The working postures among schoolchildren--a controlled intervention study on the effects of newly designed workstations.
Schoolchildren's working postures and workstations have been a neglected area in research. In today's society, the growing use of information and communication technology has contributed to increasingly sedentary lifestyles. After first sitting for several hours at school, often in stooped postures stressing musculoskeletal structures, children often continue to sit during their free time, for example, using computer or watching television. Schools have the opportunity to contribute to musculoskeletal health by means of offering workstations that meet children's anthropometric requirements.
Due to the tremendous variance in stature in puberty among children in the same school grade, workstations of just 1 size are not appropriate for all. There is evidence of a mismatch between school workstations and schoolchildren's anthropometrics; desks and chairs are too high (1,2), or desks are too high and chairs too low. (3)
An upright, neutral sitting posture, exerting least load on spinal structures, is generally believed to bring health-related benefits. Upright, neutral sitting posture during working has generally been defined by the magnitude of lumbar spine lordosis and the position of the neck during sitting. Lumbar spine lordosis during sitting has been proposed to be neutral when its magnitude is the nearest approach to the lumbosacral curve in a standing position. (4) Neck position has been proposed to be neutral, (5) or vertical, (6) when the straight-line landmark extends from the point of the lowest cervical vertebra (5) to the mastoid process.
The effects of various sitting postures on the angles of the lumbar spine were reported as early as 1950s by Keegan, (4) who found that due to varying angles between the trunk and the thigh in different sitting postures, alterations in the lumbar spine occur. Three decades later, Mandal (7) showed that a higher sitting and working position enables one to maintain a neutral posture of the lumbar spine and suggested that school desks should be at least one half and chairs one third of the person's height. More compressive loads on lumbar structures have been reported during unsupported sitting than during standing, one potential reason being the increased force level in the extensor musculature by a more flexed posture of the lumbar spine. (8) Also, Hedman and Fernie (9) found a higher level of stress on lumbar structures during kyphotic more than lordotic spine postures.
Upright, neutral sitting posture has been obtained in some school studies using a novel workstation design with more sloping desks and chairs. Compared to flat ones, desktops with a 15-degree slope and chairs with a 15-degree forward slope allow less neck flexion and a larger angle between trunk and thigh while working at the desk. (10) In adults, the lumbar spine position was less kyphotic on a high chair (135-degree angle between trunk and thigh) than on a low chair (4,11) and an angle as low as 115 degrees between trunk and thigh, together with a chair with a 15-degree forward slope, seemed to reduce lumbar flexion. (12) Further, desks with a 10-degree inclination have been reported to decrease sagittal flexion of the head and the trunk compared to working at a flat desk. (13) Contrary to the studies of Bridger et al (12) and de Wall et al (13) on adults, school workstations with more height, sloped desks, and curved chairs did not affect schoolchildren's actual sitting behavior. (14) The authors emphasized the need for proper instructions for the children and adjustment of furniture to achieve better working postures and indicated that ergonomic workstations alone do not seem to improve the sitting habits. (14)
There has been an attempt to determine the optimal adjustment of desks and chairs to meet children's anthropometric requirements and decrease static muscle loads. With an adult person seated on the chair, the desk should be adjusted appropriately in relation to the elbow position to minimize shoulder elevation and static levels of shoulder muscle load. (11) Consequently, the recommended desk height in adults is about 5 cm (11) or 5-7 cm (15) above elbow level if an arm-wrist support is available. For children, such recommendations are absent.
Working postures of schoolchildren have been studied little. According to Murphy et al, (16,17) flexed postures of the trunk and the neck and according to Saarni et al, (3) flexed or flexed/rotated back and neck postures during sitting are common using conventional workstations. The percentage of time spent with trunk and neck flexed more than 20 degrees has also been associated with low back pain and increased neck pain in taller schoolchildren. (17) In adults, occupying sedentary position for over 95% of the working time was found to be significantly associated with neck pain. (18) Ergonomically designed and individually adjusted workstations may prevent musculoskeletal symptoms in adults. (19)
This study investigates the effects of individually adjustable saddle-type chairs and desks with comfort curve to accommodate body and provide arm support (Figure 1) on schoolchildren's sitting postures in a 1-year controlled intervention. The hypothesis was that the newly designed workstations, compared to conventional ones, increase trunk-thigh angles and back and neck upright, neutral postures among participants during sitting.
