Applied Ergonomics 78 (2019) 37–53

Contents lists available at ScienceDirect

Applied Ergonomics
journal homepage: www.elsevier.com/locate/apergo

Review article

The effect of sit-stand desks on office worker behavioral and health T

outcomes: A scoping review
April J. Chambersa,∗, Michelle M. Robertsonb,c, Nancy A. Bakerd
a
Department of Bioengineering, University of Pittsburgh, Pittsburgh, PA, 15261, USA
b
Department of Psychological Sciences, Center for the Promotion of Health in the New England Workplace, University of Connecticut, Storrs, CT, 06103, USA
c
Office Ergonomics Research Committee, Framingham, MA, 01702, USA
d
Department of Occupational Therapy, Tufts University, Medford, MA, 02155, USA

ARTICLE INFO ABSTRACT

Keywords: This scoping review examines the effects of sit-stand desks (SSDs) on six domains: behavior (e.g. time sitting and
Sit-stand desk standing), physiological, work performance, psychological, discomfort, and posture. Fifty-three articles met
Standing desk criteria. We determined the percentage of significant results for each domain. Forty-seven studies were ex-
Workplace intervention perimental trials. Sample sizes ranged from six to 231 participants. Follow-up time-frames ranged from one day
to one year. Sixty-one percent of behavioral (24 studies), 37% of physiological (28 studies), 7% of work per-
formance (23 studies), 31% of psychological (11 studies), 43% of discomfort (22 studies), and 18% of posture
domain results (4 studies) were significant. We conclude that SSDs effectively change behaviors, but these
changes only mildly effect health outcomes. SSDs seem most effective for discomfort and least for productivity.
Further study is needed to examine long-term effects, and to determine clinically appropriate dosage and
workstation setup.

1. Introduction Thus, the larger concern may not just be sitting, but any sedentary
activity. Sedentary activity is defined as an activity where the MET level
Excessive sitting time has been linked to an increased likelihood of is ≤ 1.5 for sitting or ≤2.0 for passive standing (Tremblay et al., 2017).
many negative health outcomes including mortality. Those who sit from People standing at SSDs may not engage in enough active behaviors to
8 to 11 h per day have a 15% increase in mortality rate in the next 3 surpass sedentary levels. Thus, although multiple studies demonstrate
years compared to those who sit < 4 h (van der Ploeg et al., 2012). In that SSDs reduce sitting and increase standing behaviors (Commissaris
addition, obesity, type 2 diabetes, cancer, and cardiovascular disease et al., 2016), the connection with these changes and health benefits is
are more likely in those who spend excessive time sitting (Dunstan less clear.
et al., 2013). Office workers typically spend more than half of their day With the popularity of SSDs interventions, there have been multiple
sitting, making them an at risk group for developing sitting-related systematic reviews related to the use of these and other active work-
conditions (Pronk et al., 2012). Though there is research that suggests station desks for computer use, along with the role of training in im-
that sitting may be a health risk, other evidence reports little or no plementing sit-stand workstations (Commissaris et al., 2016; Benatti
association between occupational sitting and health risk (van Uffelen and Ried-Larsen, 2015; Chau et al., 2010; Chu et al., 2016; Agarwal
et al., 2010), Despite these contrasting views, office workers have be- et al., 2017; Karakolis and Callaghan, 2014; Karol and Robertson, 2015;
come a target of interventions aimed at decreasing sitting time, thereby MacEwen et al., 2015; Tew et al., 2015; Torbeyns et al., 2014; Tudor-
reducing the potentially associated health risks. One such intervention Locke et al., 2014; Wilks et al., 2006). Many of these reviews have
is sit-stand desks (SSDs). examined the ability of multiple physical workstation interventions
SSDs are built on the premise that reducing sitting time by standing such as SSDs, treadmill desks, and pedaling desks, to decrease sitting
to work during computer use has a desirable effect on health outcomes. time (Commissaris et al., 2016; Chau et al., 2010; Chu et al., 2016; Tew
Yet, this presumption may not hold true. Sitting is a sedentary behavior et al., 2015; Shrestha et al., 2015) as a surrogate for increased physical
characterized by little movement and low energy expenditure. It may activity beyond the sedentary level. Although these reviews have found
be these characteristics and not sitting, per se, that put health at risk. a positive effect of SSDs on sitting, the effect on health outcomes is less

∗
Corresponding author.
E-mail address: [email protected] (A.J. Chambers).

https://doi.org/10.1016/j.apergo.2019.01.015
Received 13 June 2018; Received in revised form 29 January 2019; Accepted 31 January 2019
0003-6870/ © 2019 Elsevier Ltd. All rights reserved.
A.J. Chambers, et al. Applied Ergonomics 78 (2019) 37–53

clear. Reviews have alternatively focused on the effect of SSDs or active 2.2. Study selection
workstations on physiological health, discomfort or work performance
(Agarwal et al., 2017; Karakolis and Callaghan, 2014; MacEwen et al., We selected quantitative studies that compared the use of SSDs for
2015; Torbeyns et al., 2014; Tudor-Locke et al., 2014). Tudor-Locke computer use to sitting-only computer workstations. The studies had to
(Tudor-Locke et al., 2014) determined energy expenditure during SSD address the use of SSDs with adults, and not children or classroom use.
use was comparable to that of a traditional desk. MacEwan (MacEwen The workstations could be evaluated in either lab or work settings, but
et al., 2015) reported that standing desks had little effect on physio- participants had to have personal workstations. Studies had to be in
logical outcomes, such as post prandial glucose or HDL cholesterol. English and contain sufficient information to complete extraction
Both Karakolis (Karakolis and Callaghan, 2014) and Agarwal (Agarwal forms. We did not limit our studies to randomized clinical trial, but
et al., 2017) examined the effect of SSD on discomfort and reported that included cohort and cross-sectional studies as long as they compared
there was some positive effect of SSDs on discomfort levels without sitting to standing. We excluded qualitative studies as well as studies
reducing productivity, but that the results were too inconclusive to that compared SSDs to active workstations, although if such a study
make any solid recommendations. Thus, although SSDs decrease sitting compared SSDs to sit conditions they were included in the review with
time without reducing work performance, their effect on important active workstation comparisons omitted. Systematic reviews, and the-
health outcomes may not be consequential. Improvements in physical oretical and expert opinion case studies were omitted.
activity levels and health-related outcomes tend to be stronger or may
only be present with active workstations (Commissaris et al., 2016;
Torbeyns et al., 2014; Tudor-Locke et al., 2014). 2.3. Data collection process
While these reviews provide important insights into the use of SSDs,
there are still gaps in our current knowledge of the effectiveness of We developed data extraction sheets to capture essential informa-
SSDs. Many of the reviews combined the results from multiple alter- tion based on the aims of our scoping review. Extractions sheets con-
native workstation types making it difficult to determine the true effect sisted of six sections: (1) overall description of study results; (2)
of a SSD (Commissaris et al., 2016; Chau et al., 2010; Chu et al., 2016; methodology; (3) sample characteristics; (4) interventions; (5) results;
Karol and Robertson, 2015; MacEwen et al., 2015; Tudor-Locke et al., (6) creditability. The results section evaluated six outcome domain
2014; Neuhaus et al., 2014a). This strategy was likely done due to areas which were further broken down into sub-categories. The six
minimal research on SSDs alone at the time of the review. However, in domains were behavior, physiological, work performance, psycholo-
the last few years an explosion of SSD studies have been published, gical, discomfort, and posture. Each of the outcomes were evaluated as
warranting an updated review. Past reviews tended to look at single either being measured through self-reports; instrumentation; and/or
outcomes; reducing sitting, reducing pain, improving physiological rating by an expert observer. Each study was reviewed independently
outcomes. No single review, to our knowledge, has looked at the mul- by at least two authors.
tiple areas that SSDs are purported to affect. There are no reviews that Behavior domain category outcomes were defined as mean time of
isolate the effect of SSDs on total worker health and performance. Thus, sitting, standing, transitioning from sit to stand, and active movements.
it is difficult to estimate and compare the full effects of SSDs across The physiological domain had multiple categories: cardiovascular, en-
multiple health outcomes. It is important to consider all aspects of docrine-related, cognitive, fatigue, edema and obesity. Specifically, the
workers health when determining the effectiveness of novel occupa- cardiovascular outcomes were defined as improvements in metabolic
tional interventions such as SSDs. equivalents (METs), rate of energy consumption, heart rate, and VO2
max. Endocrine-related measures were operationally defined as
changes in endocrine-related functions such as glucose, cholesterol and
1.1. Scoping review aim
other related derivatives. Cognitive outcomes were defined as changes
in cognitive functions such as, attention, reaction time, and memory.
In this scoping review we focus on specifying the effectiveness of
Fatigue outcomes were defined as changes in mental or physical fa-
SSDs on multiple primary outcome domains. Specifically, this review
tigue. Obesity was defined as changes related to BMI, waist cir-
aims to determine the most promising uses and implementation of SSDs
cumference, and fat ratios. The work performance domain included two
in relationship to total worker health and performance outcomes, and
categories: absenteeism/presenteeism and productivity. In the psycho-
what additional areas may need to be explored. We examined the fol-
logical domain, the categories of outcomes consisted of general health,
lowing outcomes: (1) reductions in sitting and increases in non-seden-
work satisfaction, self-efficacy/confidence, and mood. Musculoskeletal
tary behavior; (2) improvements in physiological outcomes such as
discomfort and pain were evaluated by the number of reported
cardiovascular health, endocrine-related health, cognition, fatigue,
symptom occurrences for general areas of the body. For the posture
edema, and obesity; (3) effects on work performance; (4) improvements
domain, the studies were assessed by the number of changes in postures
in psychological outcomes such as mood, self-efficacy/confidence; (5)
and workstation adjustments.
reductions in musculoskeletal discomfort; (6) improvements in com-
puting work postures.
2.4. Statistics
2. Methodology
Only results from the first follow-up visit one were included in this
2.1. Search strategy review. For completeness, results found in follow-up visits two or more
are provided in the appendix, but no statistical analysis was performed
In early 2017, we searched PubMed and Web of Science for articles as more than two visits were not included across all studies. Outcomes
that contained the following terms: sit stand desk; standing desk; height of interest for the first follow-up were counted. We then determined the
adjustable desk; sit stand workstation; sit stand computer workstation, number of these outcomes that were significant to determine the total
and office/computer work. We also searched previous systematic re- and percent of outcomes that were significant for each separate domain
views and the Human Factors & Ergonomics Conference proceedings and category. We used each study's results to develop overall de-
from 2007 to 2017 for additional papers. A second review of the lit- scriptive statistics including mean, median, standard deviation, max-
erature was completed in December 2017 using the same terms and imum, and minimum. We also determined the percentage of studies
limiting the studies to only those published in 2017 to capture any that found no significance in their results or that had all results sig-
additional articles published from January to December 2017. nificant.

