- Email: [email protected]
An assessment of seated activity and postures at five workplaces
: ........I n d U s t r i a l Ergonomics
ELSEVIER
International Journal of Industrial Ergonomics 15 (1995) 81-90
An assessment of seated activity and postures at five workplaces M. Graf, U. Guggenbi~hl, H. Krueger Institute for Hygiene and Work Physiology, Swiss Federal Institute of Technology, Clausiusstr. 21, 8092 Zurich, Switzerland Received June 30, 1993; accepted in revised form January 21, 1994
Abstract Static working positions and poor postures are both associated with the development of musculo-skelctal disorders and discomfort. An assessment of people's postural behaviour requires suitable measurement tools and scales against which the measure can be judged. Although such methods have been developed, these generally are aimed at assessing gross movements which are undertaken in an industrial setting where the workers arc not sedentary. Seated workplaces are more the norm in modern societies and sedentary work does not provide immunity from discomfort and musculo-skeletal disorders. For this reason we have developed a method for classifying and recording seated postural behaviour. This method is described along with the results of studies undcrtakcn using the method at five workplaces. The studies aimed to obtain some base data on sitting behaviour and how it related to work task and chair type. Postures were recorded at one-minute intervals during the subjects' normal work for periods up to two hours. Additionally, questionnaires regarding comfort and musculo-skeletal problems were issued to some of the subjects. Although individual variation was sizeable, the results showed clear postural behaviour differences between the various work tasks. Work tasks which have a higher incidence of musculo-skeletal disorders were found to produce less frequent and less marked postural change. From these results some preliminary conclusions about an optimal seating behaviour pattern are made. Relevance to industry Back disorders and discomfort are major sources of lost work time and worker dissatisfaction. Many of thc problems can be traced back to poor seating or workplace design. The paper elucidates postural behavioural factors which, along with the physical aspects of workplace furniture, affect people's comfort and well-being at work. Thesc postural aspects are largely determined by the work task.
Keywords: Keywords: Seating; Chairs; Posture; Musculo-skeletal disorder; Activity; Behaviour
I. Introduction W i t h t h e i n c r e a s i n g a u t o m a t i o n o f w o r k over t h e last half c e n t u r y a significant c h a n g e has t a k e n p l a c e in w o r k p l a c e activity in t h e industrialised nations. W h e r e a s once, m o s t w o r k involved m a n u a l tasks u n d e r t a k e n largely o u t d o o r s , t o d a y
m o s t w o r k activity is p e r f o r m e d inside b u i l d i n g s while s e a t e d . G e n e r a l l y m o r e p e o p l e sit much l o n g e r t o d a y t h a n even 20 y e a r s ago. In the i n d u s t r i a l i s e d c o u n t r i e s of W e s t e r n E u r o p e a n d N o r t h A m e r i c a m o r e t h a n half of t h e w o r k f o r c e is e m p l o y e d in a d m i n i s t r a t i o n , services, credit, t r a d e o r i n s u r a n c e i n d u s t r i e s ( G r i e c o , 1986) - all
0169-8141/95/$09.50 © 1995 Elsevier Science B.V. All rights reserved SSDI 0 1 6 9 - 8 1 4 1 ( 9 4 ) 0 0 0 2 7 - Z
82
M. Graf et al. /International Journal of Industrial Ergonomics 15 (1995) 81-90
sitting much of the working day. This change has profoundly altered the common physical demands made by work on the human body. The decrease in heavy physical work has not resulted in fewer musculo-skeletal disorders. These problems are still common in workplaces where no heavy loads are manipulated and where the work is generally described as light. In 1972 Magora found a high incidence of low back pain in persons whose work entailed sitting for prolonged periods or not sitting at all, Hfinting et al. (1981) found that VDT workplaces and typing were associated with physical impairments in the hands, arms, shoulders and neck, particularly where the workplaces were poorly designed, and Ong (1984) found that musculo-skeletal problems were more common amongst VDT operators than a comparative group of conventional office workers. This was attributed to poor furniture design and insufficient rest pauses, along with other factors. With these common musculo-skeletal problems, including lower back pain, shoulder and neck discomfort and upper limb disorders, there is generally no history of the classical single event overloading of the tissues involved. Other factors such as poor seated work postures and poor work design clearly also contribute to the development of these disorders.
1.1. ActiL~e seating For over a hundred years it has been recognised that our seated postures are of great importance to the maintenance of the health of our backs. Nevertheless there is still a great deal of dispute over the definition of good sitting posture. At the beginning of the century people were taught always to sit upright, with a straight back and with their knees and hips forming right angles, irrespective of the task that they were performing or the type of chair that they used. In the early sixties, it was recognised that holding this 'correct' position was not easy and the solution was sought in chair design (Grandjean and Burandt, 1962; Coermann and Rieck, 1964). The chair itself took prominence over the behaviour of the sitter. Chairs were judged favourably if they maintained the body in the previously de-
scribed 'optimum' posture. But, unfortunately, no chair design could be found that reliably and comfortably maintained the body in this posture. Work physicians then began to question whether the stiff upright posture is in fact the best sitting position. Some argued that it is simply impossible to maintain (Branton, 1969; Grandjean and Hiinting, 1977; Bendix, 1986) whereas others argued that it is biomechanically unsound (Corlett and Mananica, 1980; Mandal, 1981; Schobert, 1989). Eventually the idea dawned that no position should be continually held and that there cannot be a single ideal sitting position. In fact the degree of postural fixation, that is, lack of movement, is itself further suspected to be a cause of musculo-skeletal disorders. Cantoni et al. (1984) compared work at an old switchboard with work at a new ergonomically designed VDT switchboard and found a significant increase in the degree of postural fixity. They felt that this was a negative development. The concept of a single ideal posture has still not died completely and every year new products come onto the market which purport to ensure the optimum posture. Usually emphasis is placed on the 'correct' spinal curve which may also require the 'correct' hip position or thigh position. The concept is quite sound that the spinal disks and musculature can be unduly stressed by extreme body movements and postures. Awkward and jerky movements are acutely more dangerous to the spinal tissues than prolonged use. However, natural movements do not stress the body, in fact, they are desirable and necessary as long as they are within an acceptable range. We realize that if the body is kept in a single seated position, a specific set of postural muscles will need to work continually to maintain that posture. These muscles eventually tire because they have no opportunity to relax. Movements, on the other hand, alternate the muscles which are working and thereby create relaxation periods. Relaxation allows the replenishment of nutrient energy, oxygen and the removal of waste products from the tissues. Furthermore, the spinal disks may be dependent on physical activity because their nourishment depends on the pumping action of changing disc pressure from spinal move-
M. Graf et al. / International Journal of Industrial Ergonomics 15 (1995) 81-90
83
ment. Any seated position is therefore today recognized to be unacceptable if it is held for too long, no matter how well the spinal vertebrae are positioned and the body supported. The movement factor is particularly of importance when the task to be performed requires extended periods of sitting, as is the case in most occupations today. The term dynamic seating was coined to describe this idea of changing positions to allow the different groups of postural muscles to rest alternately. However the problem which has arisen is that the expression 'dynamic seating' has different meanings for different people. Some chair manufacturers use the term to describe a type of chair mechanism, others use it to describe chairs which do not stabilise the trunk or which offer only indirect trunk support. With these concepts the emphasis is on the chair and not on the person or how he or she sits. The emphasis should instead be kept on sitting behaviour. We suggest the term active seating is a better descriptor and we prefer its use to describe optimum seated movement behaviour. The problem which has now presented itself is that an adequate definition of good sitting behaviour, active seating, is not readily available. The principle two issues which need to be addressed are: How often should the sitting position change? And what is the optimal range of these changes? By our definition a good chair is one which supports the body in multiple desirable positions. It prevents stress by giving the sitter the opportunity to vary which groups of postural muscles are used. Extreme positions of the spine are avoided by fitting the chair and workspace to the person who is to use it and by designing work tasks to fit the anatomical and physiological limits of the body.
ment. It would seem obvious that task requirements would significantly affect sitting behaviour, that is, the person's postural range and movement frequency, but few studies have been done to document how sitting behaviour is affected by task determinants. The effect of different types of chairs on seating behaviour has also not been studied in this way so it is not known to what extent variations in chair design affect movement patterns. Part of the study was therefore designed to compare two different chair designs at the one workplace.
