizational reforms, and the following factors at baseline: age, body
mass index, smoking, physical exercise, perceived physical workload.
6.4.5. Psychosocial work factor trajectories in relation to multiple-site pain
trajectories
Trajectories of psychosocial factors at work as determinants of MSP trajectories are
presented in Table 11 (Trajectories). With the Low MSP trajectory as the reference,
adverse changes in all psychosocial factors at work, except in low skill discretion and
hurry, were associated with belonging to the Ascending MSP trajectory, with the odds
ratios varying from 2.7 to 5.5 after adjustment for covariates. Similarly, adverse changes
in all psychosocial factors, with the exception of hurry, were associated with belonging to
the High MSP trajectory (ORs 2.3 to 8.6). With the High MSP trajectory as the reference,
adverse changes in low job control and low skill discretion were inversely associated with
belonging to the Descending MSP trajectory (ORs 0.1-0.3).
These associations were in many cases stronger than the effects of the baseline levels of
the psychosocial factors on the MSP trajectories. Only poor co-worker relationships and
hurry at baseline had higher risk estimates compared with their trajectories in predicting
membership of the High vs. Low MSP trajectories. Hurry at baseline had also a slightly
higher risk estimate compared with its trajectory in predicting membership of the
Ascending vs. High MSP trajectory.
85
7. DISCUSSION
This study was intended to obtain evidence on the efficacy of PE in preventing MSDs, and
to examine its effects on psychosocial working conditions. An intensive PE intervention of
about one year's duration was conducted in municipal kitchens. The study was designed as
a CRT. The co-occurrence of musculoskeletal pain in various body sites, and the
association of MSP with psychosocial factors at work and mental stress were also
evaluated.
7.1. Main findings and comparison to earlier studies
7.1.1. Effects of the participatory ergonomics intervention on musculoskeletal
disorders and psychosocial factors at work in the CRT
It was hypothesized that optimization of physical and mental workload would have a
preventive effect on the occurrence of MSDs and that active group work, and the direct
participation of the workers in problem-solving, planning and implementing the
ergonomic changes during the intervention would have a positive effect on the
psychosocial work environment in the intervention kitchens. However, the obtained results
did not support these hypotheses.
Although the intervention model proved feasible (Pehkonen et al. 2009c) and it was well
received by both the workers and management, no effect of the intervention was found on
musculoskeletal health. In some cross-sections, a few statistically significant differences in
pain prevalence were seen, but mainly in an unanticipated direction. In the intervention
group, the occurrence of pain seemed to be stable or even rising, and especially the
number of pain-related sick leaves increased compared to the control group. No
differences in trouble caused by pain or musculoskeletal fatigue were detected.
Perceived physical workload was not affected by the intervention. When the effects of the
intervention on psychosocial factors at work were analyzed, a deterioration was observed
in several of the workers' reports about psychosocial factors. At the end of the
86
intervention, the workers in the intervention group reported more mental stress, worse job
dissatisfaction and poorer co-worker relationships compared to the workers in the control
group. The effect on job dissatisfaction still persisted at the 12-month post-intervention
follow-up. An adverse joint effect of the intervention and organizational reforms was seen
for the majority of the eight measured psychosocial factors. Unfavourable effects on
psychosocial factors at work may partly explain why the intervention did not have the
expected impact on the occurrence of MSDs.
The variety in study designs, participants, outcome measures, and intervention programs
complicate the comparison of our results to earlier findings. Since the CRT by the Dutch
researchers (Driessen et al. 2008) is still underway and results are not yet available, the
study by Morken et al. (2002) appears to be the only one with a similar study design. Their
results were in line with the present findings. The study by Morken et al. (2002) aimed at
examining the effects of a PE intervention on MSDs, psychosocial factors, and coping and
included a large number of Norwegian aluminium industry workers. The intervention
groups included only operators, only supervisors, and both. The authors found no effect of
the intervention on MSDs, job demands, job control or social support, while coping skills
increased, particularly in the operator group. Organizational restructuring took place
during the intervention period. The authors collected information from the plants about
restructuring, but did not report whether they had considered its effects in their analyses.
They did not present sample size or power calculations or whether the clustering effect
(plants, workers) was taken into account. However, the study population represented the
whole aluminium industry sector and mostly consisted of men. The workers were from the
production line, where the work is physically demanding, and the occurrence of MSDs
among workers was high. The results of that study and the results of our study support and
complement each other.
Two individually randomized trials on physically lighter sedentary work among VDU
workers have been reported. Ketola et al. (2002) found a positive short-term effect of a PE
intervention on musculoskeletal discomfort at their 2- month follow-up. However, pain or
strain levels did not differ in the intervention groups compared to those in the control
group, and no long-term effects on discomfort, pain or strain were observed at the 10-
month follow-up. In the study of Bohr (2000), the intervention groups reported less upper
87
body pain/discomfort and work stress throughout the study period than the control group
but no effects on lower body pain/discomfort were found. There was a higher turnover
than expected particularly in the intervention groups. In a subsequent study, Bohr (2002)
found no evidence that participatory methods would have been more effective than
traditional methods in training in office ergonomics with regard to positioning work
equipment correctly or to maintaining good working postures.
In a non-randomized trial involving manufacturing workers, Laing et al. (2005, 2007) had
aims and results quite similar to those of the present study. They detected minor effects of
PE on physical workload and no changes in perceived effort or pain severity levels. In the
intervention group, communication dynamics regarding ergonomics were enhanced, but
no differences between the groups were observed in overall communication between
workers, co-workers and management.
This present study seems to be the first large CRT that has evaluated the efficacy of a PE
intervention in reducing psychosocial load at work. Recently, Tsutsumi et al. (2009)
examined the effect of a participatory workplace improvement intervention on mental
health and job performance. The study was conducted in a company producing electrical
devices. They randomized 11 assembly lines to an intervention (47 workers) and a control
group (50 workers). Multilevel modelling was used to take into account that individuals
were nested in units of assembly lines. Mental health and job performance improved in the
intervention group compared to the control group. However, the overall effect was minor,
the final sample sizes for the analyses were small, and after imputation of missing values,
no significant effect was found. In addition, a non-randomized controlled trial by
Kobayashi et al. (2008) was conducted in a manufacturing company. Nine of 45
departments participated in the intervention, and the remaining 36 departments served as
the control group. Positive effects on mental health among women were found, but not
among men.
One basic aim of the participatory approach is that the workers learn new skills during the
process (Haines and Wilson 1998). Learning and internalizing of the principles of
ergonomics gives the workers the capability to independently and continuously develop
ergonomics in their daily work. In this present study, most of the changes in the
88
intervention group were targeted at work organization and methods (41%), whereas in the
control group machines, equipment and tools received most attention (52%). After the end
of the intervention, about 100 ergonomic changes had been implemented in the
intervention group during the 12-month post-intervention follow-up period, indicating that
the aim of learning was achieved.
Joint effects of the intervention and unconnected organizational reforms on
psychosocial factors at work
The adverse overall effects of the PE intervention on psychosocial factors were
accentuated in the kitchens with concomitant unconnected organizational foodservice
reforms. No independent effects of either the intervention or the organizational reforms
were detected. In some previous studies, interventions aimed at the redesign of work have
had an adverse effect on psychosocial working conditions. A reorganisation of work at an
automobile assembly plant resulted in a considerable decrease in the perception of
opportunities to influence the work and in the degree of stimulation at work (Fredriksson
et al. 2001). After the intervention, the prevalence of MSDs increased in the intervention
group, but not in the control group from the same plant. High perceived workload and
reduced occupational pride were associated with the increase in MSDs. Christmansson et
al. (1999) observed that organizational redesign of manual repetitive assembly jobs mostly
impaired the psychosocial work environment and increased the physical stress and risk of
MSDs.
Unexpected organizational reforms have been shown to interfere negatively with
interventions (Lagerstrm et al. 1998; Demure et al. 2000) and it is not uncommon that
long-lasting intervention studies may be confronted by unforeseen changes beyond the
control of the researchers (Cole et al. 2003; Silverstein and Clark 2004; Heaney and
Fujishiro 2006). In this study, the centralizing of food preparation to certain kitchens
during the organizational reforms increased the number of kitchens where food was only
distributed to the clients. As a result, work tasks probably became more monotonous in the
latter kitchens and diminished workers' possibilities to use their skills. The threat of
outsourcing the functions or restructuring the foodservice as a public utility may have
increased the employees' fear of redundancy causing uncertainty, extra tension and
competition between the workers.
89
A participatory process can have negative effects on the psychosocial aspects at work if
the process entails extra work (St-Vincent et al. 2006). In the present study, the
participation in the intervention simultaneously with the implementation of the
organizational reform may have been overly stressful to the workers. Participation in the
workshops was more active in the cities with than in those without reforms, possibly
increasing the overload of the workers in the midst of elevated work demands.
Perceived physical workload
No effect was found that the intervention would have altered perceived physical workload.
The number of portions prepared per worker has been shown to be associated with MSDs
(Shibata et al. 1991; Ono et al. 1997; Nagasu et al. 2007). According to the research
diaries, in about one third of both the intervention and control kitchens, the number of
portions increased during the study. A decrease in the number of personnel was recorded
in four intervention and eleven control kitchens. This may have led to a higher
intensification of work being done in the control kitchens.
Kitchen work is demanding, both physically and mentally. The employees work under
pressure of time and perform various parallel tasks, many of which include exposure to a
combination of risk factors of MSDs. For example, in long-lasting dishwashing, in
receiving and storing of raw material, and cooking, the individual makes frequent bending
and twisting of the trunk, lifting, repetitive movements of the hand, non-neutral wrist and
shoulder postures, and there is need for forceful hand movements. The kitchen workers
spend the working day almost exclusively walking or standing. Messing and Kilbom
(2001) conducted a small workplace field study to assess the consequences of prolonged
walking and standing among kitchen workers. They found that prolonged standing and
short-distance slow walking caused a decrease in the plantar pain-pressure threshold over
the workday. The majority of the workers had experienced foot pain during the previous 3
months. In the light of these results, it might be that standing or walking could be
associated with lower limb pain. There is evidence that exposure to both dynamic and
static repetitive motions, forceful exertion, and non-neutral body postures may cause
MSDs in one or more anatomical sites (Punnett and Wegman 2004). The high occurrence
of MSP found in this study could be connected with the pattern of loading in kitchen
work. Workload may be rather uniformly distributed on the musculoskeletal system.
90
Reducing physical workload on musculoskeletal system is not straightforward. A
reduction of biomechanical exposure on some body part may lead to an increased
exposure on another and thus only transfer the site of the problem. It is difficult to
quantify the exact levels of biomechanical exposure that are harmful to the
musculoskeletal system. Thus, at best, one only knows that the level of exposure needs to
be reduced, but not to what degree. Primary preventive interventions to reduce
biomechanical exposure may encounter difficulties in demonstrating that a reduction in
biomechanical exposure has resulted in a lower occurrence of MSDs. It has been proposed
that a reduction of at least 14% in biomechanical exposure is required to achieve any
detectable change in MSDs (Ltters and Burdof 2002; Burdorf 2010).
The results on physical risk factors based on objective expert assessment showed for some
work tasks a reduction of the level of exposure (unpublished results). A subgroup analysis
where the intervention and control groups were pooled showed that both the observed
reduction in lifting and the perceived physical workload reduction in receiving and storing
of raw material were associated with a lower risk of further shoulder symptoms during the
12-month follow-up (Pehkonen et al. 2009b).Yet, no overall effect of the intervention on
perceived physical workload was detected. One explanation for this apparent discrepancy
might be that the intervention did not produce a strong enough change on physical
workload in the intervention group compared to the annually spontaneously implemented
changes in the control group. It seems that the changes were not strong enough to
influence the overall physical workload or the health outcomes. In some kitchens,
acquisition of expensive new equipment and extensive structural changes would have been
needed. Since no extra funding was available, most of the changes were low cost
solutions.
7.1.2. The occurrence of multiple-site musculoskeletal pain
Since the number of men was low (4% in study III, 3% in study IV), the analyses
concerning the occurrence of MSP and associations of MSP with psychosocial factors at
work were conducted only in women. The cross-sectional study (III) described the overall
prevalence of pain and specifically that at multiple sites concurrently at baseline, before
the intervention started. Neck pain was the most common with seven of ten workers
91
reporting it, while every second worker reported forearm or hand pain and LBP. These
estimates are higher than usually found in the normal female population of a comparable
age range (Picavet and Schouten 2003; Bingefors and Isacson 2004; Kaila-Kangas 2007)
and among the highest published in occupational samples (Blatter and Bongers 1999;
National Research Council and Institute of Medicine 2001; Riihimki 2005a). The one-
month prevalence of LBP among Japanese female cooks working in school lunch services
was recently reported to be 75% (Nagasu et al. 2007), that is even higher compared to the
3-month prevalence (50%) desribed here.
