Postural Variability

Abstract

Objective:

This article investigates whether different interventions aimed at promoting postural change could increase body movement throughout the shift and reduce musculoskeletal discomfort.

Background:

Many researchers have reported high levels of discomfort for workers that have relatively low-level demands but whose jobs are sedentary in nature. To date, few interventions have been found to be effective in reducing worker discomfort.

Methods:

Thirty-seven call center operators were evaluated in four different workstation conditions: conventional workstation, sit-stand workstation, conventional workstation with reminder software, and sit-stand workstation with break reminder software–prompt to remind workers to take break. The primary outcome variables consisted of productivity, measured by custom software; posture changes, measured by continuous video recording; and discomfort, measured by simple survey. Each condition was evaluated over a 2-week period.

Results:

Significant reductions in short-term discomfort were reported in the shoulders, upper back, and lower back when utilizing reminder software, independent of workstation type. Although not significant, many productivity indices were found to increase by about 10%.

Conclusions:

Posture-altering workstation interventions, specifically sit-stand tables or reminder software with traditional tables, were effective in introducing posture variability. Further, postural variability appears to be linked to decreased short-term discomfort at the end of the day without a negative impact on productivity.

Applications:

An intervention that can simply induce the worker to move throughout the day, such as a sit-stand table or simple software reminder about making a large posture change, can be effective in reducing discomfort in the worker, while not adversely impacting productivity.

Keywords

sedentary static postures musculoskeletal discomfort productivity posture

Introduction

Data entry and long durations of office work at a computer terminal are common and often very detrimental to the workers (Jensen, Finsen, Sogaard, & Chistensen, 2002). Blatter and Bongers (2002) reported that more than 50% of secretarial workers indicated they worked continuously for more than 4 hours using a keyboard and mouse. Call center operators have also been documented to remain in their seats more than 95% of the work shift (Rocha, Glina, Marinho, & Nakasato, 2005). Toomingas, Forsman, Mathiassen, Heiden, and Nilsson (2012) reported that 80% of the work shift was spent seated with 9% of the workers having periods of more than 1 hour of continuous sitting. The bottom line is that office workers, particularly data entry and call center operators, will work long periods without taking a break.

Incidence rates for neck and shoulder discomfort were found to be between 10% and 60% in individuals who have extensive computer use (Ekman, Andersson, Hagberg, & Heljm, 2000; Gerr et al., 2002; Korhenon et al., 2003; Rocha et al., 2005; Wahlström, 2005). Prevalence of discomfort in the hand and wrists in call center operators has been reported at similar levels (about 40%) (Rocha et al., 2005). This body discomfort has an adverse impact on productivity of the office workers, as reported by Wahlström, Hagberg, Toomingas, and Tornqvist (2004). Another group of researchers have reported about 12% of the computer users indicated productivity reductions due to discomfort, which resulted in 10% to 20% decreases in perceived productivity (Hagberg, Tornqvist, & Toomingas, 2002). Thus, the opportunity to reduce discomfort in workers may result in increased productivity and reduce presenteeism.

Although there are many factors that contribute to the discomfort in the shoulders and upper extremities of office workers, one factor that has been strongly related to these adverse factors was sitting for long durations without a break (Ariens et al., 2001; Rocha et al., 2005). Rocha et al. (2005) found a strong relationship between symptoms in the neck/shoulder and lack of rest breaks (odds ratio [OR] = 3.17 95%; CI = 1.11–8.97). Some evidence of a dose response effect for prolonged durations of static postures and neck/shoulder discomfort was presented by Blatter and Bongers (2002), who found significant ORs for 4 to 6 hrs (OR_adj = 1.50; 95% CI = 1.11–2.02) and 6 to 8 hrs (OR_adj = 2.03; 95% CI = 1.44–2.85) as compared to 0 to 1 hrs of exposure. Jensen et al. (2002) also found a similar dose-response relationship with hours of computer work. Other factors relating to body discomfort and injuries in office workers include mouse activation, low static muscle effort due to lack of postural relief of the arms and keyboard/mouse activation, mental stress, and nonneutral postures of the body while sitting in the chair (Wahlström, 2005).

Although there have been several studies investigating changes in the work environment that target proper ergonomic position such as neutral joint postures, the results are mixed regarding the impact on discomfort and injuries (Cook, Burgess-Limerick, & Chang, 2000; Gerr et al., 2002, 2005; Ketola et al., 2002; Lewis, Krawiec, Confer, Agopsowicz, & Crandall, 2002; Rempel et al., 2006). Gerr, Marcus, and Monteilh (2004) provided extensive literature review that concluded the results linking posture to office worker discomfort are not consistent, further supporting the notion that ergonomics interventions designed to promote neutral working postures may be helpful but might be partially the entire solution. One reason why these types of ergonomic interventions may not be effective in reducing discomfort, injuries, and compensation costs might be that although the workers are in better postures, they still remain in fixed postures for long durations (e.g., no breaks). Thus, additional interventions need to be explored.

