WORKSTATION DESIGN AND POSTURAL STRESS PART 2: CASE STUDY SUMMARY An evaluation of the musculoskeletal problems associated with seated workstations was undertaken for employees of the Human Sciences Research Council. A sample of 37 workers was studied for 14 days to determine which symptoms were due to chronic disorders and which were linked to postural stress or constrained posture as a re­ sult of workstation and task design. Techniques used included two subjective comfort ratings, clinical examinations by physiotherapists, anthropometric and workstation measurement, and video recordings of subjects’ posture over time. Subjects were selected from four occupations characterised by different levels of constraint in their work pos­ ture; data-entry typists, typists, programmers and researchers. Examinations revealed that 68% of the presenting symptoms were not related to any known previous trauma or pathology. Of these problems 86% were reported to be occupationally aggravated or related. The findings con­ firmed the hypothesis that postural constraint is accompanied by an increased likelihood of developing chronic musculoskeletal disorders. * The concepts of postural stress and postural constraint and their relationship to the development of muscle strain are discussed in Part 1: Background to occupational syndromes. OPSOMMING ’n Evaluasie van die muskuloskeletale probleme verbonde aan sitwerkstasie is onderneem vir die werknemers van die Raad vir Geesteswetenskaplike Navorsing. ’n Steekproef van 37 werkers is oor n periode van 14 dae bestudeer om te bepaal watter simptome toe te skryf kan word aan kroniese afwykings, en watter aan postuurstres of -beperk- theid, veroorsaak deur werkstasie-en-taakontwerp, te wyte is. Tegnieke wat gebruik is sluit in twee subjektiewe ge- rief skattinge, kliniese ondersoeke deur fisioterapeute, antropometriese en werkstasie mates, en video opnames van proefpersone se liggamshoudings oor ’n sekere tydperk. Proefpersone is geselekteer uit vier beroepsgroepe wat kenmerkend verskll in die mate van postuurbeperking wat in die werkstasie ondervind is, naamlik datatiksters, tiksters, programmeerders en navorsers. Ondersoeke het getoon dat 68% van aanmeldingssimptome geen verband gehad het met ’n vorige trauma of pato- logie. Dit is gemeld dat 86% van die genoemde probleme beroepsverwant of -vererger is. Die bevindings het die hipotese bevestig dat beperkte lagaamshoudings gepaard gaan met ’n vergrote waarskynlikheid van die ontwikkel- ing van kroniese muskuloskeletale probleme. * Die begrippe van postuurstres en postuurbeperking en hul verhouding met die ontwikkeling van gespanne spiere word bespreek in Deel 1: Agtergrond tot werksverwante sindrome. Anthony Golding* INTRODUCTION A preliminary survey on a sample of 70 employees o f the Human Sciences Research Council (HSRC) showed that furniture was generally inappropriate for the tasks performed and that health complaints were a cause for concern. For instance, headache was experienced at least once a week by 38% of employees, backache by 37%, neckache by 40% and shoulderache by 31%. In view o f these findings it was decided to investigate in greater depth using a smaller sample o f 37 people. An aim o f the present study was to determine whether constrained work postures result in more musculoskeletal problems. To answer this question it was necessary to dif­ ferentiate between workers suffering from chronic disorders as a result of pathology or trauma and complaints with “un­ known” causes. The results, whilst not strictly generalisable, still provide indications o f what Findings might be expccted in a large-scale investigation. The degree o f postural constraint is a function o f the task and workstation. The most constrained task, that of the data- entry typist, consists o f keying in numbers with the right hand