38
A.J. Chambers, et al. Applied Ergonomics 78 (2019) 37–53

Table 1
Characteristics of Studies included in Sit/Stand Scoping Review.
Author (year) Study Design Site Sample (Baseline) 1st Follow- Intervention & Control Type of Desk CP
up
Time
Frame

Alkhajah et al. (2012) Quasi-experimental Work Total: 33; Int: 18; Con: 1 wk# Int: SSD, PEd, SubC Ergotron 30%
15 Con: RD, NoT, SubC Work-Fit-S, Single LD Sit-
Stand Workstation
Bantoft et al. (2016) RCT (Crossover) Lab Total: 45; Int: 45; Con: < 1 day Int: SSD*, NoT, ResC Not Reported 70%
45 Con: RD, NoT, ResC
Beers et al. (2008) RCT (Crossover) Lab Total: 24; Int: 24; Con: < 1 day Int: Stand*, NoT, ResC Not Reported 70%
24 Con: Sit, NoT, ResC
Britten et al. (2016) Quasi-experimental, Lab Total: 20; Int: 20; Con: < 1 day Int: Stand*, NoT, ResC Not Reported 70%
(Crossover) 20 Con: Sit, NoT, ResC
Buckley et al. (2014) Experimental (Single Field Total: 10; Int: 10 1 wk Int: SSD, NoT, ResC Ergontron 60%
Group) Con: RD, NoT, ResC Work-Fit D
Burns et al. (2017) RCT (Crossover) Lab Total: 22; Int: 22; Con: < 1 day Int: Stand, NoT, ResC Not Reported 80%
22 Con: Sit, NoT, ResC
Carr et al. (2016) Cohort Work Total: 69; SSD: 31; RD: 24 wks Int: SSD, NoT, SubC Knoll Dividend Horizon 58%
38 Con: RD, NoT, SubC
Chau et al. (2014) RCT (Crossover) Work Total: 42; Int: 42; Con: 4 wks Int: SSD, PEd, SubC Ergotron 70%
38 Con: RD, NoT, SubC Work-Fit S
Chau et al. (2016) Quasi-experimental Work Total: 31; Int: 16; Con: 1 wk# Int: SSD & email reminders, PEd, Rumba “2 Stage” Sit-Stand 60%
15 SubC Workstation
Con: RD, NoT, SubC
Coenen et al. (2017) RCT (Cluster Work Total: 231; Int: 136; 12 wks Int: SSD & multicomponent Ergotron 70%
Randomization) Con: 95 strategies, PEd, SubC Work-Fit S
Con: RD w/no add-ons, NoT, SubC
Commissaris et al. RCT (Crossover) Lab Total: 15; Int: 15; Con: < 1 day Int: SSD*, NoT, ResC Not Reported 80%
(2014) 15 Con: RD, NoT, ResC
Cox et al. (2011) RCT (Crossover) Lab Total: 31; Int: 31; Con: < 1 day Int: SSD*, NoT, ResC Not Reported 80%
31 Con: RD, NoT, ResC
Davis and Kotowski, RCT (Crossover) Work Total: 37; Int 1: 37; Int 4 wks Int: SSD w/& w/o reminder Not Reported 60%
(2014) 2: 37; Int 3: 37; Con: 37 software, NoT, SubC
Con: RD w/& w/o reminder
software, NoT, SubC
Donath et al. (2015) RCT (2 or more groups) Work Total: 38; Int: 19; Con: 12 wks Int: SSD w/prompts, NoT, SubC Office Plus Ergon 50%
19 Con: SSD w/o prompts, NoT, SubC
Dutta et al. (2014) RCT (Crossover) Work Total: 29; Int: 29; Con: 4 wks Int: SSD, NoT, SubC Ergotron 50%
29 Con: RD, NoT, SubC Work-Fit (A, S, or D)
Ebara et al. (2008) RCT (Crossover) Lab Total: 24; Int: 24; Con: < 1 day Int: Sit/Stand*, NoT, ResC NeX Desk 50%
24 Con: Sit, NoT, ResC
Finch et al. (2017) RCT (Crossover) Lab Total: 96; Int: 96; Con: < 1 day Int: Stand, NoT, ResC Not Reported 70%
Con: Sit, NoT, ResC
Foley et al. (2016) Experimental (Single Work Total: 88; Int: 88 4 wks Int: SSD & Activity Based Work, Not Reported 50%
Group) NoT, SubC
Con: RD, NoT, SubC
Gao et al. (2016a) Cohort Work Total: 45; SSD: 24; RD: 24 wks Int: SSD, NoT, SubC ISKU 67%
21 Con: RD, NoT, SubC
Gao et al., (2016b) Quasi-experimental Work Total: 24; SSD: 10; RD: 1 day Int: SSD, NoT, SubC ISKU 60%
14 Con: RD, NoT, ResC
Gao et al. (2017) RCT (Crossover) Lab Total: 18; Int: 18; Con: < 1 day Int: Stand, NoT, ResC ISKU 50%
Con: Sit, NoT, ResC
Gibbs et al. (2017a) RCT (Crossover) Lab Total: 26; Int: 25; Con: 1 day Int: Sit/Stand, NoT, ResC Float/Quickstand Humanscale 90%
Con: Sit, NoT, ResC
Gibbs et al. (2017b) RCT (Crossover) Lab Total: 18; Ints: 18; Con: < 1 day Int: Sit/Stand & Stand, NoT, ResC Float/Quickstand Humanscale 90%
18 Con: Sit, NoT, ResC
Gilson et al. (2017) RCT (Crossover) Lab Total: 20; Int: 20; Con: 4 days Int: Sit/Stand*, NoT, ResC Varidesk 60%
Con: Sit, NoT, ResC Pro Plus 48
#
Graves et al. (2015) RCT (2 or more groups) Work Total: 47; Int: 26; Con: 4 wks Int: SSD, PEd, SubC Ergontron 60%
21 Con: RD, NoT, SubC Work-Fit A
Hadgraft et al. (2017) Secondary data analysis Work Total: 231; Int: NR; Con: 12 wks# Int: SSD & multicomponent Not Reported 60%
NR strategies, PEd&AEd, SubC
Con: RD, NoT, SubC
Healy et al. (2016) RCT (2 or more groups) Work Total: 231; Int: 136; 12 wks# Int: SSD & multicomponent Ergotron 60%
Con: 95 strategies, PEd & AEd, SubC Work-Fit S
Con: RD, NoT, SubC
Healy et al. (2017) RCT (2 or more groups) Work Total: 231; Int: 7; Con: 7 12 wks# Int: SSD & multicomponent Not Reported 80%
strategies, PEd&AEd, SubC
Con: RD, NoT, SubC
RCT (2 or more groups) Work Total: 44; Int: 34; Con: 4 wks Int: SSD, NoT, SubC Not Reported 60%
10 Con: RD, NoT, SubC
Hedge et al. (2005) RCT (Crossover) Lab Total: 18; Int: 18; Con: < 1 day Int: SSD w & w/o negative tilt Not Reported 20%
18 keyboard, NoT, ResC
Con: RD w & w/o negative tilt
keyboard, NoT, ResC
(continued on next page)

39
A.J. Chambers, et al. Applied Ergonomics 78 (2019) 37–53

Table 1 (continued)

Author (year) Study Design Site Sample (Baseline) 1st Follow- Intervention & Control Type of Desk CP
up
Time
Frame

Horswill et al. (2017) Quasi-Experimental Lab Total: 16; Int: 16; Con: < 1 day Int: Stand*, NoT, ResC Not Reported 60%
16 Con: Sit, NoT, ResC
Husemann et al. (2009) RCT (2 or more groups) Lab Total: 60; Int: 30; Con: 5 days Int: Sit/Stand, NoT, ResC Not Reported 70%
30 Con: Sit, NoT, ResC
Kar and Hedge (2016) RCT (Crossover) Lab Total: 12; Int: 12; Con: < 1 day Int: Stand, NoT, ResC Not Reported 40%
12 Con: Sit, NoT, ResC
Karakolis et al. (2016) RCT (Crossover) Lab Total: 24; Int 1: 24; Int < 1 day Int: Sit/Stand & Stand, NoT, ResC Teknion Xpres 40%
2: 24; Con: 24 Con: Sit, NoT, ResC
Le and Marras, (2016) RCT (Crossover) Lab Total: 20; Int 1: 20; Int < 1 day Int: Stand*, NoT, ResC TIMOTION Technology 50%
2: 20; Con: 20 Con: Sit, NoT, ResC
Li et al. (2017) RCT (2 or more groups) Work Total: 32; Int 1: 8; Int 2: 4 wks Int: SSD w/3 different sit/stand Varidesk Pro Plus 3 or 60%
7; Int 3: 7; Con: 10 protocols, PEd, ResC Ergotron
Con: RD, NoT, SubC Work-Fit T or Strata Electric
SSD & Workstation
Lin et al. (2017) Cross-sectional Lab Total: 20 < 1 day Int: Stand, NoT, ResC Airtouch 67%
Con: Sit, NoT, ResC
MacEwen et al. (2017) RCT (2 or more groups) Work Total: 28; Int: 16; Con: 12 wks Int: SSD, NoT, SubC SC45 60%
12 Con: RD, NoT, SubC
#
Mansoubi et al. (2016) Experimental (Single Work Total: 40; Int: 40 1 wk Int: SSD, PEd, SubC Ergotron 30%
Group) Con: RD, NoT, SubC Work-Fit S
Nerhood and Thompson Cohort Work Total: NR; SSD: NR; RD: 36 wks Int: SSD, PEd&AEd, SubC Not Reported 0%
(1994) NR Con: RD, NoT, SubC
Neuhaus et al. (2014b) RCT (Cluster Field Total: 44; Int 1: 16; Int 12 wks Int: SSD & multicomponent Ergotron 80%
Randomization) 2: 14; Con: 14 strategies, PEd&AEd, SubC Work-Fit S
Con: RD, NoT, SubC
Ognibene et al. (2016) RCT (2 or more groups) Field Total: 46; Int: 25; Con: 12 wks Int: SSD, Ped, SubC Ergotron 80%
21 Con: RD, NoT, SubC Work-Fit A/Work-Fit S
Paul (1995a) Experimental (Single Lab Total: 12; Int: 12 12 wks Int: SSD, NoT, ResC Not Reported 50%
Group) Con: RD, NoT, SubC
Paul (1995b) Experimental (Single Work Total: 6; Int: 6 6 wks Int: SSD, NoT, ResC Not Reported 20%
Group) Con: RD, NoT, SubC
#
Pronk et al. (2012) Quasi-Experimental Field Total: 34; Int: 24; Con: 4 wks Int: SSD & multicomponent Ergotron 50%
10 strategies, NoT, SubC Work-Fit S/Work-Fit C
Con: RD, NoT, SubC
Robertson et al. (2013) RCT (2 or more groups) Lab Total: 22; Int: 11; Con: 3 wks Int: SSD w/training & w/o Not Reported 80%
11 training, PEd&AEd, SubC & ResC
Con: RD, PEd, SubC & ResC
Roemmich (2016) RCT (Crossover) Lab Total: 13; Int 1: 13; 1 wk# Int: SSD, PEd, SubC Not Reported 40%
Int 2: 13; Con: 13 Con: RD, NoT, SubC
Russell et al. (2016) RCT (Crossover) Lab Total: 36; Int: 36; Con: 5 days Int: Stand, NoT, ResC ACTIU mechanical elevation 70%
36 Con: Sit, NoT, ResC SSD, model MB212)
Straker et al. (2013) Cross-sectional Field Total: 131 1 day Int: SSD, NoT, SubC Not Reported 92%
Con: RD, NoT, SubC
Thorp et al. (2014b) RCT (Crossover) Field Total: 26; Int: 26; Con: 5 days Int: Sit/Stand, NoT, ResC Linak model 1600 × 800 mm 80%
26 Con: Sit, NoT, ResC
Thorp et al. (2014a) RCT (Crossover) Lab Total: 23; Int: 23; Con: 5 days Int: Sit/Stand, PEd, SubC Linak model 1600 × 800 mm 80%
23 Con: Sit, NoT, ResC
Thorp et al. (2016) RCT (Crossover) Lab Total: 23; Int: 23; Con: 5 days Int: Sit/Stand, PEd, ResC Linak model 1600 × 800 mm 70%
23 Con: Sit, NoT, ResC
Tobin et al. (2016) RCT (2 or more groups) Field Total: 52; Int: 26; Con: 5 wks Int: SSD, NoT, SubC Ergotron 70%
26 Con: RD, NoT, SubC Work-Fit

CP – Credibility Percentage; RCT – Randomized Clinical Trial; # Study had additional follow-up periods (see appendix for details); SSD – Sit-Stand Desk; RD - Regular
Desk; * additional arm, not discussed in this scoping review; Sit – study tested sitting to work, not specifically a desk; Stand – study tested standing at work, not
specifically a desk; NoT – No Training, PEd – Passive Education; AEd – Active Education; SubC – Subjects controlled timing of sit/stand; ResC – Researcher controlled
timing of sit/stand.