The premise for the interventions evaluated in the current study was that routinely engaging in postural changes during the work shift will reduce the discomfort and improve productivity. The sit-stand adjustable table may be one option to introduce postural variability without having to leave the actual work area. The premise of the sit-stand workstation is to promote a dynamic movement by allowing the worker to switch between a sitting and standing work posture throughout the day. In a simulated data entry study, Husemann, Von Mach, Borsotto, Zepf, and Scharnbacher (2009) reported that musculoskeletal complaints were lower without a detriment to productivity for the sit-stand workstation as compared to a traditional sitting workstation. Furthermore, Alkhajah et al. (2012) found sit-stand workstations reduced sitting time by about 140 minutes per day. However, previous research into sit-stand workstations has shown low levels of usage compliance—workers stop adjusting the tables after a few weeks, indicating any benefits could be lost over time (Wilks, Mortimer, & Nylen, 2006).

Another type of postural change intervention is to remind workers to take more routine breaks. A study conducted by Galinsky, Swanson, Sauter, Hurrell, and Schleifer (2000) showed that frequent (5 min break/hour) breaks significantly decreased discomfort for data entry workers as compared to a conventional break schedule (two 15 min breaks/day). A follow-up study from this group evaluated routine breaks with stretching (Galinsky et al., 2007). The results indicated that the breaks, but not stretching, had an impact on the discomfort of the workers, possibly due to noncompliance of stretching. Furthermore, the effectiveness of stretching is compounded by compliance issues and appropriateness of the stretching for specific work conditions. A review by Goodman et al. (2012) reported that the few studies investigating rest breaks as an intervention found positive results—decreased pain and static awkward postures. Bernaards, Ariëns, Simons, Knol, and Hildebrandt (2008) reported that reminder software was effective in encouraging the use of breaks. However, two additional reviews (Hoe, Urquhart, Kelsall, & Sim, 2012; Kennedy et al., 2010) reported limited positive results for rest breaks being effective in controlling musculoskeletal disorders of the neck and shoulder for computer work.

The current study investigated whether interventions aimed at postural change could encourage adoption of routine dynamic movement of the body, resulting in decreasing musculoskeletal discomfort and increased productivity. Four different office workstation conditions were evaluated including a conventional workstation, sit-stand workstation, conventional workstation with reminders to take breaks, and sit-stand workstation with reminders to switch table heights and adopt postural changes.

Methods

Participants

Thirty-seven (29 females and 8 males) call center employees were recruited from a local drug and poison information center that is open 24 hours a day, 7 days a week, and handles regional emergency calls that deal with drug overdose and poison ingestion. The gender mix was representative of the gender ratio in the call center. The participants were full-time (18 employees) and part-time (19 employees) employees at the call center. All participants had worked at the facility for at least 1 year. The participation rate was 76%, with one worker being lost to follow-up due to leaving the place of employment (the partial data were not analyzed in the results). Complete demographic and anthropometric data can be found in Table 1.

Table 1:

Demographic and Anthropometric Data for the Participants Who Enrolled and Completed the Study (Mean and Standard Deviation)

	Age (Year)	Standing Height (cm)	Body Weight (kg)	Body Mass Index (kg/cm²)	Hours Worked (Hours/Week)
Full-time
Male (5)	33.8	181.2	95.4	29.2	44.3
	(8.5)	(8.2)	(23.0)	(8.2)	(13.8)
Female (13)	39.0	165.1	71.4	26.1	33.1
	(12.9)	(6.0)	(15.6)	(5.6)	(7.9)
All (18)	37.3	170.4	79.4	27.2	35.9
	(11.5)	(10.2)	(21.1)	(6.4)	(10.9)
Part-time
Male (3)	38.5	180.8	89.0	27.3	19.3
	(20.5)	(12.4)	(4.7)	(2.3)	(10.0)
Female (16)	32.5	161.5	67.7	25.9	13.9
	(6.6)	(7.0)	(14.0)	(5.4)	(7.1)
All (19)	33.5	164.5	71.0	26.1	14.8
	(8.9)	(10.3)	(15.2)	(5.0)	(7.8)

Experimental Design and Approach

A quasi-experimental design study was adopted with each workstation evaluated for a 1-month period (2-week break-in period and a 2-week observation period) (see Figure 1 for timeline of conditions). During the entire period, video cameras remained up with appearance of being active. There were four different workstation conditions: (1) conventional workstation/existing workstation, (2) conventional workstation with reminder software, (3) sit-stand workstation, and (4) sit-stand workstation with reminders. For all conditions, the workstation consisted of a corner unit (see Figure 2) with a flat screen monitor, keyboard, mouse, phone, and reference materials (e.g., books and notes). The outcome variables included the following: productivity—handling calls and computer usage, body discomfort, and postural changes (see Outcome Variables section for actual definitions).

Figure 1.

Timeline for evaluation of the conditions: Half of the participants started in normal and half started in sit-stand; the participants switched to other nonsoftware reminder condition; reminder software was activated and half of each group was randomly placed in reminder software and sit-stand with reminder software conditions; and each participant then switched to the last reminder software condition.

Figure 2.

Participant sitting at the conventional-height work surface (a) and standing at the high-height work surface (b).

