Elsevier

Applied Ergonomics

Volume 40, Issue 5, September 2009, Pages 852-859
Applied Ergonomics

The influence of desk and display design on posture and muscle activity variability whilst performing information technology tasks

https://doi.org/10.1016/j.apergo.2008.09.004Get rights and content

Abstract

Desk design and computer display height can affect posture and muscle activation during computer use. Amplitudes of postural variables and muscle activity during computer use do not explain the results from epidemiological studies of musculoskeletal discomfort and disorders related to computer use. The purpose of this study was to assess variability of posture and muscle activity during work with two computer display heights and book/paper, in conjunction with a curved desk designed to provide forearm support and a traditional, straight desk.

18 male and 18 female participants performed 10-min tasks involving keying, mousing, reading and writing in six desk/display conditions. 3D posture and surface emg were assessed for the final 2 min of each task.

The curved desk resulted in greater postural and muscle activity variation, suggesting an advantage of this supportive surface over the straight desk. There was little difference in variability associated with the two display heights. However, greater variability of posture and muscle activity was evident with the book/paper condition. Non-touch typists had greater neck flexion variation.

The design of information technology tasks and workstations can influence the short term variation in posture and muscle activity. Variation is influenced independently of mean postures and muscle amplitudes and therefore needs to be considered to adequately assess the risk of musculoskeletal disorders.

Introduction

Computer use is becoming ubiquitous at home and work. The number of personal computers in use worldwide exceeded 900 million in 2005 and if current trends are maintained, the U.S.A. is likely to have more personal computers (PCs) in use than people in five to six years (Computer Industry Almanac, 2006). In Australia 89% of businesses used computers in the year to June 2005 (Australian Bureau of Statistics, 2006) and in the U.K. approximately 13.9 million households could access the Internet from home in early 2006 (National Statistics UK, 2006). Concerns over musculoskeletal disorders related to computer use (Bergqvist et al., 1995, Sillanpää et al., 2003) have led to the development of guidelines for workstation design. A review of guidelines for occupational loading of the musculoskeletal system – primarily the neck and shoulder regions – was conducted by Westgaard and Winkel (1996). It was found that current guidelines for physical workload mainly emphasised a reduction in the level (amplitude) of workload, while few guidelines considered the ‘time dimensions’ of exposure – i.e. the variability and duration of workload. The authors suggested that all of these components should be addressed in order to assess the risk for musculoskeletal disorders. This need for an accounting of exposure variability is starting to be addressed within the guidelines themselves. For example, one of the stated objectives of the new North American BSR/HFES 100 (a draft update of the ANSI/HFES 100 standard) is to ‘maintain user performance by allowing postural changes that minimise static loads’.

In spite of the stated need for variability of loading, and the need for evidence to inform guidelines, most research concerning workstation design has continued to focus primarily on the amplitudes of postural angles and muscle activity. Two critical aspects of workstation design are the display and the desk.

Computer display heights have generally been recommended based on reducing mean head and neck flexion and cervical extensor muscle activity. It is generally accepted that a downward gaze angle is beneficial for visual comfort (Mon-Williams et al., 1999, Sommerich et al., 2001). A consideration of postural mechanics, however, dictates that the increase in head tilt required to accommodate lower displays increases the gravitational moment, and therefore the muscular torque required to maintain this flexed posture. In agreement with this mechanical model, an increase in cervical erector spinae activity has been observed with lower display heights (Greig et al., 2005, Sommerich et al., 2001, Straker et al., 2008b, Turville et al., 1998, Villanueva et al., 1997). Greater trapezius activity may also be expected in response to an increase in the gravitational moment about the neck and some studies have reported this (Aaras et al., 1997, Sommerich et al., 2001, Turville et al., 1998, Villanueva et al., 1996) whilst others have reported a decline (Briggs et al., 2004, Turville et al., 1998). However, the relationship between display heights and musculoskeletal loading is more complex than simple gravitational moment considerations would imply (Burgess-Limerick et al., 2000, Straker et al., 2008b).

Reductions in muscular symptoms with lower displays have been reported in field studies by Fostervold et al. (2006) and Marcus et al. (2002). Current knowledge of the effects of display height on muscle activity of the neck and shoulder does not account for the reported benefits of a lower display height – a reduction in the risk of musculoskeletal disorders. It therefore appears that parameters other than EMG amplitude must be of significance for the risk of musculoskeletal disorders with changing display height.

Little is known about the impact of display height upon exposure variability. Ankrum and Nemeth (2000) suggested that a lower display may provide the opportunity to move between a range of postures that were acceptable to both ocular and musculoskeletal systems, thus avoiding postural fixity. Whilst this argument is intuitively appealing, the authors did not provide supporting data. An assessment by Turville et al. (1998) of the number of posture shifts during work with two display heights revealed no difference between conditions, however, the frequency of postural shifts increased across time for both display heights. Fostervold et al. (2006) found no difference in the number of periods during which trapezius activity was below 1% MVC for display positions of 15° and 30° below horizontal.