1.2. A i m s
2. Method
The current investigation aimed to document and comparatively evaluate the sitting behaviour of workers who were engaged in various different seated work tasks. It aimed principally to collect some base data on normal seated movements. These are required as empirical scales which can be used for comparisons of work tasks and equip-
2.1. Workplaces, tasks and equipment
1.3. Measuring seated behaHour
The methods used for studying the ergonomic effects of seating tend to be those developed for the study of heavy manual work. Postural analysis then relies on data from laboratory studies which measured disc pressure, electromyography, radiography, stature variation and subjective assessments. Descriptions of ideal sitting postures have been developed from these investigations. But how do people really sit? Observational tools such as multimoment photography and posture classification techniques have been developed for assessing heavy manual work, but the techniques available are not ideal for measuring seated activities. For a detailed description of the methods used see Grieco (1986). He emphasised the need to measure postural fixity by considering the real postural sequence, the time spent in different situations of lumbar load and the frequency of posture change. For this reason we designed a posture classification system specifically for seated activity. This is described in detail in the following section.
Five different workplaces with a variety of chair types were studied using similar methods. The working tasks studied, subject details and the chair types are shown in Table 1 and Fig. 1. The tasks varied according to how much manual work
M. Graf et aL / International Journal of Industrial Ergonomics 15 (1995) 81-90
84
Table 1 Workplaces, subjects and chair types included in the study Task
Subjects
type
Males
Females
Ages
Light assembly work Office work Listening-lecture attendance V D U work Cashier work
1 5
5 1
17-46 26-37
6 2 1
0 4 5
25-45 a C 20-40 a A and B 20-40 " B
A
Chair
As used
B
for:
Light assembly Officework
A A
C
VDU work Cashierwork
r
Lecture attendance
ii
i
L i i
*itt1
i r
i
,:i
a Estimated
is required, but they are all generally carried out while seated. All require concentration and visual attention. The VDU work involved programming tasks in a bank. The workplaces were designed in consultation with an ergonomist and fulfilled all standard ergonomic principles of size and adjustability. The cashier workstations had also been designed to incorporate modern ergonomic principles. The cashiers registered purchases in the
i
The chair has a height As for A but the seat-pan adjustable backrestand can also be tilted a heightadjustableseat backwardsThe angleof the backrestis linkedto the seat-panangle (termeda synchronised mechanism)
The chair has a backrest but both it and seat heightare not adjustable in any way
Fig. 1. The types of chairs used at the various workplaces in the study.
Shoulders straight Spine
2 IllUlU
"-""
Tr°uintik°n
1 limb
2 Imllll
3 IIWIII
4flllllll
5 iNIH
6 Hlli
I=¢"
~
~¢--'
~"
~
~
"
N
Fig. 2. Side 1 of the sitting position classification matrix. The reverse side of the matrix is used when the shoulders are not parallel to the hips (the body is twisted). Leg positions are numbered from 1 to 6. Data can be recorded manually, using column and row numbers, or barcodes can be used for automatic data entry.
M. Graf et al. /International Journal of Industrial Ergonomics 15 (1995) 81-90
grocery section of a large department store. The office work involved varying periods of VDU work interrupted by telephone calls, searching through papers, writing of notes, thought pauses, etc. The assembly work involved welding microcomponents onto an electronic circuit board. The lecture was for post-graduate education and included a short video presentation. 2.2. Posture classification system
The subjects' body and leg position was recorded once per minute using the sitting posture classification system. With this system 68 different positions, almost exclusively seated positions, are represented in a matrix (see Fig. 2). The posture of the subjects' shoulders is first recorded, whether bent sidewards or twisted relative to the hips. The next decision regards whether the lumbar spine is in a lordotic or kyphotic curve. The observers were instructed to score a kyphotic posture only when the spine was markedly curved forwards in the lumbar region. Distinction is then made between whether the upper trunk is tipped forwards relative to the hips, is over the hips (middle) or is tipped backwards. Finally the leg position (angle of the thigh relative to the floor) can be recorded. The matrix was intended to provide a comparative index rather than to quantify movements and positions. 2.3. Observation periods
The cashiers were observed for two hours each (see below), all the other subjects for one hour at their normal workplace. The position of the observed person was recorded by either direct input into a laptop computer using a barcode reading pen or by manual listing using numeric coding for the positions. Only the last fifty observations for each session were analysed, to avoid the worst of the complications arising from the subjects' awareness of being observed. For the lecture attendance, leg positions and the curve of the back were not recorded due to viewing difficulties. The VDU workers were given Type A chairs (backrest and seat non-synchronised) a few days
85
before the trial so that they had a little time to accustom themselves to them. The first observation period was done on the Type A chairs and, after a short pause, they were given their usual Type B chairs (synchronised mechanism) and observed again for a further hour. Short questionnaires containing subjective rating scales were additionally given to the VDU workers after each chair trial. These questionnaires contained rating scales for comfort and body diagrams for marking any areas of discomfort. The other subjects performed their usual work on their usual chairs during the whole observation period. The cashiers were each observed over two hours by two independent observers such that the reliability of the measuring system could be checked. Where appropriate an analysis of variance was performed on the data, followed by t-test comparisons. The results should be interpreted carefully as there is confounding by chair type in the experimental design (see Discussion section).
3. Results 3.1. Forward, middle and backward positions
The frequencies of the various sitting positions during the five tasks are shown in Fig. 3. The assembly workers, VDU workers and cashiers were found to sit principally in the forwards or middle sitting positions, rarely leaning backwards, whereas the general office workers lean back more often than forwards. The general office workers utilise their seating position options more than the other workers. The listeners at the lecture rarely adopted the middle position. They were noted to sit almost always with their upper torso weight supported either on the table in front of them or leaning on the backrest. There was unfortunately no facility in the matrix for systematically recording the use of arm or back supports. 3.2. Kyphotic and twisted postures
A comparison of the frequency of kyphotic and twisted postures can be seen in Fig. 4. Differ-
M. Graf et aL / International Journal of Industrial Ergonomics 15 (1995) 81-90
86
ences were statistically significant ( p < 0.05) for all comparisons except between assembly workers and office workers. The general office workers were frequently in kyphotic positions (approximately 80% of the recordings). A common work posture observed in this group was the 'coathanger' position, sitting on the front of the chair but leaning the shoulders on the backrest with the back falling into the hollow space at the base of the backrest. This posture was rarely observed in the other groups, who were also less likely to lean backwards. Markedly kyphotic positions were most rarely observed in the V D U and assembly groups. The cashiers were most often observed in twisted positions; however, twisting was still found
g
o g
O "6
Kyphotic Postures s0~ 40
-
30 20
E o Assembly
z
Office
VDU
Cashier
Twisted Postu res o
25f
-
i
--
2o
8 15 s
Forwards Position
tO
40 20
0
~ lO Assembly Office
~
O
g
8
Listening
VDU
Assembly
Office
VDU
Cashier
Cashier
relatively often at the other work tasks; between 15-20% of the recordings. Statistically significant differences ( p < 0.05) were found between the groups with the exception of the assembly workers and office workers.
