Ergonomics

ISSN: 0014-0139 (Print) 1366-5847 (Online) Journal homepage: http://www.tandfonline.com/loi/terg20

Associations of Office Workers’ Objectively

Assessed Occupational Sitting, Standing and
Stepping Time with Musculoskeletal Symptoms

Pieter Coenen, Genevieve N Healy, Elisabeth AH Winkler, David W Dunstan,

Neville Owen, Marj Moodie, Anthony D. LaMontagne, Elizabeth A. Eakin,
Peter O’Sullivan & Leon M Straker

To cite this article: Pieter Coenen, Genevieve N Healy, Elisabeth AH Winkler, David W Dunstan,
Neville Owen, Marj Moodie, Anthony D. LaMontagne, Elizabeth A. Eakin, Peter O’Sullivan &
Leon M Straker (2018): Associations of Office Workers’ Objectively Assessed Occupational
Sitting, Standing and Stepping Time with Musculoskeletal Symptoms , Ergonomics, DOI:
10.1080/00140139.2018.1462891

To link to this article: https://doi.org/10.1080/00140139.2018.1462891

Accepted author version posted online: 09

Apr 2018.

Submit your article to this journal

View related articles

View Crossmark data

Full Terms & Conditions of access and use can be found at

http://www.tandfonline.com/action/journalInformation?journalCode=terg20
Publisher: Taylor & Francis
Journal: Ergonomics
DOI: http://doi.org/10.1080/00140139.2018.1462891

Associations of Office Workers' Objectively Assessed Occupational Sitting, Standing and Stepping Time with
Musculoskeletal Symptoms

Pieter Coenen1,2 (p.coenen@vumc.nl), Genevieve N Healy1,3,4 (g.healy@sph.uq.edu.au), Elisabeth AH Winkler3

(e.winkler@sph.uq.edu.au), David W Dunstan3,4,5,6,7,8,9 (David.Dunstan@baker.edu.au), Neville Owen3,4,7,10
(Neville.Owen@baker.edu.au), Marj Moodie11,12 (marj.moodie@deakin.edu.au), Anthony D. LaMontagne13,14
(tony.lamontagne@deakin.edu.au), Elizabeth A. Eakin3 (e.eakin@uq.edu.au), Peter O’Sullivan1
(P.OSullivan@curtin.edu.au), Leon M Straker1 (L.Straker@curtin.edu.au)

1 School of Physiotherapy and Exercise Science, Curtin University, Perth, Australia.

2Department of Public and Occupational Health, VU University Medical Center, Amsterdam Public Health,
Amsterdam, The Netherlands
3 The University of Queensland, School of Public Health, Brisbane, Australia
4 Baker IDI Heart and Diabetes Institute, Melbourne, Australia;
5 School of Public Health and Preventive Medicine, Monash University, Melbourne, Australia
6 School of Exercise and Nutrition Sciences, Deakin University, Geelong, Australia
7 Central Clinical School, Medicine, Monash University, Melbourne, Australia
8 School of Sport Science, Exercise and Health, The University of Western Australia, Perth, Australia
9Mary MacKillop Institute for Health Research, Australian Catholic University, Melbourne, Victoria, Australia
10Swinburne University of Technology, Melbourne, VIC, Australia
11 Deakin Health Economics, Centre for Population Health Research, Deakin University, Geelong, Australia
12 Global Obesity centre, Centre for Population Health Research, Deakin University, Geelong, Australia
13Centre for Population Health Research, School of Health & Social Development, Deakin University, Geelong,

VIC, Australia.
14Melbourne School of Population and Global Health, University of Melbourne, Melbourne, VIC, Australia.

Corresponding Author
Leon Straker
School of Physiotherapy and Exercise Science, Curtin University
GPO Box U1987, Perth, WA 6845, Australia
Tel: +61 8 9266 3634
E-mail: L.Straker@curtin.edu.au

Acknowledgements
The Stand Up Victoria study was funded by a National Health and Medical Research Council (NHMRC) Project
Grant (#1002706), project funding from the Victorian Health Promotion Foundation’s Creating Healthy
Workplaces program and, and by the Victorian Government’s Operational Infrastructure Support Program.
GNH was supported by a NHMRC Career Development Fellowship (#108029). EW was supported by a NHMRC
Centre for Research Excellence Grant on Sitting Time and Chronic Disease Prevention – Measurement,
Mechanisms and Interventions (#1057608). DD was supported by a NHMRC Senior Research Fellowship
(#1078360) and by the Victorian Government’s Operational Infrastructure Support Program. NO was
supported by a NHMRC Program Grant (#569940), a NHMRC Senior Principal Research Fellowship (#1003960),
a NHMRC Centre for Research Excellence Grant (#1057608) and by the Victorian Government’s Operational
Infrastructure Support Program. MM was supported by a NHMRC Centre for Research Excellence in Obesity
Policy and Food Systems (#1041020). EE is supported by a NHMRC Senior Research Fellowship (#511001). LS
was supported by a NHMRC Senior Research Fellowship (NHMRC #1019980). We acknowledge and thank all
the participants of the Stand Up Victoria study, as well as other staff involved at the Department of Human
Services.

Word count
Number of words: 3,113

1
Number of words abstract: 150 (150 max)
Number of tables: 2
Number of figures: 2
Number of supplementary files: 2
Number of references: 45

Abstract

We examined the association of musculoskeletal symptoms (MSS) with workplace sitting,

standing and stepping time, as well as sitting and standing time accumulation (i.e., usual

bout duration of these activities), measured objectively with the activPAL3 monitor. Using

baseline data from the Stand Up Victoria trial (216 office workers, 14 workplaces), cross-

sectional associations of occupational activities with self-reported MSS (low-back, upper-

and lower-extremity symptoms in the last three months) were examined using probit

regression, correcting for clustering and adjusting for confounders.