The finding of a high occurrence of MSP is in line with previous results among
occupational samples and in the general population. The estimates for MSP among kitchen
workers were even higher than those that have been presented in the general population or
in occupational samples. About 73% of female kitchen workers reported pain in at least
two sites. The respective figures among a representative sample of Finnish adults was 33%
(Miranda et al. 2010), in the Dutch population 50%, and in Dutch women 39% (Picavet
and Schouten 2003). Recently, Solidaki et al. (2010) evaluated MSP among Greek nurses,
office workers and postal clerks. The study sample (n= 564) comprised predominantly of
women, and 66% reported pain in two or more sites. In men doing manual lifting work,
63% had pain in at least two sites (Yeung et al. 2002). However, all comparisons are
hampered by the sensitivity of pain prevalence estimates to differences in the way that the
question was worded and different reference periods.
The seven pain locations did not cluster in any simple manner in our study material. Two
thirds of the theoretically possible (2
7
=128) combinations were empirically detected.
When pain in three larger anatomical areas (neck and low back, upper limbs, lower limbs)
was studied, every third woman (36%) reported pain in all three anatomical areas and
every second of those aged 51-63 years, i.e. the co-occurrence of pain was age-related. It
is possible that degenerative changes in the musculoskeletal system could contribute to
phenomenon. According to the ACR criteria, classification of CWP includes pain in the
upper and lower extremities (both sides of the body) and axial pain that must have been
present for at least three months (Wolfe et al. 1990). Such a definition could not be applied
here, as no information was availabe on pain persistence or on pain in the left and right
lower limb separately. The observed amount of overlap in pain areas in this study is
92
clearly higher than the prevalence of CWP found in previous surveys. It has been
estimated that approximately 10% of the general population report CWP and almost
unanimously all studies have found higher rates among women compared with men, but
the mechanism responsible for the skewed gender ratio remains unknown (Gran 2003).
In Finland, as well as across all other EU member states the workforce will age rapidly
over the next 40 years, bringing with it the risk of increasing MSD prevalence. In other
words, the impact of MSDs on work disability can be anticipated to intensify rather than
diminish. Two thirds of the European workforce with MSDs reported that pain was
responsible for a significant reduction in their quality of life, and 49% were limited
because of pain in the kind of work they were able to perform (Bevan et al. 2009). MSP
poses extra challenges, because it seems to be more disabling than single-site pain both in
population-based surveys and in occupational samples.
7.1.3. Association of multiple-site musculoskeletal pain with psychosocial factors at
work and mental stress over time
Many reports recommend that consideration of pain at several sites simultaneously is
preferable to the assessment of one pain location in isolation. Nonetheless, most studies
have traditionally examined prognostic factors of single-site pain (Mallen et al. 2007) and
the risk factors of the occurrence of MSP are not well known. The need to provide
information on why pain tends to occur at multiple sites and to examine the occurrence of
MSP over time has recently been emphasized (Croft et al. 2007). Many of the
psychosocial factors at work have been found to be associated with musculoskeletal pain,
mostly in cross-sectional studies. However, in cross-sectional studies, it is not possible to
estimate the timing of determinant and outcome, as the both are assessed at the same time.
The need for conducting longitudinal studies in this area is evident (Macfarlane et al.
2009).
Since no systematic differences between the intervention and control group in our CRT
were found in MSDs, MSP or perceived physical workload, the groups were pooled for
analysis as a two-year longitudinal study (IV). There were two hypotheses, with reversed
directionality. First, the intention was to determine whether psychosocial factors at work
93
and mental stress predicted the development of MSP over time, and second, whether MSP
could predict the occurrence of psychosocial factors at work and mental stress.The
reciprocal analyses were based on the finding of Leino and Magni (1993) who examined
metal industry workers. They found that symptoms related to mental stress predicted an
increase in a sum score of musculoskeletal pain at several sites, and respectively, the pain
sum score predicted an increase in stress symptoms.
In etiological longitudinal studies, usually the relation between an exposure measured at
baseline and the occurrence or change of the outcome during the follow-up period is
examined. In this work, it was possible to make use of two different types of longitudinal
analyses in assessing relationships between the measures. One of these examined the
situation at three-month intervals (the GEE) and the other described them using the total
two-year follow-up period (trajectories). The use of trajectories also enabled the
examination of changes in psychosocial factors at work in relation to changes in MSP over
time.
The 3-month prevalence of MSP (defined as pain at 3 of seven sites) varied between 50-
61% during the two-year follow-up. According to this definition, the workers in the
category of no MSP could still have pain at 1 to 2 sites. A quarter of the workers reported
pain in at least two sites at every nine assessed time points. This is noteworthy because
having pain at one site is shown to increase the risk of developing pain at other sites and
the risk increases with the number of initial painful sites (Croft et al. 2007; Kamaleri et al.
2009). The average number of pain sites appears to be set already by age 20 and little
variation seems to occur thereafter (Croft 2009; Kamaleri et al. 2009). However, it was
found that some variation, even a decrease, in the occurrence of MSP in working age is
possible. Trajectory analysis identified four trajectories of MSP prevalence and it was
observed that 19% of the workers belonged to the trajectory with a descending pattern in
the prevalence of MSP.
The three analyses support both of the study hypotheses. First, in time-lagged analysis,
psychosocial factors at work predicted MSP reported three months later, and vice versa,
MSP predicted adverse psychosocial factors at work after adjustment for covariates. The
risk estimates were comparable in both directions with the exception of hurry at work,
94
which predicted MSP but was not predicted by MSP. Secondly, poor co-worker
relationships, mental stress and hurry at baseline predicted belonging to the High MSP
trajectory. In addition, MSP at baseline predicted belonging to the trajectories of
Ascending low job control and mental stress. Finally, the adverse development in most
psychosocial factors was associated with belonging to the High and Ascending MSP
trajectories. This reciprocal linkage complicates conclusions of the causal nature of the
relationships between these factors. Two possibilities seem to exist: this finding either
reflects two processes that are mutually and serially dependent in time or there is/are some
common underlying factor(s).
With regards to the relationships between mental stress and MSP, these current findings
basically replicate those reported by Leino and Magni (1993) with a somewhat different
measure of MSP and in a different occupational setting. As far as I am aware, no
systematic reciprocal analyses with pain symptoms and psychosocial factors at work have
been carried out before. This present study seems to be the first to examine developmental
patterns of MSP and of psychosocial factors at work. Overall, few longitudinal studies
exist in this subject area among general population or occupational samples.
Differences in study designs, samples, and measures of outcomes and determinants,
complicate the comparison of these results with those of previous studies. However, our
findings are in accordance with some earlier reports in line with our first hypothesis. Leino
and Hnninen (1995) reported that after adjustment for physical workload, poor social
relationships at work and poor job control predicted an increase of pain and
musculoskeletal clinical findings in the neck and upper limbs, low back, and lower limbs
in a 10-year follow-up. Dissatisfaction with support from colleagues or supervisors was
found to increase the risk of future episodes of forearm pain that commonly co-occurred
with other regional pain (Macfarlane et al. 2000). In this study, poor co-worker
relationships at baseline predicted belonging to the group with a constantly high
prevalence of MSP. The time-lagged analysis showed that low supervisor support and low
job control increased the risk for MSP about two-fold. It may be that in small work
communities where work is done in close co-operation, there is an emphasis on the
importance of good social relationships for well-being.
95
In these GEE-models, mental stress and hurry predicted MSP. It was also found that
mental stress and hurry at baseline predicted reporting chronic MSP, a finding in support
of earlier results. Nahit et al. (2001) conducted a cross-sectional study among newly
employed workers (n= 1081, median age 23 year) and observed that those who perceived
their work as stressful and hectic for most of the time had an increased risk of pain at
multiple sites. Later, based on the same study sample, they reported that high levels of
individual psychological distress and adverse work-related psychosocial factors predicted
the onset of musculoskeletal pain at a one-year follow-up with similar effects across
anatomical sites (Nahit et al. 2003). The median age (47 years) in the presnt sample was
higher. It could be that regardless of age, mental stress and hurry might increase the risk of
MSP.
In cross-sectional studies, women, smokers, overweight subjects and those with low
physical activity have reported a greater number of pain sites (Walker-Bone et al. 2004;
Kamaleri et al. 2008a). In the present female sample, smoking had no effect on the
development of MSP. Risk estimates for overweight were increased, but not statistically
significantly. Instead, perceived physical workload and age predicted an increase and a
constantly high prevalence of MSP (unpublished results).
Mental stress and other psychosocial factors may be linked to MSDs with many pathways.
Psychosocial factors may have a direct influence on the loading on the spine through
changes in postures, movements, and exerted forces. Psychosocial factors may trigger
chemical and physiological responses. Increased hormonal excretion or muscle tension
may influence pain perception and in the long-term lead to MSDs. Psychosocial factors
can influence awareness, and alter the ability to cope with the pain and illness and possibly
increase the likelihood of reporting of symptoms (Bongers et al. 1993; Sauter and
Swanson 1996; Davis and Heaney 2000; Hoogendoorn et al. 2000; Bongers et al. 2002).
This study raises the question about the mechanism behind the finding that the occurrence
and development of MSP over time predicts the occurrence and development of adverse
psychosocial working conditions. It is perhaps noteworthy that both measures were based
on subjective perceptions. There might be common underlying factor(s), and these will
require further study.
96
7.2. Methodological aspects
7.2.1. Validity of the CRT
The CRT (study I) is included in three systematic reviews focusing on the effectiveness of
workplace interventions for LBP and neck pain, one by Dutch researchers (Driessen et al.
2010) and the other two by a Nordic research group in the Cochrane collaboration (Aas et
al. 2009a; Aas et al. 2009b). In these reviews, the risk of bias is assessed by using the 12-
item Cochrane Back Review Group criteria list. The quality of the present study was
evaluated to be high (scores 9/12, Driessen et al. 2010). The criteria of blinding were not
fulfilled. In our view it is not possible to conduct a PE intervention in a blinded manner.
However, the baseline data on all outcomes and measures were collected before
randomization, so that the workers participating in the study or the researchers ('care
providers') were not conscious of the study arm to which the workers would be
allocated. After randomization, both the workers and researchers were aware of the
allocation. The researchers were blinded with respect to the questionnaire data during the
data collection, and data analysis was started only after all the follow-up data had been
collected.
As an additional requirement is that those assessing outcomes ('outcome assessors') should
be blinded about the study arm. The outcomes were based on self-reports collected by
questionnaires. It is conceivable that the workers in the intervention group may have been
sensitized to perceive and report their MSDs and psychosocial factors at work. The
magnitude of this possible bias is difficult to assess.
Randomization was based on an assignment algorithm generated by a computer. An
individual otherwise unconnected with the implementation of the intervention was
responsible for the randomization. Randomization was successful and good comparability
between the groups persisted throughout the study. This was true even in spite of the
organizational reforms carried out in two cities, because the study design guaranteed a
parallel proceeding in time of both study arms within each series of kitchens. Stratification
by area and type of kitchen ensured balanced clusters with respect to baseline
characteristics.
97
Data were analysed according to the intention-to-treat principle. Three kitchens dropped
out after the randomization. No information after the baseline questionnaire on these
kitchens was available, and it was not possible to make any further analyses on them. This
situation is remotely analogous to the case where the patient dies before starting the
medication/treatment. It is not generally considered that a situation such as that violates
the intention-to-treat principle. All three dropout kitchens were randomized to the
intervention group. In one kitchen, the study was considered too cumbersome. Otherwise
the reasons were coincidental. The amount of dropouts was minor in relation to the total
number of kitchens. At least 70% of the workers remained employed in the same kitchen
throughout the study. Despite the staff turnover, the groups remained similar through the
intervention phase with regard to assessed potential confounders. All these observations
indicate that selection or attrition bias did not threaten the validity of the study.
There was a relatively large sample size and sufficient statistical power. The clustered
nature of the data was taken into account in the power calculation and in the analyses of
the results. Because this intervention was targeted at the kitchen (community) level, we
used repeated cross-sectional analyses of the open population. This approach has been
recommended as being more appropriate than cohort analyses for measuring the
effectiveness of interventions in the community. The disadvantages of this approach have
also been pointed out. The analyses possibly include workers who receive limited
exposure to the intervention or that elements of the intervention diffuse to other
communities (Ukoumunne and Thompson 2001; Atienza and King 2002). However, the
present results in the open population were similar to those in the cohort analysis among
subjects who remained in the same kitchen throughout the study.
