The relationship between head and neck posture and VDT screen height in keyboard operators.

Key Words: Cervical, Computer, Ergonomics, Posture.

The use of computers has become commonplace over the past two decades. Computer operators have experienced an increased incidence of musculoskeletal disorders.[1-4] A high incidence of symptoms in video display terminal (VDT) operators has been reported.[5,6] Starr et al[5] surveyed 145 VDT users and found that 65% reported neck pain, compared with 48% in a control group of 105 subjects. Sauter et al[6] reported that 27% of VDT users surveyed reported "almost constant" cervical discomfort and that 30% reported frequent pain. A likely explanation for these findings is that use of a computer reduces the amount of task diversity and postural variation that would otherwise occur.[1,3,7-9] For example, an office function performed at a computer does not require body position changes or use of multiple tools that may be necessary without a computer.

Thus, prolonged operation of a VDT may have deleterious effects because of static postural loading and repetitive motion of the upper extremities. The risks associated with prolonged VDT use have led to the National Institute for Occupational Safety and Health (NIOSH) to recommend 15-minute breaks for every 2 hours of continuous VDT work, with the goal of reducing the cumulative stress of static posture and visual demand.[10]

Guidelines for computer workstation design and recommendations for posture while operating a computer have been proposed, with the goal of minimizing postural stress and static loading.[4,11-13] Grandjean and colleagues' study of operators' preferred settings of VDT workstations found the 95% confidence interval (CI) for screen height (floor to midscreen) to be between 95.0 and 111.0 cm (37.4-43.7 in).[8] The 95% CI for preferred viewing angle was between -4 and -14 degrees below the horizontal reference.[8] Hill and Kroemer[14] have reported a preferred mean viewing angle of -28.6 degrees. Kroemer et al[15] have reported less eye strain with a downward gaze.

Part of a physical therapist's role in treating patients with disorders resulting from VDT operation is to offer suggestions to the patients regarding their workstation design and postural habits. Based on our experience, it is often recommended that the VDT should be raised from a desktop position via the use of risers or an adjustable platform, with the goal of reducing postural stress and encouraging a more upright posture. The purpose of our study was to determine how a change in VDT height from desktop height to an elevated position affected cervical spine flexion moments and postural angles of the head and neck in the sagittal plane.

Method

Subjects

Twenty-seven subjects (3 male, 24 female), with a mean age of 36.7 years (SD = 6.0, range=25-47), participated in the study on a volunteer basis. All subjects read and signed an informed consent document approved by the Allegheny University of the Health Sciences (formerly the Medical College of Pennsylvania/Hahnemann University) Human Subjects Committee. To be included in the study, subjects were required to spend a minimum of 3 hours per day using a VDT while seated at their place of work and have no history of medical problems that may have affected their posture or spinal range of motion. For the main part of the study, data were obtained from 27 subjects. Characteristics of the subjects are given in Table 1.

Table 1.
Characteristics of Subjects (N=27)

Characteristic    [bar] X        SD                Range

Age (y)           36.7           6.0                25-47
Weight (kg)(a)    76.3 (168.2)   24.0 (52.8)   49.9-155.5 (110-342)
Height (cm)(b)   167.9 (66.5)     7.9 (3.1)    156.2-188.0 (61.5-74)
Sex
 Female            24
 Male               3

(a) Weight in pounds shown in parentheses. (b) Height in inches shown in parentheses.

Instrumentation

A Minolta Maxxum 7000 camera(*) (35-mm) with a 50-mm lens was mounted on a tripod and used to obtain lateral-view slides of subjects while they were seated at a VDT workstation performing a computer task that required use of the keyboard. Floor markers and a plumb line hung from the camera lens were used to ensure a direct lateral view of each subject. A laser pointer secured parallel to the camera lens was used to ensure that the camera lens was pointed directly at the subject, so that the lens surface was parallel to the lateral plane of the subject and the workstation. A vertical plumb line with a 25.4-cm (10-in) length reference was hung from the ceiling directly behind the subject's chair (Fig. 1). A standard slide projector was used to project images. Slide images were projected onto a flat wall at a 90-degree angle from a horizontal projector. A clear acetate grid was then placed over the image to provide vertical and horizontal reference lines. A goniometer was used directly over the images to determine the following angles: upper thoracic-plane inclination, neck inclination, head inclination, lower cervical angle, upper cervical angle, line of sight, and cervical spine moment angle (Figs. 2 and 3).