Sitting bout duration was significantly (p<0.05) associated, non-linearly, with MSS, such

that those in the middle tertile displayed the highest prevalence of upper-extremity

symptoms. Other associations were non-significant but sometimes involved large

differences in symptom prevalence (e.g., 38%) by activity. Though causation is unclear,

these non-linear associations suggest that sitting and its alternatives (i.e., standing and

stepping) interact with MSS and this should be considered when designing safe work

systems.

Practitioner summary

2
We studied associations of objectively assessed occupational activities with musculoskeletal

symptoms in office workers. Workers who accumulated longer sitting bouts reported fewer

upper-extremity symptoms. Total activity duration was not significantly associated with

musculoskeletal symptoms. We underline the importance of considering total volumes and

patterns of activity time in musculoskeletal research.

Keywords: Musculoskeletal symptoms – Office workers – Posture - Sitting – Standing

Introduction

Occupational physical activity has declined substantially over the past decades (Church et

al., 2011; Ng and Popkin, 2012), with many workers currently spending most of their

working hours sitting. With excessive sitting shown to be an important risk factor for several

health outcomes (Biswas et al., 2015; Thorp et al., 2011; Tremblay et al., 2010), workplace

sitting has been identified as an emergent occupational health issue (Straker et al., 2016).

Consequently, there has been an expanded focus in occupational health research from

primarily highly physically demanding work to also include sedentary work (Straker and

Mathiassen, 2009), in which desk-based workers are a particular group of interest (Parry

and Straker, 2013).

Recognition that sitting at work is likely to be an occupational-health hazard has

been considerable and rapid (Straker et al., 2014), with mass media and occupational health

practitioners encouraging workers to sit less, and stand and move more. There is, however,

a concern that without an adequate understanding of the various, potentially competing,

health consequences of the alternatives to sitting (such as standing and stepping), attempts

3
to reduce sitting and thereby improve some aspects of health, may expose workers to new

hazards and/or adverse health consequences. In particular, replacing sitting with standing

might shift the risk from one health outcome to another, such as improving cardio-

metabolic biomarkers but simultaneously worsening circulation problems or

musculoskeletal symptoms (MSS) (Coenen et al., 2016; Waters and Dick, 2014); or, it may

shift MSS from one region of the body to another (Roelofs and Straker, 2002). A sound

understanding of the potential influence on office workers’ musculoskeletal health of sitting

and its alternatives (e.g., standing and stepping) is therefore required to inform healthy

work practices.

Excessive occupational sitting has been suggested to be associated with a higher

prevalence of MSS, including low-back (Al-Eisa et al., 2006) and lower-extremity symptoms

(Reid et al., 2010). Computer work, which commonly occurs in a sitting posture, has been

associated with upper-extremity symptoms (Andersen et al., 2011). Occupational standing

has also been shown to be associated with MSS, including back symptoms (Andersen et al.,

2007) and lower-extremity symptoms (Reid et al., 2010), and occupational stepping has

been linked to lower-extremity symptoms (Engels et al., 1996). However, the available

evidence regarding the association of sitting and its alternatives with MSS is inconsistent,

with systematic reviews unable to elucidate conclusive evidence (da Costa and Vieira, 2010;

Roffey et al., 2010; Waersted et al., 2010).

One explanation for such inconsistency is that previous research has predominantly

relied on self-reported activity measures. In addition to errors in measuring time spent in

various activities (Burdorf, 2010), self-report tools lack the capacity to measure potentially

important aspects of the accumulation pattern of these activities. Sitting and standing

accumulation can be considered in terms of the duration each time workers remain in the

4
same posture, and conversely, how often they alternate between sitting and standing (i.e.,

regularly varying from one activity to another). The manner in which sitting and standing are

accumulated and its relationship with MSS have been discussed in the research literature

(Srinivasan and Mathiassen, 2012) and in ergonomics guidelines (European Agency for

Safety and Health at Work, 2007), with variation in activities considered as a strategy to

reduce the risk of MSS. However, associations between MSS and the pattern of

accumulation of sitting and standing time in the workplace have received limited attention.

It can thus be concluded that there has been limited use of objective measures for sitting,

standing and stepping in MSS research, in particular of the accumulation patterns of these

activities.

We explored the cross-sectional associations with low-back, and lower- and upper-

extremity musculoskeletal symptoms, of objectively assessed occupational sitting, standing

and stepping time, as well as sitting and standing accumulation patterns in a large sample of

Australian office workers. As far as we are aware, this is the first study describing the

association of objectively measured sitting, standing and stepping (as well as their

accumulation patterns) with MSS among office workers.

Material and methods

Study population

Data were drawn from Stand Up Victoria, a cluster randomised-controlled trial assessing the

impact of an intervention aimed at reducing occupational sitting time in office workers

(Dunstan et al., 2013). This research complied with the tenets of the Declaration of Helsinki

and was approved by the Institutional Review Board at Alfred Health Human Ethics

Committee (Melbourne, Australia). Participants were recruited from 14 different work sites

5
within the one organisation in the state of Victoria, Australia. Workers were eligible to take

part in the study if they were working at least 0.6 full time equivalent, were aged 18–65

years, were English speaking, had designated access to a telephone, internet and desk

within the workplace, were not pregnant, and had no health problems (including

musculoskeletal problems) that may limit their ability to regularly stand up or to stand up

for at least 10 minutes at a time. For the current analysis, baseline data (i.e., prior to

receiving the intervention) were used. A total of 246 workers were eligible to participate in

the Stand Up Victoria study with 216 providing complete data regarding occupational sitting

and its alternatives, MSS and relevant confounders. Further details about the worksites and

baseline characteristics of participants have previously been published (Hadgraft et al.,

2016).