There were very high response rates, participation in the workshops was good, and no
dropouts occurred during the intervention, indicating good compliance. If the workers
changed kitchen, they were asked not to talk about the study process. Only two workers
were transferred from an intervention kitchen to a control kitchen during the intervention.
Thus, one can state that contamination was probably minor. Co-interventions/external
changes were similar in the intervention and control groups. The most prominent external
change was the re-organization of food service in two of the participating cities during the
study. The intervention and control group were in the same phase of the changes
98
throughout the study. On average, series of four intervention and four control kitchens
entered sequentially in calendar time. Quality control based on regular meetings was
applied to ensure that there was similarity in the working methods between the researcher
teams working in the different cities. In addition, a project coordinator participated in the
workshops and observed the working of the researchers and provided them with feedback.
The inclusion of a 12-month post-intervention follow-up is also among the strengths of
this study. However, it may be that the intervention was not intensive enough. Due to
financial and time constraints, the follow-up period was limited to one year. An even
longer follow-up would have been needed to note the impacts on the musculoskeletal
system and to examine the persistence of the adverse changes in psychosocial factors at
work. Given the episodic nature of MSDs, the 3-month prevalence measures may not have
been optimal to capture a change in the frequency of bouts of MSDs. The measures of
local musculoskeletal fatigue during the past seven days were more sensitive. However, no
effect was detected using either fatigue measure.
A source of bias might arise from the fact that the workers had an active role in
implementing the ergonomic changes and the outcome assessments were based on their
subjective reports. It is conceivable that their involvement in an intervention program
increased the workers awareness of both ergonomics as well as musculoskeletal
problems. Being involved in the intervention may have encouraged the workers to develop
unrealistic expectations for the intervention to improve ergonomics and psychosocial work
environment. This might have been reflected via the increase in the adverse changes in
psychosocial factors at work.
The personal visits of a researcher to each of the kitchens encourage the workers to
complete the questionnaire probably was one of the reasons for the excellent response
rates. The researcher also documented changes in the control kitchens based on short
interviews. The visits can be interpreted as a kind of intervention for the control kitchens,
but their potential contribution to the results remains speculative. The visits may have
increased awareness of MSDs also in the control kitchens, which could have reduced
information bias regarding musculoskeletal outcomes. Documentation of the changes
during the visits may have activated the workers in the control kitchens to undertake more
99
changes, which could have diluted the overall effect. Still, the difference in the number of
changes made in the intervention kitchens and those that occurred spontaneously in the
control kitchens (402 vs. 80) is considerable. On the other hand, it has been recommended
that the groups should be dealt with as similarly as possible, except for the intervention
itself, to avoid the 'Hawthorne effect' (Westgaard and Winkel 1997; Shannon et al. 1999).
Intervention process
The intervention process has been carefully evaluated and reported (Pehkonen et al.
2009c). The process evaluation was based on research diaries, questionnaires, and focus
group interviews. The intervention model and the participatory approach as such were
found to have been successful, well accepted and perceived as motivating. The workers
ergonomics knowledge level increased and, in the workers own words, they developed an
'eye for ergonomics' with repect to their own work and that of others. Interestingly, most
of the workers felt that the intervention had had a positive effect on physical load and
musculoskeletal health, although the results did not support the hypothesis.
The workers conveyed that the feeling of togetherness within their own kitchen had
improved and that the intervention had facilitated discussion among and asking for help
from co-workers. The rotation of the workshops in different kitchens was felt to be
beneficial in enabling learning from each other's good practices. On the other hand, the
workers were dissatisfied with the support from the management and collaboration
between the kitchens, especially in one of the cities which was in the process of
implementing organizational reforms. The more satisfied the kitchens were with the
support from management, the better were the impacts of the intervention as evaluated by
the workers. The workers expected also more support from the researchers and technical
staff. A total of 113 planned changes were not completed. The most common reasons for
non-completion of changes were lack of motivation or time. The major observed obstacles
in improving kitchen ergonomics were passivity and resistance to change (Riihimki
2005b).
100
7.2.2. Clinical importance and generalization of the CRT findings
About 60% of all eligible kitchens took part in the study. Of the 80 non-participating
kitchens, some kitchens collectively refused to participate. In some cases, the decision of
refusal was administrative (e.g. made by the manager of the nursery or the headmaster of
the school), and in some kitchens at least one-third of the workers individually refused to
take part. In other cases, participation was withdrawn since there was going to be a major
reconstruction in the kitchen. The non-participating kitchens did not differ from their
participating counterparts in terms of size or location. However, no additional information
on the workers employed in the non-participating kitchens was available. Foe example, it
is possible, that physical or mental overstrain or poor social relationships at work could
have influenced the willingness to participate in a time-consuming intervention study. It
seems less likely that the workers in the non-participating kitchens would have had
different levels of pain or higher exposure to physical and/or mental workload compared
to those in the participating kitchens.
The results indicate that it is difficult to influence the heavy overall mental and physical
workload of kitchen workers without making structural changes in the kitchens and
redesigning whole work processes. There was no possibility to affect the basic design of
the kitchens or the amount of work by intervention. The ergonomic changes of the present
study mostly concerned micro-ergonomics. It may be that to obtain favourable effects on
the musculoskeletal health of the workers, a more comprehensive redesign of work
organization and processes is needed, taking more account of workers' physical and
mental resources (macro-ergonomics). The development of work should be a continuous
activity and it requires smooth collaboration between workers, management, planners,
architects, and technical staff.
The results can be generalized to other municipal kitchen workers. Although our study has
been evaluated to have a low risk of bias (Driessen et al. 2010), care must be taken in
generalizing the results beyond the target population. The amount of high quality RCTs or
non-randomized trials is insufficient to drawing conclusions of the utility of the PE
approach aimed at preventing MSDs. There is a need for further trials to elucidate this
issue.
101
7.2.3. Methodological aspects in studies of multiple-site musculoskeletal pain
In the cross-sectional study (III), both 3- and 12- month reference periods were used to
assess the prevalence of musculoskeletal pain. Different reference periods did not
materially affect the prevalence of pain. This is in accordance with previous Nordic
studies (rhede 1994). The 3-month reference period was considered more reliable due to
the shorter required span of memory, and was used in further analyses as also
recommended by rhede et al. (1994). The 3- month reference period was used also in all
other studies (I, II, IV). Illustrations showing the area of interest were used in the
questionnaire. A verbal description might leave too large a share for individual opinion on
what represents the neck, shoulder, forearm/hand, and low back. This probably enhanced
the reliability of reporting. It could be argued that a similar approach might have been
preferable for hip pain as well. On the other hand, van den Hoven et al. (2009) recently
assessed the agreement between showing the participants a manikin to locate the pain sites
and a self-administered questionnaire on musculoskeletal pain. They concluded that the
manikin gave similar findings on the prevalence of pain as written questions and they
stated it could be a good alternative for written questions only. However, the manikin
revealed higher pain prevalence rates. It could be that the use of a manikin to illustrate the
area of body site in addition to a written question may have lowered the threshold for
reporting pain in this study. One limitation in these studies was the lack of intensity and
frequency measures of pain.
The use of a trajectory analysis was a novel approach in analyzing MSP. In the field of
psychology, this method is commonly used, but it has not often been applied in the area of
musculoskeletal research. Trajectory analysis is developed for analyzing longitudinal data
and identifying different groups of individuals within a population who tend to have a
similar profile over time (trajectories). All available data points during a given period are
used. Even though the prevalence of musculoskeletal symptoms may be similar at
different time points in longitudinal studies, there is often high variation in pain reporting
over time at the individual level (Takala et al. 1992). Trajectory analysis was able to make
the maximum use of the data, examining the development of MSP as well as that of
psychosocial factors at work over time.
102
In conducting trajectory analysis, it has been recommended that measurements should be
available in at least half of the assessed time points if one wishes to reliably assess group
membership probabilities, and in at least three time points to detect trajectories of
quadratic shape (Kokko 2004). Thus, only subjects with observed values of MSP in at
least four time points of the nine possible were included. The sensitivity analysis, where
we dropped out 20 individuals with mean posterior assignment probabilities lower than
0.60, did not change the basic result of the number or shape of the MSP trajectories.
According to the mean posterior assignment probabilities, 69-92% of the subjects were
correctly assigned to the trajectories. In addition, the assignment probabilities for all
psychosocial work factor trajectories were over 0.70, as recommended (Nagin 2005).
The cut-off point for MSP is not well established. By trajectory analysis, the development
of MSP was first analyzed with MSP as a continuous variable. Forty-two per cent of the
workers were assigned to the trajectory comprised of those reporting pain in at least three
sites of seven. We then dichotomized MSP and used 3 pain sites as a cut-off point. The
results of the sensitivity analysis by using 4 pain sites as the cut-off point did not change
the basic result regarding the number and shape of the trajectories.
Although the sample was large enough to permit trajectory analysis (Kokko 2004), it was
relatively small for the assessment of risk, increasing imprecision of the estimates. Among
the limitations of this study could be the assessment of psychosocial factors at work by
single items. A single-item measure of mental stress has proved a valid measure, however
(Elo et al. 2003). All of the measured psychosocial factors at work have been identified as
risk factors for MSDs. The present study sample consisted of women in a single
occupational group. Hence, the generalizability of the results may be limited and can with
confidence be extended to other occupations where there are small work communities, a
high physical workload, and tight time constraints at work. Similar analyses among men
would be interesting and worthy of further study.
7.2.4. Self-reported outcome measures
All of the outcome measures were based on self-reports collected by questionnaires. Self-
reported symptoms collected by questionnaires have largely been basic data source in
epidemiologic research regarding MSDs. Questionnaires offer a quick and inexpensive
103
data collection on an individual basis from large samples. The use of self-reported pain
has been proposed to be the most valuable approach to measure outcome in population-
based surveys (Schierhout and Myers 1996). The International Association for the Study
of Pain defines pain as 'an unpleasent sensory and emotional experience associated with
actual or potential tissue damage or described in terms of such damage' (Merskey and
Bogduk 1994). Pain may also occur in the absence of tissue damage or any likely
pathophysiologic cause. Pain is always a subjective sensation. Yet, pain research has
recently provided new interesting findings based on brain imaging studies that have
identified within-subject relationships between regional pain activity and subjective pain
reports. However, the subjective report will probably remain the single most reliable index
for the magnitude of pain (Coghill et al. 2003).
The majority of research reports on the psychosocial work environment have been based
on self-reported assessments (Theorell and Hasselhorn 2005). The questions used in this
study were from a validated Occupational Stress Questionnaire (Elo et al. 1992; Elo et al.
1994). The problem of self-reported data is substantial in cross-sectional studies when
both psychosocial environment and health are described by self-reports and when both are
assessed concurrently. It is not clear to what extent self-reports of psychosocial work
factors reflect individual characteristics and to what extent they reflect true environmental
conditions. It has been argued that subjectivity bias may explain most of the observed
associations between psychosocial working conditions and health. Some people may have
a tendency to overreport problems in the environment and also in their own health
(negative affectivity). Similarly, some may underreport problems in both psychosocial
environment and health (denial) (Theorell and Hasselhorn 2005). Nahit et al. (2001)
discussed that one way to objectively measure an individual's psychosocial working
conditions could be to ask a co-worker or manager to do the assessment. Work unit
aggregated scores have been used to assess job strain among Finnish public sector
employees (Laine et al. 2009). Overcrowding in hospital wards has recently been used as a
proxy for mental strain of hospital staff (Virtanen et al. 2008). However, one may argue
that in some respects, the worker's own assessment is the most valid estimate, particularly
if the psychosocial work environment as perceived by the worker is being enquired. In the
present situation, it is possible that the awareness of both musculoskeletal pain and
104
psychosocial factors increased during the research period and lead to a more critical
evaluation by the workers.
A self-administered questionnaire is still the most widely used method also for assessing
physical workload (Barrero et al. 2009; Burdorf 2010). The reliability and validity of
physical workload assessments based on self-reports have been often criticized (Viikari-
Juntura et al. 1996; Buchholz et al. 2008; Burdorf 2010). In some previous studies, self-
reported physical workload has been shown to be a powerful predictor of MSDs and
physical functioning (Miranda et al. 2001; Leino-Arjas et al. 2004). Questionnaires are
also useful in identifying relative differences in exposure among occupational groups and
rank the groups according to their overall level of physical workload (Viikari-Juntura et al.
