Table 2.

PA Results, Main Findings, and Limitations

No.	Author (publishing year)	PA results		Main findings	Study limitations
		Overall	At work
1	Chauetal. (2014)	NR	(a) Net difference in stepping time (NR, p = .433) minutes/day at work from baseline to post-intervention: 13 minutes/day (p = .127) change in intervention and 2 minutes/day (p = .823) change in control group (OSPAQ) (b) Net difference in stepping time (NR, p = .453) minutes/day at work from baseline to post-intervention:11 minutes/day (p = .081) change in intervention and 3 minutes/day (p = .596) change in control (ActivPAL)	Sit-stand desk reduces overall sitting time but has no effect on step time or PA at work	• Short-term follow-up • Convenience sample • No blinding • No objective measure for frequency or duration of activity
2	Dutta et al. (2014)	(a) Increase in Alls from baseline to post-intervention in total activity: 237,729 AU/hour (p > .05) in intervention group versus 236,445 AU/hour (p > .05) in control group (Gruve)	(a) Increase in AUs in work activity from baseline to post-intervention: 229,156 AU/hour (p < .05) in intervention group versus 210,245 AU/hour (p< .05) in control group (Gruve)	Sit-stand desk used over 4 months significantly reduced sitting time and increased standing and light activity during work hours	• No blinding • Short-term follow-up • Small sample
3	Schuna et al. (2014)	(a) 1.6 minutes/hour (95% Cl = [0.5, 2.8]) increase in light PA from baseline to post-intervention between the control group and the intervention group (b)-0.1 minute/hour (95% Cl = [−0.5, 0.4]) decrease in MVPA from baseline to post-intervention between the control group and the intervention group	(a) 2.9 minutes/hour (95% Cl = [0.9, 5.0]) increase in light PA from baseline to post-intervention between the control group and the intervention group (b) −0.4 minutes/hour (95% Cl = [−0.9, 0.1]) decrease in MVPA from baseline to post-intervention between the control group and the intervention group	Shared treadmill workstations decreased sedentary time and improved low-intensity and light PA behavior but not MVPA activity during the workday and overall in overweight/obese office workers	• Only used overweight/obese workers • Short-term follow-up
4	Miyachi et al. (2015)	(a) Change in total time spent on total PA (p = NR): 544.6 (±117.5) minutes/day after intervention versus 536.1 (±117.0) minutes/day in control group (b) Change in total time spent on light PA (p = .019) minutes/day daily from baseline to post-intervention: 481.9 (±116.0) minutes/day after intervention versus 479.1 (±113.5) minutes/day in control group (c) Change in total time spent on moderate PA (p = NR) minutes/day daily from baseline to post-intervention: 58.2 (±20.7) minutes/day after intervention versus 53.4 (±17.0) minutes/day in control group (d) Change in total time spent on vigorous PA (p = NR) minutes/day daily from baseline to post-intervention: (±11.1) minutes/day after intervention versus (±11.6) minutes/day in control group	NR	Installation of sit-stand desk increases time spent on overall PA, especially PA during weekdays	• Minimal control over daily workload • Objective measure unable to determine duration of standing • Short-term intervention and follow-up • Small sample size
5	Tobin et al. (2016)	NR	(a) Net difference in stepping time by 2.1 minutes/8-hour workday (p = .761) in intervention group from baseline to post-intervention relative to control group	A significant reduction in sitting time, but no change in the amount of time participants spent stepping at work	• Short-term study period • Small sample size • Unmeasured confounders such as work or life stress • Outcomes measured during working hours only
6	Bergman et al. (2018)	(a) Increase in daily walking time (22 minutes/weekday, p = .00045) from baseline to 13 months (b) Increase in daily light activity PA time (3 minutes/weekday, p = .005) from baseline to 13 months (c) Change in daily MVPA activity time (2 minutes/weekday, p = .23) from baseline to 13 months	NR	Treadmill workstations increased daily walking time among overweight or obese office workers compared with those with a sit-stand desk. However, MVPA decreased over the study period with the largest decrease in the intervention group after 13 months	• Different companies had various health promotion programs during intervention periods • Lack of blinding • Only used overweight/obese workers • Different companies had various health promotion programs during intervention periods
7	Maylor et al. (2018)	(a) Net increase in stepping time (1.0 minute/day, p = .770) from baseline to 8 weeks	(a) Net increase in stepping time at work (12 minutes/workday, p < .001) from baseline to 8 weeks	Decrease in prolonged sitting time at workout and increase in steps per day using an individual, organization, and environmental intervention approach without the use of a sit-stand desk	• Only one worksite used • Unable to measure effect of individual interventions • Outcomes measured during working hours only
8	Pierce et al. (2019)	(a) No change in leisure PA associated with intervention (NR, p = .039)	(a) Increase in steps taken at work (NR, p < .001) from baseline to 8 weeks associated with intervention	Adding electronic adjustable height desk to the workplace was associated with increase steps at work, decrease in sitting time at work, and no change in leisure-time PA	• Small sample size • Pedometers unable to detect postural changes • Pedometers unable to capture intensity level of PA • Shorten version of the Baecke Questionnaire has not been previously assessed