Workstation conditions

The conventional workstation had a fixed height at 71.1 cm from the floor. Although the sit-stand adjustment controller was functional, the participants were instructed not to use the function when completing the conventional workstation condition (compliance was validated through video observation). For the sit-stand workstation condition, participants were instructed how to use the adjustment controller. All participants were encouraged to change the workstation height during the day as they pleased. However, instructions were provided at the beginning of the condition with no reminders after the initial interaction. The height of the workstation could be adjusted between 61.0 cm and 130.8 cm. However, the actual height of the workstation during sit-stand conditions was self-selected and was then specifically measured. The conventional workstation with software reminders utilized the fixed workstation height, but software was loaded on to the computers to track computer activity and request participants to make a postural change at 30-minute intervals. Similarly, the sit-stand workstation with software reminders utilized the adjustable workstation along with tracking software.

The reminder software, which was provided by WorkPace prompted the participants to stand up and move around or adjust the sit-stand table with a simple message (see Figure 3). The prompt would pop up on the screen every 30 minutes but did not require them to stop immediately given the potential for handling calls with life and death consequences. Instructions given to each of the participants at the start of every condition stated that the participant should use discretion when taking a break but were encouraged to follow the prompts as much as possible. One suggestion was to continue working, but to stand (or adjust table height) until time permitted a posture change. WorkPace software monitored the activity including key strokes and mouse movements, total computer usage, and compliance to postural change (e.g., no keyboard or mouse activity).

Figure 3.

Software reminder that is displayed at the appropriate time (every 30 minutes) to remind the participant to adjust their posture and/or height of the table. The prompt had a simple message with a soft lock on the screen—ignore button. The WorkPace software displayed a status bar for break time/idle time on computer (example, 52 s) and how long computer was used continuously (example, 1.43 hrs).

Each participant selected a chair that was utilized throughout the study. Numbers were placed on the back of the chairs for identification by the participants as well as allowing the research team to verify chair number through video capture.

Condition Assignment and Assessment

During the first 2 weeks for each condition, participants became familiar with the work conditions without any assessments. The break-in period also allowed the participants to establish everyday routines and minimize changes resulting purely from awareness of being observed (e.g., workers return to routines and habits). The 2-week break-in period was followed by a 2-week assessment period. During the 2-week assessment period, each participant was evaluated for productivity levels, presence of body discomfort, and any postural changes during the work shift (see dependent variables for more description of these variables). The participants were aware when the assessment period was underway, as they were required to complete the discomfort survey at the end of each shift. Figure 1 provides the timeline of the conditions.

The order of workstation conditions was counterbalanced with participants who were randomly assigned to workstation condition. At the start of the study, participants were randomly assigned to two groups: half into current workstation condition (e.g., fixed height) and half assigned to sit-stand workstation (e.g., adjustable height). After the first assessment, the groups switched workstation conditions (e.g., conventional condition switched to sit-stand condition and vice versa). At the midpoint of the study (upon completion of the first two conditions), software reminders were activated. Half of each of the original groups was randomly selected to start the conventional workstation with reminder software, whereas the other half were assigned to the sit-stand with reminder software. After the assessment period, the groups switched conditions for the final assessment. The block assessments on reminder software were necessitated by having multiple users on the same computer throughout the assessment period with no way to control the software pop-ups for nonsoftware conditions.

Given the work environment with a 24/7 service requirement and cubicle layout (two four-cubicle workstations), webcam cameras were set up to monitor each workstation continuously over the 2-week follow-up period. The webcam cameras recorded at a 50 Hz frame rate. Webcam cameras were present and visible during the entire study. Videos were digitally stored on laptops connected to the webcam cameras and then downloaded to a data server where they remained until they were viewed and analyzed by a member of the research team who documented posture changes and postural change compliance.

Recruitment of Participants

Recruitment consisted of an email announcement from the research group, which obtained email addresses of all current employees from management. The email provided study details and asked for volunteers. Once an email response was returned to the research team, one member of the research team met the individual to complete the consent process and obtain a signature on the consent form. The consent form and protocol was approved by the University of Cincinnati’s Institutional Review Board. Once a participant agreed to participate, the individual was assigned an identification number and assigned to an initial workstation group. Although complete confidentiality was not achieved because it was impossible to blind study participants from nonparticipants in the call center environment, participant-specific data were coded and kept confidential. Management remained out of the recruitment process and had limited information about who participated.

Outcome Variables

Body discomfort

Body discomfort was assessed at the end of each shift during each assessment period. A short questionnaire utilized an 11-point ordinal scale of discomfort (0 = no discomfort to 10 = unbearable discomfort) to document discomfort in the following regions: lower legs and feet, hips, knees, low back, upper back, neck, shoulders, and arms and wrists.