1996; Burdorf 2010). Direct measurements are costly and their laboriousness makes them
less feasible for use in epidemiologic studies. In a systematic review Takala et al. (2010)
recently evaluated observational methods assessing biomechanical exposures at work.
They found that though many methods exist, none of those evaluated appeared to be
superior to the others. Only some of the methods have been systematically tested for
validity, repeatability and feasibility.
7.3. Study findings in relation to the theoretical framework of the study
Karsh (2006) and Huang et al. (2002) have reviewed the theories and models of work-
related MSDs causation. These theories share many similarities. All of them emphasize
that psychological or physical exposures lead to responses that are moderated by
individual factors. Most of the models propose a feedback mechanism or cascading
effects. One common feature to all the theories is that many important aspects are not
defined, e.g. what is the magnitude and duration of an exposure that leads to certain
responses, or the length of the latency period between exposure and response. Some
theories contribute to the understanding of the relationships between psychosocial factors
and MSDs. Carayon et al. (1999) underline the role of psychophysiological mechanisms
(e.g. the effects of stress hormones, changes in regional blood flow). Workstyle is featured
in the Feuerstein (1996) model. Melin and Lundberg (1997) highlighted the influence of
demands outside the workplace. However, empirical support to the schemas of the theories
105
is limited and further research is needed to validate the multiple pathways that are
proposed.
The current study was based on the ecological model of Sauter and Swanson (1996),
which incorporates three components: psychosocial, biomechanical, and cognitive. The
cognitive component related to the attribution or interpretation of symptoms is a unique
aspect, as well as the proposition that tissue damage is not a necessary condition for the
development of symptoms. The model suggests that a complex cognitive process, which
involves the detection and attribution of symptoms, mediates the relationships between
biomechanical strain and the development of MSDs. The development of symptoms is
seen as a flexible and interpretative process, influenced by both the social context and the
subject's own experience.
The results of this study support the importance of this cognitive component. Pain and
psychosocial factors seem to be reciprocally linked together. There are individual
differences e.g. in pain sensitivity or the manner of experiencing the psychosocial working
environment. Many factors such as coping mechanisms, beliefs, motivation, past pain
history, other life experiences, personality, life situation, etc. can modify these individual
perceptions. The present results suggest that pain modifies the perceptions of psychosocial
factors, and that psychosocial factors modify pain perception. The model (Sauter and
Swanson 1996) shows these reciprocal links between MSDs, work organization and
psychological strain mediated by the cognitive process.
The perception of pain may also modify perceived physical workload. Physical activity, in
terms of biomechanical exposure at work or leisure time, may influence pain perception.
In the model (Sauter and Swanson 1996), the link between MSDs and biomechanical load
is neither direct nor reciprocal. The possible reciprocal associations between perceived
physical workload and pain experiences would be worthy of further study. If that link
would prove reciprocal, it would further strengthen the importance of individual
sensations and cognition in the development of MSDs. As a corollary, interventions
should focus on secondary and tertiary prevention, to guide people to manage and control
disorders whenever they appear.
106
8. CONCLUSIONS
While this PE intervention was successful in engaging the kitchen workers to evaluate and
improve the ergonomics of their daily work, no systematic differences in any health
outcomes were found between the intervention and control groups during the intervention
or during the 12-month post-intervention follow-up. The intervention did not reduce
perceived physical workload. The effects on psychosocial factors at work were
unfavorable. These adverse changes were mainly due to a joint effect of the intervention
and unconnected organizational reforms of foodservice in two of the participating cities.
This may partly explain why the intervention did not have any effect on the occurrence of
MSDs. The occurrence of MSP among female kitchen workers was high and more
common than single-site pain. Psychosocial factors at work and mental stress strongly
predicted MSP over time and, vice versa, MSP predicted psychosocial factors and mental
stress. Such reciprocality of the relationships suggests either two mutually dependent
processes in time, or some common underlying factor(s).
9. IMPLICATIONS OF THE RESULTS
The challenges in MSD research and in preventive work include the multifactorial
etiology, the high occurrence of many MSDs and particularly that of MSP, and the
episodic, recurrent and subjective nature of musculoskeletal pain. The workforce is ageing
in the industrialized countries. At the same time, there is pressure to extend working
careers beyond the current retirement age. Interventions that focus on maintaining
employees' work ability, controlling sick leaves, and supporting return to work are needed.
The results do not support the usefulness of the studied PE intervention in preventing
MSDs or in changing unsatisfactory psychosocial working conditions. If organizational
changes are expected to occur, it is essential to take caution with the implementation of
other workplace interventions at the same time. Developing ergonomics should be a
continuous activity in fluent collaboration with planners, architects, technical staff,
107
management, and workers themselves as the best expert in their own work. Further high
quality trials are needed to elucidate the effectiveness of ergonomics interventions. The
current study showed that conducting RCT at workplaces is feasible.
The evaluation and prevention of pain at multiple anatomical sites might be useful in
addition to considering single-site pain, both in the daily work of occupational and
healthcare professionals, in epidemiological research, and in further intervention studies.
Since MSP and psychosocial factors at work seem to be strongly linked together, parallel
assessment of both measures is recommended. As subjective perceptions, they may share
some common underlying factors, which need to be clarified in further studies. More
objective methodological approaches for assessing psychosocial factors at work and of
MSDs are also needed.
108
ACKNOWLEDGEMENTS
The present work is based on the 'ERGO-study' conducted in the Finnish Institute of
Occupational Health. The Academy of Finland, the Local Government Pensions
Institution, the Ministry of Labour, and the Finnish Work Environment Fund financially
supported the study. The last mentioned one also granted me a one year personal award.
The Finnish Cultural Foundation has awarded me a one year scholarship to complete the
thesis and for post-doctoral studies. They are all gratefully acknowledged. I also thank the
Institute for placing the facilities at my disposal.
I wish to acknowledge the collaborative kitchen workers and representatives of food
services in Espoo, Tampere, Turku and Vantaa, for participating in and their commitment
to the study. I want to express my warmest thanks to all members of the ERGO research
group. I have had the privilege of working in this excellent team that has shared its
scientific expertise and experience with me. I am in debt to many people for their
contributions during different phases of this study.
I am deeply grateful to my supervisors, the leader of ERGO study, Professor (emerita)
Hilkka Riihimki, and Docent Pivi Leino-Arjas, for their sincere interest in this work and
patiently giving me such skilful guidance. Despite their many duties, they have found time
for discussions essential to push my work forward. They both have encouraged me,
offered advice and constructive comments when needed, but also given possibilities to
think and work independently. Hilkka, your endless optimism and trust on my work were
invaluable, and Pivi, your knowledge on scientific writing is beyond compare. Both of
you introduced me to the world of epidemiology and musculoskeletal research. I have
learnt so much from you.
I wish to warmly thank all my co-authors of the original papers included in this thesis for
co-operation. I offer special thanks to Eira Viikari-Juntura for her valuable comments on
all my four papers, to Esa-Pekka Takala for his advice, encouraging attitude, and
providing help when needed, and to Antti Malmivaara for his expertise in randomized
trials. I owe much to Pertti Mutanen, Elina Nykyri, Svetlana Solovieva, and especially to
Anneli Ojajrvi, for their expert knowledge and advice in statistics during the study.
109
Anneli, your talent to take over the challenging analyses, in particular with respect to the
fourth paper, was admirable. Irmeli Pehkonen, my closest co-author and a fellow student,
has always understood me without words. Thank you for your friendship and sharing this
'journey' with me. I also wish to thank my other colleagues in the Institute for their
support.
My sincere thanks belong to the official reviewers, Professor Clas-Hkan Nygrd and
Docent Simo Taimela, for their valuable suggestions for improving the manuscript, as well
as to Ewen MacDonald for rapid revision of the language. Professor Allard van der Beek
is warmly acknowledged for accepting the appointment of being my opponent.
Finally, I am most grateful to my family and friends. I could not have completed this work
without their encouragement. My deepest thanks go to my beloved husband Veli, for his
love, inconceivable patience, and support, spurring me on throughout this work especially
during hard times. Words cannot express my feelings for our lovely children, Milla and
Miska, both of them born during these 'ERGO- years'. Their love and existence is more
meaningful to me than anything. I dedicate this work to them.
Espoo, September 2010
Eija Haukka
110
REFERENCES
Aas RW, Holte KA, Labriola M, Lund T, Tuntland H, Re C, Merkus S, Moller A, Marklund S. Workplace
interventions for low-back pain in workers (protocol). Cochrane Database of Systematic Reviews
2009a (4).
Aas RW, Holte KA, Tuntland H, Re C, Labriola M, Lund T, Moller A, Skaugen BH, Marklund S.
Workplace interventions for neck pain in workers (protocol). Cochrane Database of Systematic
Reviews 2009b (4).
Adamson G, Murphy S, Shevlin M, Buckle P, Stubbs D. Profiling schoolchildren in pain and associated
demographic and behavioural factors: a latent class approach. Pain 2007;129(3):295-303.
Aggarwal VR, McBeth J, Zakrzewska JM, Lunt M, Macfarlane GJ. The epidemiology of chronic syndromes
that are frequently unexplained: do they have common associated factors? Int J Epidemiol
2006;35(2):468-476.
Alexopoulos EC, Stathi IC, Charizani F. Prevalence of musculoskeletal disorders in dentists. BMC
Musculoskeletal Disorders 2004;5(1):16.
Ammendolia C, Kerr MS, Bombardier C. Back belt use for prevention of occupational low back pain: a
systematic review. J Manipulative Physiol Ther 2005;28(2):128-134.
Andersson H, Ejlertsson G, Leden I. Widespread musculoskeletal chronic pain associated with smoking. An
epidemiological study in a general rural population. Scand J Rehabil Med 1998;30(3):185-191.
Andersson HI. The course of non-malignant chronic pain: a 12-year follow-up of a cohort from the general
population. Eur J Pain 2004;8(1):47-53.
Arins GA, van Mechelen W, Bongers PM, Bouter LM, van der Wal G. Psychosocial risk factors for neck
pain: a systematic review. Am J Ind Med 2001;39(2):180-193.
Atienza AA, King AC. Community-based health intervention trials: an overview of methodological issues.
Epidemiol Rev 2002;24(1):72-79.
Auvinen JP, Paananen MV, Tammelin TH, Taimela SP, Mutanen PO, Zitting PJ, Karppinen JI.
Musculoskeletal pain combinations in adolescents. Spine (Phila Pa 1976) 2009;34(11):1192-1197.
Bakker EW, Verhagen AP, Van Trijffel E, Lucas C, Koes BW. Spinal mechanical load as a risk factor for
low back pain: a systematic review of prospective cohort studies. Spine 2009;34(8):E281-293.
Baldwin ML. Reducing the costs of work-related musculoskeletal disorders: targeting strategies to chronic
disability cases. J Electromyogr Kinesiol 2004;14(1):33-41.
Bambra C, Egan M, Thomas S, Petticrew M, Whitehead M. The psychosocial and health effects of
workplace reorganisation. 2. A systematic review of task restructuring interventions. J Epidemiol
Community Health 2007;61(12):1028-1037.
Barrero LH, Katz JN, Dennerlein JT. Validity of self-reported mechanical demands for occupational
epidemiologic research of musculoskeletal disorders. Scand J Work Environ Health
2009;35(4):245-260.
Battie MC, Videman T, Levalahti E, Gill K, Kaprio J. Heritability of low back pain and the role of disc
degeneration. Pain 2007;131(3):272-280.
111
Bernard B. 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 CDC (NIOSH), DHHS (NIOSH). Publication No. 97-
141. Cincinnati,1997. http://www.cdc.gov/niosh/docs/97-141/ (accessed 10 September 2010).
Bevan S, Quadrello T, McGee R, Mahdon M, Vavrovsky A, Barham L. Fit for work? Musculoskeletal
disorders in the European workforce. London: The Work Foundation, 2009.
http://www.theworkfoundation.com/Assets/Docs/Fit%20for%20Work%20pan-
European%20report.pdf (accessed 10 September 2010).
Bierma-Zeinstra SM, Koes BW. Risk factors and prognostic factors of hip and knee osteoarthritis. Nat Clin
Pract Rheumatol 2007;3(2):78-85.
Bigos SJ, Holland J, Holland C, Webster JS, Battie M, Malmgren JA. High-quality controlled trials on
preventing episodes of back problems: systematic literature review in working-age adults. Spine J
2009;9(2):147-168.