[FIGURE 1 OMITTED]
Subjects and Procedure
This study was a controlled intervention with 1-year follow-up between February 2002 and March 2003. At baseline, participants were all sixth (mean, 12 years) and eighth (mean, 14 years) graders in 2 Swedish-speaking comprehensive schools in Finnish cities. The schools were comparable with respect to neighboring cities, minority language, and social background. Participants at baseline totaled 101, 46 in the intervention school and 55 in the control school. Two participants (boys) moved from the control school to the intervention school after summer vacation in August 2002. These 2 boys were allocated to the intervention group because their exposure time to the new furniture was only 2 months shorter than that of other participants in the intervention group. Four boys were excluded on account of moving to another municipality, 3 from the control group and 1 from the intervention group. A total of 97 participants completed the study, 47 in the intervention school and 50 in the control school. At baseline, participants of both groups reported similarly levels of sports, computer use, stage of pubertal maturation, amount of schoolwork, mean grades of school reports, and absence due to illness.
Description of Workstations
In April 2002, after baseline measurements, the intervention group was equipped with new, adjustable saddle-type chairs with wheels and adjustable desks with comfort curve for the body (Easydoing Co, Rautalampi, Finland) (Figure 1). The aim was to adjust the new design workstations as optimally as possible to match the anthropometrics of each child. The space between the adjusting holes (4 cm) somewhat limited the adjustability of the chair and desk heights. The adjustment was performed starting with the bottom hole and then increasing the height 1 step at a time until the closest match to the child's anthropometrics was found. The elbow-floor heights (mean difference between desk height and elbow-floor height) were set to 5-7 cm (11,15) and the trunk-thigh angles to 115-135 degrees. (4,11) The built-in constant tilt angle of the new desks did not permit adjustment, and, consequently, the difference between the new and the conventional desks remained negligible (Table 2).
As the participants of the intervention school were able to adjust their desks and chairs, the researcher advised them not to self-adjust in order to avoid a conflict between anthropometrics and workstations. The matches between the elbow-floor height and the desk height and those between the trunk-thigh angle and the chair height of each participant were checked on average every 2 months. In addition, workstations were readjusted according to the individual anthropometrics of the child when necessary.
The control group continued using their conventional workstations throughout the intervention. Some workstations were adjustable by height. The school janitor performed some height adjustments in the beginning of the school year, but many participants were also able to adjust the desk slope (2 tilt angles).
Unlike in most Finnish schools, these schools were comparable in 2 accounts: each grade had a home classroom and used the same desks and chairs for most lessons. In the intervention school, the new workstations were placed in the home classroom. When deducting the hours regularly spent, according to the timetables, in other classrooms for craft, physical education, and home economics/domestic science, there remained an annual average of 27.3 hours per week at the most to be spent in the home classroom at the intervention school and 27.8 hours at the control school, correspondingly. In addition, other lessons occasionally held outside the home classroom were further deducted from the annual average rates. The net total exposure time was 14.3 hours in the intervention group and 17.6 hours in the control group. The difference in the exposure time between the groups was statistically significant (p = .001).
Anthropometrics and Workstation Dimensions
The height (stature), weight, and sitting height of each participant were measured at baseline and after the 1-year follow-up. The height was determined as the vertical distance from floor to top of the head and measured with participant standing without shoes, erect, and looking horizontally straight ahead. The plastic measuring instrument was wall mounted, and the participant stood with the back against the measuring instrument. Relative growth was calculated as follows: height during intervention (height at follow-up minus height at baseline) divided by height at baseline. Sitting height was determined as the vertical distance from seat surface to top of the head and measured with participant sitting erect on a flat seat, with knees bent 90 degrees and the back against the measuring instrument. Weight was determined with digital weighing scales. Height, weight, and sitting height were measured individually in the school nurse's room by the first author (L.S.), trained in physiotherapy, using plastic ruler, metal right angle, and digital scales. At baseline, the anthropometric measures did not indicate a difference between the groups.