Postural variability

The posture variability included multiple variables that were defined through video analyses. These variables were measured for each participant during every shift during the assessment periods. Each of the digitized videos was reviewed by a member of the research team. The following were the posture variables assessed (all measured in minutes): total time on shift, time without postural change, total time sitting, total time standing, total time sitting at lowered height, total time sitting at raised height, total time standing at lowered height, total time standing at raised height, number of times switching between standing/sitting, number of times switching between lower/raised, total time at lowered height, total time at raised height, and total number of postural changes per hour. A postural change was defined as a major change of the entire body posture such as standing up/sitting down, turning in chair to communicate to coworkers, or walking away from the workstation. The following were not included as a postural change: changes in posture or movements of the upper extremities only, movement in the body while in the chair (e.g., moving leg under body), and reaching for the phone or other documents. Postural changes concentrated on large movements that were easily distinguishable on the video. The research team member would identify key events (e.g., standing and sitting intervals, postural changes) along with the corresponding times using the time stamp on the video and recording them into a database for further calculations. Times between events along with frequency of events were then computed.

Productivity and computer usage

Productivity was tracked by the call center through custom software. The research team was given the data by the management of the call center. The call center reports provided productivity indices relating to average number of calls per hour, average number of calls picked up per hour, average hold time—amount of time placed on hold waiting for call center representative, and percentage of time not available for calls—relative amount of time spent on a call or entering data into computer from previous case. In addition, WorkPace software provided productivity variables that related to input devices and computer usage: total number of keystrokes, total number of mouse clicks, and total time using computer. The WorkPace software was installed on all the computers during the entire study and tracked the number of clicks on the keyboard, number of clicks of the mouse (right and left buttons), and time actively using the computer (e.g., when keyboard was being typed on, the mouse was clicked or being moved around the screen). The software provides an accurate quantitative assessment of these outcomes based on the exact time that the keyboard or mouse was being clicked or moved, with the total time on computer being the summary of keyboard and mouse usage time. Overall, the software provides an accurate measure of computer usage. Douwes, de Kraker, and Blatter (2007) estimated that WorkPace overestimates computer usage by 9%, whereas Blangsted, Hansen, and Jensen (2004) reported underestimates of 2% to 8% for computer usage.

Statistical Analysis

Descriptive statistics were computed for all the variables as a function of workstation condition. A Kruskal-Wallis one-way analysis of variance by ranks procedure was utilized to determine significant differences for the discomfort variables. Follow-up post hoc comparisons were computed for significant effects in the form of a Wilcoxon signed rank test with a Bonferroni correction to determine the source of the significant differences. For the continuous variables (productivity and video outcomes), a repeated measures analysis of variance (ANOVA) was conducted with studentized Tukey post hoc tests. For all statistical analyses, an alpha of 0.05 was utilized to determine significance.

Results

The results for the productivity and exposure outcomes are displayed in Table 2. The time worked during all four conditions was very consistent and not significantly different. This indicates that all workstation conditions had equal amounts of exposure to the office work. In general, participants averaged about 25 hrs per week and roughly 8 hrs per shift, indicating a mixture of full-time and part-time workers. Furthermore, computer-based productivity indices were also not impacted by the four different conditions, although a slight increase in keystrokes and mouse clicks was seen for the conditions with reminder software (Table 2). Over the 2-week evaluation period, the workers utilized the computer keyboard for about 16 hrs and the mouse for about 10 hrs. The number of keystrokes ranged between 40,000 and 45,000 keys with about 12,000 clicks of the mouse (single and double clicks), independent of intervention condition.

Table 2:

Summary of the Outcome Variables During Each of the Workstation Conditions (Mean and Standard Deviation)

Outcome Variable	Conventional	Sit-Stand	Conventional With Reminder Software	Sit-Stand With Reminder Software	p
Total time worked per week^a (min/week)	1,552.2 (871.8)	1,499.4 (875.4)	1,624.8 (866.4)	1,500.6 (814.8)	0.63
Average time per shift^b (min/shift)	450.6 (148.8)	472.2 (165.6)	463.2 (132.6)	486.6 (141.0)	0.41
Total time computer used per 2-week period^c (min)	943.2 (555.00)	1,030.8 (508.2)	964.2 (563.4)	993.6 (529.2)	0.45
Total time mouse used per 2-week period^c (min)	612.6 (356.4)	656.4 (324.6)	607.2 (359.4)	619.2 (348.6)	0.46
Total number of keystrokes per 2-week period^c (n)	40,370 (27,388)	45,348 (26,206)	44,154 (30,170)	43,491 (26,069)	0.53
Total number of mouse clicks per 2-week period^c (n)	11,626 (7,191)	12,659 (7,228)	12,022 (8,166)	12,727 (8,098)	0.54
Average hold time^a (min)	5.87 (1.09)	5.77 (1.17)	5.75 (1.21)	5.75 (1.10)	0.51
Percentage of time not available for calls^a (%)	17.59 (9.41)	18.03 (10.98)	16.37 (6.35)	15.81 (6.34)	0.37
Average number of calls completed^a (n/hr)	5.58^AB (0.97)	5.45^A (0.94)	5.83^B (1.01)	5.88^B (1.07)	0.002
Average time without a break^b (min)	27.88 (15.98)	27.82 (28.23)	29.96 (28.76)	28.52 (21.56)	1.00
Total number of postural changes per hour (events/hr)	0.85^A (0.74)	1.04^B (0.83)	1.07^B (0.81)	1.12^B (0.80)	0.05
Total time sitting at low height^b (min)	393.59^A (127.23)	323.45^C (129.69)	372.08^AB (110.45)	334.18^BC (114.17)	<0.0001
Total time sitting at high height^b (min)	1.08^A (7.01)	67.56^B (70.00)	7.97^A (25.37)	60.75^B (59.76)	0.02
Total time standing per 2-week period^b (min)	1.59^A (8.02)	65.75^C (70.49)	2.72^A (7.28)	44.10^B (51.08)	<0.0001
Total time sitting per 2-week period^b (min)	393.07^A (127.15)	323.29^C (129.76)	371.45^AB (110.25)	334.18^BC (113.98)	<0.0001
Number of times switching between standing and sitting^b (n)	0.50^A (1.04)	4.05^B (3.85)	1.24^A (2.42)	5.11^B (5.44)	<0.0001
Number of times switching between low and raised height^b (n)	0.05^A (0.23)	3.56^B (3.36)	0.53^A (1.61)	4.09^B (4.40)	<0.0001
Percentage of time standing^b (%)	0.30^A (1.41)	14.22^C (15.06)	0.73^A (2.08)	8.93^B (10.10)	<0.0001
Percentage of time sitting^b (%)	89.92^A (25.75)	68.90^C (18.54)	81.41^B (12.94)	69.90^C (16.18)	<0.0001
Percentage of time away from workstation^b (%)	9.78^A (25.67)	16.88^B (9.58)	17.87^BC (11.99)	21.17^C (13.19)	<0.0001
Percentage of time workstation at low height^b (%)	90.03^A (25.73)	68.93^C (18.53)	81.53^B (12.93)	69.87^C (16.27)	<0.0001
Percentage of time workstation at high height^b (%)	0.19^A (1.23)	14.47^B (14.92)	1.96^A (6.34)	12.12^B (11.95)	<0.0001