while documents are paged with the left hand. This requires the head to be inclined and twisted to the left. Typists copy from documents to wordprocessors. The absence o f docu- ment-holders means that the neck is twisted frequently. Pro­ grammers work on printouts in front which means that the terminals are placed at the far left-hand corner o f the desk, necessitating twisting and leaning. Researchers perform desk work and terminal work. Anthony Golding, MSc Ergonomics (London), BSc Psychology (CNAA). Chief Researcher, Environmental Studies Division, National Institute for Personnel Research, Johannesburg. Address: NIPR, P O Box 32410, Braamfontein 2017, Tel (011) 648-1046 Physiotherapy, May 1990, vol 46 no 2 Page 13 R ep ro du ce d by S ab in et G at ew ay u nd er li ce nc e gr an te d by th e P ub lis he r (d at ed 2 01 3. ) METHODOLOGY Subjects were selected on the basis of the relative pos­ tural constraint of their occupations. The sample included 7 data-typists, 5 typists, 8 programmers and 17 researchers (7 of whom were male). The bulk o f the subjects was thus female. The mean ages and standard deviations of the groups were: data-typists 30,83 years (7,20); typists 33,60 (7,89); pro­ grammers 21,40 (2,90); and researchers 32,67 (8,35). The mean age of the programmers was found to be significantly different from those of the other groups using Student’s / Test (p,01). TECHNIQUES AND PROCEDURE Several methods were used in evaluating workstations and health. These included measurement of anthropometric and workstation d im ensions, clinical exam ination by physiotherapists, video recording of subjects’ postural change over time, rating o f general comfort using the scale of Shackel et a l1 and rating of discomfort in different body areas using a body map2. A modified version of the scale of Roland and Morris3 was sued to rate discomfort severity, ranging from “just noticeable discomfort” (1) to “almost unbearable dis­ comfort” (5). A questionnaire incorporating these scales was administered to each subject 6 times daily for 14 consecutive days (Fig 1). N A M E D A T E _ T IM E G E N E R A L C O M F O R T RA TING P lease rate y o u r feelings n o w . M ark th e scale. ___ I feel com pletely relaxed ___ I feel perfectly com fortable ___ I feel q u ite com fo rtab le ___ I feel b arely c om fo rtab le ___ I feel u n co m fo rtab le ___ I feel restless and fidgety ___ I feel cram ped ___ I feel stiff ___ I feel num b (o r p ins a nd n eedles) ___ I feel s o re ___ I feel very painful BO DY P A RT D IS C O M F O R T P lease look at the figure. D o you feel discom fort in any p a rt o f y o u r body? Y E S/N O . If Y E S, m ark th e u n co m fo rtab le p arts o f your body o n the ch art in o r d e r o f im p o rta n c e from A to Z. F o r each body p a rt w hich y ou have m en tio n ed p lease r a te the discom fort using the scale: S C A L E : 1. J u s t n o tic e a b le d iscom fort 2. M o d e ra te d iscom fort 3. Q u ite bad discom fort 4. S ev ere discom fort 5. A lm o st u n b e a ra b le discom fort Figure 1. The subjective questionnaire. RESULTS General Comfort Ratings: Discomfort was assumed to be indicative of postural stress or fatigue. Cumulative fatigue increased over time and was found to be significant for all groups (p,01) using Page’s test o f trend. The effect is reduced by a lunch break but the fatigue continues rising afterwards. This confirms the finding o f Corlett and Bishop (ibid) that perceptions of postural discomfort are linearly related to the time o f exposure to that posture. The fatigue builds up rapidly early in the day for the data-typists. The 30-minute lunch break is inadequate as a recovery period for the data-typists and typists. Body Part Discomfort Ratings: This asked whether sub­ jects felt discomfort. If they answered “Y es” they were re­ quired to