9	Gilson et al. (2012)	NR	(a) Net difference in percentage of time at work spent in light activity PA (0.8%, 95% Cl = [−6.8, 7.9]) from baseline to post-intervention (b) Net difference in percentage of time at work spent in MVPA (−0.7%, 95% Cl = [−1.8, 2.3]) from baseline to post-intervention	Desk had no overall effect on sedentary time or PA time at work	• Armband accelerometer did not capture posture changes • Small sample size • Short measurement period • Participants shared the intervention desk
10	Gorman et al. (2013)	NR	(a) Net increase in stepping time 1.2 minutes/workday (p = .748) at work from baseline to post-intervention	Post-move to an activity-permissive workspace office worker had less time sitting and more time standing, but no change in stepping time or PA during work time	• Change to PA outside work hours was not accessed • Bias within sample population due to nature of their industry • Timing of data collection premove was not described
11	Koepp et al. (2013)	(a) Change in average daily PA (NR, p = NR) from baseline to post-intervention: 4,205 AU/day (p ≤ .001) at 12 months of intervention, 4,460 AU/day (p ≤ .001) at 6 months of intervention, and 3,353 AU/day (p ≤ .001) at baseline	(a) Change in walking time (NR, p = NR) at work from baseline to post-intervention: 70 minutes/workday (p ≤ .001) at baseline, 128 minutes/workday (p ≤ .001) at 6 months of intervention, and 109 minutes/workday (p ≤ .001) at 12 months of intervention	Overall PA and walking time at work increased over the course of a year in employees with access to a walking treadmill desk and had no significant impact on work performance	• Treadmill desk scatter throughout the office did not allow for community support among users • Small sample size • Unlike similar studies that report PA in METs or intensity categories, PA is only reported in AUs
12	Jancey et al. (2016)	NR	(a) Change in the average minutes/workday time spent doing light activity at work (p < .001) from baseline to post-intervention: 57.16 minutes/workday (95% Cl = [52, 63]) after intervention and 35.13 minutes/workday (95% Cl = [32, 39]) at baseline (b) Change in the average minutes/workday time spent doing moderate activity at work (p = .109) from baseline to post-intervention: 39.72 minutes/workday (95% Cl = [35, 45]) after intervention and 36.13 minutes/workday (95% Cl = [33, 40]) at baseline (c) Change in the average minutes/workday time spent doing vigorous activity at work(p = .658) from baseline to post-intervention: 0.29 minutes/workday (95% Cl = [0.11, 0.53]) after intervention and 0.33 minutes/workday (95% Cl = [0.16, 0.54]) at baseline	Average time spent doing light activity at work increased, time spent standing at work increased, and sitting time at work decreased	• No control group • Large loss to follow-up • Did not measure PA outside of work
13	Chauetal. (2016)	NR	(a) Net difference in walking time (−8 minutes/workday, p = .679) at work from baseline to post-intervention: −21 minutes/day (p = .144) change in intervention versus −13 minutes/day (p = .287) change in control (OSPAQ) (b) Net difference in heavy labor time (14 minutes/workday, p = .125) at work from baseline to post-intervention: 11 minutes/day (p = .396) change in intervention versus −3 minutes/day (p = .749) change in control (OSPAQ)	No changes in walking time or PA, but sit-stand desk did reduce sitting and increase standing time	• Small convenience sample • Data loss from accelerometer device malfunction • Non-adherence to wearing accelerometer device
14	Engelen et al. (2016)	(a) Decrease in number of days spent doing MVPA per week(p > .05) at work from baseline to post-intervention: 4.6 days/week in intervention versus 4.9 days/week at baseline	(a) Change in percentage of time spent walking (p > .05) at work from baseline to post-intervention: 11% of workday at baseline versus 10% of workday in intervention (OSPAQ) (b) No change in percentage of time spent on heavy labor at work from baseline to post-intervention: 0% of workday in intervention and 0% of workday at baseline (OSPAQ)	Workers sat less and new building provided more opportunities for incidental activity	• No objective measure of PA • Small sample size • Large loss to follow-up • Study did not compare the four baseline sites for differences at baseline
15	Mansoubi etal. (2016)	NR	(a) No change in stepping time at work from baseline to post-intervention (ActivPAL) (b) No change in overall proportion of time spent in light activity on workdays from baseline to post-intervention (ActiGraph) (c) No change in overall proportion of time spent in MVPA on workdays from baseline to post-intervention (ActiGraph)	The use of sit-to-stand desk decreased sedentary time at work, increased light activity time at work, and had no effect on leisure-time moderate PA	• Short-term study period • Small sample size • Convenience sample
16	Eyler et al. (2018)	NR	(a) Increase in average steps/day (NR, p = .99) from baseline to post-intervention in all groups: 1,591 steps/day (p < .001) change in movers, 928 steps/day (p = .09) change in non-movers, and 1,756 steps/day (p = .008) change in control	PA increased in all study groups; it is unclear whether building had an effect on PA	• Wellness challenge to track PA for cash incentives started at time of post-data collection • Self-selection bias in those who volunteered to wear accelerometer • Unable to capture sample demographics • Did not compare sitting and activity patterns at baseline for differences between groups