[Figure 1 to 3 ILLUSTRATION OMITTED]

Procedure and Data Analysis

A VDT[dagger] was positioned on a computer tabletop (96.5 cm from floor to midscreen) or in an elevated position on a 12.7-cm (5-in) platform resting on the tabletop. The keyboard rested on a platform 67.3 cm (26.5 in) from the floor. The height of the tabletop surface upon which the VDT rested was 73.7 cm (29 in) from the floor. The computer desk used was chosen because it represented a typical workstation desk height, based on our experience. The VDT had a rectangular screen measuring 26.7x19.7 cm (10.5x7.75 in). The 35-mm camera mounted on a tripod was positioned 213.4 cm (84 in) from the subject and was adjusted so that the camera height approximated the shoulder height of the subject.

Markers were applied to the subjects in order to allow identification of anatomical landmarks. Small adhesive paper markers were placed on the subjects 10 mm (0.4 in) anterior to the supratragal notch of the ear (the center of mass of the head and neck)[16] and over the infraorbital notch. The T-1 spinous process was palpated by first identifying C-7 and then counting down one adjacent spinous process inferiorly. The location of C-7 was confirmed by palpating for the movement of C-6 and the relative lack of movement at C-7 during cervical extension.[17] Cylindrical paper markers with adhesive ends were placed on the spinous process of T-1 and on the suprasternal notch such that they projected in a perpendicular direction from the body surface.

A preliminary reliability study of the measurement procedure was performed with nine subjects. Each subject was seated at the workstation in an adjustable-height office chair, with the height adjusted to the subject's preference. With the VDT positioned on the desktop, the subject adjusted the screen tilt angle to his or her preference. The VDT's base allowed for upward or downward tilt of the face of the screen, ranging from 1 degree of downward tilt to 12 degrees of upward tilt. The screen tilt was measured with an inclinometer[double dagger] and recorded. The subject was instructed to assume his or her typical, relaxed sitting posture. The subject then performed a computer task requiring keyboard use and screen observation for 10 minutes. After the first 5 minutes, slides were taken at 60-second intervals for a 5-minute period. After a 5-minute rest period, the procedure was repeated with the same chair height, screen height, and screen tilt.

Intraclass correlation coefficients (ICC[3,1]) were used to determine measurement reliability. The formula used to calculate the ICCs was based on a repeated-measures analysis-of-variance design, with the tested rater considered the only rater of interest.[18] Standard deviations and standard errors of measurement (SEMs) were also calculated for head and neck angles and flexion moment on the cervical spine. Using the SEM values, 95% CIs were also calculated (95% CI = [+ or -] 2 SEMs). For the main study, paired t tests were used to determine whether there was a change in postural angles and flexion moment when the VDT was raised from the desktop to an elevated position. All statistical procedures were performed using SPSS for Windows, version 6.0.[sections]

The ICC values, SDs, SEMs, and 95% CIs are listed in Table 2. The ICCs indicate good ([is greater than] .75) measurement reliability[18] for all head and neck angles and flexion moments, with the exception of upper thoracic-plane inclination (ICC=.68). This low ICC value was influenced by subject 2, who assumed a more upright trunk posture during the second sitting, thus creating a much smaller upper thoracic-plane inclination angle of 24 degrees, versus 34 degrees during the first test period. Because only nine subjects were used in the preliminary measurement reliability study, the large variance demonstrated by subject 2 in this particular angle had a large effect on the ICC value. The ICC value for the same angle, excluding subject 2 from the calculation, was .88. Most subjects demonstrated good reliability with their upper thoracic inclination when asked to assume their natural, relaxed sitting posture in the office chair used in the experiment. With the exception of subject 2, change in this angle ranged from 0.4 to 3.6 degrees. Subject 2 demonstrated a large change in this angle (10[degrees]), suggesting that she was inconsistent in the amount of trunk flexion she assumed when asked to assume a natural, relaxed sitting posture in the chair provided.