mark the location of discomfort on a body manikin and the severity of pain using the 5 point rating scale. The frequency of “Yes” responses changed throughout the day with the range at Time 1 being 5-25%, at Time 4 (before lunch) 25-50%, and at Time 6 (end o f the day) 30-40%. Ratings for the severity o f discomfort have to be judged against its frequency. At Time 1 this varied from 1-1,8, at Time 4 from 1,8-2,7, and at Time 6 from 1,4-3,5. Clearly some subjects were in considerable pain since these were mean values. Body parts in discomfort: The number of occasions on which discomfort was registered is expressed as a percentage of the number of times the questionnaire was administered. Figure 2 shows the neck and lower back to be the main sites of discomfort. There is also great variability between the occupations. Problems with the right shoulder/arm were unique in the data-typist profile. Low back, neck and shoulder discomfort was confirmed in the clinical examination and the video assessment of postural change; largely indicative of deficiencies in workstation design. 50 40 - 30 - 20 - 10 - 0 Clinical examination: A computerised subjective as­ sessment was followed by an objective examination (based on the Maitland principle), including assessment of physical posture; range of body movements; neurological aspects of sensations, reflexes and muscle power; and used palpation to determine soft tissue spasm, thickening, pain and stiffness. Figure 3 reveals which of the reported symptoms were related to previous trauma and/or pathology and which were not (Clinical). Overall 68% of problems clinically verified had no known cause. Neckache was experienced by 78% o f subjects. O f these 65% had no known cause. All of the subjects who complained of upper backache and 62% o f subjects who complained o f low back pain had no known previous trauma or pathology. Breakdown by group showed an average of 3 presenting symptoms for data-typists, 2 for typists, 1,85 for p ercen t 4 4 r 42 16 13 12 T-- ------ 1-- ------ 1-- ------ T Neck L /B a c k U /B a c k R /S h L / S h H ead R /T h ig h L /T h ig h Body a r e a s Figure 2. Body parts in discomfort. Bladsy 14 Fisioterapie, Mei 1990, deel 46 no 2 R ep ro du ce d by S ab in et G at ew ay u nd er li ce nc e gr an te d by th e P ub lis he r (d at ed 2 01 3. ) Especially if space and budget are limited A range of exercise patterns are offered. • Knee Extension/Flexion • Ankle Plantar/Dorsiflexion (Gastroc & Soleus) • Hip Abduction/Adduction • Hip Extension/Flexion • Shoulder Abduction/Adduction • Shoulder Extension/Flexion Clinical Emergencie/ Telephone: Johannesburg (011)613 # / l-f/Cape To*9 {02/ )89-f046JPietermaritzburg <0331 )94-8V77 Cnr. Outspan A Fortune Sir eels, Ntr. 6 Old Mutual Industrial Park, City Deep Postal Enquiries to: P.O. Box II0H3, Johanntsbitrg 2000, Physiotherapist R26 000 per annum (neg dependent on experience) plus free housing C D M (Pty) Lid operates an o p enc ast d iam ond m ine on the W est coast o f N a m ib ia O u r em ployees an d their families live in O ranjem und, a m odem , attractive tow n boasting a central shopping com plex, hospital, nursery school, prim ary school an d excellent sporting and recreational facilities. W orking w ith another P hysiotherapist an d one assistant, you will find the w ork varied and interesting, ranging from general rehabilitation program m es to chest physiotherapy and ante-natal care. T his is an excellent p ost for a newly qualified T h erapist or one w ith 1 - 2 years' experience to develop skills an d gain or im prove expertise in general practice. It is essential th a t you are registered w ith the S A S.W.A , , NAMIBIA M edical and D ental Council. T h e successful candidate should be able to converse in both official languages. Successful candidates w h o w ould have expatriate status in