Methods

As previously described in detail (Dunstan et al., 2013), participants took part in a

measurement protocol in which sitting and its alternatives were objectively assessed and

self-administered questionnaires were completed using an online Lime-Service tool

(www.limeservice.com).

Objectively measured occupational sitting and its alternatives

The activPAL3 activity monitor (PAL Technologies Limited, Glasgow, UK) was used to

measure sitting, standing and stepping time, as well as sitting and standing accumulation

patterns. The monitor records the commencement and completion of each bout of sitting or

lying (here termed sitting), standing, and stepping. Accuracy is high for assessment of time

spent in, as well as accumulation patterns of these activities (Kozey-Keadle et al., 2011;

6
Lyden et al., 2012; Ryan et al., 2006). For example, the monitor has been shown to have a

low bias percentage for assessing breaks from sitting time (0.3%) (Lyden et al., 2012), while

sitting time is known to be underestimated by only 2.8% by the activPAL3 (Kozey-Keadle et

al., 2011).

The monitor was secured with medical tape to the anterior mid-line of the right thigh, about

a third of the way down between the hip and knee. Participants were asked to wear the

monitor for 24 hours per day, for seven consecutive days. Self-completed daily logs were

used to record wake and sleep times and work hours. A customized program (SAS version

≥9.3; SAS Institute Inc., Cary, NC) was used to calculate sitting, standing, and stepping

measures, for worn, waking time, and restricted to work time for occupational activity

measures. Data from the self-completed logs were used to differentiate activities at work,

compared to activities not at work. Here, only work for the study organisation was

considered, which comprised nearly all reported work (≥98%) and was seldom reported to

occur outside of the workplace (<5%). Hence, the terms “workplace” and “at work” are used

interchangeably. Valid days for occupational activity had removals for <20% of work hours;

valid days overall had removals for <20% of waking hours and had ≥10 hours waking wear

time. Time spent sitting, standing and stepping during work hours were totalled for each

day, averaged across valid work days and standardized to minutes per 8-hour workday. Non-

work activity was estimated as the absolute amount of total activity minus workplace

activity.

An established measure called “usual bout duration” (Stephens et al., 2014) was

used to indicate how long participants usually sat continuously at a time at work, with half

of all occupational sitting occurring in bouts of this duration or longer. Usual bout duration

was estimated by non-linear curve-fitting techniques as the midpoint of the sitting

7
accumulation curve (Stephens et al., 2014). We applied this same method to estimate usual

standing bout duration. The method approximated the observed midpoint, despite the

different distributions of sitting and standing bouts (Supplementary file 1).

Musculoskeletal symptoms

MSS were assessed using the 27-item Nordic Musculoskeletal Questionnaire (Kuorinka et al.,

1987), which is repeatable and sensitive to change (Palmer et al., 1999). The questionnaire

was modified to ask about the last three months rather than 12 months (Dickinson et al.,

1992). MSS in the neck, shoulders, elbows, wrists, low-back, hips, knees and ankles were

assessed and collapsed into three body areas: low-back, lower-extremities (hips, knees and

ankles combined) and upper-extremities (neck, shoulders, elbows and wrists combined).

MSS were examined in terms of presence/absence of symptoms (i.e., ache, pain or

discomfort) in each of these three body areas.

Potential confounding variables

A questionnaire also assessed personal and lifestyle factors (e.g., age, gender and smoking

behaviour), mental demands at work and physical health symptoms, including fatigue.

Physical assessments included measurement of waist circumference and standing height

using standard measures. Further details on the measures, including the reliability and/or

validity of the scales, are available elsewhere (Dunstan et al., 2013).

Data analyses

Statistical analyses were conducted using STATA (StataCorp. 2013. Stata Statistical Software:

Release 13. College Station, TX). Significance was set at p<0.05 (two-tailed). In all analyses,

8
linearized variance estimation was used (survey commands) to correct for the clustered

sampling. The associations of occupational activities with MSS in the three body areas were

examined using multivariable probit regression models, modelling each outcome and each

activity separately. Models were adjusted for age and gender, and for other potential

confounders (smoking, height, waist circumference, sitting not at work, standing not at

work, stepping not at work, mental demands at work, and fatigue) if they showed an

association with the outcome at p<0.2 in backward stepwise selection procedure (Mickey

and Greenland, 1989). Occupational activities have sometimes shown nonlinear associations

with MSS (Coenen et al., 2015), thus activities were examined both continuously and

categorised in tertiles, to test for both possible linear and non-linear associations. Adjusted

mean prevalences, with pairwise comparisons, were reported, with Sidak’s multiple

comparison adjustment used for the pairwise comparisons.

Results

The mean [sd] age of participating workers was 45.4 [9.3] years with 150 (69%) females

(Table 1). The reported symptoms during the last three months for low-back, lower-

extremities, and upper extremities were 68%, 69% and 83% respectively. As previously

reported (Healy et al., 2016), in an 8-hour workday, workers spent on average 79% of their

time (mean 379.7 [sd 44.4] minutes) sitting, 14% (67.2 [37.6] minutes) standing and 7%

(33.1 [14.2] minutes) stepping. On average, workers accumulated half of their workplace

sitting and standing time in bouts of ≥33.1 [14.9] and ≥1.7 [1.2] minutes at a time,

respectively.