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;8(5):435-450.
Blatter B, Bongers PM. Work related neck and upper limb symptoms (RSI): High risk occupations and risk
factors in the Dutch working population. TNO-report 4070117/r9800293. Hoofddorp: TNO
Work,1999. http://www.arbobondgenoten.nl/arbothem/lichblst/rsi/enquete2000/tnorapport.pdf
(accessed 10 September 2010).
Bohr PC. Efficacy of office ergonomics education. Journal of Occupational Rehabilitation 2000;10(4):243-
255.
Bohr PC. Office ergonomics education: a comparison of traditional and participatory methods. Work
2002;19(2):185-191.
Bonate P. Analysis of pretest-posttest designs. USA: Chapman & Hall/CRC, 2000:105-110.
Bongers PM. Are ergonomic interventions effective and worth the cost in preventing or reducing MSDs?.
Proceedings of the 17th World Congress on Ergonomics, International Ergonomics Association
(IEA); August 9-14, 2009, Beijing, China [CD-ROM].
Bongers PM, de Winter CR, Kompier MA, Hildebrandt VH. Psychosocial factors at work and
musculoskeletal disease. Scand J Work Environ Health 1993;19(5):297-312.
Bongers PM, Ijmker S, van den Heuvel S, Blatter BM. Epidemiology of work related neck and upper limb
problems: Psychosocial and personal risk factors (Part I) and effective interventions from a bio
behavioural perspective (Part II). J Occup Rehabil 2006;16(3):272-295.
Bongers PM, Kremer AM, ter Laak J. 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;41(5):315-342.
Boocock MG, McNair PJ, Larmer PJ, Armstrong B, Collier J, Simmonds M, Garrett N. Interventions for the
prevention and management of neck/upper extremity musculoskeletal conditions: a systematic
review. Occup Environ Med 2007;64(5):291-303.
112
Bos EH, Krol B, Van Der Star A, Groothoff JW. The effects of occupational interventions on reduction of
musculoskeletal symptoms in the nursing profession. Ergonomics 2006;49(7):706-723.
Braunholtz DA, Edwards SJ, Lilford RJ. Are randomized clinical trials good for us (in the short term)?
Evidence for a "trial effect". J Clin Epidemiol 2001;54(3):217-224.
Brewer S, van Eerd D, Amick BC III, Irvin E, Daum KM, Gerr F, Moore JS, Cullen K, Rempel D.
Workplace interventions to prevent musculoskeletal and visual symptoms and disorders among
computer users: a systematic review. J Occup Rehabil 2006;16(3):325-358.
Brown O. Participatory ergonomics (PE). In: Stanton N, Hedge A, Brookhuis K, Salas E, Hendrick H, eds.
Handbook of human factors and ergonomics methods. USA: CRC Press, 2005:81-1-81-7.
Buchholz B, Park JS, Gold JE, Punnett L. Subjective ratings of upper extremity exposures: inter-method
agreement with direct measurement of exposures. Ergonomics 2008;51(7):1064-1077.
Buckle P. Ergonomics and musculoskeletal disorders: overview. Occup Med (Lond) 2005;55(3):164-167.
Buckle PW, Devereux JJ. The nature of work-related neck and upper limb musculoskeletal disorders. Appl
Ergon 2002;33(3):207-217.
Burdorf A. The role of assessment of biomechanical exposure at the workplace in the prevention of
musculoskeletal disorders. Scand J Work Environ Health 2010;36(1):1-2.
Burdorf A, Sorock G. Positive and negative evidence of risk factors for back disorders. Scand J Work
Environ Health 1997;23(4):243-256.
Buring JE. Special issues related to randomized trials of primary prevention. Epidemiol Rev 2002;24(1):67-
71.
Burton AK, Balague F, Cardon G, Eriksen HR, Henrotin Y, Lahad A, Leclerc A, Muller G, van der Beek
AJ. Chapter 2. European guidelines for prevention in low back pain: November 2004. Eur Spine J
2006;15 Suppl 2:S136-168.
Campbell M, Grimshaw J, Steen N. Sample size calculations for cluster randomised trials. Changing
Professional Practice in Europe Group (EU BIOMED II Concerted Action). J Health Serv Res
Policy 2000;5(1):12-16.
Campbell MK, Elbourne DR, Altman DG. CONSORT statement: extension to cluster randomised trials.
Bmj 2004;328(7441):702-708.
Carayon P, Smith MJ, Haims MC. Work organization, job stress, and work-related musculoskeletal
disorders. Hum Factors 1999;41(4):644-663.
Carnes D, Parsons S, Ashby D, Breen A, Foster NE, Pincus T, Vogel S, Underwood M. Chronic
musculoskeletal pain rarely presents in a single body site: results from a UK population study.
Rheumatology (Oxford) 2007;46(7):1168-1170.
Chen SM, Liu MF, Cook J, Bass S, Lo SK. Sedentary lifestyle as a risk factor for low back pain: a
systematic review. Int Arch Occup Environ Health 2009;82(7):797-806.
Christmansson M, Friden J, Sollerman C. Task design, psycho-social work climate and upper extremity pain
disorders - effects of an organisational redesign on manual repetitive assembly jobs. Appl Ergon
1999;30(5):463-472.
113
Coghill RC, McHaffie JG,Yen YF. Neural correlates of interindividual differences in the subjective
experience of pain. Proc Natl Acad Sci U S A 2003;100(14):8538-8542.
Cole D, Rivilis I. Individual factors and musculoskeletal disorders. In: Marras W, Karwowski W, eds.
Fundamentals and assessment tools for occupational ergonomics. The occupational ergonomics
handbook. 2nd edn. New York: Taylor & Francis, 2006:19-1-19-10.
Cole D, Rivilis I, van Eerd D, Cullen K, Kramer D. Effectiveness of participatory ergonomic interventions: a
systematic review. Overview. Toronto: Institute for Work & Health, 2005.
Cole DC,van Eerd D, Bigelow P, Rivilis I. Integrative interventions for MSDs: nature, evidence, challenges
& directions. J Occup Rehabil 2006;16(3):359-374.
Cole DC, Wells RP, Frazer MB, Kerr MS, Neumann WP, Laing AC. Methodological issues in evaluating
workplace interventions to reduce work-related musculoskeletal disorders through mechanical
exposure reduction. Scand J Work Environ Health 2003;29(5):396-405.
Ct P, van der Velde G, Cassidy JD, Carroll LJ, Hogg-Johnson S, Holm LW, Carragee EJ, Haldeman S,
Nordin M, Hurwitz EL, Guzman J, Peloso PM. The burden and determinants of neck pain in
workers: results of the Bone and Joint Decade 2000-2010 Task Force on Neck Pain and Its
Associated Disorders. Spine (Phila Pa 1976) 2008;33(4 Suppl):S60-74.
Croft P. The question is not "have you got it"? But "how much of it have you got"? Pain 2009;141(1-2):6-7.
Croft P, Dunn KM, Von Korff M. Chronic pain syndromes: you can't have one without another. Pain
2007;131(3):237-238.
D'Souza JC, Franzblau A, Werner RA. Review of epidemiologic studies on occupational factors and lower
extremity musculoskeletal and vascular disorders and symptoms. J Occup Rehabil 2005;15(2):129-
165.
Davies HT, Crombie IK, Macrae WA. Where does it hurt? Describing the body locations of chronic pain.
Eur J Pain 1998;2(1):69-80.
Davis KG, Heaney CA.The relationship between psychosocial work characteristics and low back pain:
underlying methodological issues. Clin Biomech (Bristol, Avon) 2000;15(6):389-406.
de Zwart BC, Broersen JP, Frings-Dresen MH, van Dijk FJ. Repeated survey on changes in musculoskeletal
complaints relative to age and work demands. Occupational and environmental medicine
1997;54(11):793-799.
Demure B, Mundt KA, Bigelow C, Luippold RS, Ali D, Liese B. Video display terminal workstation
improvement program: II. Ergonomic intervention and reduction of musculoskeletal discomfort. J
Occup Environ Med 2000;42(8):792-797.
Donner A, Klar N. Design and analysis of cluster randomization trials in health research. London: Arnold,
2000.
Driessen MT, Anema JR, Proper KI, Bongers PM, van der Beek AJ. Stay@Work: Participatory Ergonomics
to prevent low back and neck pain among workers: design of a randomised controlled trial to
evaluate the (cost-) effectiveness. BMC Musculoskelet Disord 2008;9:145.
114
Driessen MT, Proper KI, van Tulder MW, Anema JR, Bongers PM, van der Beek AJ. The effectiveness of
physical and organisational ergonomic interventions on low back pain and neck pain: a systematic
review. Occupational and environmental medicine 2010;67(4):277-285.
Egan M, Bambra C, Thomas S, Petticrew M, Whitehead M, Thomson H. The psychosocial and health
effects of workplace reorganisation. 1. A systematic review of organisational-level interventions
that aim to increase employee control. J Epidemiol Community Health 2007;61(11):945-954.
Elo A-L, Leppnen A, Lindstrm K, Ropponen T. OSQ-occupational stress questionnaire: users
instructions (reviews 19). Helsinki: Finnish Institute of Occupational Health, 1992.
Elo A-L, Leppnen A, Jahkola A.Validity of a single-item measure of stress symptoms. Scand J Work
Environ Health 2003;29(6):444-451.
Elo A-L, Leppnen A, Lindstrm K, Ropponen T. The Occupational Sress Questionnaire. In: Zenz C,
Dickerson O, Horvath E, eds. Occupational Medicine. 3rd edn. St.Louis: MO Mosby, 1994:1234-
1237.
EPIQ. A Generic Appraisal Tool for Epidemiology (GATE). 2004.
http://user.meduni-graz.at/andrea.berghold/EBM/Gate_Notes.pdf (accessed 10 September 2010).
European Agency for Safety and Health at Work. E-facts 24. Musculoskeletal disorders (MSDs) in
HORECA. Publishing date 11.1. 2008. http://osha.europa.eu/en/publications/e-facts/efact24
(accessed 10 September 2010).
Evanoff BA, Bohr PC, Wolf LD. Effects of a participatory ergonomics team among hospital orderlies. Am J
Ind Med 1999;35(4):358-365.
Fejer R, Hartvigsen J, Kyvik KO. Heritability of neck pain: a population-based study of 33,794 Danish
twins. Rheumatology (Oxford) 2006;45(5):589-594.
Feuerstein M.Work style: definition, empirical support, and implications for prevention, evaluation, and
rehabilitation of occupational upper-extremity disorders. In: Moon SD, Sauter SL, eds. Beyond
Biomechanics: psychosocial aspects of musculoskeletal disorders in office work. Bristol, PA:
Taylor and Francis, 1996:177-206.
Feuerstein M, Marshall L, Shaw WS, Burrell LM. Multicomponent intervention for work-related upper
extremity disorders. Journal of Occupational Rehabilitation 2000;10(1):71-83.
Finnish Centre for Pensions. Statistical yearbook of pensioners in Finland, 2007. Helsinki: Edita Prima Oy,
2008.
Forma P. Incidence of disability pensions among municipal occupational groups from 1995 to 2001 In:
Forma P, Blomster P, Halmeenmki T, Peltonen H, Tiilikka T. Articles about the municipal
pensions system (in Finnish). Helsinki: The Local Government Pensions Institution, 2004:54-71.
Franceshi S, Plummer M. Intervention trials. In: Ahrens W, Pigeot I, eds. Handbook of epidemiology.
Berlin: Springer, 2005;345-370.
Fredriksson K, Bildt C, Hagg G, Kilbom A. The impact on musculoskeletal disorders of changing physical
and psychosocial work environment conditions in the automobile industry. International Journal of
Industrial Ergonomics 2001;28(1):31-45.
115
Gatty CM, Turner M, Buitendorp DJ, Batman H. The effectiveness of back pain and injury prevention
programs in the workplace. Work 2003;20(3):257-266.
Goldenhar LM, LaMontagne AD, Katz T, Heaney C, Landsbergis P. The intervention research process in
occupational safety and health: an overview from the National Occupational Research Agenda
Intervention Effectiveness Research team. J Occup Environ Med 2001;43(7):616-622.
Goldenhar LM, Schulte PA. Methodological issues for intervention research in occupational health and
safety. Am J Ind Med 1996;29(4):289-294.
Gran JT. The epidemiology of chronic generalized musculoskeletal pain. Best Pract Res Clin Rheumatol
2003;17(4):547-561.
Green SB. Design of randomized trials. Epidemiol Rev 2002;24(1):4-11.