Table 2.
Results of Repeated-Measures Analysis of Variance

Variable                  ICC            SD            SEM

Line of sight             .98     8.3 [degrees]    1.31 [degrees]
Head inclination          .99     9.4 [degrees]    0.94 [degrees]
Upper cervical angle      .96     8.8 [degrees]    1.76 [degrees]
Lower cervical angle      .98     8.4 [degrees]    1.19 [degrees]
Neck inclination          .86     6.6 [degrees]    2.47 [degrees]
Upper thoracic-plane
 inclination              .68     5.3 [degrees]    3.0 [degrees]
Moment angle              .84     6.0 [degrees]    2.4 [degrees]
Flexion moment            .99     2.45 ft-lb       0.245 ft-lb

Variable                   95% CI

Line of sight           [+ or -] 2.62 [degrees]
Head inclination        [+ or -] 1.88 [degrees]
Upper cervical angle    [+ or -] 3.52 [degrees]
Lower cervical angle    [+ or -] 2.38 [degrees]
Neck inclination        [+ or -] 4.94 [degrees]
Upper thoracic-plane
 inclination            [+ or -] 6.00 [degrees]
Moment angle            [+ or -] 4.80 [degrees]
Flexion moment          [+ or -] 0.49 ft-lb

For the main part of the study, a random selection of desktop or elevated screen positions was made. The subjects adjusted the chair height and screen tilt angle to their preference. The chair heights, measured from the floor to the midpoint of the depth of the seat pan, ranged from 41.9 to 52.1 cm (16.5-20.5 in). The chair used in this study had an available height range of 41.9 to 53.3 cm (16.5-21 in). Screen tilts were measured and recorded. Subjects were instructed to assume their natural, relaxed sitting posture while they performed a computer task requiring keyboard use and screen observation for 10 minutes. This task consisted of typing from memory. After the first 5 minutes, slides were taken at 60-second intervals for a 5-minute period. Subjects were then given a 10-minute break, during which the screen height was changed. Subjects then returned to the workstation with the same seat height as chosen earlier. The screen tilt was again adjusted to the subjects' preference, and this value was recorded. The computer task and photography were then repeated as described.

Thus, five slides for each screen position per subject were produced.

At a later time, the slides were projected onto a flat vertical wall. The size of the projected image was approximately 76x51 cm (30x20 in). An acetate grid was placed over the image, and a goniometer was used (Fig. 2) to measure the following angles (Fig. 3):

1. Line of sight: formed by a line connecting the center of the eye to the top edge of the VDT screen and a line connecting the tragus and the infraorbital notch (the Frankfurt plane[14]). This angle was considered positive when the line connecting the center of the eye to the screen was above the Frankfurt plane and negative when it fell below the Frankfurt plane.[14]

2. Upper thoracic-plane inclination: formed by a line connecting the T-1 spinous process to the suprasternal notch and the horizontal reference.

3. Neck inclination: formed by a line connecting the T-1 spinous process to the tragus and the horizontal reference.

4. Head inclination: formed by a line connecting the tragus to the infraorbital notch and the horizontal reference. Positive angles indicated that the line connecting the tragus and the infraorbital notch was above the horizontal reference, and negative values indicated that this line was below the horizontal reference.

5. Lower cervical angle: formed by a line connecting T-1 to the tragus and a line connecting T-1 to the suprasternal notch.

6. Upper cervical angle: formed by a line connecting T-1 to the tragus and a line connecting the tragus to the infraorbital notch.

7. Moment angle: formed by a line connecting T-1 to the center of mass of the head[16] and the horizontal reference.

The length from T-1 to the center of mass of the head was also measured and recorded. The top edge of the VDT screen was chosen as a landmark of convenience. Although a computer user's gaze may be directed to different points on the screen during VDT operation, one consistent point on the screen was required to measure the line of sight. Although this measurement may not correspond exactly to the true line of sight during VDT operation, it remains useful as a reference to determine relative change in line of sight between the two screen positions.

Data from the preliminary study were used to determine the reliability of the measurements. The five results for each angle from the first test period were averaged and compared with the mean angles obtained during the second test period. Using a line from the head's center of mass (identified by sighting the marker placed 10 mm anterior to the supratragal notch of the ear[16]) to T-1 to describe neck length (NL), the weight of the head and neck (7.9% of body weight [BW]),[16] and the mean flexion moment angle on the cervical spine ([Theta]), the cervical flexion moment (FM) was calculated for each sitting with the following equation (Fig. 4):

[Figure 4 ILLUSTRATION OMITTED]

(1) FM = (.079)(BW)(cos [Theta])(NL)

The calculated flexion moments from the two test periods were compared.