Desks which provide for forearm support have also been shown to reduce the incidence of musculoskeletal discomfort and disorders (Cook et al., 2004, Marcus et al., 2002, Rempel et al., 2006). However, studies which have examined the effect of forearm support on muscle activity amplitudes have provided mixed results, and therefore cannot adequately explain this protective effect. A reduction in activity of the neck/shoulder muscles with forearm support was reported by Aaras et al. (1997) and Karlqvist et al. (1999). In contrast, the use of a curved desk designed to facilitate forearm support was shown by Straker et al. (2008b) to increase rather than decrease the mean amplitude of upper trapezius activity, when compared to a traditional, straight desk. Cook et al. (2004) found a reduction of trapezius and anterior deltoid activity with the utilisation of wrist support but not forearm support, when compared to a ‘floating’ posture. In one of the few studies to have considered variables other than EMG amplitude, Aaras et al. (1997) reported that the number of periods and total duration when trapezius activity was below 1% MVC increased when the forearms were supported on the desk.

From the preceding studies it can be seen that there is growing evidence that a focus on EMG amplitudes does not necessarily capture the risk of musculoskeletal disorders, and that assessment of variation in exposure over time may be required, as discussed by Mathiassen (2006). A recent study by Delisle et al. (2006) highlights the value of considering exposure variability. This study compared support from chair armrests (with and without the use of an adjustable workstation) and support from the desk surface during computer work. There was no difference in the amplitude of trapezius activity, as measured using the Amplitude Probability Distribution Function (APDF). However, parameters of the Exposure Variation Analysis (Mathiassen and Winkel, 1991) proved more sensitive for assessment of the effects of differing forms of support. There was greater variation of muscle activity with the corner workstation (desk-based support) than with the use of armrests. The authors suggested that a single height surface may have afforded greater opportunity for postural changes than the multiple levels provided by chair rests and desk.

Aside from new, computer-based tasks, many adults are exposed to information technology (IT) tasks using old, book/paper-based technology. Prior reports have demonstrated higher mean posture and muscle activity loads using old IT (Straker et al., 2008a, Straker et al., 2008b). However, it was suggested that old IT may be more variable, which may offset the increased risk associated with higher muscle activity amplitudes.

The aim of this study was to examine the influence of desk and display design on the variability of posture and muscle activity whilst performing IT tasks using new and old technology.

Section snippets

Study design

The study used a 2 × 3 within-subjects design, with desk and display conditions forming the independent variables. The first factor, desk, had two levels: 1) ‘traditional’straight desk set at 3 cm below seated participant's elbow height with 0° shoulder flexion and forearms unsupported, and 2) ‘horseshoe’ partially wrapped around curved desk surface located 3 cm above elbow height, enabling full forearm support with some shoulder flexion (see Fig. 1). The second factor, display, comprised three

Results

3D spinal and upper limb postural angles and muscle activity mean amplitudes for the desk ×display conditions have been described previously (Straker et al., 2008a, Straker et al., 2008b). Mean gaze angles for the six desk ×display conditions are recorded in Table 2.

Desk

As previously reported, the curved desk was associated with greater mean scapula elevation (4–7°) and protraction (2–3°), together with more arm flexion (6–13°) and abduction (12–17°) (Straker et al., 2008a) and small increases in CES (2–4%) and UT (4–7%) mean muscle activity (Straker et al., 2008b). Based on mean amplitudes alone, the straight desk would be recommended over the curved desk. However, the curved desk was designed to provide support and there is evidence to show that support is

Conclusion

Prior guidelines for IT workstation design have been largely based on mean exposure amplitudes. The evidence from studies of mean amplitudes associated with desk and display designs has not been able to explain the associations with musculoskeletal disorders identified in epidemiological studies. This suggested that measures of variation may be critical to adequately characterise risk. This paper reports the influence of desk and display designs on posture and muscle activity variation. A curved

Conflict of interest

The authors had no conflicts of interest.

Acknowledgements

The authors would like to thank the participants, the Australian National Health and Medical Research Council for funding this research (project 229011), Kevin Netto and Jemma Coleman for data collection and processing, and Paul Davey for programming.

References (35)

  • G.P.Y. Szeto et al.

    A comparison of symptomatic and asymptomatic office workers performing monotonous keyboard work: 2) Neck and shoulder kinematics

    Manual Therapy

    (2005)
  • K.L. Turville et al.

    The Effects of video display terminal height on the operator: a comparison of the 15° and 40° recommendations

    Applied Ergonomics

    (1998)
  • R.H. Westgaard et al.

    Guidelines for occupational musculoskeletal load as a basis for intervention: a critical review

    Applied Ergonomics

    (1996)
  • A. Aaras et al.

    Postural load during VDU work: a comparison between various work postures

    Ergonomics

    (1997)
  • D.R. Ankrum et al.