2.5. Creditability bias and increased trustworthiness of the results.

During data extraction we evaluated each study for credibility of

research methods using two different sets of “yes/no” criteria. Each 3. Results
study was assessed by one reviewer. Experimental designs were as-
sessed using 10 questions and observational studies (cohort and cross 3.1. Study descriptions
sectional) using 12 (for exact questions see appendix). Criteria were
based on those suggested by Jewell et al. (Jewell, 2018). The score of The literature search resulted in 911 articles of which 536 had re-
“yes” was only assigned if the article explicitly provided information on levant titles. We identified an additional 13 through a hand search of
the criteria. The “yes” results were summed and a percent score was review articles for a total of 549 relevant titles. We removed 353 du-
calculated (Table 1). A high credibility percentage indicates that the plicates leaving 196 articles. After abstract review, 53 studies remained.
study used best practice research methodology with reduced risk for Studies were removed for the following reasons: they were not related
to a completed study (study protocol only); they were on children; they

40
A.J. Chambers, et al. Applied Ergonomics 78 (2019) 37–53

scores greater than 80% suggesting excellent credibility while 62% had
scores between 50 and 79%, suggesting moderate to good credibility.
Of the studies with the highest credibility, slightly more than half were
lab based studies. The study with a 0% in the credibility score (Nerhood
and Thompson, 1994) was an early published abstract, and lacked
sufficient detail to assign any of the credibility scores.

3.2. Overall outcome domains

Table 2 provides a breakdown of the results of the study by domain

and sub category, and includes any results from multiple follow-ups.
The most commonly reported domain was physiological, followed
closely by behavioral, work performance, and discomfort. A limited
number of studies included outcomes in the domains of psychological
and posture (Fig. 2). The greatest number of significant results were in
the behavior domain. Discomfort, physiological, and psychological re-
ported moderate to mild significance. Minimal significance was noted
in posture and work performance domains (Fig. 3).

3.3. Behavior

A total of 24 studies reported results related to the behavioral do-

main (Fig. 2). Almost all of these included sit time and stand time. Only
eight studies included the number of transitions while 18 studies re-
ported on changes in active movement (Fig. 4). These behavioral
measures were primarily recorded using objective measurements such
as accelerometer-based tools. Approximately one third of studies used
self-reported measures of behavioral variables (Table 2). Overall, 61%
of results for the behavioral domain were significant (Fig. 3). Twenty-
nine percent of studies had no significant results, while 58% had all
their outcomes significant. Significant reductions in sit time and in-
creases in stand time were found in 76% and 89% of reported outcome
Fig. 1. Literature search input terms and results. measures, respectively. Significant changes in transitions were found in
54% of reported outcome measures. A significant change in active
were qualitative; they did not address a computer workstation; SSDs movements was found in only 30% of all reported outcome measures
were part of larger restructuring of the work environment, and the (Fig. 4). The majority of studies that included transitions or active
contribution of the desks could not be teased out; there was no sit only movement found no significance in any of their outcome measures.
comparison group (Fig. 1). When compared to the regular desk, the average amount of time people
Table 1 provides details of all studies. Overall, the majority of stu- decreased their sitting time when using a SSD ranged from 0.1 to 3.6 h
dies, 35 in total, were randomized controlled trials, 23 of which were per day, the average amount of time they increased standing time
crossover studies. The remaining studies were composed of seven quasi- ranged from 0.5 to 3.1 h, and the average amount of time they in-
experimental, five experimental, three cohort, two cross-sectional, and creased active movement ranged from 0 to 0.6 h (Table 2). The median
one secondary data analysis. Study data was recorded in a work setting, average times for sit, stand, and active movement were 1.3, 1.4, and 0.1
21 studies, laboratory setting, 25 studies, and in the field, 7 studies. respectively.
Studies tended to be small in size and short in duration. Sample sizes
ranged from six to 231 participants. Twenty-six studies had less than 30 3.4. Physiological
participants and 18 studies reported between 30 and 60 participants.
Only eight studies recorded data from over 60 participants. Nine studies A total of 28 studies had results related to the physiological domain
had more than one follow-up. Follow-up time frames were limited to (Fig. 2). For the sub-categories of physiological outcomes, 14 studies
one day or less in 19 studies and one month or less in an additional 13 assessed cardiovascular outcomes, 8 endocrine-related, 9 fatigue, 7
studies. A 5–12 week time period was used in 13 studies. Only eight cognitive, 4 obesity, and 1 edema (Table 2, Fig. 5). Physiological
studies used a time period of greater than 12 weeks. The majority of measures were primarily completed using instrumented measures and
studies recruited healthy working age adults for participation. One set rater observations, except for fatigue outcomes which were measured
of three studies included obese adults (Thorp et al., 2014a, 2014b, by self-report (Table 2). Overall, 37% of all results were significant
2016). A variety of SSDs were used. Thirty-four studies focused on (Fig. 3). No significant results were found in 52% of studies while 24%
testing SSDs while the others were interested in comparing a sitting had all outcomes significant. Of the categories, 61% of cardiovascular,
condition to a standing condition. Very few studies included training or 13% of endocrine-related, 63% of fatigue, 3% of cognitive, 0% of
multi-component interventions. Six studies included an active training obesity and 100% of edema results were significant (Fig. 5). The edema
component with some including health coaching. Seventeen studies results favored sitting, suggesting that foot and leg edema increased
involved passive education, involving brochures. Nine studies included during standing.
additional components to their interventions such as email reminders.
Sit-stand time and number of transitions was controlled by the parti- 3.5. Work performance
cipants in 26 of the studies.
Credibility scores ranged from 0% to 92% (Table 1). The majority of Twenty-three studies reported results related to the work perfor-
studies had moderate to excellent credibility: 23 percent of studies had mance domain (Fig. 2). Measures of productivity were included in 21
studies while only five studies included a measure of absenteeism/

41
A.J. Chambers, et al. Applied Ergonomics 78 (2019) 37–53

Table 2
Measurement Type (in italics), number of outcome measured and number of significant results for each study in the Sit/Stand Scoping Review for the Domains of
Behavior, Work Performance, Discomfort, Physiological, Psychological, and Posture at first follow-up.
Author (Year) Behav Work Perf Discom Physiol Psychol Post

Alkhajah et al. (2012) MT:IM MT:SR MT:SR MT:IM

Sit: 1,1c (↓2.4 h)Δ Ab/P: 0,1 Dis: 0,10d End: 1,5d
Std: 1,1 (↑2.2 h) Prod: 0,1 Ob: 0,5d
AMvt: 1,1 (↑0.1 h) MT:SR
Tran: 1,1 Ftg: 0,1d
Bantoft et al. (2016) MT:RO MT:SR
Cog: 0,9 Mood: 0,2
Beers et al. (2008) MT:IM MT:SR MT:IM
Prod: 0,3 Dis: 1,1a CV: 3,3
MT:SR
Ftg: 1,1a
Britten et al. (2016) MT:IM
Prod: 0,3
Buckley et al. (2014) MT:IM
CV: 1,2
End: 1,4
Burns et al. (2017) MT:IM
CV: 0,4
Carr et al. (2016) MT:IM MT:IM
Sit: 1,1 (↓1.1 h) CV: 0,5
Std: 1,1 (↑1.0 h) Ob: 0,6
AMvt: 0,2 (↑0.2 h)
Tran: 0,1
Chau et al. (2014) MT:IM;SR
Sit: 1,2 (↓1.1 h)
Std: 2,2 (↑1.0 h)
AMvt: 0,3 (↑0.1 h)
Chau et al. (2016) MT:IM;SR MT:RO MT:SR MT:SR
Sit: 0,2 (↓0.7 h) Ab/P: 0,2 Ftg: 0,1 Mood: 0,1
Std: 2,2 (↑1.3 h) Prod: 0,4
AMvt: 0,4 (↑0 h)
Coenen et al. (2017) MT:IM
Sit: 2,2 (−)
Std: 2,2 (−)
Commissaris et al. (2014) MT:IM MT:IM
Prod: 1,5a Cog: 0,10
Cox et al. (2011) MT:IM
CV: 10,16
Davis and Kotowski (2014) MT:IM MT:IM MT:SR
Sit: 9,12 (↓1.2 h) Prod: 0,30 Dis: 8,27
Std: 4,6 (↑1.1 h)
AMvt: 3,3 (−)
Tran: 5,6
Donath et al. (2015) MT:IM MT: IM
Sit: 0,1 (↓0.1 h) Cog: 0,1
Std: 0,1 (↑0.7 h)
Dutta et al. (2014) MT:IM;SR MT:SR MT:SR
Sit: 2,2 (↓1.7 h) Ab/P: 0,3 Ftg: 0,1
Std: 1,1 (−) Prod: 0,1
AMvt: 1,1 (−)
Ebara et al. (2008) MT:SR MT:SR MT:SR
Prod: 0,1 Dis: 2,14a Ftg: 0,1
Finch et al. (2017) MT:RO MT:SR MT:SR
Prod: 0,5 Dis: 1,2a Mood: 3,15
Foley et al. (2016) MT:IM;SR MT:SR MT:SR
Sit: 1,2 (↓1.1 h) Prod: 0,1 Dis: 1,8
Std: 1,1 (↑0.9 h)
AMvt: 2,5 (↑0.2 h)
Gao et al. (2016a) MT:SR MT:SR MT:SR
Sit: 1,2 (↓1.0 h) Prod: 1,1 Dis: 1,3
Std: 2,2 (↑0.5 h)
Gao et al. (2016b) MT:SR MT:SR MT:SR
Sit: 2,2 (↓1.7 h) Prod: 0,1 Dis: 1,3
Std: 2,2 (↑1.5 h)
Gao et al. (2017) MT:IM
CV: 11,11
End: 2,12
Ftg: 3,4
Gibbs et al. (2017a) MT:IM MT:IM
Sit: 1,1 (↓3.6 h) CV: 3,7
Std: 1,1 (↑3.1 h)
AMvt: 0,3 (↑0 h)
Gibbs et al. (2017b)
(continued on next page)

42
A.J. Chambers, et al. Applied Ergonomics 78 (2019) 37–53

Table 2 (continued)