MiddlePosition
40 30
"6 20 10 0
3.3. Mouement frequency Assembly Office
~
O
z
Fig. 4. The relative frequency o f kyphotic and twisted postures for the different workp]aces. Data was not collected for the listeners due to visibility difficulties. Column height indicates the mean number o f observations made during the measurement period. Bars indicate the standard errors. The maximum for each column was 50.
g 3o
.,Q
o
S o
Listening
VDU
Cashier
BackwardsPosition
~,40 .~30 I O 20
The relative frequency of position change can be seen in Fig. 5. The office workers and cashiers changed position most frequently, the listeners least. This result needs to be evaluated in the light of the type of the postures that are adopted, as the degree of movement is important.
"5
10
3.4. Leg position Assembly Office
Listening
VDU
Cashier
Fig. 3. The frequency of the various trunk positions for the different workplaces. Columns indicate the m e a n n u m b e r of observations made during the m e a s u r e m e n t period. Bars indicate the standard errors. The maximum n u m b e r of observations for each workplace was 50 during the measured period.
The position of the thighs is important because this may affect the spinal posture (Mandal, 1981). In Fig. 6 it can be seen that the only thigh position that the cashiers adopt while sitting is position 3. The clearance for their knees at their
M. Graf et al. / International Journal of Industrial Ergonomics 15 (1995) 81-90
Frequency of Position Change o
20~
8 lo ~ 5 ~ o z
Assembly
Office
Listening
VDU
Cashier
Fig. 5. The relative frequency of position change for the different workplaces. A change of position was recorded when the trunk posture (forward, middle or backward) was altered from the previous recording. Columns indicate the mean number of changes made during the measurement period (maximum 50). Bars indicate the standard errors.
workstations is very narrow, due to the transport belt installed in the counter, and it does not permit the knees to be crossed, thus virtually eliminating the possibility for positions 1 and 2. The general office workers showed the greatest flexibility again with leg position. The assembly workers did not lift their knees above the
Assembly Work
C
|° o50
horizontal but instead moved forward on the seat to tip them downwards. The V D U workers did not do this, but it was seen in the office workers. The assembly workers were the only subjects who were observed to stand during part of the observation period (position 6). 3.5. The tiltable seats
The comparison between the two different types of chair used by the V D U workers revealed no significant differences ( p < 0.3) in the observed sitting behaviour for position, lumbar curve or movement frequency. There was, however, a significant difference in the acceptance ratings of the two chairs. The results revealed overwhelming support for the subject's usual chairs, that is, the chairs with the synchronised backrest and seat angle mechanism. All subjects preferred them. They reported feeling more comfortable on them, better supported and they felt more free to change position.
VDU Work e0
50 40
83o
30
,~2O t0
o
z
Z
20 10
Cashier Work
Office Work e--
50
50
4O
4O
30
87
2O
82o
10
10
/
0 z
Fig. 6. The frequency of the various thigh positions at the differen workplaces. A total of 50 recordings were taken during the measurement periods. However, some data is missing due to visual obstructions which occurred in the workplaces.
88
M. Graf et al. / International Journal of lndustrial Ergonomics 15 (1995) 81-90
3.6. Reliability of the observation method No significant differences were found between the ratings from the two observers used simultaneously for the cashiering study ( p < 0.001). In fact the inter-observer reliability was over 98% on all factors. As a comparative tool the matrix therefore proved to be quite reliable. A further study is planned to validate the quantitative aspects of this method.
4. Discussion
It was concluded that task demands have a significant effect on sitting position and posture. General office workers with more varied tasks are, furthermore, more able to make use of multiple sitting options than those whose work is restricted to VDU work. Seated inactivity eventually produces discomfort. It is unlikely that this could be engineered out with a better distribution of pressure, because of the need to maintain posture by some muscular effort, even in the best of chairs. But how detrimental is the inactivity to health? The epidemiological data previously described indicates that prolonged seating is linked to musculo-skeletal disorders particularly in the cervical region (see also Grieco, 1986). Coupled with this is the continual mechanical strain on the supporting tissues of the spine and trunk and the nutritional requirements of these structures. Grieco describes a process in which protracted work, such as that done by the postural muscles in inactive seating, results in reduced blood flow which eventually results in inflammation which in time leads to a scaring reaction of both the muscle and surrounding tissues. Pain is furthermore both a consequence of this process and a factor in maintaining it as pain tends to produce further muscular tension and immobility. Furthermore, as has been mentioned, the spinal discs themselves rely on movement for their nutritional exchange as they have no internal blood supply. This is also the case with other postural supporting tissue such as the articular cartilage along the spine. Kraemer (1985) has argued that there is a threshold load limit for disc nutrition processes
and that immobile postures, where the lumbar spine and arms are well supported, prevent this load from being reached. The resulting decrease in nutrition eventually leads to disc disease. From these and other studies (Andersson et al., 1974; Grandjean and Burandt, 1962) the possibility of relieving spinal loading by bracing of the upper body appears to be very important in seating behaviour. Where the work task permitted, this seating option always appears to be used. The listeners were able to reduce the load on their spines and muscles by almost continual support of their upper body on their desks and backrests. The office workers also relied heavily on the backrest; however, the frequency of the coathanger posture in this group calls into question its benefit. These kyphotic postures provide relief for the upper torso but are associated with increased loading of the lumbar spine, which may offset the gain. This group did not have seats which could be tipped backwards, or angle adjustable backrests. In 1962 Grandjean and Burandt used multimoment photography to record the postural behaviour of general office workers and also found that they most frequently leant backwards and most rarely leant forwards. From the results obtained in this study, this behaviour does not seem to be typical for more intensive VDU work. The difference, however, may be due to the type of chair used. Theoretically the synchronised backrest-seat mechanism on the VDU workers' chairs should support the body better in the backwards position. It is furthermore designed to permit easier position change between the backwards and the middle positions; however, the backwards posture was seldom adopted for the VDU task. The VDU workers were significantly less likely to show kyphotic or twisted postures than the office workers. Differences in seating behaviour between the chair types may be found in tasks where the backwards position is more frequently used and kyphotic postures are common, such as in general office work. This question needs to be further investigated. It is, however, possible that the reason for the high comfort and acceptance rating of the synchronised chairs is due to the very short time which was available for the subjects to accustom them-
M. Graf et al. /International Journal of Industrial Ergonomics 15 (1995) 81-90