The method used to calculate moment assumed the axis of rotation to be at the T-1 spinous process. In reality, the axis for motion is probably closer to the body of T-1. In previous work,[17] Black et al used data obtained from a three-dimensional digitizing method to estimate the center of the T-1 vertebral body. Using data from that study, we compared the moment angle calculated using the T-1 spinous process point with the moment angle calculated using the T-1 vertebral body point and found the relationship to be quite linear (r =.95). This relationship can be expressed in the form:

(2) [[Theta].sub.T-1body] = 0.652 [[Theta].sub.T-1spinous process] + 43.5

where [Theta] = moment angle (in degrees). Because we were only concerned with the relative comparisons in moment between the high and low screen conditions and not the actual values, we considered our method to be appropriate.

In the main study, the five results for each angle for each screen position were averaged to determine representative mean angles for each of the two screen positions. Cervical flexion moments were calculated as described. Five repeated measurements were used for each screen position to account for postural variation over time, assuming that subjects would not rigidly maintain a static posture during the test period. We noted, however, that any changes in the observed angles were typically small (ie, 1 [degree] or 2 [degrees]).

Results

Table 3 gives the mean values, standard deviations, and ranges for each dependent variable in each screen position. The results of the t tests, including two-tailed significance values and 95% CIs, are shown in Table 4. In addition to head and neck angles and flexion moments, a paired t test was performed on the screen tilt angle. Differences (P [is less than] .05) occurred with screen tilt, line of sight, upper cervical angle, lower cervical angle, head inclination, neck inclination, and moment angle. Screen tilt angle was less when the VDT was in the elevated position (mean change = 6.0 [degrees]). Line of sight values were less negative in the elevated position, indicating; less downward gaze (mean change = 8.5 [degrees]). The upper cervical angle became greater with the VDT in the elevated position, indicating extension of the head relative to the neck (mean change = 2.3 [degrees]). The lower cervical angle became greater with the VDT in the elevated position, indicating extension of the neck relative to the thorax (mean change = 1.0 [degrees]). Head inclination became greater with the screen in the elevated position, indicating posterior rotation of the cranium relative to the horizontal (mean change = 3.1 [degrees]). Neck inclination also became greater when the VDT was in the elevated position, indicating that the cervical spine became more vertically oriented (mean change = 0.9 [degrees]). An increase in moment angle was seen with the screen in the elevated position, indicating that the line connecting T-1 with the center of mass of the head and neck became more vertically oriented (mean change = 0.8 [degrees]). No differences occurred with flexion moment (mean change = 0.025 ft-lb) or upper thoracic-plane inclination (mean change = 0.1 [degrees]).

Table 3.
Postural Angles, Flexion Moment, and Screen Tilt

                                     [bar] X              SD

Variable                         Low(a)   High(b)   Low(a)   High(b)

Screen tilt angle ([degrees])       9.6     3.6      2.8       2.6
Line of sight ([degrees])         -13.8    -5.4      7.2       7.8
Head inclination ([degrees])        5.3     8.4      6.9       7.1
Upper cervical angle
 ([degrees])                      139.1   141.4      9.8      11.1
Lower cervical angle
 ([degrees])                       71.7    72.7      7.6       7.3
Neck inclination ([degrees])       47.5    48.4      6.3       6.2
Upper thoracic-plane
 inclination ([degrees])           24.5    24.5      5.1       5.2
Moment angle ([degrees])           46.0    46.8      6.1       6.0
Flexion moment (ft-lb)              5.0     5.0      2.0       2.1

                                             Range

Variable                             Low(a)           High(b)

Screen tilt angle ([degrees])      2.5 to 12.5      -1.0 to 12.0
Line of sight ([degrees])        -29.0 to 5.3      -20.4 to 17.2
Head inclination ([degrees])     -14.8 to 17.8     -12.4 to 22.4
Upper cervical angle
 ([degrees])                     121.4 to 162.8    121.2 to 171.4
Lower cervical angle
 ([degrees])                      52.4 to 86.6      56.6 to 90.0
Neck inclination ([degrees])      32.2 to 57.0      32.8 to 57.4
Upper thoracic-plane
 inclination ([degrees])          15.4 to 34.2      16.0 to 37.8
Moment angle ([degrees])          32.2 to 55.0      32.8 to 55.4
Flexion moment (ft-lb)             2.9 to 12.1       2.7 to 12.3

(a) Screen in desktop position (96.5 cm [38 in] from floor to midscreen).