Elbow height (seated) was the vertical distance from seat surface to the olecranon tip (under the elbow) and measured with the arm at the side (vertically adducted) and elbow flexion at 90 degrees. The measurements were performed in the participants' home classroom during a lesson by the first and fifth authors (L.S. and A.K.), using plastic ruler and wooden measuring board. Elbow-floor height (seated) was the sum of elbow height and chair height. The difference between the new desk height and the elbow-floor height was compared to the difference between the conventional desk height and the elbow-floor height to ensure optimal difference defined according to the guidelines and proposals. (11,15)
Using wooden and plastic rulers, the desks and chairs were measured in the classrooms by the first and fifth authors (L.S. and A.K.) while participants sat at their workstations. The desk height was the vertical distance from floor to desk front edge (left side). Desk tilt (sin [alpha] = a/b) was calculated using the depth (b) and the height difference of the front and back edges of the desk (a). The height of the conventional chair was determined as the vertical distance from floor to the highest point on the rear end of the seat. The height of the saddle-type chair was calculated by decreasing 1 cm from the highest point on the rear end of the seat (curved, with sideward buckle).
Children's anthropometrics and dimensions of desks and chairs were measured at baseline. During follow-up, elbow height, elbow-floor height, desk height, chair height, and difference between desk height and elbow-floor height were measured at April 2002, August 2002, October 2002, December 2002, and February 2003 in the intervention group and at August 2002, December 2002, and February 2003 in the control group.
Posture Analysis by Video Recordings
Twenty-one schoolchildren from the intervention and 21 from the control group were selected at random from among the participants (N = 101) for video recordings in 2002 and 2003 to be performed in home classrooms during morning or afternoon lessons. During lessons, participants had mathematics, languages, religious education, chemistry, or history. A video camera was positioned on either side, sagittally toward the participant, with field of view from the thigh to the top of the head. The recording time varied due to participants' dynamic moving on and off the chair or to other children blocking the view.
For posture analysis, the modified Ovako Working posture Analyzing System (OWAS) was used, (20) with observation intervals set at 15 seconds. OWAS categories were modified to school sitting postures to include (a) back: straight, flexed, rotated, or flexed and rotated; (b) upper limbs: neither limb supported, 1 supported, or both supported on desk; (c) buttocks and lower limbs: buttocks resting on rear of seat, buttocks resting on front of seat, or standing or walking; and (d) neck: straight, flexed, or rotated.
All postures were recorded in relation to the upright sitting posture, with the back and neck positions defined as straight with flexion [less than or equal to] 20 degrees and rotation [less than or equal to] 45 degrees. The variables used in the analyses were back and neck postures during sitting. The average length of each video recording obtained of 1 participant was 36.5 minutes during 1 lesson. The average number of observations for each child was 133.
The comparison of anthropometrics and workstation dimensions between the intervention group and the control group is presented as means and standard deviations. To analyze working postures, the model fitted for each of the outcome measures was [Y.sub.I] = U + [b.sub.I] + [c.sub.I]t +dz + c for the intervention group and [Y.sub.c] = U + [b.sub.c] + [C.sub.c]t + dz + [epsilon] for the control group. In the model, U was a random intercept term associated with a study participant, [b.sub.I] and [b.sub.c] were intercept terms in intervention and control groups, correspondingly, and [c.sub.I] and [c.sub.c] were slope terms associated with time (t) effects in both groups. Further, d was a coefficient associated with covariate z, and was a residual term. The main hypothesis of interest was whether the intervention effect in time was similar in both groups, ie, [H.sub.O]: [c.sub.I] = [c.sub.c].
The linear mixed-effects model used here processes differences in the postures between the groups during follow-up, also taking into account individual variation between the participants and the correlation structure within the participant's measurements.
Relative growth (height during intervention divided by height at baseline), a potential confounding factor, may have affected the sitting postures. Relative growth was tested in linear mixed-effects models. Statistical analysis to determine the differences for the development of the main outcome variables in time between the control and the intervention groups was performed using linear mixed-effects models and tested by t and F tests given by R-program LME. (21) The basic analysis was performed by the SPSS 11.5 for Windows. The relative growth as a confounding factor had no significant effect on sitting postures.
After written information to the parents and written and verbal information to the children, written informed assent and consent were obtained from the participants and parents, respectively. Written permission was also obtained from the school headmasters after verbal information about the research plan. The Ethical Committee of the Hospital District of Pirkanmaa approved the protocol.
The descriptive statistics of the schoolchildren at baseline and during follow-up are shown in Table 1. The groups did not differ significantly between mean height, weight, and sitting height. The new workstations for the intervention group were significantly higher than the conventional ones for the control group. The measurements of trunk-thigh angles showed no difference between the groups at baseline, but the saddle-type chairs allowed the thighs to incline significantly more downward than the conventional chairs during intervention. Variation in desk tilt was large within the control group (0-16 degrees) and narrow within the intervention group (0-1 degrees). Desk tilt differed significantly between the groups at baseline but not during follow-up (Table 2).