selves to the different chairs, or due to aesthetic considerations. The Type A chairs were not as stylishly designed. An investigation of synchronised chairs in more general office situations is underway. Some of the twisting postures of the cashiers may be due to the limitation in their leg movements. The cashiers have swivel chairs and therefore, in theory, do not need to twist their bodies. A small push with a foot would theoretically suffice to turn the whole body and avoid these awkward twisting movements. It is, however, more probable that this activity reflects a biomechanical bias. It requires less effort to twist the upper body than to lift and push with the leg. It can be seen that the cashiers and office workers change their position most frequently; however, the changes by the cashiers are basically between only two very similar positions. Their posture is therefore much more static than the office workers who move between all three positions. In this light the frequency o f position change alone cannot be seen to accurately reflect the fixed or actice nature of the work. Information is also needed about the degree of movement. The listeners move less often but the change is more significant in that they swap from forwards to backwards. VDU workers and cashiers are particularly at risk for musculo-skeletal problems compared to the other groups (Krueger et al., 1988; Hfinting et al., 1981; Ong, 1984) and there is some indication from these results that the reason may be linked more to the restriction in their movements than to the frequency of the position change. Larger movements may be of more value than frequent small changes. How often should the sitting position change? Movements may be dictated either by the task requirements or by a physiologically driven need for position change. They can therefore be a measure of discomfort rather than mobility. If it is true that people seek to minimise their physical work load then in an ideally comfortable chair at an ideally designed workplace no postural changes should take place but, as has been discussed, this is physiologically undesirable. Is there a mechanism which drives people to change position or
89
must postural activity be designed into their tasks? As yet a suitable range of movements has not been clearly defined, nor has the border between dangerous and beneficial movement types or the optimum frequency of the change. From this study there is some indication of where the danger area may lie. The VDU workers move the least and are the most prone to musculo-skeletal problems. It can therefore be assumed that they do not move enough. But can there be too much movement? Very frequent movement probably indicates discomfort or instability. We need to define a suitable range. In this study there were too few reports of discomfort to draw a maximum limit. What is the optimal sequence of the changes?, or rather, what range is necessary? From this study it seems that the use of all three body positions is necessary - or at least the alternative to move between the forwards and the backwards postures. Active seating is where the sitter utilises multiple postures and positions at his or her workplace. This behaviour is influenced by the task, the furniture and the training of the person doing the sitting. The implication that we draw from our results is that we should not further restrict movements at the workplace by designing chairs which only support the postures most frequently seen. We should, however, still avoid mismatches of chairs and workplaces. A backwards sloping chair is inappropriate for a predominantly forward situated task, but a tiltable chair increases their options. No chair can guarantee back health at the workplace. The first line of action should always be job redesign rather than seat redesign. Education of the worker, particularly in the operation of their chairs and in the need to vary their postures, also plays a role.
5. Recommendations
Good seating depends on opportunities for significant changes in position. Extreme positions of the spine are avoided by fitting the chair and workspace to the person who is to use it and by designing work tasks to fit the anatomical and
90
M. Graf et al. / International Journal of Industrial Ergonomics 15 (l 995) 81-90
p h y s i o l o g i c a l limits o f t h e body. A g o o d c h a i r is o n e w h i c h s u p p o r t s t h e b o d y in m u l t i p l e d e s i r a b l e p o s i t i o n s a n d p o s t u r e s . It p r e v e n t s u n d u e stress by g i v i n g t h e s i t t e r t h e o p p o r t u n i t y to v a r y w h i c h g r o u p s o f p o s t u r a l m u s c l e s a r e u s e d . It s h o u l d not fix h i m o r h e r i n t o o n e p a r t i c u l a r p o s t u r e . An ergonomic assessment of a workplace generally includes a subjective assessment of comfort, b u t as single m e a s u r e s u s i n g r a t i n g scales a r e not completely reliable indicators of long term c o m f o r t a n d p h y s i o l o g i c a l suitability, p o s t u r a l b e havioural studies may be more revealing. These should check the frequency and range of posture c h a n g e . L e g p o s i t i o n s h o u l d also n o t b e i g n o r e d .
Acknowledgements W e wish to t h a n k G i r o f l e x E n t w i c k l u n g s A G , K o b l e n z , for t h e t e c h n i c a l a n d f i n a n c i a l s u p p o r t t h a t m a d e this s t u d y p o s s i b l e .
References Andersson, B.J.G., Ortengren, R., Nachemson, A. and Elfstr6m, G., 1974. Lumbar disc pressure and myoelectric back muscle activity during sitting. I. Studies on an experimental chair. Scandinavian Journal of Rehabilitation Medicine, 3:104-114. Bendix, T., 1986. Chair and table adjustments for seated work. In: N. Corlett, J. Wilson and I. Manenica (Eds.), The Ergonomics of Working Postures. Taylor and Francis, London, pp. 355-362. Branton, P., 1969. Behaviour, body mechanics and discomfort. Ergonomics, 12(2): 316-327. Coermann, R. and Rieck, A., 1964. Line verfeinerte Methode zur Bestimmung des Komfortgrades yon Sitzen. Interna-
tionale Zeitschrift fiir angewandte Physiologie, 20:376 397. Cantoni, S., Colombini, D., Occhipinti, E., Grieco, A., Frigo, C. and Pedotti, A., 1984. Posture analysis and evaluation at the old and new work place of a telephone company. In: E. Grandjean (Ed.), Ergonomics and Health in Modern Offices. Taylor and Francis, London, pp. 456-464. Corlett, E.N. and Manenica, I., 1980. The effects and measurement of working postures. Applied Ergonomics, 11( 1): 7-16. Grieco, A., 1986. Sitting posture: An old problem and a new one. Ergonomics, 29(3): 345-362. Grandjean, E. and Burandt, U., 1962. Das Sitzverhalten von Biiroangestellten. Industrielle Organisation, 31: 243-250. Grandjean, E. and Hiinting, W., 1977. Ergonomics of posture Review of various problems of standing and sitting posture. Applied Ergonomics, 8(3): 135-140. Hiinting, W., LS_ubli, Th. and Grandjean, E., 1981. Postural and visual loads at VDT workplaces. I. Constrained postures. Ergonomics, 24(12): 917-931. Kraemer, J., 1985. Dynamic characteristics of the vertebral column, effects of prolonged loading. Ergonomics, 28: 95 -97. Krueger, H., Hinnen, U. Guggenbflhl, U. and L~iubli, Th., 1988. Musculoskeletel disorders at different work places. In: A.S. Adams, R.R. Hall, B.J. McPhee and M.S. Oxenburgh (Eds.). Designing a Better World. International Ergonomics Association 10th International Congress. Vol. 2, Ergonomics Society of Australia, Sydney, pp. 445-447. Magora, A., 1972, Investigation of the relation between low back pain and occupation. 3 physical requirements: Sitting, standing and weight lifting. Industrial Medicine, 41(12): 5 9. Mandal, A.C., 1981. The seated man: Homo sedens. The seated work position theory and practice, Applied Ergonomics, 12(1): 19-26. Ong, C.N., 1984. VDT work place design and physical fatigue: A case study in Singapore. In: E. Grandjean (Ed.), Ergonomics and Health in Modern Offices. Taylor and Francis, London, pp. 484-494. Schobert, H., 1989. Orthopaedie des Sitzens. Springer Verlag, Berlin.