Note. Different alpha characters indicate significant difference. A ≠ B ≠ C.

Based on productivity data provided by facility.

Based on video analyses.

Based on WorkPace software.

Productivity data provided by the call centers’ productivity reports indicated that most of the productivity indices were not affected by workstation condition, including average number of calls picked up per hour, average hold time, and percentage of time not available for calls (Table 2). A couple of variables relating to productivity were found to be significantly impacted by workstation condition. First, although the number of phone calls answered per hour increased with the reminder software, the change was only significant among those using sit-stand workstations. Second, the number of calls completed was not different between conventional and sit-stand. Differences in calls completed were documented for the reminder software (increase of 10%) and sit-stand with reminder software (decrease of 5%) as compared to conventional condition.

The video analyses evaluated the time spent sitting, standing, and away from the desk as well as the number of times workers transitioned between sit-stand and low-high workstations (see Table 2). Figure 4 provides a summary of how the different conditions influenced the amount of sitting. During the conventional workstation (with and without reminder software), the percentage of time standing was virtually nothing, whereas sitting approached 90%. The reminder software did reduce the percentage of time sitting by 8%. The sit-stand condition increased the amount of time standing by approximately 10% to 15%, with and without reminder software, respectively. The amount of time sitting was also significantly reduced by about 20% (for both with and without reminder software). The percentages of time for using the low and high workstations were mirror images of the sitting and standing percentages, potentially indicating an influence of the sit-stand intervention. The remaining time was spent away from the work area, which actually increased with the reminder software. The conventional workstation resulted in about 10% of the time away from the desk, whereas the sit-stand and reminder software conditions resulted in 15% of the time not using the computer. The greatest amount of time away from the workstation was for the sit-stand with reminder software (about 21% of the time). This time away allowed for breaks and completion of other activities outside the work area.

Figure 4.

Percentage of time spent sitting, standing, and away from workstation for the four conditions: (1) conventional, (2) sit-stand, (3) conventional with reminder software, and (4) sit-stand with reminder software.

As expected, the conditions with sit-stand tables had more times transitioning between sitting and standing (about four times per shift). The addition of the reminder software resulted in an increase of approximately one switch per shift more. The number of postural changes per shift was impacted (p < 0.05) by condition, with the conventional condition having the least—about seven per 8-hr shift. Both of the sit-stand conditions (with and without reminder software) had approximately eight postural changes per 8-hr shift, whereas the reminder software with conventional workstation had nine postural changes per 8-hr shift. The reminder software appeared to be effective in encouraging postural changes throughout the shift. However, the data from the WorkPace tracking software indicated a 62% compliance of the break prompts (e.g., computer was idle).

The body discomfort (measured at end of shift) across the 2-week intervention period was significantly lower for the shoulders, lower back, and upper back generally in the direction of sit-stand compared to conventional and reminder software compared to no reminder software (see Table 3). The decrease in discomfort levels ranged between 22% and 46% (Figure 5). There was no difference (p > 0.05) among the workstation conditions for discomfort in the neck, elbows, hands and wrists, hips, knees, and lower legs and feet.