Author (Year) Behav Work Perf Discom Physiol Psychol Post

MT:RO MT:SR MT:RO MT:SR

Prod: 0,2 Dis: 0,1 CV: 6,6 WSat: 0,1
MT:SR
Ftg: 0,1
Gilson et al. (2017) MT:IM
End: 0,1
Cog: 0,1
Graves et al. (2015) MT:SR MT:SR MT:IM
Sit: 1,1 (↓1.5 h) Dis: 0,3 CV: 0,4
Std: 1,1 (↑1.4 h) End: 1,3
AMvt: 0,1 (↑0.1 h)
Hadgraft et al. (2017) MT:SR
SC/SE: 2,2
Healy et al. (2016) MT:IM
Sit: 4,4 (↓1.7 h)
Std: 2,2 (↑1.6 h)
AMvt: 0,2 (↑0 h)
Healy et al. (2017)# MT:IM
CV: 0,2
End: 0,9
Ob: 0,4
MT:SR MT:SR MT:SR
Sit: 1,1 (↓1.3 h) Prod: 1,1 Dis: 16,28
Std: 1,1 (↑1.0 h)
Hedge et al. (2005) MT:RO MT:RO MT:SR MT:RO
AMvt: 2,8 (−) Prod: 0,2 Dis: 2,4a Pos: 2,4a
Horswill et al. (2017) MT:IM
CV: 4,4
Cog: 1,1
Husemann et al. (2009) MT:IM MT:SR
Prod: 0,2 Mood: 0,3
Kar and Hedge (2016) MT:RO MT:SR
Prod: 1,2 Dis: 2,8b
Karakolis et al. (2016) MT:IM MT:SR MT:RO
Prod: 0,4 Dis: 22,24b Pos: 3,48
Le and Marras (2016) MT:IM MT:SR MT:RO
Prod: 0,5 Dis: 2,2a Pos: 2,3
Li et al. (2017) MT:IM;SR
Sit: 3,3 (↓2.1 h)
Std: 3,3 (↑1.7 h)
AMvt: 0,9 (↑0.2 h)
Tran: 0,1
Lin et al. (2017) MT:SR MT:RO
Dis: 2,3b Pos: 5,11b
MacEwen et al. (2017) MT:IM MT:IM
Sit: 2,2 (↓2.6 h) CV: 0,3
Std: 2,2 (↑2.7 h) End: 0,10
AMvt: 1,1 (−) Ob: 0,5
Tran: 1,1
Mansoubi et al. (2016) MT:IM
Sit: 4,4 (↓0.8 h)
Std: 2,2 (↑2.4 h)
AMvt: 6,6 (↑0.6 h)
Nerhood and Thompson (1994) MT:SR MT:SR
Ab/P: 0,1 Dis: 8,15
Neuhaus et al. (2014b) MT:IM MT:SR MT:SR
Sit: 0,2 (↓0.6 h) Ab/P: 0,1 Dis: 0,5
Std: 0,1 (↑0.6 h)
AMvt: 0,3 (↑0 h)
Tran: 0,1
Ognibene et al. (2016) MT:SR MT:SR
Dis: 2,3 WSat: 1,1
Paul (1995a) MT:SR
WSat: 0,4
Mood: 4,5
Paul (1995b) MT:IM
Edm: 1,1a
Pronk et al. (2012) MT:SR MT:SR MT:SR MT:SR
Sit: 1,1 (↓1.9 h) Dis: 1,2 Ftg: 1,1 Mood: 5,6;
AMvt: 0,1 (↓0 h) SC/SE: 0,1
Robertson et al. (2013) MT:RO MT:IM MT:SR
Std: 1,1 (−) Prod: 1,3 Dis: 7,7
Roemmich (2016) MT:IM MT:SR
CV: 1,1 Mood: 0,6
Russell et al. (2016)
(continued on next page)

43
A.J. Chambers, et al. Applied Ergonomics 78 (2019) 37–53

Table 2 (continued)

Author (Year) Behav Work Perf Discom Physiol Psychol Post

MT:IM
Cog: 0,10
Straker et al. (2013) MT:IM
Sit: 1,2 (↓0.4 h)
AMvt: 1,1 (−)
Tran: 0,1
Thorp et al. (2014b) MT:IM
End: 1,3
Thorp et al. (2014a) MT:SR MT:SR MT:SR MT:SR
Prod: 1,2a Dis: 1,9 Ftg: 5,5 WSat: 0,1
Mood: 0,1
Thorp et al. (2016) MT:IM
CV: 7,8
Tobin et al. (2016) MT:IM MT:SR
Sit: 1,1 (↓1.7 h) Cog: 0,2
Std: 1,1 (1.6 h)
AMvt: 0,2 (↑0 h)
Tran: 0,1

Behav - Behaviors; Work Perf – Work Performance; Discom – Discomfort; Physiol – Physiological; Psychol – Psychological; Post – Posture.
SR - Self Report; IM - Instrument Measure; RO – Rater Observation; FU – Follow-up; Std – Stand; Tran - Transitions between sit and stand; AMvt – Active Movement;
Prod – Productivity; Ab/P – Absenteeism/Presenteeism; CV – Cardiovascular, End – Endocrine-related; Ftg – Fatigue; Cog – Cognitive; Ob – Obesity; Edm – Edema;
SC/SE – Self-Confidence/Self-Efficacy; WSat – Work Satisfaction.
Δ - The numbers in parentheses are the number of hours difference between the sit desk and sit/stand desk for sitting, standing and active movement. The arrow
indicates if the time for the sit/stand desk was lower or higher than the sit desk, a (−) indicates the number could not be calculated.
a
favored sit position.
b
mixed results.
c
Follow-up information – the first number is the number of significant results, the second number is the number of outcome measures addressed in this category.
d
Outcome not measured at Follow-up 1 only Follow-up 2.

Fig. 2. Total number of studies that included outcome measures by domain.

presenteeism (Fig. 6). Work performance measures were primarily (Fig. 2). Work satisfaction, self-confidence/self-efficacy, and mood were
collected through self-report (Table 2). Only 7% of results reported in included in four, two, and eight studies, respectively (Fig. 6, Table 2).
the work performance domain were significant (Fig. 3). Seventy-seven Psychological outcomes were always measured using self-report
percent of studies had no significant results. There were no significant (Table 2). Overall, 31% of results for the psychological domain were
results for the absenteeism/presenteeism category, while only 8% of the significant (Fig. 3). Significant improvements in work satisfaction were
productivity category outcomes were significant (Fig. 6). Interestingly, noted in 14% of these outcomes. Self-confidence/efficacy had the
two studies, out of 21, reported that productivity results favored sitting greatest number of significant results with 67% of outcomes significant
not standing (Thorp et al., 2014a; Commissaris et al., 2014) (Table 2). in the two studies that included this measure. Mood was only sig-
nificant in 31% of the studies (Fig. 6). The majority of studies that in-
3.6. Psychological cluded psychological outcomes found no significance in any of their
outcome measures.
Eleven studies reported results related to the psychological domain

44
A.J. Chambers, et al. Applied Ergonomics 78 (2019) 37–53

Fig. 3. Percentage of significant outcome measures across all studies by domain for follow-up one.

3.7. Discomfort total of 66 outcome measures were found (Table 2). Unfortunately, as
with many of the other outcomes, what was examined varied, ranging
A total of 22 studies had results related to the discomfort domain from full body positioning to body angles of the neck, back, shoulders,
(Fig. 2). Discomfort was always measured using self-report, either by wrist or more. Of these, only 18% of results for the posture domain were
individual body parts or by body sections (Table 2). Overall 43% of significant (Fig. 2). The majority of studies that included posture out-
results were significant (Fig. 3). No significant results were found in comes showed limited significance in their outcome measures.
21% of studies while 13% had all outcomes significant. In five studies,
the results favored sitting and in three the results were mixed where
some outcomes favored sitting and some standing (Table 2). 4. Discussion

Overall, this scoping review synthesized sit-stand workstation in-

3.8. Posture tervention studies, and related implementation strategies, to assess
their effect on six primary outcome domains: behavior; physiological;
Only four studies reported results related to the posture domain work performance; discomfort; psychological; and posture. Fifty-three
which were measures with both observed and objective measures studies were included that recorded data from work, laboratory and
(Fig. 2, Table 2). Even though only four studies reported on posture, a field settings. Studies tended to be small in size and short in duration

Fig. 4. Percentage of significant outcome measures by category in the behavior domain for follow-up one. Total number of studies that included outcome measures in
each category is provided in square brackets.

45
A.J. Chambers, et al. Applied Ergonomics 78 (2019) 37–53

Fig. 5. Percentage of significant outcome measures by category in the physiological domain for follow-up one. Total number of studies that included outcome
measures in each category is provided in square brackets. CV -Cardiovascular.

Fig. 6. Percentage of significant outcome measures by cate-

gory in the work performance and psychological domains for
follow-up one. Total number of studies that included out-
come measures in each category is provided in square
brackets. Ab/P – Absenteeism/Presenteeism and Productivity
are categories in the work performance domain. WSat – Work
Satisfaction, SE/SC – Self-Confidence/Self-Efficacy and Mood
are categories in the psychological domain.

while collecting primarily from healthy, normal weight, younger, sedentary behavior. In fact, our review finds that although greater than
working adults. 75% of outcomes in those studies that examined sitting and standing
behaviors improved for the better, only 30% of outcomes measured in
the active movement categories were significant, and only 8 out of the
4.1. Behavior
18 studies that examined active movement had any significant results.
Active movement was typically defined across studies as an increased
SSDs are frequently touted as improvements over sitting desks be-
level of activity compared to sitting or standing, such as stepping or
cause they are thought to reduce sedentary behavior. When compared
walking. The average amount of time people increased active move-
to the regular desk, the average amount of time people decreased their
ment ranged from 0 to 0.6 h with a median of 0.1 h. These results
sitting time when using a SSD ranged from 0.1 to 3.6 h per day and the
suggest that sit and stand time are not accurate surrogate measures for
average amount of time they increased standing time ranged from 0.5
increases in physical activity behavior. It also demonstrates that SSDs
to 3.1 h with median average times for sit and stand of 1.3 and 1.4 h,
are minimally efficient at increasing physical activity.
respectively. Decreased sitting and, the flip side, increased standing, are
viewed as measures of decreased sedentary time while on the job.
Unfortunately, both fall below the MET cutoff for sedentary behavior 4.2. Physiological
(Tremblay et al., 2017). So, although SSDs decrease sitting and improve
standing time, their use does not necessarily represent a decrease in Physiological outcomes were of great interest with 28 studies