ELSEVIER
International Journal of Industrial Ergonomics 15 (1995) 81-90
An assessment of seated activity and postures at five workplaces M. Graf, U. Guggenbi~hl, H. Krueger Institute for Hygiene and Work Physiology, Swiss Federal Institute of Technology, Clausiusstr. 21, 8092 Zurich, Switzerland Received June 30, 1993; accepted in revised form January 21, 1994
Abstract Static working positions and poor postures are both associated with the development of musculo-skelctal disorders and discomfort. An assessment of people's postural behaviour requires suitable measurement tools and scales against which the measure can be judged. Although such methods have been developed, these generally are aimed at assessing gross movements which are undertaken in an industrial setting where the workers arc not sedentary. Seated workplaces are more the norm in modern societies and sedentary work does not provide immunity from discomfort and musculo-skeletal disorders. For this reason we have developed a method for classifying and recording seated postural behaviour. This method is described along with the results of studies undcrtakcn using the method at five workplaces. The studies aimed to obtain some base data on sitting behaviour and how it related to work task and chair type. Postures were recorded at one-minute intervals during the subjects' normal work for periods up to two hours. Additionally, questionnaires regarding comfort and musculo-skeletal problems were issued to some of the subjects. Although individual variation was sizeable, the results showed clear postural behaviour differences between the various work tasks. Work tasks which have a higher incidence of musculo-skeletal disorders were found to produce less frequent and less marked postural change. From these results some preliminary conclusions about an optimal seating behaviour pattern are made. Relevance to industry Back disorders and discomfort are major sources of lost work time and worker dissatisfaction. Many of thc problems can be traced back to poor seating or workplace design. The paper elucidates postural behavioural factors which, along with the physical aspects of workplace furniture, affect people's comfort and well-being at work. Thesc postural aspects are largely determined by the work task.
Keywords: Keywords: Seating; Chairs; Posture; Musculo-skeletal disorder; Activity; Behaviour
I. Introduction W i t h t h e i n c r e a s i n g a u t o m a t i o n o f w o r k over t h e last half c e n t u r y a significant c h a n g e has t a k e n p l a c e in w o r k p l a c e activity in t h e industrialised nations. W h e r e a s once, m o s t w o r k involved m a n u a l tasks u n d e r t a k e n largely o u t d o o r s , t o d a y
m o s t w o r k activity is p e r f o r m e d inside b u i l d i n g s while s e a t e d . G e n e r a l l y m o r e p e o p l e sit much l o n g e r t o d a y t h a n even 20 y e a r s ago. In the i n d u s t r i a l i s e d c o u n t r i e s of W e s t e r n E u r o p e a n d N o r t h A m e r i c a m o r e t h a n half of t h e w o r k f o r c e is e m p l o y e d in a d m i n i s t r a t i o n , services, credit, t r a d e o r i n s u r a n c e i n d u s t r i e s ( G r i e c o , 1986) - all
0169-8141/95/$09.50 © 1995 Elsevier Science B.V. All rights reserved SSDI 0 1 6 9 - 8 1 4 1 ( 9 4 ) 0 0 0 2 7 - Z
82
M. Graf et al. /International Journal of Industrial Ergonomics 15 (1995) 81-90
sitting much of the working day. This change has profoundly altered the common physical demands made by work on the human body. The decrease in heavy physical work has not resulted in fewer musculo-skeletal disorders. These problems are still common in workplaces where no heavy loads are manipulated and where the work is generally described as light. In 1972 Magora found a high incidence of low back pain in persons whose work entailed sitting for prolonged periods or not sitting at all, Hfinting et al. (1981) found that VDT workplaces and typing were associated with physical impairments in the hands, arms, shoulders and neck, particularly where the workplaces were poorly designed, and Ong (1984) found that musculo-skeletal problems were more common amongst VDT operators than a comparative group of conventional office workers. This was attributed to poor furniture design and insufficient rest pauses, along with other factors. With these common musculo-skeletal problems, including lower back pain, shoulder and neck discomfort and upper limb disorders, there is generally no history of the classical single event overloading of the tissues involved. Other factors such as poor seated work postures and poor work design clearly also contribute to the development of these disorders.
1.1. ActiL~e seating For over a hundred years it has been recognised that our seated postures are of great importance to the maintenance of the health of our backs. Nevertheless there is still a great deal of dispute over the definition of good sitting posture. At the beginning of the century people were taught always to sit upright, with a straight back and with their knees and hips forming right angles, irrespective of the task that they were performing or the type of chair that they used. In the early sixties, it was recognised that holding this 'correct' position was not easy and the solution was sought in chair design (Grandjean and Burandt, 1962; Coermann and Rieck, 1964). The chair itself took prominence over the behaviour of the sitter. Chairs were judged favourably if they maintained the body in the previously de-
scribed 'optimum' posture. But, unfortunately, no chair design could be found that reliably and comfortably maintained the body in this posture. Work physicians then began to question whether the stiff upright posture is in fact the best sitting position. Some argued that it is simply impossible to maintain (Branton, 1969; Grandjean and Hiinting, 1977; Bendix, 1986) whereas others argued that it is biomechanically unsound (Corlett and Mananica, 1980; Mandal, 1981; Schobert, 1989). Eventually the idea dawned that no position should be continually held and that there cannot be a single ideal sitting position. In fact the degree of postural fixation, that is, lack of movement, is itself further suspected to be a cause of musculo-skeletal disorders. Cantoni et al. (1984) compared work at an old switchboard with work at a new ergonomically designed VDT switchboard and found a significant increase in the degree of postural fixity. They felt that this was a negative development. The concept of a single ideal posture has still not died completely and every year new products come onto the market which purport to ensure the optimum posture. Usually emphasis is placed on the 'correct' spinal curve which may also require the 'correct' hip position or thigh position. The concept is quite sound that the spinal disks and musculature can be unduly stressed by extreme body movements and postures. Awkward and jerky movements are acutely more dangerous to the spinal tissues than prolonged use. However, natural movements do not stress the body, in fact, they are desirable and necessary as long as they are within an acceptable range. We realize that if the body is kept in a single seated position, a specific set of postural muscles will need to work continually to maintain that posture. These muscles eventually tire because they have no opportunity to relax. Movements, on the other hand, alternate the muscles which are working and thereby create relaxation periods. Relaxation allows the replenishment of nutrient energy, oxygen and the removal of waste products from the tissues. Furthermore, the spinal disks may be dependent on physical activity because their nourishment depends on the pumping action of changing disc pressure from spinal move-
M. Graf et al. / International Journal of Industrial Ergonomics 15 (1995) 81-90
83
ment. Any seated position is therefore today recognized to be unacceptable if it is held for too long, no matter how well the spinal vertebrae are positioned and the body supported. The movement factor is particularly of importance when the task to be performed requires extended periods of sitting, as is the case in most occupations today. The term dynamic seating was coined to describe this idea of changing positions to allow the different groups of postural muscles to rest alternately. However the problem which has arisen is that the expression 'dynamic seating' has different meanings for different people. Some chair manufacturers use the term to describe a type of chair mechanism, others use it to describe chairs which do not stabilise the trunk or which offer only indirect trunk support. With these concepts the emphasis is on the chair and not on the person or how he or she sits. The emphasis should instead be kept on sitting behaviour. We suggest the term active seating is a better descriptor and we prefer its use to describe optimum seated movement behaviour. The problem which has now presented itself is that an adequate definition of good sitting behaviour, active seating, is not readily available. The principle two issues which need to be addressed are: How often should the sitting position change? And what is the optimal range of these changes? By our definition a good chair is one which supports the body in multiple desirable positions. It prevents stress by giving the sitter the opportunity to vary which groups of postural muscles are used. Extreme positions of the spine are avoided by fitting the chair and workspace to the person who is to use it and by designing work tasks to fit the anatomical and physiological limits of the body.
ment. It would seem obvious that task requirements would significantly affect sitting behaviour, that is, the person's postural range and movement frequency, but few studies have been done to document how sitting behaviour is affected by task determinants. The effect of different types of chairs on seating behaviour has also not been studied in this way so it is not known to what extent variations in chair design affect movement patterns. Part of the study was therefore designed to compare two different chair designs at the one workplace.