Table 3:

Summary of the Body Discomfort at the End of the Shift During Each of the Workstation Conditions (Mean and Standard Deviation)

Outcome Variable	Conventional	Sit-Stand	Conventional With Reminder Software	Sit-Stand With Reminder Software	p
Neck	1.71 (1.79)	1.19 (1.50)	1.21 (1.61)	1.55 (2.10)	0.10
Shoulders	2.14^AB (2.01)	1.42^BC (1.56)	1.16^C (1.52)	1.20^C (1.66)	0.02
Elbows	0.55 (1.05)	0.24 (0.60)	0.43 (0.92)	0.44 (0.91)	0.33
Hands and wrists	1.49 (1.82)	0.98 (1.49)	0.89 (1.24)	1.00 (1.52)	0.11
Upper back	2.28^B (2.04)	1.33^A (1.59)	1.44^A (1.61)	1.37^A (2.07)	0.004
Lower back	2.18^A (2.31)	1.37^B (2.11)	1.70^C (1.77)	1.41^B (1.73)	0.04
Hips	0.51 (1.06)	0.51 (1.25)	0.80 (1.18)	0.63 (1.29)	0.17
Knees	0.78 (1.32)	0.71 (1.20)	0.83 (1.31)	0.83 (1.75)	0.59
Lower legs and feet	1.01 (1.40)	0.99 (1.80)	0.88 (1.14)	0.93 (1.58)	0.63

Note. Different alpha characters indicate significant difference. A ≠ B ≠ C.

Figure 5.

Discomfort at the end of the shift for the shoulder, upper back, and lower back for the four intervention conditions: (1) conventional, (2) sit-stand, (3) conventional with reminder software, and (4) sit-stand with reminder software.

Discussion

Although the study population was relatively small and the observation periods short, the results indicated sit-stand tables and break reminder software produce routine postural changes that could impact the discomfort of sedentary office workers. The effect of sit-stand or break reminders appeared to be independent as the synergistic impact on the outcome variables was not significantly different from the individual effects. The breaks/postural changes included in the current study were different than previously investigated by Galinsky et al. (2000, 2007) (e.g., 5 min/hr) and were generally designated (soft break every 30 min) due to the critical nature of the work. This type of break pattern allowed the user to have control of the breaks and to focus on making a large body movement—change in position of the entire body such as standing up, walking away from the workstation, and turning away from the workstation to communicate with other coworkers. The purpose of the break was to change body posture by altering the sit-stand height or completing productive tasks, which may have lead to the lack of adverse impact on the productivity indices.

As seen by others (Galinsky et al., 2000, 2007; Husemann et al., 2009), breaks and using an adjustable-height table had no impact on productivity of the workers. Actually, in the current study, the indices of productivity improved in most cases by about 10% (not statistically significant). It appeared that the postural changes resulted in the worker feeling better without any negative impact on productivity. One potential explanation might be that when workers felt better, they were more productive, even though they were away from their workstations more or paused routinely throughout the shift. Future work needs to better understand how postural changes impact productivity.

The impact of sit-stand on time sitting was slightly lower than what was reported by Alkhajah et al. (2012), where they reported a decrease of about 140 min (or 29%) in sitting time. The current study found reductions in sitting to be between 15% and 18%. However, the number of postural changes was increased by 40% for the sit-stand conditions, indicating they did more major body changes within their chair rather than standing up entirely.

It appears the frequency of switches in the current study between sit and stand postures were found to be lower than previous work of Toomingas et al. (2012). Toomingas et al. (2012) found approximately 10 changes/hr, whereas sit-stand conditions resulted in four to five per shift and reminder software alone was 1.25 per shift in the current study. These lower numbers of switches also point to dynamic posture changes while seated. For the conventional condition, the average number of switches was less than 0.5 per shift, indicating long durations of sitting. So although there was an increase in the number of times that the workers changed between sit and stand for the intervention studies as compared to the conventional workstation, the increases did not equal an average of 2/hr or close to the 10/hr found by Toomingas et al. (2012). The culture, type of emergency situation, and relatively short break-in period (e.g., 2 weeks for each condition) may have contributed to the lower number of actual changes between sitting and standing. Although it appeared from the video analyses that the workers complied with the protocols, more rigid breaks could have had more promising results.

To provide adequate context of the results, the following considerations and limitations need to be discussed. First, the participant population was a sample of convenience that was not selected randomly and was selected from one facility with a given culture and work demands. The sample of convenience was a result of funding considerations and the intensive evaluation of the workers (e.g., monitoring for 2 weeks, 24 hrs and 7 days a week for each condition). Although the sample came from one call center, the specific one selected offered some unique opportunities such as monitoring continuously for long periods of time (e.g., 2 weeks for each observation period), observing more than one person at a time (e.g., multiple users in multiple cubicle arrangement), and observing workers who were well-educated individuals (e.g., all had a college degree in order to provide the emergency services). Although the study was completely voluntary with limited encouragement from management, there were some confidentiality issues (e.g., some did not complete confidentiality of participation) that may have influenced or biased who volunteered for the study. However, the potential bias of a small participant pool and nonparticipation was low, given the overall response from the workers was high (e.g., more than 75% participation). Furthermore, the work environment was relatively unique to this call center in that the calls carried some level of urgency and potential life-threatening risk. As a result, the operators were not able to take breaks immediately on several occasions.