Grimshaw J, Campbell M, Eccles M, Steen N. Experimental and quasi-experimental designs for evaluating
guideline implementation strategies. Fam Pract 2000;17 Suppl 1:S11-16.
Hagberg M. Clinical assessment of musculoskeletal disorders in workers exposed to hand-arm vibration. Int
Arch Occup Environ Health 2002;75(1-2):97-105.
Hagberg M, Silverstein B, Wells R, Smith M, Hendrick H, Carayon P, Prusse M. In: Kuorinka I, Forcier L,
eds. Work related musculoskeletal disorders (WMSDs): a reference book for prevention. London:
Taylor & Francis, 1995.
Hagen EM, Svensen E, Eriksen HR, Ihlebaek CM, Ursin H. Comorbid subjective health complaints in low
back pain. Spine 2006;31(13):1491-1495.
Haims MC, Carayon P. Theory and practice for the implementation of 'in-house', continuous improvement
participatory ergonomic programs. Appl Ergon 1998;29(6):461-472.
Haines H, Wilson J. Development of a framework for participatory ergonomics. Health and Safety
executive. Contract Research Report 174/1998. London: HSE Books, 1998.
http://www.hse.gov.uk/research/crr_pdf/1998/crr98174.pdf (accessed 10 September 2010)
Hamberg-van Reenen HH, Ariens GA, Blatter BM, van Mechelen W, Bongers PM. A systematic review of
the relation between physical capacity and future low back and neck/shoulder pain. Pain
2007;130(1-2):93-107.
Hansson T, Westerholm P, (red). Arbete och besvr i rrelseorganen. En vetenskaplig vrdering av frgor
om samband. Stockholm: Arbete och Hlsa, Arbetslivsinstitutet 2001;12.
Harkness EF, Macfarlane GJ, Nahit E, Silman AJ, McBeth J. Mechanical injury and psychosocial factors in
the work place predict the onset of widespread body pain: a two-year prospective study among
cohorts of newly employed workers. Arthritis Rheum 2004;50(5):1655-1664.
Harkness EF, Macfarlane GJ, Nahit ES, Silman AJ, McBeth J. Mechanical and psychosocial factors predict
new onset shoulder pain: a prospective cohort study of newly employed workers. Occup Environ
Med 2003;60(11):850-857.
Hartvigsen J, Lings S, Leboeuf-Yde C, Bakketeig L. Psychosocial factors at work in relation to low back
pain and consequences of low back pain; a systematic, critical review of prospective cohort studies.
Occup Environ Med 2004;61(1):e2.
116
Hartvigsen J, Nielsen J, Kyvik KO, Fejer R, Vach W, Iachine I, Leboeuf-Yde C. Heritability of spinal pain
and consequences of spinal pain: a comprehensive genetic epidemiologic analysis using a
population-based sample of 15,328 twins ages 20-71 years. Arthritis Rheum 2009;61(10):1343-
1351.
Hartvigsen J, Petersen HC, Frederiksen H, Christensen K. Small effect of genetic factors on neck pain in old
age: a study of 2,108 Danish twins 70 years of age and older. Spine (Phila Pa 1976)
2005;30(2):206-208.
Heaney C, Fujishiro K. Psychosocial characteristics of work as targets for change. In: Marras WS,
Karwowski W, eds. Interventions, controls, and applications in occupational ergonomics. The
occupational ergonomics handbook. 2nd edn. New York: Taylor & Francis, 2006;25-1-25-14.
Hignett S, Wilson JR, Morris W. Finding ergonomic solutions - participatory approaches. Occup Med
(Lond) 2005;55(3):200-207.
Hogg-Johnson S, van der Velde G, Carroll LJ, Holm LW, Cassidy JD, Guzman J, Ct P, Haldeman S,
Ammendolia C, Carragee E, Hurwitz E, Nordin M, Peloso P. The burden and determinants of neck
pain in the general population: results of the Bone and Joint Decade 2000-2010 Task Force on Neck
Pain and Its Associated Disorders. Spine (Phila Pa 1976) 2008;33(4 Suppl):S39-51.
Hooftman WE, van Poppel MN, van der Beek AJ, Bongers PM, van Mechelen W. Gender differences in the
relations between work-related physical and psychosocial risk factors and musculoskeletal
complaints. Scand J Work Environ Health 2004;30(4):261-278.
Hoogendoorn WE, van Poppel MN, Bongers PM, Koes BW, Bouter LM. Physical load during work and
leisure time as risk factors for back pain. Scand J Work Environ Health 1999;25(5):387-403.
Hoogendoorn WE, van Poppel MN, Bongers PM, Koes BW, Bouter LM. Systematic review of psychosocial
factors at work and private life as risk factors for back pain. Spine 2000;25(16):2114-2125.
Hooper MM. Tending to the musculoskeletal problems of obesity. Cleve Clin J Med 2006;73(9):839-845.
Huang GD, Feuerstein M, Sauter SL. Occupational stress and work-related upper extremity disorders:
concepts and models. Am J Ind Med 2002;41(5):298-314.
Huang J, Ono Y, Shibata E, Takeuchi Y, Hisanaga N. Occupational musculoskeletal disorders in lunch
centre workers. Ergonomics 1988;31(1):65-75.
Ijzelenberg W, Burdorf A. Impact of musculoskeletal co-morbidity of neck and upper extremities on
healthcare utilisation and sickness absence for low back pain. Occup Environ Med
2004;61(10):806-810.
Imada AS. The rationale and tools of participatory ergonomics. In: Noro K, Imada A.S, eds. Participatory
ergonomics. London: Taylor & Francis, 1991:30-51.
Irving DB, Cook JL, Menz HB. Factors associated with chronic plantar heel pain: a systematic review. J Sci
Med Sport 2006;9(1-2):11-22; discussion 23-14.
IWH. Factors for success in participatory ergonomics. Institute for Work & Health, 2008.
http://www.iwh.on.ca/sbe/factors-for-success-in-participatory-ergonomics (accessed 10 September
2010).
117
Jellema P, van Tulder MW, van Poppel MN, Nachemson AL, Bouter LM. Lumbar supports for prevention
and treatment of low back pain: a systematic review within the framework of the Cochrane Back
Review Group. Spine (Phila Pa 1976) 2001;26(4):377-386.
Jensen LK. Hip osteoarthritis: influence of work with heavy lifting, climbing stairs or ladders, or combining
kneeling/squatting with heavy lifting. Occup Environ Med 2008a;65(1):6-19.
Jensen LK. Knee osteoarthritis: influence of work involving heavy lifting, kneeling, climbing stairs or
ladders, or kneeling/squatting combined with heavy lifting. Occup Environ Med 2008b;65(2):72-
89.
Jones BL, Nagin DS. Advances in group-based trajectory modeling and an SAS procedure for estimating
them. Sociological Methods & Research 2007;35(4):542-571.
Jones GT, Silman AJ, Macfarlane GJ. Predicting the onset of widespread body pain among children.
Arthritis Rheum 2003;48(9):2615-2621.
Jni P, Altman DG, Egger M. Systematic reviews in health care: Assessing the quality of controlled clinical
trials. Bmj 2001;323(7303):42-46.
Kaila-Kangas L, ed. Musculoskeletal disorders and diseases in Finland - results of the Health 2000 Survey.
Publications of the National Public Health Institute B 25. Helsinki, 2007.
Kalichman L, Hunter DJ. The genetics of intervertebral disc degeneration. Familial predisposition and
heritability estimation. Joint Bone Spine 2008;75(4):383-387.
Kamaleri Y, Natvig B, Ihlebaek CM, Benth JS, Bruusgaard D. Number of pain sites is associated with
demographic, lifestyle, and health-related factors in the general population. Eur J Pain
2008a;12(6):742-748.
Kamaleri Y, Natvig B, Ihlebaek CM, Benth JS, Bruusgaard D. Change in the number of musculoskeletal
pain sites: A 14-year prospective study. Pain 2009;141(1-2):25-30.
Kamaleri Y, Natvig B, Ihlebaek CM, Bruusgaard D. Does the number of musculoskeletal pain sites predict
work disability? A 14-year prospective study. Eur J Pain 2008b.
Kamaleri Y, Natvig B, Ihlebaek CM, Bruusgaard D. Localized or widespread musculoskeletal pain: does it
matter? Pain 2008c;138(1):41-46.
Karsh B. Theories of work-related musculoskeletal disorders: implications for ergonomic interventions.
Theoretical Issues in Ergonomics Science 2006;7(1):71-88.
Karsh B, Moro F, Smith M. The efficacy of workplace ergonomic interventions to control musculoskeletal
disorders:a critical analysis of the peer-reviewed literature. Theoretical Issues in Ergonomics
Science 2001;2(1):23-96.
Kauppila LI. Atherosclerosis and disc degeneration/low-back pain--a systematic review. Eur J Vasc
Endovasc Surg 2009;37(6):661-670.
Ketola R, Toivonen R, Hkknen M, Luukkonen R, Takala EP, Viikari-Juntura E. Effects of ergonomic
intervention in work with video display units. Scand J Work Environ Health 2002;28(1):18-24.
Kobayashi Y, Kaneyoshi A, Yokota A, Kawakami N. Effects of a worker participatory program for
improving work environments on job stressors and mental health among workers: a controlled trial.
J Occup Health 2008;50(6):455-470.
118
Kogi K. Participatory methods effective for ergonomic workplace improvement. Appl Ergon
2006;37(4):547-554.
Kokko K. An applied viewpoint to the trajectory analysis method (in Finnish). Psykologia 2004;39:424-433.
Kristensen TS. Intervention studies in occupational epidemiology. Occup Environ Med 2005;62(3):205-210.
Kuorinka I, Jonsson B, Kilbom A, Vinterberg H, Biering-Sorenson F, Anderson G, Jorgensen K.
Standardised Nordic Questionnaires for the analysis of musculoskeletal symptoms. Appl Ergon
1987;18:233-237.
Kuorinka I. Tools and means of implementing participatory ergonomics. International Journal of Industrial
Ergonomics 1997;19(4):267-270.
Lagerstrm M, Josephson M, Pingel B, Tjernstrom G, Hagberg M, Grp MS. Evaluation of the
implementation of an education and training programme for nursing personnel at a hospital in
Sweden. International Journal of Industrial Ergonomics 1998;21(1):79-90.
Laine S, Gimeno D, Virtanen M, Oksanen T, Vahtera J, Elovainio M, Koskinen A, Pentti J, Kivimki M.
Job strain as a predictor of disability pension: the Finnish Public Sector Study. J Epidemiol
Community Health 2009;63(1):24-30.
Laing AC, Cole DC, Theberge N, Wells RP, Kerr MS, Frazer MB. Effectiveness of a participatory
ergonomics intervention in improving communication and psychosocial exposures. Ergonomics
2007;50(7):1092-1109.
Laing AC, Frazer MB, Cole DC, Kerr MS, Wells RP, Norman RW. Study of the effectiveness of a
participatory ergonomics intervention in reducing worker pain severity through physical exposure
pathways. Ergonomics 2005;48(2):150-170.
Larsson B, Sund AM. Emotional/behavioural, social correlates and one-year predictors of frequent pains
among early adolescents: influences of pain characteristics. Eur J Pain 2007;11(1):57-65.
Leboeuf-Yde C, Nielsen J, Kyvik KO, Fejer R, Hartvigsen J. Pain in the lumbar, thoracic or cervical
regions: do age and gender matter? A population-based study of 34,902 Danish twins 20-71 years
of age. BMC Musculoskelet Disord 2009;10:39.
Leino-Arjas P, Solovieva S, Riihimki H, Kirjonen J, Telama R. Leisure time physical activity and
strenuousness of work as predictors of physical functioning: a 28 year follow up of a cohort of
industrial employees. Occup Environ Med 2004;61(12):1032-1038.
Leino P, Magni G.Depressive and distress symptoms as predictors of low back pain, neck-shoulder pain, and
other musculoskeletal morbidity: a 10-year follow-up of metal industry employees. Pain
1993;53(1):89-94.
Leino PI, Hnninen V. Psychosocial factors at work in relation to back and limb disorders. Scand J Work
Environ Health 1995;21(2):134-142.
Leppnen A. Improving the mastery of work and the development of the work process in paper production.
Relations Industrielles/ Industrial Relations 2001;56(3):579-609.
Liang K-Y, Zeger SL. Longitudinal data analysis using generalized linear models. Biometrika 1986;73:13-
22.