1.2. A i m s
2. Method
The current investigation aimed to document and comparatively evaluate the sitting behaviour of workers who were engaged in various different seated work tasks. It aimed principally to collect some base data on normal seated movements. These are required as empirical scales which can be used for comparisons of work tasks and equip-
2.1. Workplaces, tasks and equipment
1.3. Measuring seated behaHour
The methods used for studying the ergonomic effects of seating tend to be those developed for the study of heavy manual work. Postural analysis then relies on data from laboratory studies which measured disc pressure, electromyography, radiography, stature variation and subjective assessments. Descriptions of ideal sitting postures have been developed from these investigations. But how do people really sit? Observational tools such as multimoment photography and posture classification techniques have been developed for assessing heavy manual work, but the techniques available are not ideal for measuring seated activities. For a detailed description of the methods used see Grieco (1986). He emphasised the need to measure postural fixity by considering the real postural sequence, the time spent in different situations of lumbar load and the frequency of posture change. For this reason we designed a posture classification system specifically for seated activity. This is described in detail in the following section.
Five different workplaces with a variety of chair types were studied using similar methods. The working tasks studied, subject details and the chair types are shown in Table 1 and Fig. 1. The tasks varied according to how much manual work
M. Graf et aL / International Journal of Industrial Ergonomics 15 (1995) 81-90
84
Table 1 Workplaces, subjects and chair types included in the study Task
Subjects
type
Males
Females
Ages
Light assembly work Office work Listening-lecture attendance V D U work Cashier work
1 5
5 1
17-46 26-37
6 2 1
0 4 5
25-45 a C 20-40 a A and B 20-40 " B
A
Chair
As used
B
for:
Light assembly Officework
A A
C
VDU work Cashierwork
r
Lecture attendance
ii
i
L i i
*itt1
i r
i
,:i
a Estimated
is required, but they are all generally carried out while seated. All require concentration and visual attention. The VDU work involved programming tasks in a bank. The workplaces were designed in consultation with an ergonomist and fulfilled all standard ergonomic principles of size and adjustability. The cashier workstations had also been designed to incorporate modern ergonomic principles. The cashiers registered purchases in the
i
The chair has a height As for A but the seat-pan adjustable backrestand can also be tilted a heightadjustableseat backwardsThe angleof the backrestis linkedto the seat-panangle (termeda synchronised mechanism)
The chair has a backrest but both it and seat heightare not adjustable in any way
Fig. 1. The types of chairs used at the various workplaces in the study.
Shoulders straight Spine
2 IllUlU
"-""
Tr°uintik°n
1 limb
2 Imllll
3 IIWIII
4flllllll
5 iNIH
6 Hlli
I=¢"
~
~¢--'
~"
~
~
"
N
Fig. 2. Side 1 of the sitting position classification matrix. The reverse side of the matrix is used when the shoulders are not parallel to the hips (the body is twisted). Leg positions are numbered from 1 to 6. Data can be recorded manually, using column and row numbers, or barcodes can be used for automatic data entry.
M. Graf et al. /International Journal of Industrial Ergonomics 15 (1995) 81-90
grocery section of a large department store. The office work involved varying periods of VDU work interrupted by telephone calls, searching through papers, writing of notes, thought pauses, etc. The assembly work involved welding microcomponents onto an electronic circuit board. The lecture was for post-graduate education and included a short video presentation. 2.2. Posture classification system
The subjects' body and leg position was recorded once per minute using the sitting posture classification system. With this system 68 different positions, almost exclusively seated positions, are represented in a matrix (see Fig. 2). The posture of the subjects' shoulders is first recorded, whether bent sidewards or twisted relative to the hips. The next decision regards whether the lumbar spine is in a lordotic or kyphotic curve. The observers were instructed to score a kyphotic posture only when the spine was markedly curved forwards in the lumbar region. Distinction is then made between whether the upper trunk is tipped forwards relative to the hips, is over the hips (middle) or is tipped backwards. Finally the leg position (angle of the thigh relative to the floor) can be recorded. The matrix was intended to provide a comparative index rather than to quantify movements and positions. 2.3. Observation periods
The cashiers were observed for two hours each (see below), all the other subjects for one hour at their normal workplace. The position of the observed person was recorded by either direct input into a laptop computer using a barcode reading pen or by manual listing using numeric coding for the positions. Only the last fifty observations for each session were analysed, to avoid the worst of the complications arising from the subjects' awareness of being observed. For the lecture attendance, leg positions and the curve of the back were not recorded due to viewing difficulties. The VDU workers were given Type A chairs (backrest and seat non-synchronised) a few days
85
before the trial so that they had a little time to accustom themselves to them. The first observation period was done on the Type A chairs and, after a short pause, they were given their usual Type B chairs (synchronised mechanism) and observed again for a further hour. Short questionnaires containing subjective rating scales were additionally given to the VDU workers after each chair trial. These questionnaires contained rating scales for comfort and body diagrams for marking any areas of discomfort. The other subjects performed their usual work on their usual chairs during the whole observation period. The cashiers were each observed over two hours by two independent observers such that the reliability of the measuring system could be checked. Where appropriate an analysis of variance was performed on the data, followed by t-test comparisons. The results should be interpreted carefully as there is confounding by chair type in the experimental design (see Discussion section).
3. Results 3.1. Forward, middle and backward positions
The frequencies of the various sitting positions during the five tasks are shown in Fig. 3. The assembly workers, VDU workers and cashiers were found to sit principally in the forwards or middle sitting positions, rarely leaning backwards, whereas the general office workers lean back more often than forwards. The general office workers utilise their seating position options more than the other workers. The listeners at the lecture rarely adopted the middle position. They were noted to sit almost always with their upper torso weight supported either on the table in front of them or leaning on the backrest. There was unfortunately no facility in the matrix for systematically recording the use of arm or back supports. 3.2. Kyphotic and twisted postures
A comparison of the frequency of kyphotic and twisted postures can be seen in Fig. 4. Differ-
M. Graf et aL / International Journal of Industrial Ergonomics 15 (1995) 81-90
86
ences were statistically significant ( p < 0.05) for all comparisons except between assembly workers and office workers. The general office workers were frequently in kyphotic positions (approximately 80% of the recordings). A common work posture observed in this group was the 'coathanger' position, sitting on the front of the chair but leaning the shoulders on the backrest with the back falling into the hollow space at the base of the backrest. This posture was rarely observed in the other groups, who were also less likely to lean backwards. Markedly kyphotic positions were most rarely observed in the V D U and assembly groups. The cashiers were most often observed in twisted positions; however, twisting was still found
g
o g
O "6
Kyphotic Postures s0~ 40
-
30 20
E o Assembly
z
Office
VDU
Cashier
Twisted Postu res o
25f
-
i
--
2o
8 15 s
Forwards Position
tO
40 20
0
~ lO Assembly Office
~
O
g
8
Listening
VDU
Assembly
Office
VDU
Cashier
Cashier
relatively often at the other work tasks; between 15-20% of the recordings. Statistically significant differences ( p < 0.05) were found between the groups with the exception of the assembly workers and office workers.