46
A.J. Chambers, et al. Applied Ergonomics 78 (2019) 37–53

reporting in this domain. While there is strong evidence that SSDs de- the studies with significant differences in productivity reported that
crease sitting time, sit and stand time are likely not accurate surrogate standing was superior to sitting (Gao et al., 2016a; Hedge and Ray,
measures for health related benefits related to changes in physiological 2004; Kar and Hedge, 2016; Robertson et al., 2013) while two favored
outcomes. Mild evidence exists to support cardiovascular improvements sitting (Thorp et al., 2014a; Commissaris et al., 2014). Additionally,
with SSD use. However, it is difficult to distinguish energy expenditure others have found that standing did not reduce performance and re-
gained from alternating positions while using SSDs and energy ex- sulted in improved mouse function (Sangachin et al., 2016). One
penditure related to standing. Several studies found no significant weakness of the assessment of productivity was the statistical methods
change in energy expenditure when using SSDs (Burns et al., 2017; Carr used to identify “no difference” in productivity levels. Many of the
et al., 2016; Cox et al., 2011). Those that did, found minimal increases studies that examined productivity interpreted a non-significant result
in energy expenditure ranging from 4.2 kcal/h to 10 kcal/h (Thorp as indicating that workstation type did not affect performance. How-
et al., 2016; Beers et al., 2008; Gao et al., 2017; Gibbs et al., 2017b; ever, a non-significant result based on inferential testing cannot be used
Horswill et al., 2017). It is important to note that the greatest hourly to interpret “no difference” between groups. To better assess the effect
change in energy expenditure was seen in an overweight and obese of SSDs on productivity, studies will need to use non-inferiority meth-
population (Thorp et al., 2016). Heart rate was commonly found to be odology (Walker and Nowacki, 2011).
higher, 7.5 bpm to 13.7 bpm, when using a SSDs (Beers et al., 2008; Cox
et al., 2011; Gao et al., 2017; Gibbs et al., 2017b; Horswill et al., 2017). 4.4. Psychological
Changes in blood pressure and VO2 were equivocal as several studies
found no change when using a SSD (Carr et al., 2016; Graves et al., SSDs are perceived as potentially improving mood and work sa-
2015; Healy et al., 2017; MacEwen et al., 2017) while others found an tisfaction though very few studies, 11 in total, included self-reported
improvement (Cox et al., 2011; Gao et al., 2017; Horswill et al., 2017). outcomes related to the psychological domain. Comparing these out-
Thus, SSDs do not consistently improve cardiovascular biomarkers, and comes is challenging since all psychological surveys varied from study
when improvements are documented, they are generally not clinically to study. Additionally, these outcomes were recorded in both lab and
important. work environments with follow-up periods of 30 min to 12 weeks ex-
SSDs also had limited effects on endocrine-related biomarkers. Eight posure. Even with these differences, the majority of studies that in-
studies included endocrine-related outcomes including glucose, trigly- cluded psychological outcomes found no significant effect of using a
cerides, insulin, or cholesterol. No significant change in endocrine-re- SSDs. Ognibene et al. (2016) was the only study to note a significant
lated outcomes was found in the majority of studies in healthy adults changes in work satisfaction. It should be noted though, that this was an
(Thorp et al., 2014b; Gao et al., 2017; Gilson et al., 2017; Graves et al., interaction effect of pain on ability to concentration in a population
2015; Healy et al., 2017; MacEwen et al., 2017). Interestingly, a ben- with low back pain. Mood was largely uninfluenced by SSDs with the
eficial change in glucose was noted in an obese population (Thorp et al., exception of two studies in which improvements in mood states in-
2014b). In the four studies that examined the effects of SSDs on obesity, cluding happiness, stress, sluggishness, alertness, and energy were re-
no significant changes were found in healthy adults (Alkhajah et al., ported (Pronk et al., 2012; Paul, 1995b).
2012; Carr et al., 2016), overweight adults (Healy et al., 2017), or
adults with abdominal obesity (MacEwen et al., 2017). These results 4.5. Discomfort
support the limited clinical importance of the cardiovascular results
found above for reducing weight. Standing significantly improved musculoskeletal discomfort in
SSDs had few effects on other physiological variables. Of the seven about half the measures. The body part that most often showed re-
studies that included cognitive outcomes, only one study had a sig- ductions in pain on standing was the low back, with eight (Thorp et al.,
nificant change in function, as measured by the Stroop word color test, 2014a; Nerhood and Thompson, 1994; Davis and Kotowski, 2014; Foley
and this study, admittedly, may have been biased by a strong learning et al., 2016; Gao et al., 2016b; Hedge and Ray, 2004; Ognibene et al.,
effect (Horswill et al., 2017). SSDs are often cited as a way to improve 2016; Robertson et al., 2013) out of the 17 studies that directly mea-
energy and reduce fatigue at work. Of the nine studies that measured sured low back pain (Pronk et al., 2012; Thorp et al., 2014a; Nerhood
fatigue four found no change in energy level or sleepiness when using a and Thompson, 1994; Alkhajah et al., 2012; Davis and Kotowski, 2014;
SSD (Chau et al., 2016; Dutta et al., 2014; Ebara et al., 2008; Gibbs Ebara et al., 2008; Foley et al., 2016; Gao et al., 2016a, 2016b; Graves
et al., 2017b). While two studies (Pronk et al., 2012; Thorp et al., et al., 2015; Hedge and Ray, 2004; Karakolis et al., 2016; Le and
2014a), noted less fatigue during standing, two others reported in- Marras, 2016; Lin et al., 2017; Neuhaus et al., 2014b; Ognibene et al.,
creased self-reported tiredness and increased potential for leg muscle 2016; Robertson et al., 2013) reporting significant reductions. No other
fatigue during standing (Beers et al., 2008; Gao et al., 2017). One body part had as consistent reductions in discomfort among multiple
possible explanation for these conflicting results is the differences in studies. Several studies reported increases in pain with standing. These
dosage across studies. Prolonged static postures, sitting or standing, can were most often in the lower extremities (Kar and Hedge, 2016; Le and
cause low-level muscle fatigue. Frequent alternating between postures, Marras, 2016) and in the case of Lin (Lin et al., 2017), the users re-
which occurs when using SSDs, may help reduce muscle fatigue and in ported that they had double the discomfort in their low back after
turn decrease self-reported fatigue (Thorp et al., 2014a). Additional standing than while sitting. Ognibene et al. looked specifically on the
research is necessary to determine if effective sit-stand dosage can re- effects of using SSDs on people with chronic low back pain (Ognibene
duce worker fatigue. One study included a measure of edema and found et al., 2016). They reported that after 12 weeks those in the interven-
that increased foot swelling was present during standing (Paul, 1995b). tion were reporting significantly less back pain and significantly less
This result is not a new finding, as lower extremity swelling has been impact of pain on daily activities. These results suggest that SSDs has
commonly noted during occupational standing (Cham and Redfern, some potential as an effective way to address low back pain. However,
2001). the best postures and dosage to promote reduced discomfort are un-
known.
4.3. Work performance
4.6. Posture
One argument against SSDs is that standing may reduce the pro-
ductivity of workers. Twenty-one of the studies included productivity Standing as an intervention to improve pain requires proper
outcomes, and only 8% found significant differences between sitting standing posture to prevent additional discomfort including, but not
and standing or sitting and combination of sitting and standing. Four of limited to, neutral postures of the neck, back, and upper extremities.

47
A.J. Chambers, et al. Applied Ergonomics 78 (2019) 37–53

Few studies looked at posture, a total of four. Unfortunately, as with intervention consists of a new piece of equipment, tool or workstation
many of the other outcomes, what was examined varied, resulting in to support and encourage new working postures along with providing
limited information available to inform decisions about SSD postures. training on how to use the new device and why it is important to adopt
One study reported that working in a sit-stand paradigm was found to these new working postures to reduce discomfort and improve perfor-
be associated with reduced lumbar flexion during sitting compared to mance. Other traditional ergonomic interventions that have been
sitting alone. This would imply that SSDs could potentially reduce pain shown to improve comfort or reduce injury risks, such as anti-fatigue
or injury risk in the low back (Karakolis et al., 2016). A recent study mats, foot rests, and wrist supports, were absent from the current ex-
also found increased upper body posture variability during SSD use amination of SSDs.
compared to sitting (Barbieri et al., 2019). It is possible that alternating This review extracted information to examine the impact of SSDs on
position improves musculoskeletal health and posture. However, ad- total worker health. While current research suggests that only some
ditional research is necessary to further investigate these limited results domains benefit from SSD use, it is important in future research to
as it was also found that extended standing time with a SSD eventually continue to consider all aspects of worker health. Limited research has
decreased posture variability (Barbieri et al., 2019). Interestingly, not been done in the posture and psychological domains, and few results
all significant results favored SSDs or standing. More neutral wrist an- were found. We must include these measures in future studies as they
gles, a crucial factor in preventing the development of occupational are associated with other aspects of worker health such as discomfort or
injuries like carpal tunnel syndrome, were reported during sitting workplace behaviors. A better understanding in these areas may help
compared to standing with SSDs (Hedge and Ray, 2004; Lin et al., explain SSD usage and behaviors. While many aspects of worker health
2017). Therefore, SSDs exhibit a potentially beneficial response, in re- have been explored in relation to SSDs, the use of SSDs in special po-
duced lumbar flexion, and negative responses, in excess wrist extension, pulations, such as middle-aged, obese, pain, or gender-specific popu-
that could potentially prevent or cause musculoskeletal pain or injury. lations is largely unexplored. SSDs likely effect these populations dif-
Many of the studies that did examine posture examined body parts in ferently which should be considered in future work as well.
isolation, back posture or wrist postures, rather than overall postures Standing may be a potentially sedentary behavior, so it is necessary
during SSD use. The lack of research into posture, only 4 studies, and to evaluate the effect that adding standing to a computer task has on the
conflicting pain, discomfort, or injury potential, indicates that addi- actual health outcomes of interest. In general, prolonged static postures,
tional research is needed to develop better guidelines on proper SSD including both sitting and standing, are associated with poor health and
ergonomics including desk setup and posture. increased injury risk and discomfort. SSDs benefits may lie in their
ability to encourage alteration of postures and reductions in time spent
4.7. Limitations in any one static posture. However, the optimal sit to stand dosage is
unknown and likely varies across populations and occupations. Half the
This review had several limitations that are worth noting. studies in this scoping review had the amount of standing to sitting time
Demographics of the populations in the reviewed studies were difficult controlled by the researcher, and, except for those completed by Li (Li
to summarize. Unfortunately the majority of studies reviewed did not et al., 2017) and Paul (1995a) these were all laboratory based studies.
accurately report demographics of their populations or only included Thus, dosing was almost always self-selected at work. In general, stu-
recruitment parameters. Follow-up one time frames ranged from one dies that reported amounts of time in sitting and standing when using
day to one year. The strength of previously reported outcome measures SSDs found a wide range of time: from 1 h of standing per day (Neuhaus
likely critically depends on the parameters of a SSDs intervention, in- et al., 2014b) to 5 h (MacEwen et al., 2017). Dosing in research con-
cluding length of use, follow-up time frame, dosage, compliance, and trolled study was also very broad and ranged from 5 min (Burns et al.,
training. Follow-up and intervention characteristics varied greatly 2017) to 4 h (Buckley et al., 2014). Li et al. were the only group to test 3
across the literature, which should be considered and corrected in fu- different sit/stand protocols, with a 2:1, 1:1 and a 1:2 sit:stand hourly
ture work. Studies examined the impact of SSDs using different proto- protocol tested over a 4-week period (Li et al., 2017). Recently, Bao and
cols which resulted in comparisons of sitting to standing or sitting to Lin tested different sit stand schedules (Bao and Lin, 2018). In both
alternating sitting and standing. Few studies were concerned about this studies workers reported that they preferred different schedules or an
difference yet it may impact our understanding of the potential benefits unstructured approach to determine sitting and standing durations (Li
of SSDs. The manner in which measures were collected also effects their et al., 2017; Bao and Lin, 2018). Additional research is needed to assess
strength and the quality of the studies reviewed. It is possible that the benefits of alternating posture using SSDs, what those postures
better objective measures in future work may enhance the strength of should be, and incorporating the proper knowledgeable dosage for both
outcomes in different domains. Several domains were included in this sitting and standing. This research would benefit from longer follow-
review in an attempt to represent total work health. Other aspects of ups with better tracking of SSD usage.
lifestyle behaviors that were outside the scope of this review, such as
leisure activity levels, may also be associated with occupational sitting
behaviors and SSD use. 5. Conclusion

4.8. Future directions The recent interest in SSDs has resulted in a surge in available lit-
erature on their potential benefits in the workplace. Obvious behavior
The majority of studies in our scoping review used a small, healthy changes were found across numerous publications, mostly of short in-
population with a short follow-up time period. While behaviors were tervention duration and small healthy populations, with decreased time
changed in these short term follow-ups, compliance and health benefits spent sitting and increased time spent standing. Though additional re-
over a longer time period are largely unknown. Few studies discussed search is necessary to determine the appropriate dosage of sitting and
the importance of training as part of the introduction of the new standing. Unfortunately, modest cardiometabolic health benefits were
workstations, although research indicates that training is a key aspect noted when using SSDs. Generally, SSDs did not reduce work perfor-
of changing behaviors (Verbeek, 1991). Compliance of use is usually mance or improve psychological health. SSDs were most effective at
minimal if training and reminders are not provided to encourage user's reducing discomfort. Additional research is necessary to determine the
to knowledgeably exert control over their workstations (Wilks et al., effect of SSDs on posture as this measure was not commonly included,
2006; Robertson et al., 2013). Identifying best training methods to in- and yet is important in long term worker health. Further research is
corporate SSDs may be a key aspect for improving health benefits, as it needed to examine long-term effects, and to determine clinically ap-
can address dosage and correct positioning. Typically, an ergonomic propriate dosage and workstation setup.