(b) Screen in elevated position (109.2 cm [43 in] from floor to midscreen

Table 4.
Results of Paired t Tests for Effect of Change in VDT
Screen Height

                                         Mean
Variable                              Difference        t        df

Screen tilt                         -6.0 [degrees]     -9.50     26
Line of sight                       -8.5 [degrees]    -14.19     26
Head inclination                    -3.1 [degrees]     -5.46     26
Upper cervical angle                -2.3 [degrees]     -3.24     26
Lower cervical angle                -1.0 [degrees]     -2.28     26
Neck inclination                    -0.9 [degrees]     -2.61     26
Upper thoracic-plane inclination    -0.1 [degrees]     -0.18     26
Moment angle                        -0.8 [degrees]     -2.43     26
Flexion moment                       0.025 ft-lb        0.75     26

Variable                              p(a)      95% CI [degrees])

Screen tilt                          .000        -7.3, -4.7
Line of sight                        .000        -9.7, -7.2
Head inclination                     .000        -4.2, -1.9
Upper cervical angle                 .003        -3.8, -0.9
Lower cervical angle                 .031        -1.9, -0.1
Neck inclination                     .015        -1.7, -0.2
Upper thoracic-plane inclination     .858        -1.0, +0.9
Moment angle                         .022        -1.4, -0.1
Flexion moment                       .458        -0.04, +0.09

(a) Two-tailed test for significance.

Discussion

Effects of Changing VDT Height

The angular change in the line of sight was greater than the change in any of the head and neck angles. We believe that the subjects in our study used visual accommodation (ie, rotation of the eyes within their sockets) to adjust to the change in screen height. The mean line of sight was -13.8 degrees in the low screen position and -5.4 degrees in the elevated screen position. Grandjean et al[8] reported preferred viewing angles are between -4 and -14 degrees. The line-of-sight results from our study and those of Grandjean and colleagues cannot be directly compared because of differences in the measurement of viewing angles. Grandjean et al used a horizontal reference as well as a center-screen reference, whereas we used the Frankfurt-plane reference. The Human Factors Society has suggested that viewing angles above a horizontal reference may increase muscle stress and spinal pressure.[12] Hill and Kroemer[14] reported a preferred mean viewing angle of -28.6 degrees when the Frankfurt plane was fixed in a horizontal position. Pheasant[4] suggests that the preferred zone for visual displays is from 0 to 30 degrees below a horizontal reference. It has also been reported that visual stress is reduced with downward gaze.[15] Consideration of preferred viewing angles suggests that shorter subjects may want to avoid elevated screen positions because such positions may result in a line of sight above a horizontal reference.

In our study, subjects adjusted the screen tilt to their preference and demonstrated a tendency to tilt the screen upward from vertical when the screen was in the low position. Change in line of sight and change in screen tilt were used by most subjects to accommodate for the change in screen height. In our experience, most VDT users have adjustable-tilt screens in their work environment. By allowing the subjects to adjust screen tilt in this study, the workstation and work posture were more representative of real work situations. If adjustable tilt is not available, the effects of screen height may be different.

In addition to changes in line of sight and screen tilt, most head and neck angles showed a statistically significant change. The only measured angle that did not show a change was upper thoracic-plane inclination. This finding suggests that trunk posture was relatively stable between screen heights. The upper thoracic-plane inclination may be thought of as the base of the cervical spine and the head. Thus, the base remained stable and head and neck angles changed when the screen height was manipulated.

Although there were differences for all angles other than upper thoracic-plane inclination, the actual differences (see 95% CIs in Tab. 4) were small. For example, a t-test value of -5.46 with a two-tailed significance of .0001 was calculated for head inclination. The 95% CI, however, for change in this angle between screen heights is -4.2, -1.9 degrees. Applying the same consideration to the other head and neck angles, it can be seen that the angular changes were rather small when considering the range of motion possible. For example, the amount of sagittal-plane rotation possible in the lower cervical spine is approximately 66 degrees.[19] The SEMs and CIs (Tab. 2) suggest that some of the statistically significant angular changes may be within a range attributable to measurement error, which further brings into question their meaningfulness. Even a small change in angles may create changes in the forces involved that are clinically or physiologically consequential. Therefore, we considered the flexion moment on the cervical spine as a more useful measure of postural stress.