Unadjusted prevalence, and the amount of activity occurring in each tertile, are

presented in Supplemental file 2. The amount of time spent sitting, standing and stepping at

9
work (per 8-hour workday) did not show statistically significant associations with the

prevalence of MSS (Table 2; Figure 1). Some associations were large however, such as an

approximately 28% lower prevalence of lower-extremity symptoms per hour spent sitting at

work (95% CI: [-63% 6%], p=0.102) and conversely a 38% higher prevalence of lower-

extremity symptoms per hour spent standing at work (95% CI: [-2% 77%], p=0.062). Though

not statistically significant, the tendency was for the prevalence of lower-extremity and low-

back symptoms to decrease with additional sitting, and to increase with additional standing

or stepping (Figure 1). By contrast, upper-extremity symptoms tended to either remain

stable across activities or display non-linear trends with the highest symptom prevalence

among workers in the middle tertile of sitting, standing and stepping (Figure 1).

Upper-extremity symptoms were significantly associated with how long at a time

participants accumulated their workplace sitting, though not linearly. The highest upper-

extremity symptom prevalence occurred in the middle tertile, while the lowest symptom

prevalence was seen in those with the most prolonged sitting accumulation pattern (Table

2; Figure 2), with an estimated 18% difference in prevalence (95% CI: [2% 34%], p=0.029)

between these two groups. This same pattern and a similar degree of difference (17%) was

evident for lower-extremity symptoms; however, the association did not quite reach

statistical significance (p=0.061). Low-back symptoms were not significantly associated with

usual sitting bout duration and did not show this pattern.

The results for workplace standing accumulation did not resemble those of sitting

accumulation as there was no large or statistically significant difference in MSS by how long

at a time participants usually stood at work. Further, the direction of associations indicated

that participants in the middle tertile of standing bout duration tended to show the lowest

(rather than the highest) symptom prevalence.

10
Discussion

Interpretation of findings

This study used accurate objective measures of activity volume and accumulation patterns

to explore cross-sectional associations of occupational sitting, standing and stepping with

MSS (i.e., symptoms in the low-back, lower- and upper-extremities) among office workers.

This is the first study to document accumulation patterns of sitting and standing from an

MSS point of view.

We observed that the manner in which workers accumulated their workplace sitting

was significantly associated, non-linearly, with the prevalence of upper-extremity

symptoms. The lowest prevalence was among those accumulating their sitting in the longest

periods at a time. A similar but non-significant association was seen for the prevalence of

lower-extremity symptoms and a slightly weaker trend in the same direction was observed

for low-back symptoms. Whilst there are multiple, sometimes contradictory, mechanisms

proposed for why prolonged sitting could be associated with an increased risk of MSS

(including muscle fatigue (Balasubramanian et al., 2009), tissue compression (de Looze et

al., 2003), non-neutral postures such as slumping or slouching (O'Sullivan et al., 2011),

swelling of the lower limbs due to blood pooling (Chester et al., 2002), and a lack variation in

movement (Srinivasan and Mathiassen, 2012)), there are no clear proposed mechanisms for

how prolonged sitting could decrease the risk of MSS. Therefore it may be that the direction

of the association is reversed; i.e. people with MSS are prompted to change postures more

often than those without MSS. However, cross-sectional associations provide limited

elucidation of the direction of the associations (whether MSS cause the behaviour or vice

11
versa). Longitudinal and/or intervention research will help to better understand temporal

directions, causation and mechanisms.

Despite using accurate objective measurement, and allowing for possible linear or

non-linear relationships, the present study did not identify significant associations between

the volume of workplace time spent sitting, standing and stepping, and MSS. Our findings in

office workers are in contrast to the few previous findings with objective measures. For

example, the Danish NOMAD study of blue-collar workers recruited from a variety of non-

office occupations, including construction workers, cleaners and garbage collectors (Gupta

et al., 2015; Hallman et al., 2015), measured sitting with two monitors (worn on the trunk

and thigh). They observed that more occupational sitting was associated with a higher

intensity of low-back symptoms (Gupta et al., 2015) and that occupational sitting had non-

linear associations with neck and shoulder symptoms (Hallman et al., 2015). The complex

mechanisms explaining associations of sitting and alternatives of sitting with MSS may

however be different at one end of the physical work demand spectrum (such as in our

office-based workers) compared to the other end (such as in blue-collar occupations).

Our findings were consistent with the conclusions of systematic reviews of studies

using subjective activity measures (da Costa and Vieira, 2010; Roffey et al., 2010; Waersted

et al., 2010) in not being able to find consistent associations of sitting, standing and/or

stepping with MSS. However, important effects may have been missed in the current study

as confidence intervals were wide and some of the non-significant effects were of a

meaningful magnitude, with tendencies for the prevalence of lower-extremity and low-back

symptoms to decrease with more sitting, and to increase with more standing or stepping. In

the changing work environment that is characterised by a decline in occupational physical

activity over the past decades (Ng and Popkin, 2012), and consequently a growing number

12
of initiatives encouraging workers to sit less, and stand and move more, our findings add to

the knowledge on musculoskeletal impact of sitting and its alternatives (e.g., standing and

stepping). In order to inform healthy work practices, such knowledge should, however, be

balanced with that of other (non-musculoskeletal) health consequences of these activities.

Strengths and limitations

Key study strengths were the high-quality, objective measures of both volumes of activities

and accumulation patterns of sitting and standing and the consideration of non-linear

associations across multiple body regions. A relatively large proportion of our study sample

reported on MSS with three-month prevalence of 68%, 69% and 83% for symptoms in the

low-back, lower- and upper-extremity, respectively. This is however reasonably in line with

earlier findings of musculoskeletal symptoms being highly prevalent among office workers

(Griffiths et al., 2012), e.g. with a 12-month prevalence of musculoskeletal symptoms in any

region of 76% being reported (Cook et al., 2000).