At baseline, upright neck postures were significantly more common (p = .03) in the control group compared with the intervention group. Regarding back postures at baseline, there was no difference between the groups. The proportion of time schoolchildren sat with their back and neck in upright posture ([less than or equal to] 20-degree flexion and/or [less than or equal to] 45-degree rotation) increased more in the intervention group compared to the controls during follow-up (Figure 2). However, the new workstations failed to bring about better sitting postures among all participants, instead, some postures deteriorated during follow-up, although similar development was seen in the control group.
[FIGURE 2 OMITTED]
In Table 3, the time effect on back posture in both groups was significantly higher in the intervention group (p = .01). The time effect on neck posture was positive in the intervention group and negative in the control group. The difference was statistically significant (p = .02). The positive estimates of the time effect on sitting postures are interpreted as a course toward a higher percentage of time in upright sitting postures and the negative estimate toward a lower percentage. Interpretation of the estimates of the model parameters is the effect in unit when all other factors are held fixed.
The new design of the individually adjustable saddle chairs and desks with comfort curve and arm support increased the upright, neutral back and neck postures during sitting at school lessons compared to conventional workstations. This is an improvement from the musculoskeletal point of view. The increase was the result of proper adjustments and the new workstation design. The new workstations force proper elbow support and increase the angle between trunk and thigh, thus enabling a more neutral lumbar position. Despite the somewhat limited adjustability of the new design desks and chairs, optimal relationship between anthropometrics and workstations was mostly obtained. Moreover, the adjusting mechanism was "user-friendly" compared to the conventional workstations. Some participants in the control group were able to adjust their tiltable desk slopes between 2 positions. For the intervention group, self-adjustment of the desk slope was discouraged. The calculated mean desk slope angles differed significantly between the groups at baseline but not during follow-up.
The schools in this study, being composed of a minority language group in their city, were comparable in their cultural features; also, there was continuity of education within the same school complex and grounds for at least grades 1-9. At both schools, most lessons were held in home classrooms. At the intervention school, for practical reasons, the new workstations were placed in the home classroom only, while conventional workstations were maintained in the other classrooms. Consequently, of the total exposure time, only 52.4% of all sitting hours were spent at the new workstations, which may have diluted the effects.
In the intervention group, desks and chairs were adjusted according to the anthropometric dimensions of the participants. It was noted, however, that on a few occasions some participants, or other students, had done readjustments. To ensure maintenance of correct adjustments and to allow for growth, the desks and chairs were adjusted by the first author (L.S.) on average every 2 months. The individual response to the neutral posture of the lumbar and cervical curve showed substantial variation. Despite the new design of workstations, schoolchildren still have their individual sitting and working habits during lessons. Consequently, our posture analysis allowed a certain degree of individual sitting posture variation owing to the neutral posture angle definition tolerances ([less than or equal to] 20 degrees). Moreover, it is possible that additional instructions for the intervention group of the optimal sitting postures might have enabled those showing a poor sitting posture to avail themselves better of the new design.
A working posture video analysis is an appropriate method to evaluate postures of the back and the neck, position of the buttocks on the seat, and position of the upper limbs. (16,18,22) It should be noted that the potential bias associated with subjects being aware that they are being observed may have appeared during the early intervention, video analysis in particular, but it is unlikely that it would extend over the whole 1-year follow-up. The modified OWAS method was used to measure the frequency of sitting postures of the participants at 15-second intervals. In this study, the 15-second intervals and the posture categories were chosen according to earlier studies. (16,18,22) This procedure ensured that the sampling frequency was quite high, and only a few observations were missed during 1 lesson, the school data obtained can be considered unique. Another posture-analyzing system, portable ergonomic observation (PEO) method (22) provides information about posture intensity, duration, and frequency. However, only the duration of posture was missing from OWAS compared to PEO. As described by Murphy et al, (16) the PEO method also seemed to miss some observations compared to video analysis (percentage of time spent in each posture), especially concerning neck postures.
This study showed that adjustable saddle-type chairs and desks with comfort curve contributed to better working postures compared to conventional workstations. The authors propose that these results should be taken into account by decision makers in schools and adopted as part of a healthy school environment, which might lead to improved awareness and development of ergonomically more beneficial school workstations among furniture designers and manufacturers.
(1.) Parcells C, Stommel M, Hubbard R. Mismatch of classroom workstation and student body dimensions. J Adolesc Health. 1999;24:265-273.