Second, the evaluation was a relatively short-term assessment with only a 2-week break-in period and a 2-week follow-up assessment period. As a result, the health outcomes were limited to discomfort responses rather than actual workplace injuries (e.g., reportable back discomfort, hand/wrist problems, and shoulder problems). The long-term ramifications of postural variability through sit-stand and reminder software remain unexplored. The results provide a more immediate impact of the interventions without any perspective to more long-term impact.

Third, the postural assessment was limited to global movements such as moving away from the computer, transitioning between sitting and standing, and leaving the work area. With the focus on large dynamic movements, the postural changes did not account for microbreaks or small extremity breaks, which could have had some impact on the results. With the length of assessment and time-consuming nature of the video analyses, it was not feasible to quantify the small movement level.

Fourth, the study design was not able to utilize a completely randomized control design due to multiple users sharing a limited number of workstations and cross-contamination concerns. The logistics of having reminder software for some participants while others were not to have reminder software was not feasible, so the study was broken into two phases: no reminder software and reminder software. Although not ideal, participants were randomized within these groups, minimizing the bias associated with condition ordering effects. Anecdotal data confirmed the issue as nonparticipants complained about the software popping up on the screen, which was resolved by showing them how to ignore the reminders.

Overall, the limitations discussed provide some context when interpreting the results of the current study. However, none of them were felt to provide significant issues with respect to the validity and generalizability of the results. However, the concerns above do indicate additional work is needed to better understand the underlying mechanisms of postural changing interventions such as investigating (1) muscular responses and changes in fatigue, (2) changes in inflammatory responses in these regions, and (3) more detailed information about postural changes through the use of quantitative tools such as goniometers. Future work also needs to expand these results to a long-term prospective study that would investigate musculoskeletal injuries rather than discomfort.

Conclusion

The study results have practical implications for sedentary office workers who take limited breaks. Based on the overall results, the implementation of sit-stand workstations and/or break reminder software influenced the number of gross postural changes and/or movements of the body that a worker will take throughout the shift. A result of the large body changes was a decreased level of discomfort at the end of the day without having any adverse impact on the productivity indices. However, the results should be viewed as preliminary as the long-term impact of these interventions is not fully understood. The study shows promise that reminders and sit-stand workstations leads to short-term reduction of body discomfort without adverse impact on productivity.

Key Points

The adoption of sit-stand workstations or reminder software produced significant changes in the postural response during the shifts.

Sit-stand workstations or reminder software reduced the discomfort in the shoulders, upper back, and lower back that the individuals suffered at the end of the shifts.

Introduction of postural changing modalities did not adversely impact the productivity of the workers.

Footnotes

Acknowledgements

We would like to thank Balaji Sharma, Donald Herrmann, and Anita P. Krishnan for their contribution and hard work in analyzing and processing the video data. Also, we would like to thank the Office Ergonomics Research Committee for partially funding the project.

Kermit G. Davis is an associate professor at the University of Cincinnati in the College of Medicine, Department of Environmental Health, where he also directs the Low Back Biomechanics and Workplace Stress Laboratory. He received his PhD in occupational ergonomics from the Ohio State University, College of Engineering, Department of Industrial and Systems Engineering. He is a certified professional ergonomist.

Susan E. Kotowski is an assistant professor at the University of Cincinnati in the College of Allied Health Sciences. She is also director of the Gait and Movement Analysis Lab. She received her PhD in occupational ergonomics and safety from the University of Cincinnati, College of Medicine. She is also a certified professional ergonomist.

References

Alkhajah

T. A.

Reeves

M. M.

Eakin

E. G.

Winkler

E. A.

Owen

Healy

G. N.

(2012), Sit-stand workstations: A pilot intervention to reduce office sitting time. American Journal of Preventative Medicine, 43, 298–303.

Ariens

Bongers

Douwes

Miedema

Hoogendoorn

van der Wal

Bouter

van Mechelen

(2001). Are neck flexion, neck rotation, and sitting at work risk factors for neck pain? Results of a prospective cohort study. Occupational and Environmental Medicine, 58, 200–207.

Bernaards

C. M.

Ariëns

G. A.

Simons

Knol

D. L.

Hildebrandt

V. H.

(2008). Improving work style behavior in computer workers with neck and upper limb symptoms. Journal of Occupational Rehabilitation, 18(1), 87–101.

Blangsted

A. K.

Hansen

Jensen

(2004). Validation of a commercial software package for quantiﬁcation of computer use. International Journal of Industrial Ergonomics, 34, 237–241.

Blatter

B. M.

Bongers

P. M.

(2002). Duration of computer use and mouse use in relation to musculoskeletal disorders of neck or upper limb. International Journal of Industrial Ergonomics, 30, 295–306.

Cook

Burgess-Limerick

Chang

(2000). The prevalence of neck and upper extremity musculoskeletal symptoms in computer mouse users. International Journal of Industrial Ergonomics, 26, 347–356.

Douwes

de Kraker

Blatter

B. M.

(2007). Validity of two methods to assess computer use: Self-report by questionnaire and computer use software. International Journal of Industrial Ergonomics, 37, 425–431.