A limitation was that the study did not power a-priori on this research question and

showed evidence of being underpowered, with wide 95% confidence intervals

encompassing meaningful effects. The characterisation of MSS did not include the extent to

which the symptoms bothered participants, or the intensity of the pain associated with the

symptoms. Similarly, the measure of activities did not consider posture details such as

lumbar flexion, while the MSS questionnaire and the activity assessment were not

contemporaneous. Moreover, the MSS questions used for the current analysis only

differentiates participants with pain from those without. Our study does, however, not

provide insight into other symptom modalities, e.g. intensity, duration or impact of

13
symptoms. Future work on the association of sitting, standing and stepping with MSS should

therefore focus on these modalities in order to provide more insight into the association.

Although workers from 14 different worksites were recruited, they were all office

workers from a single organisation and findings may therefore not be generalisable to other

worker groups, especially to those performing highly physically demanding work. Rather,

the results from our study provide an important counterpoint to studies addressing

physically demanding work (Gupta et al., 2015; Hallman et al., 2015), by examining activities

within settings predominated by sitting and typical office-based tasks (e.g., computer work).

Conclusion

These findings, which varied depending on measure and body region, underline the

complexity of studying associations of sitting, standing and stepping time with MSS in office

workers and highlight the potential utility of objective measures to consider accumulation

patterns and not just the total volumes of these activities. Although the associations of

activities with MSS remain unclear (with non-significant but substantial non-linear symptom

prevalence differences by activity), the interactions between MSS and activities should be

considered in future studies and have the potential to inform the design safer office-work

systems.

Conflicts of interest

There were no conflicts of interest reported by the authors.

14
References

Al-Eisa, E., Egan, D., Deluzio, K., Wassersug, R., 2006. Effects of pelvic asymmetry and low
back pain on trunk kinematics during sitting: a comparison with standing. Spine 31, E135-
143.
Andersen, J.H., Fallentin, N., Thomsen, J.F., Mikkelsen, S., 2011. Risk factors for neck and
upper extremity disorders among computer users and the effect of interventions: an
overview of systematic reviews. PloS One 6, e19691.
Andersen, J.H., Haahr, J.P., Frost, P., 2007. Risk factors for more severe regional
musculoskeletal symptoms: a two-year prospective study of a general working population.
Arthritis Rheum 56, 1355-1364.
Balasubramanian, V., Adalarasu, K., Regulapati, R., 2009. Comparing dynamic and
stationary standing postures in an assembly task. Int J Ind Ergonom 39, 649-654.
Biswas, A., Oh, P.I., Faulkner, G.E., Bajaj, R.R., Silver, M.A., Mitchell, M.S., Alter, D.A.,
2015. Sedentary time and its association with risk for disease incidence, mortality, and
hospitalization in adults: a systematic review and meta-analysis. Ann Intern Med 162,
123-132.
Burdorf, A., 2010. The role of assessment of biomechanical exposure at the workplace in the
prevention of musculoskeletal disorders. Scand J Work Environ Health 36, 1-2.
Chester, M.R., Rys, M.J., Konz, S.A., 2002. Leg swelling, comfort and fatigue when sitting,
standing, and sit/standing. Int J Ind Ergonom 29, 289-296.
Church, T.S., Thomas, D.M., Tudor-Locke, C., Katzmarzyk, P.T., Earnest, C.P., Rodarte,
R.Q., Martin, C.K., Blair, S.N., Bouchard, C., 2011. Trends over 5 decades in U.S.
occupation-related physical activity and their associations with obesity. PLoS One 6,
e19657.
Coenen, P., Mathiassen, S.E., Kingma, I., Boot, C.R., Bongers, P.M., Dieën van, J.H., 2015.
The effect of the presence and characteristics of an outlying group on exposure-outcome
associations. Scand J Work Environ Health 41, 65-74.
Coenen, P., Willenberg, L., Parry, S., Shi, J., Maher, C., Healy, G., Dunstan, D., Straker, L.,
2016. Associations of occupational standing with musculoskeletal symptoms: a systematic
review with meta-analysis. Br J Sports Med 10.1136/bjsports-2016-096795.
Commonwealth Government of Australia, 2011. Work Health and Safety Act, Act No. 137,
in: Commonwealth Government of Australia (Ed.), Canberra, Australia.
Cook, C., Burgess-Limerick, R., Chang, S., 2000. The prevalence of neck and upper
extremity musculoskeletal symptoms in computer mouse users. Int J Ind Ergonom 26,
347-356.
da Costa, B.R., Vieira, E.R., 2010. Risk factors for work-related musculoskeletal disorders: A
systematic review of recent longitudinal studies. Am J Ind Med 53, 285-323.
de Looze, M.P., Kuijt-Evers, L.F., van Dieën, J., 2003. Sitting comfort and discomfort and
the relationships with objective measures. Ergonomics 46, 985-997.
Dickinson, C.E., Campion, K., Foster, A.F., Newman, S.J., O'Rourke, A.M., Thomas, P.G.,
1992. Questionnaire development: an examination of the Nordic Musculoskeletal
questionnaire. Appl Ergon 23, 197-201.
Dunstan, D., Wiesner, G., Eakin, E.G., Neuhaus, M., Owen, N., LaMontagne, A.D., Moodie,
M., Winkler, E.A., Fjeldsoe, B.S., Lawler, S., Healy, G.N., 2013. Reducing office workers'
sitting time: rationale and study design for the Stand Up Victoria cluster randomized trial.
BMC Public Health 13, 1057.
Engels, J.A., van der Gulden, J.W., Senden, T.F., van 't Hof, B., 1996. Work related risk
factors for musculoskeletal complaints in the nursing profession: results of a questionnaire
survey. Occup Environ Med 53, 636-641.