    Head and Neck Posture at Computer Workstations - What is Neutral?, 14th Triennial Congress of the International Ergonomics Association

    (2000)
  • Australian Bureau of Statistics

    8129.0 Business use of information technology 2004–05

    (2006)
  • P. Beighton et al.

    Hypermobility of Joints

    (1983)
  • Cited by (33)

    • Linking head and neck posture with muscular activity and perceived discomfort during prolonged smartphone texting

      2021, International Journal of Industrial Ergonomics
      Citation Excerpt :

      Elbow flexion-extension angle (EA): angle between the line connecting the marker placed on the acromion and that on the epicondyle, and the line connecting the marker on the epicondyle and that on the wrist. For all the angles and for each posture condition, the values over time were purged of the mean value taken in the reference condition; from the resulting distribution, the median value (P50NA, P50TA, P50GA, P50EA for the four described angles) and the posture range (PRNA, PRTA, PRGA, PREA) taken as the difference between the 10th and the 90th percentile, were calculated (Straker et al., 2009). To capture information on the viewing distance for each participant, the distance d between the cyclops and the smartphone was also extracted.

    • Lumbar Movement Dysfunction Based on Movement Control Impairment Classification System in Those Who Do and Do Not Develop Transient Low Back Pain During Prolonged Sitting

      2020, Journal of Manipulative and Physiological Therapeutics
      Citation Excerpt :

      Today, many people use computers and smart phones to work, study, and self-monitor their health, and the number of people who sit and work long is increasing.1-3

    • Trapezius muscle activity variation during computer work performed by individuals with and without neck-shoulder pain

      2019, Applied Ergonomics
      Citation Excerpt :

      EVA has been shown in studies of muscle activity, in particular in the upper trapezius, to be sensitive in various tasks outside computer work (Attebrant et al., 1997; Bao et al., 1996; Delisle et al., 2005; Fjellman-Wiklund et al., 2004; Januario et al., 2018; Mathiassen and Winkel, 1996; Reynolds et al., 2014). Specifically with regard to upper trapezius muscle activity during computer work, EVA has been shown to be sensitive to variation in task (Ciccarelli et al., 2013; Jensen et al., 1999; Mathiassen et al., 2003; Straker et al., 2009; Wærsted and Westgaard, 1997), demographic groups (pain, age and pregnant vs. non-pregnant) (Dumas et al., 2008; Hägg and Åström, 1997; Maslen and Straker, 2009), and configuration of work environment (Dumas et al., 2009; Samani et al., 2009a, 2009b). Of these studies, Hägg and Åström (1997) is the only study to have compared workers with and without pain using EVA.

    • Using evidence to support the design of submarine control console workstations

      2019, Applied Ergonomics
      Citation Excerpt :

      Two moderate quality studies found that chairs that were backward tilted reduced head repositioning ability (Wong et al., 2006) and poorly designed chairs increased neck muscle load (Konarska et al., 2005). Eight studies (9 articles) were assessed, and an overall moderate level of evidence was found for work surface height impacting on musculoskeletal symptoms (Aaras et al., 1997; Baker and Moehling, 2013; Fogleman and Lewis, 2002; Gerr et al., 2000; Hsu and Wang, 2003; Marcus et al., 2002; Shin and Hegde, 2010; Straker et al., 2008, 2009a). Low keyboard height resulted in head and eye discomfort (Fogleman and Lewis, 2002), postural changes (Baker and Moehling, 2013), back discomfort (Hsu and Wang, 2003), reduced shoulder elevation (Straker et al., 2008), increased wrist flexion (Gerr et al., 2000), and increased risk of developing musculoskeletal symptoms (Marcus et al., 2002).

    • Spinal kinematics during smartphone texting – A comparison between young adults with and without chronic neck-shoulder pain

      2018, Applied Ergonomics
      Citation Excerpt :

      50th percentile (%ile) APDF which is the median displacement, and the difference between 90th%ile and 10th%ile APDF (APDF range) representing the range of the posture changes, were computed (Jonsson et al., 2007; Mathiassen, 2006; Straker et al., 2009; Szeto et al., 2005). An increased APDF range means a greater amplitude of change in postural angles and therefore reflects the increased size of movement variability (Arvidsson et al., 2006; Mathiassen, 2006; Straker et al., 2009; Wahlström et al., 2010). In addition, the frequency of postural changes regarding “how often” the posture varied was also examined.

    • Work-Related Injuries of Radiologists and Possible Ergonomic Solutions: Recommendations From the ACR Commission on Human Resources

      2017, Journal of the American College of Radiology
      Citation Excerpt :

      The keyboard and mouse are particularly important in minimizing stress to the hands and wrists. The keyboard and mouse should be placed in a convenient location and at a comfortable height, with plenty of desk space available around them and few obstructions, to allow fluid movements [29]. Ideally, these devices should be thin and flat to reduce wrist extension.

    View all citing articles on Scopus
    View full text