MiddlePosition
40 30
"6 20 10 0
3.3. Mouement frequency Assembly Office
~
O
z
Fig. 4. The relative frequency o f kyphotic and twisted postures for the different workp]aces. Data was not collected for the listeners due to visibility difficulties. Column height indicates the mean number o f observations made during the measurement period. Bars indicate the standard errors. The maximum for each column was 50.
g 3o
.,Q
o
S o
Listening
VDU
Cashier
BackwardsPosition
~,40 .~30 I O 20
The relative frequency of position change can be seen in Fig. 5. The office workers and cashiers changed position most frequently, the listeners least. This result needs to be evaluated in the light of the type of the postures that are adopted, as the degree of movement is important.
"5
10
3.4. Leg position Assembly Office
Listening
VDU
Cashier
Fig. 3. The frequency of the various trunk positions for the different workplaces. Columns indicate the m e a n n u m b e r of observations made during the m e a s u r e m e n t period. Bars indicate the standard errors. The maximum n u m b e r of observations for each workplace was 50 during the measured period.
The position of the thighs is important because this may affect the spinal posture (Mandal, 1981). In Fig. 6 it can be seen that the only thigh position that the cashiers adopt while sitting is position 3. The clearance for their knees at their
M. Graf et al. / International Journal of Industrial Ergonomics 15 (1995) 81-90
Frequency of Position Change o
20~
8 lo ~ 5 ~ o z
Assembly
Office
Listening
VDU
Cashier
Fig. 5. The relative frequency of position change for the different workplaces. A change of position was recorded when the trunk posture (forward, middle or backward) was altered from the previous recording. Columns indicate the mean number of changes made during the measurement period (maximum 50). Bars indicate the standard errors.
workstations is very narrow, due to the transport belt installed in the counter, and it does not permit the knees to be crossed, thus virtually eliminating the possibility for positions 1 and 2. The general office workers showed the greatest flexibility again with leg position. The assembly workers did not lift their knees above the
Assembly Work
C
|° o50
horizontal but instead moved forward on the seat to tip them downwards. The V D U workers did not do this, but it was seen in the office workers. The assembly workers were the only subjects who were observed to stand during part of the observation period (position 6). 3.5. The tiltable seats
The comparison between the two different types of chair used by the V D U workers revealed no significant differences ( p < 0.3) in the observed sitting behaviour for position, lumbar curve or movement frequency. There was, however, a significant difference in the acceptance ratings of the two chairs. The results revealed overwhelming support for the subject's usual chairs, that is, the chairs with the synchronised backrest and seat angle mechanism. All subjects preferred them. They reported feeling more comfortable on them, better supported and they felt more free to change position.
VDU Work e0
50 40
83o
30
,~2O t0
o
z
Z
20 10
Cashier Work
Office Work e--
50
50
4O
4O
30
87
2O
82o
10
10
/
0 z
Fig. 6. The frequency of the various thigh positions at the differen workplaces. A total of 50 recordings were taken during the measurement periods. However, some data is missing due to visual obstructions which occurred in the workplaces.
88
M. Graf et al. / International Journal of lndustrial Ergonomics 15 (1995) 81-90
3.6. Reliability of the observation method No significant differences were found between the ratings from the two observers used simultaneously for the cashiering study ( p < 0.001). In fact the inter-observer reliability was over 98% on all factors. As a comparative tool the matrix therefore proved to be quite reliable. A further study is planned to validate the quantitative aspects of this method.
4. Discussion
It was concluded that task demands have a significant effect on sitting position and posture. General office workers with more varied tasks are, furthermore, more able to make use of multiple sitting options than those whose work is restricted to VDU work. Seated inactivity eventually produces discomfort. It is unlikely that this could be engineered out with a better distribution of pressure, because of the need to maintain posture by some muscular effort, even in the best of chairs. But how detrimental is the inactivity to health? The epidemiological data previously described indicates that prolonged seating is linked to musculo-skeletal disorders particularly in the cervical region (see also Grieco, 1986). Coupled with this is the continual mechanical strain on the supporting tissues of the spine and trunk and the nutritional requirements of these structures. Grieco describes a process in which protracted work, such as that done by the postural muscles in inactive seating, results in reduced blood flow which eventually results in inflammation which in time leads to a scaring reaction of both the muscle and surrounding tissues. Pain is furthermore both a consequence of this process and a factor in maintaining it as pain tends to produce further muscular tension and immobility. Furthermore, as has been mentioned, the spinal discs themselves rely on movement for their nutritional exchange as they have no internal blood supply. This is also the case with other postural supporting tissue such as the articular cartilage along the spine. Kraemer (1985) has argued that there is a threshold load limit for disc nutrition processes
and that immobile postures, where the lumbar spine and arms are well supported, prevent this load from being reached. The resulting decrease in nutrition eventually leads to disc disease. From these and other studies (Andersson et al., 1974; Grandjean and Burandt, 1962) the possibility of relieving spinal loading by bracing of the upper body appears to be very important in seating behaviour. Where the work task permitted, this seating option always appears to be used. The listeners were able to reduce the load on their spines and muscles by almost continual support of their upper body on their desks and backrests. The office workers also relied heavily on the backrest; however, the frequency of the coathanger posture in this group calls into question its benefit. These kyphotic postures provide relief for the upper torso but are associated with increased loading of the lumbar spine, which may offset the gain. This group did not have seats which could be tipped backwards, or angle adjustable backrests. In 1962 Grandjean and Burandt used multimoment photography to record the postural behaviour of general office workers and also found that they most frequently leant backwards and most rarely leant forwards. From the results obtained in this study, this behaviour does not seem to be typical for more intensive VDU work. The difference, however, may be due to the type of chair used. Theoretically the synchronised backrest-seat mechanism on the VDU workers' chairs should support the body better in the backwards position. It is furthermore designed to permit easier position change between the backwards and the middle positions; however, the backwards posture was seldom adopted for the VDU task. The VDU workers were significantly less likely to show kyphotic or twisted postures than the office workers. Differences in seating behaviour between the chair types may be found in tasks where the backwards position is more frequently used and kyphotic postures are common, such as in general office work. This question needs to be further investigated. It is, however, possible that the reason for the high comfort and acceptance rating of the synchronised chairs is due to the very short time which was available for the subjects to accustom them-
M. Graf et al. /International Journal of Industrial Ergonomics 15 (1995) 81-90