119
Lincoln AE, Vernick JS, Ogaitis S, Smith GS, Mitchell CS, Agnew J. Interventions for the primary
prevention of work-related carpal tunnel syndrome. Am J Prev Med 2000;18(4 Suppl):37-50.
Linton SJ. Occupational psychological factors increase the risk for back pain: a systematic review. J Occup
Rehabil 2001;11(1):53-66.
Linton SJ, van Tulder MW. Preventive interventions for back and neck pain problems: what is the evidence?
Spine 2001;26(7):778-787.
Lis AM, Black KM, Korn H, Nordin M. Association between sitting and occupational LBP. Eur Spine J
2007;16(2):283-298.
Ltters F, Burdof A. Are changes in mechanical exposure and musculoskeletal health good performance
indicators for primary interventions? Int Arch Occup Environ Health 2002;75(8):549-561.
Macfarlane GJ, Hunt IM, Silman AJ. Role of mechanical and psychosocial factors in the onset of forearm
pain: prospective population based study. Bmj 2000;321(7262):676-679.
Macfarlane GJ, Pallewatte N, Paudyal P, Blyth FM, Coggon D, Crombez G, Linton S, Leino-Arjas P,
Silman AJ, Smeets RJ, van der Windt D. Evaluation of work-related psychosocial factors and
regional musculoskeletal pain: results from a EULAR Task Force. Ann Rheum Dis
2009;68(6):885-891.
Mallen CD, Peat G, Thomas E, Dunn KM, Croft PR. Prognostic factors for musculoskeletal pain in primary
care: a systematic review. Br J Gen Pract 2007;57(541):655-661.
Marras WS, Cutlip RG, Burt SE, Waters TR.National occupational research agenda (NORA) future
directions in occupational musculoskeletal disorder health research. Appl Ergon 2009;40(1):15-22.
Martimo KP, Verbeek J, Karppinen J, Furlan AD, Takala EP, Kuijer PP, Jauhiainen M, Viikari-Juntura E.
Effect of training and lifting equipment for preventing back pain in lifting and handling: systematic
review. Bmj 2008;336(7641):429-431.
McBeth J, Harkness EF, Silman AJ, Macfarlane GJ.The role of workplace low-level mechanical trauma,
posture and environment in the onset of chronic widespread pain. Rheumatology (Oxford)
2003;42(12):1486-1494.
Medical Research Council. Cluster randomized trials: methodological and ethical considerations. MRC
clinical trials series. London, 2002:1-16.
http://www.mrc.ac.uk/Utilities/Documentrecord/index.htm?d=MRC002406 (accessed 10
September 2010).
Melin B, Lundberg U. A biopsychosocial approach to workstress and musculoskeletal disorders. J
Psychophysiol 1997;11(3):238-247.
Merskey H, Bogduk N. Part III: Pain terms, A Current List with Definitions and Notes on Usage. In:
Merskey H, Bogduk N, eds. Classification of Chronic Pain, IASP Task Force on Taxonomy.
Seattle; IASP Press, 1994:209-213. 3http://www.iasp-
pain.org/AM/Template.cfm?Section=Classification_of_Chronic_Pain&Template=/CM/HTMLDisp
lay.cfm&ContentID=2687#Pain (accessed 10 September 2010)
Messing K, Kilbom A. Standing and very slow walking: foot pain-pressure threshold, subjective pain
experience and work activity. Appl Ergon 2001;32(1):81-90.
120
Mikkelsson M, Kaprio J, Salminen JJ, Pulkkinen L, Rose RJ. Widespread pain among 11-year-old Finnish
twin pairs. Arthritis Rheum 2001;44(2):481-485.
Miranda H, Kaila-Kangas L, Helivaara M, Leino-Arjas P, Haukka E, Liira J, Viikari-Juntura E.
Musculoskeletal pain at multiple sites and its effects on work ability in a general working
population. Occup Environ Med 2010;67(7):449-455.
Miranda H, Punnett L, Viikari-Juntura E, Helivaara M, Knekt P. Physical work and chronic shoulder
disorder.Results of a prospective population-based study. Ann Rheum Dis 2008a;67(2):218-223.
Miranda H, Viikari-Juntura E, Martikainen R, Riihimaki H. A prospective study on knee pain and its risk
factors. Osteoarthritis and cartilage / OARS, Osteoarthritis Research Society 2002;10(8):623-630.
Miranda H, Viikari-Juntura E, Martikainen R, Takala EP, Riihimki H. A prospective study of work related
factors and physical exercise as predictors of shoulder pain. Occup Environ Med 2001;58(8):528-
534.
Miranda H, Viikari-Juntura E, Punnett L, Riihimki H. Occupational loading, health behavior and sleep
disturbance as predictors of low-back pain. Scand J Work Environ Health 2008b;34(6):411-419.
Molano SM, Burdorf A, Elders LA. Factors associated with medical care-seeking due to low-back pain in
scaffolders. Am J Ind Med 2001;40(3):275-281.
Morken T, Moen B, Riise T, Hauge SHV, Holien S, Langedrag A, Olson HO, Pedersen S, Saue ILL, Seljebo
GM, Thoppil V. Effects of a training program to improve musculoskeletal health among industrial
workers - effects of supervisors role in the intervention. International Journal of Industrial
Ergonomics 2002;30(2):115-127.
Morken T, Riise T, Moen B, Hauge SH, Holien S, Langedrag A, Pedersen S, Saue IL, Seljebo GM, Thoppil
V. Low back pain and widespread pain predict sickness absence among industrial workers. BMC
Musculoskelet Disord 2003;4(1):21.
Murphy LR, Sauter SL. Work organization interventions: state of knowledge and future directions. Soz
Praventivmed 2004;49(2):79-86.
Mntyselk P, Kumpusalo E, Ahonen R, Kumpusalo A, Kauhanen J, Viinamki H, Halonen P, Takala J.
Pain as a reason to visit the doctor: a study in Finnish primary health care. Pain 2001;89(2-3):175-
180.
Mntyselk PT, Kumpusalo EA, Ahonen RS, Takala JK. Direct and indirect costs of managing patients with
musculoskeletal pain-challenge for health care. Eur J Pain 2002;6(2):141-148.
Nagamachi M. Requisites and practices of participatory ergonomics. International Journal of Industrial
Ergonomics 1995;15:371-377.
Nagasu M, Sakai K, Ito A, Tomita S, Temmyo Y, Ueno M, Miyagi S. Prevalence and risk factors for low
back pain among professional cooks working in school lunch services. BMC Public Health
2007;7:171.
Nagin DS. Analyzing developmental trajectories: a semiparametric, group-based approach. Psychological
Methods 1999;4(2):139-157.
Nagin DS. Group-based modeling of development. Cambridge, MA: Harvard University Press, 2005.
121
Nahit ES, Hunt IM, Lunt M, Dunn G, Silman AJ, Macfarlane GJ. Effects of psychosocial and individual
psychological factors on the onset of musculoskeletal pain: common and site-specific effects. Ann
Rheum Dis 2003;62(8):755-760.
Nahit ES, Pritchard CM, Cherry NM, Silman AJ, Macfarlane GJ. The influence of work related psychosocial
factors and psychological distress on regional musculoskeletal pain: a study of newly employed
workers. J Rheumatol 2001;28(6):1378-1384.
National Research Council Panel on Musculoskeletal Disorders and the Workplace. Magnitude of
occupationally-related musculoskeletal disorders. In: Marras WS, Karwowski W, eds.
Fundamentals and assessment tools for occupational ergonomics. Occupational ergonomics
handbook. 2nd edn. New York: Taylor & Francis, 2006;2-1-2-21.
National Research Council and Institute of Medicine. Musculoskeletal Disorders and the workplace: low
back and upper extremities. Washington, DC: National Academy Press, 2001.
http://www.nap.edu/catalog.php?record_id=10032 (accessed 10 September 2010).
Natvig B, Eriksen W, Bruusgaard D. Low back pain as a predictor of long-term work disability.
Scandinavian Journal of Public Health 2002;30(4):288-292.
Neumann WP, Eklund J, Hansson B, Lindbeck L. Effect assessment in work environment interventions: a
methodological reflection. Ergonomics 2010;53(1):130-137.
Norlund A, Waddel G. Cost of back pain in some OECD countries. In: Nachemson A, Jonson E, eds. Neck
and back pain: The scientific evidence of causes, diagnosis and treatment. Philadelphia: Lippincott
Williams & Wilkins. 2000:421-425.
Nyman T, Grooten WJ, Wiktorin C, Liwing J, Norrman L. Sickness absence and concurrent low back and
neck-shoulder pain: results from the MUSIC-Norrtalje study. Eur Spine J 2007;16(5):631-638.
Ono Y, Nakamura R, Shimaoka M, Hiruta S, Hattori Y, Ichihara G, Kamijima M, Takeuchi Y. Epicondylitis
among cooks in nursery schools. Occup Environ Med 1998;55(3):172-179.
Ono Y, Shimaoka M, Hiruta S, Takeuchi Y. Low back pain among cooks in nursery schools. Industrial
Health 1997;35(2):194-201.
Paananen MV, Auvinen JP, Taimela SP, Tammelin TH, Kantomaa MT, Ebeling HE, Taanila AM, Zitting
PJ, Karppinen JI. Psychosocial, mechanical, and metabolic factors in adolescents' musculoskeletal
pain in multiple locations: A cross-sectional study. European journal of pain (London, England)
2010;14(4):395-401.
Palmer KT, Harris EC, Coggon D. Carpal tunnel syndrome and its relation to occupation: a systematic
literature review. Occupational medicine (Oxford, England) 2007a;57(1):57-66.
Palmer KT, Harris EC, Coggon D. Compensating occupationally related tenosynovitis and epicondylitis: a
literature review. Occupational medicine (Oxford, England) 2007b;57(1):67-74.
Papageorgiou AC, Silman AJ, Macfarlane GJ. Chronic widespread pain in the population: a seven year
follow up study. Ann Rheum Dis 2002;61(12):1071-1074.
Pehkonen I, Ketola R, Ranta R, Takala EP. A video-based observation method to assess musculoskeletal
load in kitchen work. Int J Occup Saf Ergon 2009a;15(1):75-88.
122
Pehkonen I, Miranda H, Haukka E, Luukkonen R, Takala EP, Ketola R, Leino-Arjas P, Riihimki H,
Viikari-Juntura E. Prospective study on shoulder symptoms among kitchen workers in relation to
self-perceived and observed work load. Occup Environ Med 2009b;66(6):416-423.
Pehkonen I, Takala EP, Ketola R, Viikari-Juntura E, Leino-Arjas P, Hopsu L, Virtanen T, Haukka E,
Holtari-Leino M, Nykyri E, Riihimki H. Evaluation of a participatory ergonomic intervention
process in kitchen work. Appl Ergon 2009c;40(1):115-123.
Perki-Mkel M, Hirvonen M, Elo AL, Ervasti J, Huuhtanen P, Kandolin I, Kauppinen K, Kauppinen T,
Ketola R, Lindstrm K, Manninen P, Mikkola J, Reijula K, Riala R, Salminen S, Toivanen M,
Viluksela M. Work and health survey 2006, Statistics. Helsinki; Finnish Institute of Occupational
Health, 2006.
Picavet HS, Schouten JS. Musculoskeletal pain in the Netherlands: prevalences, consequences and risk
groups, the DMC(3)-study. Pain 2003;102(1-2):167-178.
Podniece Z. Work-related musculoskeletal disorders: prevention report. Luxembourg: European Agency for
Safety and Health at Work, 2008.
http://osha.europa.eu/en/publications/reports/en_TE8107132ENC.pdf (accessed 10 September
2010)
Punnett L, Pruss-Utun A, Nelson DI, Fingerhut MA, Leigh J, Tak S, Phillips S. Estimating the global burden
of low back pain attributable to combined occupational exposures. Am J Ind Med 2005;48(6):459-
469.
Punnett L, Wegman DH. Work-related musculoskeletal disorders: the epidemiologic evidence and the
debate. Journal of Electromyography and Kinesiology 2004;14(1):13-23.
Riihimki H. Musculoskeletal disorders. In: Ahrens W, Pigeot I, eds. Handbook of epidemiology. Berlin:
Springer, 2005a: 1443-72.
Riihimki H. Participatory ergonomic intervention at work place: randomized controlled trial and
etnographic study. In: Javanainen M, ed. GEMS of the Health Promotion Research Programme.
Tampere: Juvenes Print, 2005b; 91-100.
Rivilis I, Cole DC, Frazer MB, Kerr MS, Wells RP, Ibrahim S. Evaluation of a participatory ergonomic
intervention aimed at improving musculoskeletal health. Am J Ind Med 2006;49(10):801-810.