48
A.J. Chambers, et al. Applied Ergonomics 78 (2019) 37–53

Funding Acknowledgment

None. This work was supported by the National Institute for Occupational
Safety and Health [grant number K01-OH010759].

Appendix

Appendix Table 1
Criteria for credibility review for experimental and non-experimental studies.

Experimental Criteria

E1 Random Groups were randomly allocated to treatment or no treatment

Allocation
E2 Allocation Allocation to group was concealed
Concealment
E3 Groups A comparison of group characteristics was made, and they were similar, or difference were controlled in analysis OR The reviewer compared the two groups and
Similar at they seem similar in basic characteristics
Baseline
E4 Outcome Those completing assessment of outcome measures were blinded to group status
Assessor
Blinded
E5 Reliable Authors provide evidence of reliability OR the measures are instruments well known to be reliable (standard measures in the field)
Measures
E6 Competent Assessors were trained and checked for accuracy periodically OR measures were instrumented
Assessors
E7 Fidelity Authors indicate that the intervention/control were checked periodically to see if protocols for treatment were being followed appropriately
E8 Appropriate Control group matched hypothesis and ruled out plausible alternative explanations appropriate
Control
Group
E9 Attrition Attrition rate was less than 20% for each group
< 20%
E10 Intention to Authors indicate that people were analyzed in the groups to which they were assigned.
Treat
Analysis
Non-experimental Criteria
NE1 Response Rate 75% or greater of those approached took part in the study
> 75%
NE2 More than 1 Outcomes were tracked in more than 1 group to ensure a comparator
Group
NE3 Groups A comparison of group characteristics was made, and they were similar, or difference were controlled in analysis OR The reviewer compared the two
Similar at groups and they seem similar in basic characteristics, OR If the groups were not similar the authors controlled for the difference
Baseline
NE4 Variables Specification of variables for the study were provided
Operationally
Defined
NE5 Outcome Those completing assessment of outcome measures were blinded to group status
Assessors
Blinded
NE6 Reliable Authors provide evidence of reliability OR the measures are instruments well known to be reliable (standard measures in the field)
Measures
NE7 Competent Assessors were trained and checked for accuracy periodically OR measures were instrumented
Assessors
NE8 Standardized Data was collected using standardized forms and protocols
Data
Collection
NE9 Intervention Results clearly followed from the intervention (sufficient measurment time, baseline taken before intervention initiated)
Preceded
Outcome
NE10 Attrition Attrition rate was less than 20% for each group
< 20%
NE11 Confounding Authors discussed possible confounding and how they accounted for it
Accounted for
NE12 Intention to Authors indicate that people were analyzed in the groups that they were assigned to
Treat Analysis

Criteria was scored: 0 = No/not reported; 1 = Yes.

Appendix Table 2
Individual experimental study scores on each credibility item.

Study E1 E2 E3 E4 E5 E6 E7 E8 E9 E10 Total Credibility %

Alkhajah et al. (2012) 0 0 0 1 0 0 0 1 1 0 3 30%

Bantoft et al. (2016) 1 0 1 0 1 0 1 1 1 1 7 70%
Beers et al. (2008) 0 0 1 0 1 1 1 1 1 1 7 70%
(continued on next page)

49
A.J. Chambers, et al. Applied Ergonomics 78 (2019) 37–53

Appendix Table 2 (continued)

Study E1 E2 E3 E4 E5 E6 E7 E8 E9 E10 Total Credibility %

Britten et al. (2016) 1 0 1 0 1 0 1 1 1 1 7 70%

Buckley et al. (2014) 0 0 1 0 1 1 1 0 1 1 6 60%
Burns et al. (2017) 1 0 1 0 1 1 1 1 1 1 8 80%
Chau (2014) 1 0 1 0 1 1 0 1 1 1 7 70%
Chau et al. (2016) 0 0 1 1 1 1 0 1 0 1 6 60%
Coenen et al. (2017) 1 0 1 0 1 1 0 1 1 1 7 70%
Commissaris et al. (2014) 1 0 1 0 1 1 1 1 1 1 8 80%
Cox et al. (2011) 1 0 1 0 1 1 1 1 1 1 8 80%
Davis (2014) 1 0 1 0 0 0 1 1 1 1 6 60%
Donath et al. (2015) 1 0 1 0 1 0 0 1 1 0 5 50%
Dutta et al. (2014) 1 0 1 0 1 0 0 1 1 0 5 50%
Ebara et al. (2008) 0 0 0 0 1 0 1 1 1 1 5 50%
Finch et al. (2017) 1 1 1 0 1 0 0 1 1 1 7 70%
Foley et al. (2016) 0 0 0 0 1 0 1 1 1 1 5 50%
Gao (2016b) 0 0 0 0 1 1 1 1 1 1 6 60%
Gao et al. (2017) 1 0 1 0 1 0 0 1 1 0 5 50%
Gibbs (2017a) 1 0 1 1 1 1 1 1 1 1 9 90%
Gibbs (2017b) 1 0 1 1 1 1 1 1 1 1 9 90%
Gilson et al. (2017) 1 0 1 0 1 1 1 1 0 0 6 60%
Graves et al. (2015) 1 0 1 0 1 0 0 1 1 1 6 60%
Hadgraft et al. (2017) 1 0 1 0 1 0 1 1 0 1 6 60%
Healy et al. (2016) 1 0 1 0 1 0 1 1 0 1 6 60%
Healy et al. (2017) 1 0 1 0 1 1 1 1 1 1 8 80%
Hedge (2004) 1 0 0 0 1 1 1 1 0 1 6 60%
Hedge et al. (2005) 1 0 0 0 1 0 0 0 0 0 2 20%
Horswill et al. (2017) 0 0 1 0 1 1 0 1 1 1 6 60%
Husemann et al. (2009) 1 0 1 0 1 0 1 1 1 1 7 70%
Kar (2016) 1 0 0 0 1 0 0 1 1 0 4 40%
Karakolis et al. (2016) 0 0 0 0 1 0 1 0 1 1 4 40%
Le (2016) 0 0 0 0 1 0 1 1 1 1 5 50%
Li et al. (2017) 1 0 1 0 1 0 0 1 1 1 6 60%
MacEwen et al. (2017) 1 0 1 1 1 0 0 1 1 0 6 60%
Mansoubi et al. (2016) 0 0 0 0 1 0 0 0 1 1 3 30%
Neuhaus (2014) 1 1 1 0 1 1 1 1 0 1 8 80%
Ognibene et al. (2016) 1 0 1 0 1 1 1 1 1 1 8 80%
Paul (1995a) 0 0 1 0 1 0 0 1 1 1 5 50%
Paul (1995b) 0 0 1 0 0 0 0 0 0 1 2 20%
Pronk et al. (2012) 0 0 0 0 1 1 1 1 0 1 5 50%
Robertson et al. (2013) 1 0 1 0 1 1 1 1 1 1 8 80%
Roemmich (2016) 1 0 0 0 1 0 0 1 0 1 4 40%
Russell et al. (2016) 1 0 1 0 1 0 1 1 1 1 7 70%
Thorp (2014a) 1 0 1 0 1 1 1 1 1 1 8 80%
Thorp (2014b) 1 0 1 0 1 1 1 1 1 1 8 80%
Thorp et al. (2016) 1 0 0 0 1 1 1 1 1 1 7 70%
Tobin (2016) 1 0 1 0 1 0 1 1 1 1 7 70%

E1 - Random Allocation; E2 - Allocation Concealment; E3 - Groups Similar at Baseline; E4 - Outcome Assessor Blinded; E5 - Reliable Measures; E6 - Competent
Assessors; E7- Fidelity; E8 - Appropriate Control Group; E9 - Attrition < 20%; E10 - Intention to Treat Analysis.

Appendix Table 3
Individual non-experimental study scores on each credibility item.

Study NE1 NE2 NE3 NE4 NE5 NE6 NE7 NE8 NE9 NE10 NE11 NE12 Total Credibility %

Carr et al. (2016) 0 1 0 1 0 1 1 1 0 1 0 1 7 58%

Gao (2016a) 0 1 1 1 0 1 0 1 1 0 1 1 8 67%
Lin et al. (2017) 0 0 1 1 0 1 1 1 1 1 0 1 8 67%
Nerhood (1994) 0 0 0 0 0 0 0 0 0 0 0 0 0 0%
Straker et al. (2013) 1 1 1 1 0 1 1 1 1 1 1 1 11 92%

NE1 - Response Rate > 75%; NE2 - More than 1 Group; NE3 - Groups Similar at Baseline; NE4 - Variables Operationally Defined; NE5 - Outcome Assessors Blinded;
NE6 - Reliable Measures; NE7 - Competent Assessors; NE8 - Standardized Data Collection; NE9 - Intervention Preceded Outcome.
NE10 - Attrition < 20%; NE11 - Confounding Accounted for; NE12 - Intention to Treat Analysis.