(2.) Panagiotopoulou G, Christoulas K, Papanckolaou A, Mandroukas K. Classroom workstation dimensions and anthropometric measures in primary school. Appl Ergon. 2004; 35:121-128.
(3.) Saarni L, Nygard C-H, Kaukiainen A, Rimpela A. Are the desks and chairs appropriate? Ergonomics. In press.
(4.) Keegan J. Alterations of the lumbar curve related to posture and seating. J Bone Joint Surf. 1953;35-A:589-603.
(5.) Harms-Ringdhal K, Ekholm J, Schuldt K, Nemeth G, Arborelius U. Load moments and myoelectric activity when the cervical spine is held in full flexion and extension. Ergonomics. 1986;29:1539-1552.
(6.) Chaffin DB, Andersson GBJ, Martin BJ. Occupational Biomechanics. 3rd ed. New York, NY: John Wiley & Sons, Inc; 1999.
(7.) Mandal A. The correct height of school workstation. Hum Factors. 1982;24:257-269.
(8.) Callaghan J, McGill M. Low back joint loading and kinematics during standing and unsupported sitting. Ergonomics. 2001;44: 280-294.
(9.) Hedman T, Fernie G. Mechanical response of the lumbar spine to seated postural loads. Spine. 1997;22:734-743.
(10.) Marschall M, Harrington A, Steele J. Effect of work station design on sitting posture in young children. Ergonomics. 1995; 38:1932-1940.
(11.) Bendix T, Krohn L, Jessen F, Aaras A. Trunk posture and trapezius muscle load while working in standing, supportedstanding, and sitting positions. Spine. 1985;10:433-439.
(12.) Bridger R, Von Eisenhart-Rothe C, Henneberg M. Effects of seat slope and hip flexion on spinal angles in sitting. Spine. 1989;31:679-688.
(13.) de Wall M, van Riel M, Snijders C. The effect on sitting posture of a desk with a 10[degrees] inclination for reading and writing. Ergonomics. 1991;34:575-584.
(14.) Linton S, Hellsing A-L, Halme T, Akerstedt K. The effects of ergonomically designed school workstation on pupils' attitudes, symptoms and behaviour. Appl Ergon. 1994;25:299-304.
(15.) Finnish Institute of Occupational Health. Dimensions of the workplace. Ergon Bull. 1986;4:3-11.
(16.) Murphy S, Buckle P, Stubbs D. The use of the portable ergonomic observation method (PEO) to monitor the sitting posture of schoolchildren in the classroom. Appl Ergon. 2002;33: 365-370.
(17.) Murphy S, Buckle P, Stubbs D. Classroom posture and self-reported back and neck pain in schoolchildren. Appl Ergon. 2004;35:113-120.
(18.) Ariens G, Bongers P, Douwes M, et al. Are neck flexion, neck rotation, and sitting at work risk factors for neck pain? Results of a prospective cohort study. Occup Environ Med. 2001;58: 200-207.
(19.) Ketola R, Toivonen R, Hakkanen M, Luukkonen R, Takala E-P, Viikari-Juntura E. Effects of ergonomic intervention in work with video display units. Scand J Work Environ Health. 2002;28:18-24.
(20.) Karhu O, Kansi P, Kuorinka I. Correcting working postures in industry: a practical method for analysis. Appl Ergon. 1977;8: 199-201.
(21.) Pinheiro JC, Bates DM. Mixed-Effects Models in S and S-PLUS. New York, NY: Springer-Verlag; 2000.
(22.) Fransson-Hall C, Gloria R, Kilbom A, Winkel J. A portable ergonomic observation method (PEO) for computerized online recording of postures and manual handling. Appl Ergon. 1995;26:93-100.
Address correspondence to: Lea Saarni, Researcher, (firstname.lastname@example.org), School of Public Health, University of Tampere, Medisiinarinkatu 3, Tampere 33014, Finland.
The study was supported by the Academy of Finland and the Yrjo Jahnsson Foundation.
LEA SAARNI, MSc (a)
CLAS-HAKAN NYGARD, PhD (b)
ARJA RIMPELA, MD, PhD, MSc (c)
TAPIO NUMMI, PhD (d)
ANNELI KAUKIAINEN, PhD (e)
(a) Researcher, (email@example.com), School of Public Health, University of Tampere, Medisiinarinkatu 3, Tampere 33014, Finland.