SWA/NAM IBIA, will be required to en ter into a 2 y ea r contract renew able by m utual ag ree m en t Benefits include: An end o f contract gratuity, generous leave, suitable m arried accom m odation, subsidised board for single em ployees, 13th cheque, prim ary and subsidised secondary schoolding, m em bership o f the D e Beers M edical Benefit Society an d assistance w ith relocation expenses. Applicants should write, giving full details to: The Senior Personnel Manager, CDM (Pty) Ltd, P 0 Box 35, Oranjemund, SW A/Namibia 9000. For informal enquiries contact Vanessa Line, Senior Physiotherapist, on (06332) 2353. CDM (Proprietary) Limited ARv.ARK THE BEST P tO P li Fun i HE JOB S 2 126 R ep ro du ce d by S ab in et G at ew ay u nd er li ce nc e gr an te d by th e P ub lis he r (d at ed 2 01 3. ) p ro g ra m m e rs an d 1,8 for re se a rc h ers. In the subjective assess­ m en t 86% o f p roblem s w ere re p o rte d to be occupationally aggravated (97% for the neck an d 67% for th e low back). T h e p hysio th erap ists’ re p o rts on each subject suggest th a t c o n ­ stra in e d w orking po stu re is likely to be at th e root o f m any of th e problem s, especially in th e case o f neck sym ptom s w here in the m ajority o f cases th e re was asso ciated trap ezii and rh o m b o id spasm which is indicative o f static loading a n d /o r repetitive m ovem ents. 15 10 5 0 CLINICAL (68%) TRAUMA (22%) PATHOLOGY (7%) TR & PATH (3%) S y m p to m s [ZH N eck m U /B a c k [ZH L / B a c k BS H e a d Figure 3. Classification of clinical symptoms. Table 1. Comparison of anthropometric and workstation dimensions Gomearison Student’s t test d a Work surface height versus Elbow height 2,22 <0.05 37 Seat height versus Popliteal height 3,28 <0.01 37 Seat depth versus Buttock-popliteal 2,45 <0.05 23 Anthropometric and workstation dimensions: T h e m ajor p oints n o ted h e re w ere th a t w ork surfaces w ere higher than elbow height, leading to h u n ch ed sh o u ld ers an d neck tension; seat height higher th an popliteal height, with re s u l­ tan t ischaem ia in th e thighs; seat d e p th g re a te r th an buttock- popliteal length for p ro g ram m ers an d re se a rc h ers w here o ld er chairs w ere in use, ren d e rin g im possible p ro p e r use of th e backrest. Basically th e w orkstations w ere such th at som e 84% o f subjects w ere forced to a d o p t p o o r p o stu res in som e way in o rd e r to work. DATA-TYPISTS TYPISTS PROGRAMMERS RES. (VDT) RES. (DESK) 120 100 8 0 6 0 40 20 0 5 10 15 fr e q u e n cy (seconds) frequency (se conds) L. 1 POSTURE SHIFT HEAD MOVEMENT Figure 4. Postural change Assessment of work posture: V ideo was u sed to exam ine th e frequency o f p o stu ral change, am o u n t o f tim e sp en t in differen t p o stu re s an d th e rela tio n sh ip betw een w ork p o stu re an d physical sym ptom s. F ig u re 4 show s th e freq u en cy o f shifts in p o stu re an d h e a d m ovem ents. It is evident how th e task p e rfo rm e d d e te rm in e s th e w ork p o stu re. T h e sp e e d o f data- e n try typing an d em phasis on visual inform ation retrieval lead to a co m p arativ e im m obility o f this g ro u p . Shifts in p o stu re o ccu r only once every 102 seco n d s an d h e a d m ovem ents every 13 seconds.. P o stu re s w ere c o d e d a c c o rd in g to th e p o stu re o f the h ead , tru n k a n d arm s. T h e relative fixity o f im m obility of p o stu re may b e derived by co m p arin g the p e rc e n ta g e o f tim e sp e n t in th e most fre q u e n t positio n for th e various g roups. F ig u re 5 reveals th e tre n d o f p o stu ra l co n strain t. T h e data- typists’ typical position was with th e head b e n t dow n and tw isted to the left, re a d in g from th e so u rce d o c u m e n ts and typing with th e right han d w ith th e tru n k ag ainst th e b