15
European Agency for Safety and Health at Work, 2007. Office Ergonomics. European
Agency for Safety and Health at Work, Bilbao, Spain.
Griffiths, K.L., Mackey, M.G., Adamson, B.J., Pepper, K.L., 2012. Prevalence and risk
factors for musculoskeletal symptoms with computer based work across occupations.
Work 42, 533-541.
Gupta, N., Christiansen, C.S., Hallman, D.M., Korshøj, M., Carneiro, I.G., Holtermann, A.,
2015. Is objectively measured sitting time associated with low back pain? A cross-
sectional investigation in the NOMAD study. PLoS One 10, e0121159.
Hadgraft, N.T., Healy, G.N., Owen, N., Winkler, E.A., Lynch, B.M., Sethi, P., Eakin, E.G.,
Moodie, M., LaMontagne, A.D., Wiesner, G., Willenberg, L., Dunstan, D.W., 2016.
Office workers' objectively assessed total and prolonged sitting time: Individual-level
correlates and worksite variations. Prev Med Rep 4, 184-191.
Hallman, D.M., Gupta, N., Mathiassen, S.E., Holtermann, A., 2015. Association between
objectively measured sitting time and neck-shoulder pain among blue-collar workers. Int
Arch Occup Environ Health 88, 1031-1042.
Healy, G.N., Eakin, E.G., Owen, N., LaMontagne, A.D., Moodie, M., Winkler, E.A.H.,
Fjeldsoe, B., Wiesner, G., Willenberg, L., Dunstan, D.W., 2016. A cluster RCT to reduce
office workers' sitting time: impact on activity outcomes. Med Sci Sports Exerc 48, 1787-
1797.
Kozey-Keadle, S., Libertine, A., Lyden, K., Staudenmayer, J., Freedson, P.S., 2011.
Validation of wearable monitors for assessing sedentary behavior. Med Sci Sports Exerc
43, 1561-1567.
Kuorinka, I., Jonsson, B., Kilbom, A., Vinterberg, H., Biering-Sorensen, F., Andersson, G.,
Jorgensen, K., 1987. Standardised Nordic questionnaires for the analysis of
musculoskeletal symptoms. Appl Ergon 18, 233-237.
Lerner, D., Amick, B.C., Rogers, W.H., Malspeis, S., Bungay, K., Cynn, D., 2001. The work
limitations questionnaire. Med Care 39, 72-85.
Lyden, K., Kozey Keadle, S.L., Staudenmayer, J.W., Freedson, P.S., 2012. Validity of two
wearable monitors to estimate breaks from sedentary time. Med Sci Sports Exerc 44,
2243-2252.
Mickey, R.M., Greenland, S., 1989. The impact of confounder selection criteria on effect
estimation. Am J Epidemiol 129, 125-137.
Ng, S.W., Popkin, B.M., 2012. Time use and physical activity: a shift away from movement
across the globe. Obes Rev 13, 659-680.
O'Sullivan, P.B., Smith, A.J., Beales, D.J., Straker, L.M., 2011. Association of
biopsychosocial factors with degree of slump in sitting posture and self-report of back pain
in adolescents: a cross-sectional study. Phys Ther 91, 470-483.
Palmer, K., Smith, G., Kellingray, S., Cooper, C., 1999. Repeatability and validity of an
upper limb and neck discomfort questionnaire: the utility of the standardized Nordic
questionnaire. Occup Med 49, 171-175.
Parry, S., Straker, L., 2013. The contribution of office work to sedentary behaviour associated
risk. BMC Public Health 13, 296.
Reid, C.R., Bush, P.M., Karwowski, W., Durrani, S.K., 2010. Occupational postural activity
and lower extremity discomfort: A review. Int J Ind Ergonom 40, 247-256.
Roelofs, A., Straker, L., 2002. The experience of musculoskeletal discomfort amongst bank
tellers who just sit, just stand or sit and stand at work. Ergonomics SA 14, 11-29.
Roffey, D.M., Wai, E.K., Bishop, P., Kwon, B.K., Dagenais, S., 2010. Causal assessment of
occupational standing or walking and low back pain: results of a systematic review. Spine
J 10, 262-272.

16
Ryan, C.G., Grant, P.M., Tigbe, W.W., Granat, M.H., 2006. The validity and reliability of a
novel activity monitor as a measure of walking. Br J Sports Med 40, 779-784.
Srinivasan, D., Mathiassen, S.E., 2012. Motor variability in occupational health and
performance. Clin Biomech 27, 979-993.
Stephens, S.K., Winkler, E.A., Trost, S.G., Dunstan, D.W., Eakin, E.G., Chastin, S.F., Healy,
G.N., 2014. Intervening to reduce workplace sitting time: how and when do changes to
sitting time occur? Br J Sports Med 48, 1037-1042.
Straker, L., Coenen, P., Dunstan, D.W., Gilson, N., Healy, G.N., 2016. Sedentary work -
Evidence on an emergent work health and safety issue. Safe Work Australia, Canberra,
Australia.
Straker, L., Healy, G.N., Atherton, R., Dunstan, D.W., 2014. Excessive occupational sitting is
not a "safe system of work": time for doctors to get chatting with patients. Med J Aust
201, 138-140.
Straker, L., Mathiassen, S.E., 2009. Increased physical work loads in modern work--a
necessity for better health and performance? Ergonomics 52, 1215-1225.
Thorp, A.A., Owen, N., Neuhaus, M., Dunstan, D.W., 2011. Sedentary behaviors and
subsequent health outcomes in adults a systematic review of longitudinal studies, 1996-
2011. Am J Prev Med 41, 207-215.
Tremblay, M.S., Colley, R.C., Saunders, T.J., Healy, G.N., Owen, N., 2010. Physiological
and health implications of a sedentary lifestyle. Appl Physiol Nutr Metab 35, 725-740.
Waersted, M., Hanvold, T.N., Veiersted, K.B., 2010. Computer work and musculoskeletal
disorders of the neck and upper extremity: a systematic review. BMC Musculoskelet
Disord 11, 79.
Waters, T.R., Dick, R.B., 2014. Evidence of health risks associated with prolonged standing
at work and intervention effectiveness. Rehabil Nurs 40, 148-165.