17	Zhuetal. (2018)	NR	(a) No change to light PA or MVPA at the workplace	Sit-stand desk paired with motivational support decreased sitting time, increased standing time, and increased low-intensity PA in the workplace and is sustainable for 18 months	• Small sample • High attrition at 18 months
18	Candido et al. (2019)	NR	(a) Increase in average steps/day: 300 steps/day (p = NR) after relocation to new office environment	The average steps per day increased after moving to a active building design environment	• Selection bias in those who volunteered to wear Fitbit • PA data only collected from 20 volunteers
19	Dutta et al. (2019)	NR	(a) 24,748 AU/hour (95% Cl = [7,150, 42,347]) increase at work after 1 year follow-up from baseline	Overall PA during the workday remained about 12% higher, and sitting time remained reduced after 1 year of sit-stand desk relative to baseline	• Small sample size • Blinding not achievable • Low retention from original study
20	Malaeb et al. (2019)	(a) Increase in total PA step count in intervention group compared with baseline (p < .01)	NR	Treadmill desk usage over 2 weeks increased overall step count and had positive body composition results	• Small sample size • All participants had BMI >25 • Short intervention period • Accelerometer counts not converted to METs or minutes of activity
21	Wahlstrom et al. (2019)	NR	(a) Change in walking time (NR, p = .001) minutes/workday from baseline to 18-month follow-up: 39 minutes/workday (95% Cl = [35, 43]) at baseline and 47 minutes/workday (95% Cl = [44, 52]) at 18-month follow-up in flex office; 42 minutes/workday (95% Cl = [38, 46]) at baseline and 41 minutes/workday (95% Cl = [40, 46]) at 18-month follow-up in cell office (b) Change in MVPA time (NR, p < .001) minutes/workday from baseline to 18-month follow-up: 19 minutes/workday (95% Cl = [15, 22]) at baseline and 27 minutes/workday (95% Cl = [23, 30]) at 18-month follow-up in flex office; 16 minutes/workday (95% Cl = [13, 19]) at baseline and 19 minutes/workday (95% Cl = [15, 22]) at 18-month follow-up in cell office	Greatest increase in walking time, number of steps, and MVPA time during the workday compared with baseline occurred in flex offices. No changes in sitting time occurred	• All employees had sit-stand Desk before and after move regardless of office type • Unbalanced distribution of gender and managers between the two groups • Utilization of provided health and wellness hour not measured • Seasonal differences in moves were not accounted for
22	Wallmann- Sperlich et al. (2019)	NR	(a) Change in minutes/workday walking (p = .33) from baseline to 7 months post-intervention: 70.2 minutes/workday (95% Cl = [30.7,109.7]) at baseline and 84 minutes/workday (95% Cl = [50.1, 117.9]) at post-intervention measure	Offices that were active and biophilic designed increased walking time and standing time and reduced sitting time during the workday	• Small sample size • High attrition • No objective PA measurement • PA intensity levels not measured
23	McGann et al. (2015)	NR	(a) Light-intensity PA was 4.6% in Building 1, 2.6% in Building 2, and 3.3% in Building 3. Building 1 has best quality staircases and corridors	The building with the best quality staircase and participants had the highest mean step count per day and highest mean level of moderate PA	• Did not discuss the participants workstations or desk • Did not provide PA measures per building
24	Carr et al. (2016)	NR	(a) Difference in average time walking hours/day at work in employees with sit-stand desk versus walking desk: 0.7 hours/day in sit desk and 0.9 hours/day in sit-stand desk (p = .22)	Office employees with long-term access to sit-stand desk stood 60 minutes during the workday than employees with only a sit desk	• Small sample size • Limited generalizability • PA intensity levels not measured
25	Lindberg et al. (2018)	NR	(a) Workers in open bench seating exhibited 31.83% higher rates of PA than those in private offices (225.51 mG; 95% Cl = [137, 314]) (b) Workers in open bench seating exhibited 20.16% higher rates of PA than those in cubicles (185 mG; 95% Cl = [67, 304])	Workers in open bench seating types had higher rates of PA than those in cubicles or private offices	• Cannot confirm causal relationship between office type and PA
26	Renaud et al. (2018)	(a) Meeting the guidelines for PA showed a positive trend with sit-stand desk users: 30% in non-users, 30% in monthly to weekly users, and 35% in daily users	(a) Walking hours/week at work showed a positive trend with sit-stand desk users: 2.3 hours/week in non-users, 2.2 hours/week in monthly to weekly users, and 3.2 hours/week in daily users	Positive trend in overall PA, access trend of walking time at work, and decreased sitting time at work in those with long-term access to sit-to-stand desk	• Recall bias • Social desirability • Did not measure how long employees had used a sit-stand desk • Limited PA measurement

Note. PA = physical activity; NR = not reported; OSPAQ = Occupational Sitting and Physical Activity Questionnaire; AUs = activity units; Cl = confidence interval; MVPA = moderate- to vigorous-physical activity; BMI = body mass index; mG = milli-Gs (g-force); METs= Metabolic equivalent of task.