In our opinion, the values of the head and neck angles cannot be directly interpreted as postural stress on the cervical spine. Angular measurements serve only to describe a position in space. To determine whether the change in angles resulted in a change in postural stress on the cervical spine, a biomechanical model of stress, such as cervical spine moment, needs to be considered. Changes in cervical rotation or upper thoracic-plane inclination can potentially affect the moment on the cervical spine. The moment angle used to calculate flexion moment was affected by postural angles of the head and neck and described the position of the head's center of mass relative to its base of support.

In our study, moment was considered to be a useful variable for describing postural stress on the cervical spine. Because the head's center of mass (a point 1.02 cm [0.4 in] anterior to the supratragal notch) was located anterior to the first thoracic vertebra in all subjects when they were seated at the workstation, a flexion moment on the cervical spine was created. Both contractile and noncontractile tissues in the cervical spine need to provide a counterbalancing force to keep the head's center of mass from moving farther anteriorly. Thus, increasing amounts of flexion moment may increase postural stress on posterior cervical tissues. There was no change in flexion moment when the screen height was changed. Although there were changes in other variables, including most head and neck angles, line of sight, and screen tilt, the composite of these changes did not result in a meaningful change in the flexion moment on the cervical spine between the two screen positions.

Calculation of Pearson product-moment correlation coefficients between flexion moment and postural angles showed a high correlation between moment angle and flexion moment (r = - .70 in the low screen position and r = - .76 in the elevated screen position). The paired t test showed no differences in moment between VDT screen heights, although differences were found for moment angle. The equation for flexion moment considered neck length. With this factor included in the calculation, there was no difference in the flexion moment between screen heights. It was observed that there was a small decrease in measured neck length when the screen was in the low position. This finding probably accounts for why no change was seen for the calculated moment, despite a change in moment angle.

Limitations

Because subjects recruited for this study spent a minimum of 3 hours of a typical workday at a VDT workstation and were instructed to assume their natural, relaxed sitting posture, we assumed that their postural habits were well established and that their posture during the study was representative of how they would accommodate to changes in screen height if the changes occurred in their work setting. Our study, however, was performed in a laboratory setting, and the subjects were aware of the fact that they were being observed and photographed. There is some possibility that their postural habits and means of accommodating to a change in screen height may be different in a real work situation.

The use of eyeglasses was not considered in our study. Subjects wearing corrective lens, especially bifocals, may have accommodated differently to the change in screen height. Several subjects in our study wore corrective lens, but this variable was not included in the data analysis. Generalizability of the results of our study to real work situations is also limited by the duration of time that subjects were observed. The method used in our study did not consider what would happen to postural angles and flexion moments over periods of time longer than 10 minutes. All subjects spent at least 3 hours of a typical workday using a computer. Whether their postural angles and flexion moments would remain consistent over a period of several hours needs to be tested experimentally. The results of our study cannot be generalized to longer periods of VDT operation. The results of our study indicate that for 10-minute periods of time, VDT screen height does not have a meaningful effect on cervical spine flexion moments. Further study is needed to determine the effects of prolonged periods of computer use on postural angles and flexion moments. The method used in our study would not easily lend itself to recording postural angles over prolonged periods of time. The use of a video image computer processing system such as that described and used by Green et al[20] would be more conducive to recording posture over prolonged periods.

The model used in our study considered the cervical spine as a single functional unit. Cervical segmental mobility issues and biomechanical stresses on individual motion segments were not considered. The measurement system used in our study was not sensitive to discrete changes in individual motion segments of the cervical spine that may have occurred with the change in screen height.

Conclusions

The results of our study demonstrate that although many postural angles of the head and neck show a statistically significant change when the VDT screen is moved from desktop height to an elevated position, the flexion moment on the cervical spine does not change between the two positions. Considering flexion moment on the cervical spine as a biomechanical indicator of postural stress, it does not appear that the elevated screen height position used in our study offers any advantage compared with desktop screen position in terms of reducing postural stress on the cervical spine during short (10-minute) periods of VDT operation. Screen tilt and line-of-sight changes were the primary means of accommodation to the change in VDT screen position. For short periods of VDT operation, the two screen heights tested do not appear to have a meaningful effect on flexion moment on the cervical spine. The best determinant of screen height within this range of heights may be subject preference.

Further study to determine the relationship between postural stress and a wider range of screen positions than those used in our study is warranted. The effects of longer periods of VDT operation on postural angles and cervical flexion moments also need to be studied.