17
Table 1. Descriptive statistics of the study population.

Variable Mean (SD) or n (%)

n 216
Female, n (%) 150 (69%)
Age (years) 45.4 (9.3)
Height (cm) 167.0 (9.6)
Waist circumference (cm) 93.2 (14.9)
Low-back symptoms present, n (%) 146 (68%)
Lower-extremity symptoms present, n (%) 149 (69%)
Upper-extremity symptoms present, n (%) 180 (83%)
Sitting at work (minutes/8-hour workday) 379.7 (44.4)
Standing at work (minutes/8-hour workday) 67.2 (37.6)
Stepping at work (minutes/8-hour workday) 33.1 (14.2)
Usual sitting bout duration (minutes) 33.1 (14.9)
Usual standing bout duration (minutes) 1.7 (1.2)
Sitting not at work (minutes/day) 492.6 (118.8)
Standing not at work (minutes/day) 278.6 (98.4)
Stepping not at work (minutes/day) 117.3 (42.4)
Currently smoke, n (%) 40 (18%)
Mental work demands score (0-100)1 17.0 (16.0)
Fatigue score (0-28)2 13.8 (5.3)
Data are n (%) or mean (with standard deviation; SD) with standard deviation calculated by linearized variance
estimation to account for workplace clustering.
1Mental work demands were assessed using the Work Limitations Questionnaire (Lerner et al., 2001). Score

are composed of 9 items, ranging from 0% (least demanding) to 100% (most demanding).
2Fatigue score was measured using a checklist (Commonwealth Government of Australia, 2011) to assess

physical health symptoms (including fatigue). Scores range from 0 (least fatigue) to 28 (most fatigue).

18
Table 2. Adjusted mean prevalence of musculoskeletal symptoms (low-back, lower-extremity and upper-
extremity) with occupational activity and usual bout duration, and compared between tertiles of activitya

Lower-extremity Upper-extremity
Low-back symptoms symptoms symptoms
Difference (95% Difference (95% Difference (95%
CI)b p CI)b p CI)b p
h/day -0.19 (-0.56 0.18) 0.287 -0.28 (-0.63 0.06) 0.102 0.09 (-0.24 0.41) 0.576
T2 v 1 -0.12 (-0.31 0.06) 0.259 -0.05 (-0.19 0.1) 0.793 0.06 (-0.05 0.16) 0.420
Sitting T3 v 1 -0.12 (-0.37 0.13) 0.526 -0.10 (-0.40 0.20) 0.762 -0.03 (-0.13 0.08) 0.893
T3 v 2 0.01 (-0.18 0.20) >0.999 -0.05 (-0.32 0.22) 0.934 -0.08 (-0.22 0.05) 0.328
Overall p - 0.270 - 0.640 - 0.325
h/day 0.26 (-0.17 0.69) 0.219 0.38 (-0.02 0.77) 0.062 -0.03 (-0.47 0.41) 0.887
T2 v 1 0.13 (-0.07 0.33) 0.279 0.13 (-0.10 0.35) 0.382 0.04 (-0.1 0.17) 0.845
Standing T3 v 1 0.16 (-0.09 0.42) 0.282 0.17 (-0.12 0.47) 0.354 0.01 (-0.14 0.15) 0.999
T3 v 2 0.03 (-0.15 0.21) 0.945 0.04 (-0.14 0.23) 0.881 -0.03 (-0.1 0.04) 0.603
Overall p - 0.218 - 0.334 - 0.494
h/day 0.13 (-0.95 1.21) 0.799 0.34 (-0.60 1.29) 0.447 -0.58 (-1.48 0.31) 0.182
T2 v 1 0.06 (-0.06 0.18) 0.454 0.04 (-0.11 0.19) 0.833 0.03 (-0.11 0.18) 0.921
Stepping T3 v 1 0.01 (-0.31 0.34) 0.999 0.02 (-0.20 0.23) 0.995 -0.07 (-0.23 0.09) 0.612
T3 v 2 -0.05 (-0.34 0.24) 0.956 -0.03 (-0.18 0.13) 0.958 -0.10 (-0.22 0.02) 0.117
Overall p 0.387 - 0.695 - 0.174
10 min -0.07 (-0.19 0.05) 0.236 -0.00 (-0.17 0.16) 0.949 -0.02 (-0.18 0.14) 0.841
Usual T2 v 1 -0.01 (-0.17 0.15) 0.997 0.09 (-0.05 0.23) 0.273 0.09 (-0.06 0.24) 0.292
sitting
T3 v 1 -0.11 (-0.27 0.05) 0.217 -0.07 (-0.27 0.12) 0.690 -0.09 (-0.28 0.11) 0.562
bout
duration T3 v 2 -0.10 (-0.29 0.09) 0.433 -0.17 (-0.34 0.01) 0.061 -0.18 (-0.34 -0.02) 0.029
Overall p - 0.203 - 0.072 - 0.049
min 0.00 (-0.14 0.14) 0.959 0.10 (-0.06 0.25) 0.195 0.13 (-0.03 0.30) 0.107
Usual T2 v 1 -0.11 (-0.31 0.09) 0.378 -0.03 (-0.30 0.24) 0.988 -0.04 (-0.30 0.23) 0.977
standing
T3 v 1 -0.06 (-0.23 0.10) 0.659 0.06 (-0.12 0.23) 0.765 0.09 (-0.06 0.24) 0.318
bout
duration T3 v 2 0.05 (-0.11 0.21) 0.803 0.09 (-0.11 0.28) 0.558 0.13 (-0.10 0.35) 0.371
Overall p - 0.363 - 0.337 - 0.129
a Probit models with linearized variance estimation adjusted for potential confounders: age, gender, mental