selves to the different chairs, or due to aesthetic considerations. The Type A chairs were not as stylishly designed. An investigation of synchronised chairs in more general office situations is underway. Some of the twisting postures of the cashiers may be due to the limitation in their leg movements. The cashiers have swivel chairs and therefore, in theory, do not need to twist their bodies. A small push with a foot would theoretically suffice to turn the whole body and avoid these awkward twisting movements. It is, however, more probable that this activity reflects a biomechanical bias. It requires less effort to twist the upper body than to lift and push with the leg. It can be seen that the cashiers and office workers change their position most frequently; however, the changes by the cashiers are basically between only two very similar positions. Their posture is therefore much more static than the office workers who move between all three positions. In this light the frequency o f position change alone cannot be seen to accurately reflect the fixed or actice nature of the work. Information is also needed about the degree of movement. The listeners move less often but the change is more significant in that they swap from forwards to backwards. VDU workers and cashiers are particularly at risk for musculo-skeletal problems compared to the other groups (Krueger et al., 1988; Hfinting et al., 1981; Ong, 1984) and there is some indication from these results that the reason may be linked more to the restriction in their movements than to the frequency of the position change. Larger movements may be of more value than frequent small changes. How often should the sitting position change? Movements may be dictated either by the task requirements or by a physiologically driven need for position change. They can therefore be a measure of discomfort rather than mobility. If it is true that people seek to minimise their physical work load then in an ideally comfortable chair at an ideally designed workplace no postural changes should take place but, as has been discussed, this is physiologically undesirable. Is there a mechanism which drives people to change position or
89
must postural activity be designed into their tasks? As yet a suitable range of movements has not been clearly defined, nor has the border between dangerous and beneficial movement types or the optimum frequency of the change. From this study there is some indication of where the danger area may lie. The VDU workers move the least and are the most prone to musculo-skeletal problems. It can therefore be assumed that they do not move enough. But can there be too much movement? Very frequent movement probably indicates discomfort or instability. We need to define a suitable range. In this study there were too few reports of discomfort to draw a maximum limit. What is the optimal sequence of the changes?, or rather, what range is necessary? From this study it seems that the use of all three body positions is necessary - or at least the alternative to move between the forwards and the backwards postures. Active seating is where the sitter utilises multiple postures and positions at his or her workplace. This behaviour is influenced by the task, the furniture and the training of the person doing the sitting. The implication that we draw from our results is that we should not further restrict movements at the workplace by designing chairs which only support the postures most frequently seen. We should, however, still avoid mismatches of chairs and workplaces. A backwards sloping chair is inappropriate for a predominantly forward situated task, but a tiltable chair increases their options. No chair can guarantee back health at the workplace. The first line of action should always be job redesign rather than seat redesign. Education of the worker, particularly in the operation of their chairs and in the need to vary their postures, also plays a role.
5. Recommendations
Good seating depends on opportunities for significant changes in position. Extreme positions of the spine are avoided by fitting the chair and workspace to the person who is to use it and by designing work tasks to fit the anatomical and
90
M. Graf et al. / International Journal of Industrial Ergonomics 15 (l 995) 81-90
p h y s i o l o g i c a l limits o f t h e body. A g o o d c h a i r is o n e w h i c h s u p p o r t s t h e b o d y in m u l t i p l e d e s i r a b l e p o s i t i o n s a n d p o s t u r e s . It p r e v e n t s u n d u e stress by g i v i n g t h e s i t t e r t h e o p p o r t u n i t y to v a r y w h i c h g r o u p s o f p o s t u r a l m u s c l e s a r e u s e d . It s h o u l d not fix h i m o r h e r i n t o o n e p a r t i c u l a r p o s t u r e . An ergonomic assessment of a workplace generally includes a subjective assessment of comfort, b u t as single m e a s u r e s u s i n g r a t i n g scales a r e not completely reliable indicators of long term c o m f o r t a n d p h y s i o l o g i c a l suitability, p o s t u r a l b e havioural studies may be more revealing. These should check the frequency and range of posture c h a n g e . L e g p o s i t i o n s h o u l d also n o t b e i g n o r e d .
Acknowledgements W e wish to t h a n k G i r o f l e x E n t w i c k l u n g s A G , K o b l e n z , for t h e t e c h n i c a l a n d f i n a n c i a l s u p p o r t t h a t m a d e this s t u d y p o s s i b l e .
References Andersson, B.J.G., Ortengren, R., Nachemson, A. and Elfstr6m, G., 1974. Lumbar disc pressure and myoelectric back muscle activity during sitting. I. Studies on an experimental chair. Scandinavian Journal of Rehabilitation Medicine, 3:104-114. Bendix, T., 1986. Chair and table adjustments for seated work. In: N. Corlett, J. Wilson and I. Manenica (Eds.), The Ergonomics of Working Postures. Taylor and Francis, London, pp. 355-362. Branton, P., 1969. Behaviour, body mechanics and discomfort. Ergonomics, 12(2): 316-327. Coermann, R. and Rieck, A., 1964. Line verfeinerte Methode zur Bestimmung des Komfortgrades yon Sitzen. Interna-
tionale Zeitschrift fiir angewandte Physiologie, 20:376 397. Cantoni, S., Colombini, D., Occhipinti, E., Grieco, A., Frigo, C. and Pedotti, A., 1984. Posture analysis and evaluation at the old and new work place of a telephone company. In: E. Grandjean (Ed.), Ergonomics and Health in Modern Offices. Taylor and Francis, London, pp. 456-464. Corlett, E.N. and Manenica, I., 1980. The effects and measurement of working postures. Applied Ergonomics, 11( 1): 7-16. Grieco, A., 1986. Sitting posture: An old problem and a new one. Ergonomics, 29(3): 345-362. Grandjean, E. and Burandt, U., 1962. Das Sitzverhalten von Biiroangestellten. Industrielle Organisation, 31: 243-250. Grandjean, E. and Hiinting, W., 1977. Ergonomics of posture Review of various problems of standing and sitting posture. Applied Ergonomics, 8(3): 135-140. Hiinting, W., LS_ubli, Th. and Grandjean, E., 1981. Postural and visual loads at VDT workplaces. I. Constrained postures. Ergonomics, 24(12): 917-931. Kraemer, J., 1985. Dynamic characteristics of the vertebral column, effects of prolonged loading. Ergonomics, 28: 95 -97. Krueger, H., Hinnen, U. Guggenbflhl, U. and L~iubli, Th., 1988. Musculoskeletel disorders at different work places. In: A.S. Adams, R.R. Hall, B.J. McPhee and M.S. Oxenburgh (Eds.). Designing a Better World. International Ergonomics Association 10th International Congress. Vol. 2, Ergonomics Society of Australia, Sydney, pp. 445-447. Magora, A., 1972, Investigation of the relation between low back pain and occupation. 3 physical requirements: Sitting, standing and weight lifting. Industrial Medicine, 41(12): 5 9. Mandal, A.C., 1981. The seated man: Homo sedens. The seated work position theory and practice, Applied Ergonomics, 12(1): 19-26. Ong, C.N., 1984. VDT work place design and physical fatigue: A case study in Singapore. In: E. Grandjean (Ed.), Ergonomics and Health in Modern Offices. Taylor and Francis, London, pp. 484-494. Schobert, H., 1989. Orthopaedie des Sitzens. Springer Verlag, Berlin.