50
A.J. Chambers, et al. Applied Ergonomics 78 (2019) 37–53

Appendix Table 4
Timing, number of follow-up visits, and results for studies that had multiple follow-up points

Author (year) Follow-up Behavior Work Performance Discomfort Physiological Psychological Posture
Time Frame
(s)

Alkhajah et al. FU 1: 1 wk; Sit - FU 1: 1,1*; FU 2: 1,1 Ab/P - FU 1: 0,1; FU 2: FU 2: 0,10X End- FU 2: 1,5X
(2012) FU 2: Std - FU 1: 1,1; FU 2: 1,1 0,1 Ftg - FU 2: 0,1X
12 wks Tran - FU 1: 1,1; FU 2: 1,1 Prod - FU 1: 0,1; FU 2: Ob - FU 2: 0,5X
AMvt - FU 1: 1,1; FU 2: 0,1 0,1
Chau et al. FU 1: 1 wk; Sit - FU 1: 0,2; FU 2: 0,2; FU 3: Ab/P - FU 1: 0,2; FU 2: Ftg - FU 1: 0,1; FU 2: 0,1; FU Mood - FU 1: 0,1; FU 2: 0,1; FU
(2016) FU 2: 4 wks; 0,2 0,2; 3: 0,1 3: 0,1
FU 3: Std - FU 1: 2,2 FU 3: 0,2
19 wks AMvt - FU 1:0,4; FU 2: 0,4; FU Prod - FU 1: 0,4; FU 2:
3: 0,4 0,4;
FU 3: 0,4
Graves et al. FU 1: 4 wks; Sit - FU 1: 1,1 FU 2: 0,3X CV - FU 2: 0,4X
(2015) FU 2: 8 wks FU 2: 1,1 End - FU 2: 1,3X
Std - FU 1: 1,1
FU 2: 1,1
AMvt - FU 1: 0,1
FU 2: 0,1
Hadgraft et al. FU 1: SC/SE - FU 1: 2,2; FU 2: 2,2
(2017) 12 wks;
FU 2:
52 wks
Healy et al. FU 1: Sit - FU 1: 4,4
(2016) 12 wks; FU 2: 4,4
FU 2: Std - FU 1: 2,2
52 wks FU 2: 2,2
AMvt - FU 1: 0,2
FU 2: 0,2
Healy et al. FU 1: CV – FU 1: 0,2; FU 2: 0,2
(2017) 12 wks; End - FU 1: 0,9; FU 2: 2,9
FU 2: Ob - FU 1: 0,4; FU 2: 0,4
52 wks
Mansoubi et al., FU 1: 1 wk; Sit - FU 1: 4,4; FU 2: 4,4; FU 3:
2016 FU 2: 6 wks; 3,4
FU 3: Std - FU 1: 2,2
12 wks AMvt - FU 1: 6,6
Pronk et al. FU 1: 4 wks; Sit - FU 1: 1,1 FU 1: 1,2 Ftg - FU 1: 1,1 Mood - FU 1: 5,6;
(2012) FU 2: 6 wks AMvt - FU 1: 0,1 FU 2: 2,6
SC/SE - FU 1: 0,1
Roemmich FU 1: 1 wk;
(2016) FU 2:
24 wks;
FU 3:
52 wks

SR - Self Report; IM - Instrument Measure; RO – Rater Observation; FU – Follow-up; Std – Stand; Tran - Transitions between sit and stand; AMvt – Active Movement;
Prod – Productivity; Ab/P – Absenteeism/Presenteeism; CV – Cardiovascular, End – Endocrine-related; Ftg – Fatigue; Cog – Cognitive; Ob – Obesity; Edm – Edema;
SC/SE – Self-Confidence/Self-Efficacy; WSat – Work Satisfaction.
$
favored sit position.
&
mixed results.
*Follow-up information – the first number is the number of significant results, the second number is the number of outcome measures addressed in this category.
X
Outcome not measured at Follow-up1 only Follow-up 2.

References Benatti, F.B., Ried-Larsen, M., 2015. The effects of breaking up prolonged sitting time: a
review of experimental studies. Med. Sci. Sports Exerc. 47 (10), 2053–2061. https://
doi.org/10.1249/MSS.0000000000000654.
Agarwal, S., Steinmaus, C., Harris-Adamson, C., 2017. Sit-stand workstations and impact Britten, L., Shire, K., Coats, R.O., Astill, S.L., 2016. The effect of standing desks on manual
on low back discomfort: a systematic review and meta-analysis. Ergonomics 1–36. control in children and young adults. Gait Posture 48, 42–46. https://doi.org/10.
https://doi.org/10.1080/00140139.2017.1402960. 1016/j.gaitpost.2016.04.027.
Alkhajah, T.A., Reeves, M.M., Eakin, E.G., et al., 2012. Sit-stand workstations: a pilot Buckley, J.P., Mellor, D.D., Morris, M., Joseph, F., 2014. Standing-based office work
intervention to reduce office sitting time. Am. J. Prev. Med. 43 (3), 298–303. https:// shows encouraging signs of attenuating post-prandial glycaemic excursion. Occup.
doi.org/10.1016/j.amepre.2012.05.027. Environ. Med. 71 (2), 109–111. https://doi.org/10.1136/oemed-2013-101823.
Bantoft, C., Summers, M.J., Tranent, P.J., et al., 2016. Effect of standing or walking at a Burns, J., Forde, C., Dockrell, S., 2017. Energy expenditure of standing compared to
workstation on cognitive function: a randomized counterbalanced trial. Hum. Factors sitting while conducting office tasks. Hum. Factors 59 (7), 1078–1087. https://doi.
58 (1), 140–149. https://doi.org/10.1177/0018720815605446. org/10.1177/0018720817719167.
Bao, S., Lin, J.H., 2018. An investigation into four different sit-stand workstation use Carr, L.J., Swift, M., Ferrer, A., Benzo, R., 2016. Cross-sectional examination of long-term
schedules. Ergonomics 61 (2), 243–254. https://doi.org/10.1080/00140139.2017. access to sit-stand desks in a professional office setting. Am. J. Prev. Med. 50 (1),
1353139. 96–100. https://doi.org/10.1016/j.amepre.2015.07.013.
Barbieri, D.F., Srinivasan, D., Mathiassen, S.E., Oliveira, A.B., 2019. Variation in upper Cham, R., Redfern, M.S., 2001. Effect of flooring on standing comfort and fatigue. Hum.
extremity, neck and trunk postures when performing computer work at a sit-stand Factors 43 (3), 381–391. https://doi.org/10.1518/001872001775898205.
station. Appl. Ergon. 75, 120–128. Chau, J.Y., der Ploeg, H.P., van Uffelen, J.G., et al., 2010. Are workplace interventions to
Beers, E.A., Roemmich, J.N., Epstein, L.H., Horvath, P.J., 2008. Increasing passive energy reduce sitting effective? A systematic review. Prev. Med. 51 (5), 352–356. https://
expenditure during clerical work. Eur. J. Appl. Physiol. 103 (3), 353–360. https:// doi.org/10.1016/j.ypmed.2010.08.012.
doi.org/10.1007/s00421-008-0713-y. Chau, J.Y., Daley, M., Dunn, S., et al., 2014. The effectiveness of sit-stand workstations for