work demands score, fatigue score (low back); age, gender, waist circumference, mental work demands score,
fatigue score (lower-extremity symptoms); and, smoking, mental work demands and fatigue score (upper-
extremity symptoms).
b Coefficient (95% Confidence Interval, CI) for per h/day for linear association, or for comparing tertiles (T),

pairwise comparison of marginal means with significance adjustment for multiple comparisons (Sidak method).
T1 = least time/shortest accumulation; T3=most time/most prolonged accumulation with cut points described
in supplementary file 2.

19
Figures

Figure 1. Adjusted mean prevalence of musculoskeletal symptoms with increasing

workplace time spent sitting (a), standing (b) and stepping (c)a

a Adjusted mean prevalence (95% CI), adjusting for: age, gender, mental work demands
score, fatigue score (low back); age , gender, waist circumference, mental work demands
score, fatigue score (lower-extremity symptoms); and, smoking, mental work demands and
fatigue score (upper-extremity symptoms).

Figure 2. Adjusted mean prevalence of musculoskeletal symptoms with accumulating

workplace sitting (a) and standing (b) in increasingly long boutsa

a Adjusted mean prevalence (95% CI), adjusting for: age, gender, mental work demands
score, fatigue score (low back); age , gender, waist circumference, mental work demands
score, fatigue score (lower-extremity symptoms); and, smoking, mental work demands and
fatigue score (upper-extremity symptoms).

20
21
Supplementary file 1. Calculation of ‘usual bout durations’

The appropriate distributions to assume for standing are not known. Active (non-sitting) bouts have
been argued to follow a log-normal distribution (Chastin & Granat, 2010). The distribution of
standing bouts (with duration rounded up to the nearest second) are plotted in Figure S1, alongside
a log-normal distribution. The approach used to calculate usual sitting bout duration (W50) was as
the midpoint of the cumulative distribution function for sitting bouts, which have been shown to

22
follow a power-law distribution (Chastin & Granat, 2010; Chastin et al., 2015; Stephens et al., 2014).
In this approach, the outcome (y, cumulative proportion of standing time accrued in bouts of
duration ≤ t) is modelled in a non-linear regression (Marquadt method) as a function of bout
duration (t), usual bout duration (W50%) and the free parameter (n) in the form of:
𝑛
𝑡
𝑦 = 𝑛 𝑛
𝑡 + (𝑊50)

The same approach to determining the accumulation midpoint was used to estimate the standing
bout accumulation midpoint, as a generic curve-fitting method. We evaluated the appropriateness
of the curve-fitting technique as an approximation for the midpoint by plotting observed data
against the predicted curve and its 95% confidence intervals. The observed data closely followed the
fitted curve, especially around the midpoint. The method appeared to perform acceptably in
approximating the mid-point of standing time accumulation (just over one minute). All bouts are
shown in Figure S2 and only the short bouts of less than five minutes in Figure S3.

Figure S1: Frequency distribution of standing bouts (log-log plot).

23
Figure S2: Standing time accumulation in bouts of each duration and shorter: observed and as
estimated by the curve fitting method

24
Figure S3: Standing time accumulation in bouts of each duration and shorter: observed and as
estimated by the curve fitting method, only bouts ≤ 5 minutes shown.

Supplemental file 2. Unadjusted prevalence of MSS by tertiles of workplace activity and amount of activity in
each tertile a
Tertile Mean [Minimum, n Low-back Lower- Upper-
Maximum] value extremity extremity
1 330.3 [199.7 366.9] 70 53 (76%) 57 (81%) 53 (76%)
Sitting 2 386.2 [367.5 402.2] 73 46 (63%) 64 (88%) 49 (67%)
3 420.6 [402.3 452.6] 73 47 (64%) 59 (81%) 47 (64%)
1 34.9 [13.3 46.4] 73 43 (59%) 60 (82%) 43 (59%)
Standing 2 61.1 [46.7 75.2] 73 51 (70%) 62 (85%) 51 (70%)
3 107.3 [75.5 253.6] 70 52 (74%) 58 (83%) 55 (79%)
1 18.6 [5.7 25.7] 71 46 (65%) 60 (84%) 48 (68%)
Stepping 2 31.1 [25.8 37.4] 71 52 (73%) 64 (90%) 51 (72%)
3 48.8 [37.5 88.5] 74 48 (65%) 56 (76%) 50 (68%)
1 17.9 [8.0 24.1] 68 49 (72%) 55 (81%) 47 (69%)
Usual sitting bout
2 30.0 [24.5 37.1] 73 51 (70%) 65 (89%) 57 (78%)
duration
3 50.0 [37.2 77.9] 75 46 (61%) 60 (80%) 45 (60%)
1 0.9 [0.4 1.2] 71 51 (72%) 63 (89%) 49 (69%)
Usual Standing bout 2 74
1.4 [1.2 1.7] 47 (63%) 61 (82%) 46 (62%)
duration
3 2.8 [1.7 12.9] 71 48 (68%) 56 (79%) 54 (76%)
a n and prevalence excludes participants without corresponding confounder data for the adjusted models.

25