Prolonged static loading and the lack of task diversity may contribute to physiological and mechanical stresses on tissues. Although ergonomically correct workstations and unconstrained postures are important in minimizing biomechanical stress on the VDT operator, exercises and increased task diversity may be just as important in presenting cumulative stress and overuse syndromes. More research that investigates the effects of postural variation, exercise, and task diversity of VDT operators is needed.

(*) Minolta Corp. 101 Williams Dr, Ramsey, NJ 07446.

([dagger]) Gateway 2000, 610 Gateway Dr, North Sioux City, ND 57049-2000.

([double dagger]) Fabrication Enterprises Inc, 50 S Buckhout St, Irvington, NY 10533.

([sections]) SPSS Inc, 444 N Michigan Ave, Chicago, IL. 60611.

References

[1] Grieco A. Sitting posture: an old problem and a new one. Ergonomics 1986;29:345-362.

[2] Arndt R. Working posture and musculoskeletal problems of video display terminal operators: review and reappraisal. Am Ind Hyg Assoc J. 1983;44:437-446.

[3] McPhee B. Deficiencies in the ergonomic design of keyboard work and upper limb and neck disorders in operators. J Hum Ergol (Tokyo). 1982;11:31-36.

[4] Pheasant S. Ergonomics, Work, and Health. Gaithersburg, Md: Aspen Publishers Inc; 1991.

[5] Starr SJ, Thompson CR, Shute SJ. Effects of video display terminals on telephone operators. Hum Factors. 1982;24:699-711.

[6] Sauter S, Schleifer L, Knutson S. Workposture, workstation design, and musculoskeletal discomfort in a VDT data entry task. Hum Factors. 1991;33:151-167

[7] Shute SJ, Starr SJ. Effects of adjustable furniture on VDT users. Hum Factors. 1984;26: 157-170.

[8] Grandjean E, Hunting W, Piderman M. VDT workstation design: preferred settings and their effects. Hum Factors. 1983;25:161-175.

[9] Grandjean E. Postures and the design of VDT workstations. Behaviour and Information Technology 1984;3:301-311.

[10] Johnson B, Melius J. Review of NIOSH's VDT Studies and Recommendations: NIOSH Publications on Video Display Terminals. Morgantown, WV: National Institute for Occupational Safety and Health; 1987.

[11] Chaffin D, Andersson G. Occupational Biomechanics. New York, NY: John Wiley & Sons Inc; 1991.

[12] American National Standard for Human Factors Engineering of Visual Display Terminal Workstations. Santa Monica, Calif: The Human Factors Society Inc; 1988.

[13] Enwemeka CS, Bonet IM, Ingl JA, et al. Postural correction in persons with neck pain, part 1: a survey of neck positions recommended by physical therapists. J Orthop Sports Phys Ther. 1988;65:1173-1176.

[14] Hill SG, Kroemer KH. Preferred declination of the line of sight. Hum Factors. 1986;28:127-134.

[15] Kroemer KHE, Kroemer HB, Kroemer-Elbert KE. Ergonomics. Englewood Cliffs, NJ: Prentice Hall; 1993.

[16] Dempster WT. Space requirements of the seated operator: geometrical, kinematic, and mechanical aspects of the body with special reference to the limbs. Wright Air Development Center Technological Report. 1955:55-159.

[17] Black KM, McClure P, Polansky M. The influence of different sitting positions on cervical and lumbar posture. Spine 1996;21:65-70.

[18] Portney L, Watkins M. Foundations of Clinical Research: Applications to Practice. East Norwalk, Conn: Appleton and Lange; 1993:509-516.

[19] White A, Panjabi M. Clinical Biomechanics of the Spine. Philadelphia, Pa: JB Lippincott Co; 1978:61-90.

[20] Green RA, Briggs CA, Wrigley TV. Factors related to working posture and its assessment among keyboard operators. Applied Ergonomics. 1991;22:29-35.

DM Kietrys, PT, is Physical Therapist, Rehabilitation and Fitness Services Inc, 2503 Lombard St, Philadelphia, PA 19146 (USA). Address all correspondence to Mr Kietrys.

PW McClure, PhD, PT, OCS, is Assistant Professor, Department of Physical Therapy (MS 502), Allegheny University of the Health Sciences, Philadelphia, Pa.

GK Fitzgerald, PT, OCS, is Assistant Professor, Department of Physical Therapy, University of Delaware, Newark, Del.

This study was approved by the Allegheny University of the Health Sciences Human Subjects Committee.

This article was submitted January 24, 1997, and was accepted October 1, 1997.