Ekman

Andersson

Hagberg

Heljm

E. W.

(2000). Gender differences in musculoskeletal health of computer and mouse users in the Swedish workforce. Occupational Medicine, 50, 608–613.

Galinsky

Swanson

Sauter

Dunkin

Hurrell

Schleifer

(2007). Supplementary breaks and stretching exercises for data entry operators: A follow-up field study. American Journal of Industrial Medicine, 50, 519–527.

10.

Galinsky

Swanson

Sauter

Hurrell

Schleifer

(2000). A field study of supplementary rest breaks for data entry operators. Ergonomics, 43, 622–638.

11.

Gerr

Marcus

Ensor

Kleinbaum

Cohen

Edwards

Genrty

Ortiz

D. J.

Monteilh

(2002). A prospective study of computer users: I. Study design and incidence of musculoskeletal symptoms and disorders. American Journal of Industrial Medicine, 41, 221–235.

12.

Gerr

Marcus

Monteilh

(2004). Epidemiology of musculoskeletal disorders among computer users: Lesson learned from the role of posture and keyboard use. Journal of Electromyography and Kinesiology, 14, 25–31.

13.

Gerr

Marcus

Monteilh

Hannan

Ortiz

Kleinbaum

(2005). A randomised controlled trial of postural interventions for prevention of musculoskeletal symptoms among computer users. Occupational and Environmental Medicine, 62, 478–487.

14.

Goodman

Kovach

Fisher

Elsesser

Bobinski

Hansen

(2012). Effective interventions for cumulative trauma disorders of the upper extremity in computer users: Practice models based on systematic review. Work: A Journal of Prevention, Assessment and Rehabilitation, 42(1), 153–172.

15.

Hagberg

Tornqvist

E. W.

Toomingas

(2002). Self-reported reduced productivity due to musculoskeletal symptoms: Associations with workplace and individual factors among white-collar computer users. Journal of Occupational Rehabilitation, 12, 151–162.

16.

Hoe

V. C.

Urquhart

D. M.

Kelsall

H. L.

Sim

M. R.

(2012). Ergonomic design and training for preventing work-related musculoskeletal disorders of the upper limb and neck in adults. Cochrane Database of Systematic Reviews, 8, CD008570.

17.

Husemann

Von Mach

C. Y.

Borsotto

Zepf

K. I.

Scharnbacher

(2009). Comparisons of musculoskeletal complaints and data entry between a sitting and a sit-stand workstation paradigm. Human Factors, 51, 310–320.

18.

Jensen

Finsen

Sogaard

Chistensen

(2002). Musculoskeletal symptoms and duration of computer and mouse use. International Journal of Industrial Ergonomics, 30, 265–275.

19.

Kennedy

C. A.

Amick

B. C.

III Dennerlein

J. T.

Brewer

Catli

Williams

. . . Rempel

(2010). Systematic review of the role of occupational health and safety interventions in the prevention of upper extremity musculoskeletal symptoms, signs, disorders, injuries, claims and lost time. Journal of Occupational Rehabilitation, 20(2), 127–162.

20.

Ketola

Toivonen

Häkkänen

Takala

E. P.

Viikari-Juntura

, & Expert Group in Ergonomics. (2002). Effects of ergonomic intervention in work with video display units. Scandinavian Journal of Work, Environment, and Health, 28, 18–24.

21.

Korhenon

Ketol

Toivonen

Luukkoenen

Häkkänen

Viikari-Junura

(2003). Work related and individual predictors for incident neck pain among office employees working with video display units. Occupational and Environmental Medicine, 60, 475–482.

22.

Lewis

R. J.

Krawiec

Confer

Agopsowicz

Crandall

(2002). Musculoskeletal disorder worker compensation costs and injuries before and after an office ergonomics program. International Journal of Industrial Ergonomics, 29, 95–99.

23.

Rempel

D. M.

Kause

Goldberg

Benner

Hudes

Goldner

G. U.

(2006). A randomised controlled trial evaluating the effects of two workstation interventions on upper body pain and incident musculoskeletal disorders among computer operators. Occupational and Environmental Medicine, 63, 300–306.

24.

Rocha

L. E.

Glina

D. M. R.

Marinho

M. D. F.

Nakasato

(2005). Risk factors for musculoskeletal symptoms among call center operations of a bank in Sao Paulo, Brazil. Industrial Health, 43, 637–646.

25.

Toomingas

Forsman

Mathiassen

S. E.

Heiden

Nilsson

(2012). Variation between seated and standing/walking postures among male and female call centre operators. BMC Public Health, 12. Retrieved March 2, 2012, from http://www.biomedcentral.com/1471-2458/12/154

26.

Wahlström

(2005). Ergonomics, musculoskeletal disorders and computer work. Occupational Medicine, 55, 168–176.

27.

Wahlström

Hagberg

Toomingas

Tornqvist

E. W.

(2004). Perceived muscular tension, job strain, physical exposure and associations with neck pain among VDU users. A prospective cohort study. Occupational and Environmental Medicine, 61, 523–528.

28.

Wilks

Mortimer

Nylen

(2006). The introduction of sit-stand worktables: Aspects of attitudes, compliance and satisfaction. Applied Ergonomics, 37, 359–365.