(b) Professor, (firstname.lastname@example.org), School of Public Health, University of Tampere, Tampere, Finland.
(c) Professor, (email@example.com), School of Public Health, University of Tampere, Tampere, Finland.
(d) Lecturer, (firstname.lastname@example.org), Department of Statistics, University of Tampere, Tampere, Finland.
(e) Special Researcher, (email@example.com), Finnish Institute of Occupational Health, Tampere, Finland.
Table 1. Means and Standard Deviations (SD) of Height, Weight, and Sitting Height Baseline Intervention Control Group p Value * Group(n = 47) (n = 50) Height (cm) Mean 163.6 164.3 NS SD 9.1 11.2 Weight (kg) Mean 54.7 57.8 NS SD 10.2 12.5 Sitting height (cm) Mean 84.5 85.8 NS SD 4.9 5.4 Follow-Up Intervention Control Group p Value* Group (n = 47) (n = 50) Height (cm) Mean 168.3 167.9 NS SD 8.7 10.3 Weight (kg) Mean 58.2 62.4 NS ([dagger]) SD 8.8 12.4 Sitting height (cm) Mean 86.6 87.4 NS SD 4.2 5.1 * t Test for equality of means between groups. ([dagger]) n = 49. NS, Not significant. Table 2. Means and Standard Deviations (SD) of Desk Height, Angle of Desk Slope, Trunk-Thigh Angle, Chair Height, Elbow Height, Elbow-Floor Height, and the Difference Between Elbow-Floor Height and Desk Height Baseline Intervention Control School School p (n = 47) (n = 50) Value * Desk height (cm) Mean 71.8 74.6 .000 SD 1.7 3.5 Angle of desk slope (degree) Mean 0.0 2.1 .000 SD 0.2 3.5 Trunk-thigh angle (degree) ([section]) Mean 96.3 98.8 NS SD 3.9 5.4 Chair height (cm) Mean 43.3 43.7 NS SD 1.4 1.0 Elbow height (cm) Mean 16.7 17.3 NS SD 2.7 2.4 Elbow-floor height (cm) Mean 60.0 61.0 NS SD 3.5 2.3 Difference between desk height and elbow-floor height (cm) ([parallel]) Mean 11.8 13.6 .037 SD 4.7 3.7 Follow-up Intervention Control School School p (n = 47) (n = 50) Value * Desk height (cm) Mean 96.3 73.9 .001 SD 5.0 3.1 Angle of desk slope (degree) Mean 4.1 3.9 NS ([dagger]) ([double dagger]) SD 0.7 2.4 Trunk-thigh angle (degree) ([section]) Mean 125.0 99.8 .001 SD 5.8 7.0 Chair height (cm) Mean 68.2 43.8 .001 SD 4.5 2.0 Elbow height (cm) Mean 20.9 20.3 NS SD 2.7 1.9 Elbow-floor height (cm) Mean 89.1 64.1 .001 SD 5.5 2.8 Difference between desk height and elbow-floor height (cm) ([parallel]) Mean 7.2 9.7 .001 SD 2.9 3.9 * t Test for equality of means between groups. ([dagger]) n = 46. ([double dagger]) n = 49. ([section]) n = 20 in both groups, measured from video recordings. ([parallel]) The optimal difference is suggested to be 5 cm (11) or 5-7 cm. (15) NS, Not significant. Table 3. Estimates and Tests for the Fitted Mixed Models for Sitting Posture With Relative Growth as a Confounding Factor Time Effect Time Effect on Intervention on Control Group, Estimate Group, Estimate (t Value) (t Value) Sitting posture * Back posture 50.38 (7.88) 26.73 (4.18) Neck posture 14.26 (2.09) -9.26 (-1.35) F Test for Time Effects Between Groups, Estimate p Value Sitting posture * Back posture 6.84 .012 Neck posture 5.92 .019 * n = 21 in both intervention and control groups.
|Printer friendly Cite/link Email Feedback|
|Title Annotation:||Research Article|
|Author:||Saarni, Lea; Nygard, Clas-Hakan; Rimpela, Arja; Nummi, Tapio; Kaukiainen, Anneli|
|Publication:||Journal of School Health|
|Date:||May 1, 2007|
|Previous Article:||Association of relative backpack weight with reported pain, pain sites, medical utilization, and lost school time in children and adolescents.|
|Next Article:||Chronic pain in the classroom: teachers' attributions about the causes of chronic pain *.|