51
A.J. Chambers, et al. Applied Ergonomics 78 (2019) 37–53

changing office workers' sitting time: results from the Stand@Work randomized Horswill, C., Scott, H.M., Voorhees, D.M., 2017. Effect of a novel workstation device on
controlled trial pilot. Int. J. Behav. Nutr. Phys. Act. 11, 127. https://doi.org/10. promoting non-exercise activity thermogenesis (NEAT). Work 58, 447–454. https://
1186/s12966-014-0127-7. doi.org/10.3233/WOR-172640.
Chau, J.Y., Sukala, W., Fedel, K., et al., 2016. More standing and just as productive: Husemann, B., Von Mach, C.Y., Borsotto, D., et al., 2009. Comparisons of musculoskeletal
effects of a sit-stand desk intervention on call center workers' sitting, standing, and complaints and data entry between a sitting and a sit-stand workstation paradigm.
productivity at work in the Opt to Stand pilot study. Prev Med Rep 3, 68–74. https:// Hum. Factors 51 (3), 310–320. https://doi.org/10.1177/0018720809338173.
doi.org/10.1016/j.pmedr.2015.12.003. Jewell, D., 2018. Guide to Evidence-Based Physical Therapy. Jones & Bartlett Learning,
Chu, A.H., Ng, S.H., Tan, C.S., et al., 2016. A systematic review and meta-analysis of Burlington, MA.
workplace intervention strategies to reduce sedentary time in white-collar workers. Kar, G., Hedge, A., 2016. Effects of sitting and standing work postures on short-term
Obes. Rev. 17 (5), 467–481. https://doi.org/10.1111/obr.12388. typing performance and discomfort. Proc. Hum. Factors Ergon. Soc. Annu. Meet. 60
Coenen, P., Healy, G.N., Winkler, E.A.H., et al., 2017. Pre-existing low-back symptoms (1), 460–464. https://doi.org/10.1177/1541931213601104.
impact adversely on sitting time reduction in office workers. Int. Arch. Occup. Karakolis, T., Callaghan, J.P., 2014. The impact of sit-stand office workstations on worker
Environ. Health 90 (7), 609–618. https://doi.org/10.1007/s00420-017-1223-1. discomfort and productivity: a review. Appl. Ergon. 45 (3), 799–806. https://doi.org/
Commissaris, D.A., Konemann, R., Hiemstra-van Mastrigt, S., et al., 2014. Effects of a 10.1016/j.apergo.2013.10.001.
standing and three dynamic workstations on computer task performance and cog- Karakolis, T., Barrett, J., Callaghan, J.P., 2016. A comparison of trunk biomechanics,
nitive function tests. Appl. Ergon. 45 (6), 1570–1578. https://doi.org/10.1016/j. musculoskeletal discomfort and productivity during simulated sit-stand office work.
apergo.2014.05.003. Ergonomics 59 (10), 1275–1287. https://doi.org/10.1080/00140139.2016.1146343.
Commissaris, D., Huysmans, M.A., Mathiassen, S.E., et al., 2016. Interventions to reduce Karol, S., Robertson, M.M., 2015. Implications of sit-stand and active workstations to
sedentary behavior and increase physical activity during productive work: a sys- counteract the adverse effects of sedentary work: a comprehensive review. Work 52
tematic review. Scand. J. Work. Environ. Health 42 (3), 181–191. https://doi.org/10. (2), 255–267. https://doi.org/10.1111/obr.12388.
5271/sjweh.3544. Le, P., Marras, W.S., 2016. Evaluating the low back biomechanics of three different office
Cox, R.H., Guth, J., Siekemeyer, L., et al., 2011. Metabolic cost and speech quality while workstations: seated, standing, and perching. Appl. Ergon. 56, 170–178. https://doi.
using an active workstation. J. Phys. Activ. Health 8 (3), 332–339. https://doi.org/ org/10.1016/j.apergo.2016.04.001.
10.1123/jpah.8.3.332. Li, I., Mackey, M.G., Foley, B., et al., 2017. Reducing office workers' sitting time at work
Davis, K.G., Kotowski, S.E., 2014. Postural variability: an effective way to reduce mus- using sit-stand protocols: results from a pilot randomized controlled trial. J. Occup.
culoskeletal discomfort in office work. Hum. Factors 56 (7), 1249–1261. https://doi. Environ. Med. 59 (6), 543–549. https://doi.org/10.1097/JOM.0000000000001018.
org/10.1177/0018720814528003. Lin, M.Y., Barbir, A., Dennerlein, J.T., 2017. Evaluating biomechanics of user-selected
Donath, L., Faude, O., Schefer, Y., Roth, R., Zahner, L., 2015. Repetitive daily point of sitting and standing computer workstation. Appl. Ergon. 65, 382–388. https://doi.
choice prompts and occupational sit-stand transfers, concentration and neuromus- org/10.1016/j.apergo.2017.04.006.
cular performance in office workers: an RCT. Int. J. Environ. Res. Public Health 12 MacEwen, B.T., MacDonald, D.J., Burr, J.F., 2015. A systematic review of standing and
(4), 4340–4353. https://doi.org/10.3390/ijerph120404340. treadmill desks in the workplace. Prev. Med. 70, 50–58. https://doi.org/10.1016/j.
Dunstan, D.W., Wiesner, G., Eakin, E.G., et al., 2013. Reducing office workers' sitting ypmed.2014.11.011.
time: rationale and study design for the Stand up Victoria cluster randomized trial. MacEwen, B.T., Saunders, T.J., MacDonald, D.J., Burr, J.F., 2017. Sit-stand desks to re-
BMC Public Health 13, 1057. https://doi.org/10.1186/1471-2458-13-1057. duce workplace sitting time in office workers with abdominal obesity: a randomized
Dutta, N., Koepp, G.A., Stovitz, S.D., et al., 2014. Using sit-stand workstations to decrease controlled trial. J. Phys. Activ. Health 14 (9), 710–715. https://doi.org/10.1123/
sedentary time in office workers: a randomized crossover trial. Int. J. Environ. Res. jpah.2016-0384.
Public Health 11 (7), 6653–6665. https://doi.org/10.3390/ijerph110706653. Mansoubi, M., Pearson, N., Biddle, S.J.H., Clemes, S.A., 2016. Using sit-to-stand work-
Ebara, T., Kubo, T., Inoue, T., et al., 2008. Effects of adjustable sit-stand VDT workstations stations in offices: is there a compensation effect? Med. Sci. Sports Exerc. 48 (4),
on workers' musculoskeletal discomfort, alertness and performance. Ind. Health 46 720–725. https://doi.org/10.1249/MSS.0000000000000802.
(5), 497–505. https://doi.org/10.2486/indhealth.46.497. Nerhood, H.L., Thompson, S.W., 1994. Adjustable sit-stand workstations in the office.
Finch, L.E., Tomiyama, A.J., Ward, A., 2017. Taking a stand: the effects of standing desks Proc. Hum. Factors Ergon. Soc. Annu. Meet. 38 (10), 668–672. https://doi.org/10.
on task performance and engagement. Int. J. Environ. Res. Public Health 14 (8). 1177/154193129403801028.
https://doi.org/10.3390/ijerph14080939. Neuhaus, M., Eakin, E.G., Straker, L., et al., 2014a. Reducing occupational sedentary time:
Foley, B., Engelen, L., Gale, J., et al., 2016. Sedentary behavior and musculoskeletal a systematic review and meta-analysis of evidence on activity-permissive work-
discomfort are reduced when office workers trial an activity-based work environ- stations. Obes. Rev. 15 (10), 822–838. https://doi.org/10.1111/obr.12201.
ment. J. Occup. Environ. Med. 58 (9), 924–931. https://doi.org/10.1097/JOM. Neuhaus, M., Healy, G.N., Dunstan, D.W., et al., 2014b. Workplace sitting and height-
0000000000000828. adjustable workstations: a randomized controlled trial. Am. J. Prev. Med. 46 (1),
Gao, Y., Nevala, N., Cronin, N.J., Finni, T., 2016a. Effects of environmental intervention 30–40. https://doi.org/10.1016/j.amepre.2013.09.009.
on sedentary time, musculoskeletal comfort and work ability in office workers. Eur. J. Ognibene, G.T., Torres, W., von Eyben, R., Horst, K.C., 2016. Impact of a sit-stand
Sport Sci. 16 (6), 747–754. https://doi.org/10.1080/17461391.2015.1106590. workstation on chronic low back pain: results of a randomized trial. J. Occup.
Gao, Y., Cronin, N.J., Pesola, A.J., Finni, T., 2016b. Muscle activity patterns and spinal Environ. Med. 58 (3), 287–293. https://doi.org/10.1097/JOM.0000000000000615.
shrinkage in office workers using a sit-stand workstation versus a sit workstation. Paul, R.D., 1995a. Effects of office layout and sit-stand adjustable furniture: a field study.
Ergonomics 59 (10), 1267–1274. https://doi.org/10.1080/00140139.2016.1139750. Proc. Hum. Factors Ergon. Soc. Annu. Meet. 39 (10), 422–426. https://doi.org/10.
Gao, Y., Silvennoinen, M., Pesola, A.J., et al., 2017. Acute metabolic response, energy 1177/154193129503900704.
expenditure, and emg activity in sitting and standing. Med. Sci. Sports Exerc. 49 (9), Paul, R.D., 1995b. Foot swelling in VDT operators with sitting and sit-stand workstations.
1927–1934. https://doi.org/10.1249/MSS.0000000000001305. Proc. Hum. Factors Ergon. Soc. Annu. Meet. 39 (10), 568–572. https://doi.org/10.
Gibbs, B.B., Kowalsky, R.J., Perdomo, S.J., et al., 2017a. Effect of alternating standing and 1177/154193129503901006.
sitting on blood pressure and pulse wave velocity during a simulated workday in Pronk, N.P., Katz, A.S., Lowry, M., Payfer, J.R., 2012. Reducing occupational sitting time
adults with overweight/obesity. J. Hypertens. 35 (12), 2411–2418. https://doi.org/ and improving worker health: the Take-a-Stand Project, 2011. Prev. Chronic Dis. 9,
10.1097/HJH.0000000000001463. E154 doi: 10.5888.pcd9.110323.
Gibbs, B.B., Kowalsky, R.J., Perdomo, S.J., et al., 2017b. Energy expenditure of deskwork Robertson, M.M., Ciriello, V.M., Garabet, A.M., 2013. Office ergonomics training and a
when sitting, standing or alternating positions. Occup. Med. (Lond.) 67 (2), 121–127. sit-stand workstation: effects on musculoskeletal and visual symptoms and perfor-
https://doi.org/10.1093/occmed/kqw115. mance of office workers. Appl. Ergon. 44 (1), 73–85. https://doi.org/10.1016/j.
Gilson, N.D., Hall, C., Renton, A., et al., 2017. Do sitting, standing, or treadmill desks apergo.2012.05.001.
impact psychobiological indicators of work productivity? J. Phys. Activ. Health 14 Roemmich, J.N., 2016. Height-adjustable desks: energy expenditure, liking, and pre-
(10), 793–796. https://doi.org/10.1123/jpah.2016-0712. ference of sitting and standing. J. Phys. Activ. Health 13 (10), 1094–1099. https://
Graves, L.E.F., Murphy, R.C., Shepherd, S.O., et al., 2015. Evaluation of sit-stand work- doi.org/10.1123/jpah.2015-0397.
stations in an office setting: a randomised controlled trial. BMC Public Health 15, Russell, B.A., Summers, M.J., Tranent, P.J., et al., 2016. A randomised control trial of the
1145. https://doi.org/10.1186/s12889-015-2469-8. cognitive effects of working in a seated as opposed to a standing position in office
Hadgraft, N.T., Willenberg, L., LaMontagne, A.D., et al., 2017. Reducing occupational workers. Ergonomics 59 (6), 737–744. https://doi.org/10.1080/00140139.2015.
sitting: workers' perspectives on participation in a multi-component intervention. Int. 1094579.
J. Behav. Nutr. Phys. Act. 14 (1), 73. https://doi.org/10.1186/s12966-017-0530-y. Sangachin, M., Gustafson, W., Cavuoto, L., 2016. Effect of active workstation use on
Healy, G.N., Eakin, E.G., Owen, N., et al., 2016. A cluster randomized controlled trial to workload, task performance, and postural and physiological responses. IIE Trans
reduce office workers' sitting time: effect on activity outcomes. Med. Sci. Sports Occupat Ergon Hum Factors 4, 67–81. https://doi.org/10.1080/21577323.2016.
Exerc. 48 (9), 1787–1797. https://doi.org/10.1249/MSS.0000000000000972. 1184196.
Healy, G.N., Winkler, E.A.H., Eakin, E.G., et al., 2017. A cluster RCT to reduce workers' Shrestha, N., Ijaz, S., Kukkonen-Harjula, K.T., et al., 2015. Workplace interventions for
sitting time: impact on cardiometabolic biomarkers. Med. Sci. Sports Exerc. 49 (10), reducing sitting at work. Cochrane DB Syst Rev(1). https://doi.org/10.1002/
2032–2039. https://doi.org/10.1249/MSS.0000000000001328. 14651858.CD010912.pub3.
Hedge, A., Ray, E.J., 2004. Effects of an electronic height-adjustable worksurface on Straker, L., Abbott, R.A., Heiden, M., et al., 2013. Sit-stand desks in call centres: asso-
computer worker musculoskeletal discomfort and productivity. Proc. Hum. Factors ciations of use and ergonomics awareness with sedentary behavior. Appl. Ergon. 44
Ergon. Soc. Annu. Meet. 48 (8), 1091–1095. https://doi.org/10.1177/ (4), 517–522. https://doi.org/10.1016/j.apergo.2012.11.001.
154193120404800803. Tew, G.A., Posso, M.C., Arundel, C.E., McDaid, C.M., 2015. Systematic review: height-
Hedge, A., Jagdeo, J., Agarwal, A., Rockey-Harris, K., 2005. Sitting or standing for adjustable workstations to reduce sedentary behaviour in office-based workers.
computer work — does a negative-tilt keyboard tray make a difference? Proc. Hum. Occup Med-Oxford 65 (5), 357–366. https://doi.org/10.1093/occmed/kqv044.
Factors Ergon. Soc. Annu. Meet. 49 (8), 808–811. https://doi.org/10.1177/ Thorp, A.A., Kingwell, B.A., Owen, N., Dunstan, D.W., 2014a. Breaking up workplace
154193120504900804. 2005. sitting time with intermittent standing bouts improves fatigue and musculoskeletal

52
A.J. Chambers, et al. Applied Ergonomics 78 (2019) 37–53

discomfort in overweight/obese office workers. Occup. Environ. Med. 71 (11), Tudor-Locke, C., Schuna, J.M., Frensham, L.J., Proenca, M., 2014. Changing the way we
765–771. https://doi.org/10.1136/oemed-2014-102348. work: elevating energy expenditure with workstation alternatives. Int. J. Obes. 38
Thorp, A.A., Kingwell, B.A., Sethi, P., et al., 2014b. Alternating bouts of sitting and (6), 755–765.
standing attenuate postprandial glucose responses. Med. Sci. Sports Exerc. 46 (11), van der Ploeg, H.P., Chey, T., Korda, R.J., et al., 2012. Sitting time and all-cause mortality
2053–2061. https://doi.org/10.1249/MSS.0000000000000337. risk in 222,497 Australian adults. Arch. Intern. Med. 172 (6), 494–500. https://doi.
Thorp, A.A., Kingwell, B.A., English, C., et al., 2016. Alternating sitting and standing org/10.1001/archinternmed.2011.2174.
increases the workplace energy expenditure of overweight adults. J. Phys. Activ. van Uffelen, J.G., Wong, J., Chau, J.Y., et al., 2010. Occupational sitting and health risks:
Health 13 (1), 24–29. https://doi.org/10.1123/jpah.2014-0420. a systematic review. Am. J. Prev. Med. 39 (4), 379–388. https://doi.org/10.1016/j.
Tobin, R., Leavy, J., Jancey, J., 2016. Uprising: an examination of sit-stand workstations, amepre.2010.05.024.
mental health and work ability in sedentary office workers, in Western Australia. Verbeek, J., 1991. The use of adjustable furniture: evaluation of an instruction pro-
Work 55 (2), 359–371. https://doi.org/10.3233/WOR-162410. gramme for office workers. Appl. Ergon. 22 (3), 179–184.
Torbeyns, T., Bailey, S., Bos, I., Meeusen, R., 2014. Active workstations to fight sedentary Walker, E., Nowacki, A.S., 2011. Understanding equivalence and noninferiority testing. J.
behaviour. Sports Med. 44 (9), 1261–1273. https://doi.org/10.1007/s40279-014- Gen. Intern. Med. 26 (2), 192–196. https://doi.org/10.1007/s11606-010-1513-8.
0202-x. Wilks, S., Mortimer, M., Nylen, P., 2006. The introduction of sit-stand worktables; aspects
Tremblay, M.S., Aubert, S., Barnes, J.D., et al., 2017. Sedentary behavior research net- of attitudes, compliance and satisfaction. Appl. Ergon. 37 (3), 359–365. https://doi.
work (SBRN) - terminology consensus project process and outcome. Int. J. Behav. org/10.1038/ijo.2013.223.
Nutr. Phys. Act. 14 (1), 75. https://doi.org/10.1186/s12966-017-0525-8.

53