Rivilis I, van Eerd D, Cullen K, Cole DC, Irvin E, Tyson J, Mahood Q. Effectiveness of participatory
ergonomic interventions on health outcomes: a systematic review. Appl Ergon 2008;39(3):342-358.
Rothman K, Greenland S. Modern epidemiology. 2nd edn. Philadelphia: Lippincott - Raven Publishers,
1998.
Saastamoinen P, Leino-Arjas P, Laaksonen M, Martikainen P, Lahelma E. Pain and health related
functioning among employees. Journal of Epidemiology and Community Health 2006;60(9):793-
798.
Sauter S, Swanson N. An ecological model of musculoskeletal disorders in office work. In: Moon SD,
Sauter SL, eds. Beyond biomechanics: psychosocial aspects of musculoskeletal disorders in office
work. London: Taylor & Francis, 1996:3-21.
123
Schierhout GH, Myers JE. Is self-reported pain an appropriate outcome measure in ergonomic-
epidemiologic studies of work-related musculoskeletal disorders? Am J Ind Med 1996;30(1):93-98.
Schulte PA, Goldenhar LM, Connally LB. Intervention research: science, skills, and strategies. Am J Ind
Med 1996;29(4):285-288.
Shannon HS, Robson LS, Guastello SJ. Methodological criteria for evaluating occupational safety
intervention research. Safety Science 1999;31:161-179.
Shibata E, Ono Y, Huang J, Hisanaga N, Takeuchi Y, Shimaoka M, Hiruta S. Ergonomics problems among
kitchen workers in nurseries. In: Kumashiro M, Megaw ED, eds. Towards Human Work: Solutions
to Problems in Occupational Health and Safety. London: Taylor & Francis, 1991: 263-268.
Shiri R, Karppinen J, Leino-Arjas P, Solovieva S, Varonen H, Kalso E, Ukkola O, Viikari-Juntura E.
Cardiovascular and lifestyle risk factors in lumbar radicular pain or clinically defined sciatica: a
systematic review. Eur Spine J 2007;16(12):2043-2054.
Shiri R, Karppinen J, Leino-Arjas P, Solovieva S, Viikari-Juntura E. The Association Between Obesity and
Low Back Pain: A Meta-Analysis. Am J Epidemiol 2010a;171(2):135-154.
Shiri R, Karppinen J, Leino-Arjas P, Solovieva S, Viikari-Juntura E. The association between smoking and
low back pain: a meta-analysis. Am J Med 2010b;123(1):87. e87-87.e35.
Shiue HS, Lu CW, Chen CJ, Shih TS, Wu SC, Yang CY, Yang YH, Wu TN. Musculoskeletal disorder
among 52,261 Chinese restaurant cooks cohort: result from the National Health Insurance Data. J
Occup Health 2008;50(2):163-168.
Silverstein B, Clark R. Interventions to reduce work-related musculoskeletal disorders. J Electromyogr
Kinesiol 2004;14(1):135-152.
Solidaki E, Chatzi L, Bitsios P, Markatzi I, Plana E, Castro F, Palmer K, Coggon D, Kogevinas M. Work-
related and psychological determinants of multisite musculoskeletal pain. Scand J Work Environ
Health 2010;36(1):54-61.
St-Vincent M, Bellemare M, Toulouse G, Tellier C. Participatory ergonomic processes to reduce
musculoskeletal disorders: summary of a Quebec experience. Work 2006;27(2):123-135.
Straker L, Burgess-Limerick R, Pollock C, Egeskov R. A randomized and controlled trial of a participative
ergonomics intervention to reduce injuries associated with manual tasks: physical risk and
legislative compliance. Ergonomics 2004;47(2):166-188.
Strazdins L, Bammer G. Women, work and musculoskeletal health. Soc Sci Med 2004;58(6):997-1005.
Streiner DL, Norman GR. Health measurement scales. A practical guide to their development and use. 2nd
edn. Oxford: Oxford University Press, 1995:35
Taimela S, Kaila-Kangas L, Nykyri E, Helivaara M. Self-rated disability due to musculoskeletal disorders
at work and during leisure time. In: Kaila-Kangas L, ed. Musculoskeletal disorders and diseases in
Finland. Results of the Health 2000 Survey. Publications of the National Public Health Institute
B25. Helsinki:Hakapaino Oy, 2007.
Takala EP, Pehkonen I, Forsman M, Hansson GA, Mathiassen SE, Neumann WP, Sjgaard G, Veiersted
KB, Westgaard RH, Winkel J. Systematic evaluation of observational methods assessing
biomechanical exposures at work. Scand J Work Environ Health 2010;36(1):3-24.
124
Takala EP, Viikari-Juntura E, Moneta GB, Saarenmaa K, Kaivanto K. Seasonal variation in neck and
shoulder symptoms. Scand J Work Environ Health 1992;18(4):257-261.
Theberge N, Granzow K, Cole D, Laing A. Negotiating participation: understanding the "how" in an
ergonomic change team. Appl Ergon 2006;37(2):239-248.
Theorell T, Hasselhorn HM. On cross-sectional questionnaire studies of relationships between psychosocial
conditions at work and health--are they reliable? Int Arch Occup Environ Health 2005;78(7):517-
522.
The Social Insurance Institution. Statistical Yearbook of the Social Insurance Institution 2007. Official
Statistics of Finland. Helsinki: Vammalan kirjapaino Oy, 2008.
Tsutsumi A, Nagami M, Yoshikawa T, Kogi K, Kawakami N. Participatory intervention for workplace
improvements on mental health and job performance among blue-collar workers: a cluster
randomized controlled trial. J Occup Environ Med 2009;51(5):554-563.
Tveito TH, Hysing M, Eriksen HR. Low back pain interventions at the workplace: a systematic literature
review. Occup Med (Lond) 2004;54(1):3-13.
Ukoumunne OC, Thompson SG. Analysis of cluster randomized trials with repeated cross-sectional binary
measurements. Statistics in medicine 2001;20(3):417-433.
Vahtera J, Kivimki M, Liira J. The increasing trend in sickness absence among municipality workers is
declining (in Finnish). 2008. Available from
http://www.ttl.fi/Internet/Suomi/Organisaatio/Osaamiskeskukset/Tyoyhteisot+ja+organisaatiot/Sair
auspoissaolojen+kasvu (accessed 20 July 2010).
van den Hoven LH, Gorter KJ, Picavet HS. Measuring musculoskeletal pain by questionnaires: The manikin
versus written questions. Eur J Pain 2009.
van der Klink JJ, Blonk RW, Schene AH, van Dijk FJ. The benefits of interventions for work-related stress.
Am J Public Health 2001;91(2):270-276.
van der Molen HF, Sluiter JK, Hulshof CT, Vink P, Frings-Dresen MH. Effectiveness of measures and
implementation strategies in reducing physical work demands due to manual handling at work.
Scand J Work Environ Health 2005a;31 Suppl 2:75-87.
van der Molen HF, Sluiter JK, Hulshof CT, Vink P, van Duivenbooden C, Holman R, Frings-Dresen MH.
Implementation of participatory ergonomics intervention in construction companies. Scand J Work
Environ Health 2005b;31(3):191-204.
van der Windt DA, Thomas E, Pope DP, de Winter AF, Macfarlane GJ, Bouter LM, Silman AJ.
Occupational risk factors for shoulder pain: a systematic review. Occup Environ Med
2000;57(7):433-442.
van Eerd D, Cole D, Irvin E, Mahood Q, Keown K, Theberge N, Village J, St-Vincent M, Cullen K,
Widdrington H. Report on process and implementation of participatory ergonomic interventions: a
systematic review, volume 1. Toronto: Institute for Work and Health,
2008.http://www.iwh.on.ca/sys-reviews/implementation-of-pe-interventions (accessed 10
September 2010).
125
van Poppel MN, Hooftman WE, Koes BW. An update of a systematic review of controlled clinical trials on
the primary prevention of back pain at the workplace. Occup Med (Lond) 2004;54(5):345-352.
van Poppel MN, Koes BW, Smid T, Bouter LM. A systematic review of controlled clinical trials on the
prevention of back pain in industry. Occup Environ Med 1997;54(12):841-847.
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;35(1):19-36.
van Tulder M, Furlan A, Bombardier C, Bouter L. Updated method guidelines for systematic reviews in the
cochrane collaboration back review group. Spine (Phila Pa 1976) 2003;28(12):1290-1299.
Verhagen AP, de Vet HC, de Bie RA, Kessels AG, Boers M, Bouter LM, Knipschild PG. The Delphi list: a
criteria list for quality assessment of randomized clinical trials for conducting systematic reviews
developed by Delphi consensus. J Clin Epidemiol 1998;51(12):1235-1241.
Viikari-Juntura E, Rauas S, Martikainen R, Kuosma E, Riihimki H, Takala EP, Saarenmaa K. Validity of
self-reported physical work load in epidemiologic studies on musculoskeletal disorders. Scand J
Work Environ Health 1996;22(4):251-259.
Viikari-Juntura E, Shiri R, Solovieva S, Karppinen J, Leino-Arjas P, Varonen H, Kalso E, Ukkola O. Risk
factors of atherosclerosis and shoulder pain--is there an association? A systematic review. Eur J
Pain 2008;12(4):412-426.
Virtanen M, Pentti J, Vahtera J, Ferrie JE, Stansfeld SA, Helenius H, Elovainio M, Honkonen T, Terho K,
Oksanen T, Kivimki M. Overcrowding in hospital wards as a predictor of antidepressant treatment
among hospital staff. Am J Psychiatry 2008;165(11):1482-1486.
Volinn E. Do workplace interventions prevent low-back disorders? If so, why?: a methodologic
commentary. Ergonomics 1999;42(1):258-272.
Walker-Bone K, Cooper C. Hard work never hurt anyone--or did it? A review of occupational associations
with soft tissue musculoskeletal disorders of the neck and upper limb. Ann Rheum Dis
2005;64(8):1112-1117.
Walker-Bone K, Reading I, Coggon D, Cooper C, Palmer KT. The anatomical pattern and determinants of
pain in the neck and upper limbs: an epidemiologic study. Pain 2004;109(1-2):45-51.
Walker-Bone KE, Palmer KT, Reading I, Cooper C. Soft-tissue rheumatic disorders of the neck and upper
limb: prevalence and risk factors. Semin Arthritis Rheum 2003;33(3):185-203.
Wearing SC, Hennig EM, Byrne NM, Steele JR, Hills AP. Musculoskeletal disorders associated with
obesity: a biomechanical perspective. Obes Rev 2006;7(3):239-250.
Weevers HJ, van der Beek AJ, Anema JR, van der Wal G, van Mechelen W. Work-related disease in general
practice: a systematic review. Fam Pract 2005;22(2):197-204.
Westgaard RH, Winkel J. Ergonomic intervention research for improved musculoskeletal health: A critical
review. International Journal of Industrial Ergonomics 1997;20(6):463-500.
WHO. Identification and control of work-related diseases. WHO Technical Report Series 714. Geneva:
World Health Organization, 1985.
Wilson J, Haines H. Participatory ergonomics. In: Salvendy, G, ed. Handbook of human factors and
ergonomics. 2nd edn. New York: Wiley, 1997:490-513.
126
Wolfe F, Smythe HA, Yunus MB, Bennett RM, Bombardier C, Goldenberg DL, Tugwell P, Campbell SM,
Abeles M, Clark P, et al. The American College of Rheumatology 1990 Criteria for the
Classification of Fibromyalgia. Report of the Multicenter Criteria Committee. Arthritis Rheum
1990;33(2):160-172.
Yeung SS, Genaidy A, Deddens J, Alhemood A, Leung PC. Prevalence of musculoskeletal symptoms in
single and multiple body regions and effects of perceived risk of injury among manual handling
workers. Spine 2002;27(19):2166-2172.
Zubieta JK, Heitzeg MM, Smith YR, Bueller JA, Xu K, Xu Y, Koeppe RA, Stohler CS, Goldman D. COMT
val158met genotype affects mu-opioid neurotransmitter responses to a pain stressor. Science
2003;299(5610):1240-1243.
Zwerling C, Daltroy LH, Fine LJ, Johnston JJ, Melius J, Silverstein BA. Design and conduct of occupational
injury intervention studies: a review of evaluation strategies. Am J Ind Med 1997;32(2):164-179.
rhede E. Nordic cooperation in research on the work environment. Scand J Work Environ Health
1994;20(1):65-66.