ack rest. T ypists tu rn th eir h e a d s from th e sc re e n to the so u rc e d o c u ­ m ent in quick succession. R e s e a rc h e rs using V D T s a re fairly c o n stra in e d b u t d o not use th e term in als constantly. T h e P ro g ra m m e rs’ w ork layout was u n usual with a V D T o c c u ­ pying th e far left c o rn e r o f th e desk. T erm in al w ork thus re q u ire d twisting th e neck. H u n c h e d sh o u ld ers w ere also observed in 6 o f th e sub jects as a resu lt o f high d esk height. T h ese factors a re th o u g h t to b e largely resp o n sib le for th e neck p ro b lem s re p o rte d by this g ro u p in th e clinical exam in­ ation. frequency % □ □ HEAD □ □ TRUNK W& ARMS TREND Figure 5. Duration of postures (group median body position) T o en d eav o u r to elim in ate som e o f th e bias th a t must inevitably c re e p into o b servations m ad e by o n e p e rso n it was a rra n g e d for th e co n su ltan t ph y sio th erap ist to view th e video films a n d give an in d e p e n d e n t ju d g e m e n t on th e sites of p ro b a b le sym ptom s a c c o rd in g to th e w ork p o stu re s o f su b ­ jects. T h e physio th erap ist w atch ed 22 video films se lected at ran d o m . C om m on instan ces o f b a d p o stu re w ere sitting witht back su p p o rt, desks bein g to o high o r ch airs to low, w h ere the h e a d was ro ta te d , an d o th e r c ra m p e d positions. T h e id e n ti­ fied sites o f p ro b a b le sym ptom s w ere la te r c o m p a re d with the results o f th e clinical exam ination. In 16 o f th e 22 cases the c o rre sp o n d e n c e b etw een th e two was exact. T h a t is, w h ere a p ro b a b le site was identified as low back, th en this sym ptom was rev ealed in th e exam ination. C learly th e use of th e “expert eye” o f th e physio th erap ist may be a p o ten tial sub stitu te for lengthy video analysis. Relationships between measures: T h e previous p a r a ­ g ra p h is a good exam ple o f w hat th e study set o u t to achieve. T h a t is, to verify th e accu racy o f th e d ifferen t m eth o d s being Bladsy 16 Fisioterapie, M e i 1990, d ee 1 46 no 2 R ep ro du ce d by S ab in et G at ew ay u nd er li ce nc e gr an te d by th e P ub lis he r (d at ed 2 01 3. ) programmers and 1,8 for researchers. In the subjective assess­ ment 86% of problems were reported to be occupationally aggravated (97% for the neck and 67% for the low back). The physiotherapists’ reports on each subject suggest that con­ strained working posture is likely to be at the root of many of the problems, especially in the case of neck symptoms where in the majority of cases there was associated trapezii and rhomboid spasm which is indicative of static loading and/or repetitive movements. su b jects | 20 15 10 5 0 CLINICAL (68%) TRAUMA (22%) PATHOLOGY (7%) TR & PATH (3%) S ym p tom s I I N eck U /B a c k L _̂! L / B a c k ISM3 H e a d Figure 3. Classification of clinical symptoms. Table 1. Comparison of anthropometric and workstation dimensions Comparison Student's t test d a Work surface height versus Elbow height 2,22 •0.05 37 Seat height versus Popliteal height 3,28 <0.01 37 Seat depth versus Buttock-popliteal 2,45 <0.05 23 Anthropometric and workstation dimensions: The major points noted here were that work surfaces were higher than elbow height, leading to hunched shoulders and neck tension; seat height higher than popliteal height, with resul­ tant ischaemia in the thighs; seat depth greater than buttock- popliteal length for programmers and researchers where older chairs were in use, rendering impossible proper use of the backrest. Basically the workstations were such that some 84% of subjects were forced to adopt poor postures in some way in order to work. DATA-TYPISTS TYPISTS PROGRAMMERS RES. (VDT) RES. (DESK) 120 100 8 0 6 0 4 0 20 0 5 10 15 fre q u e n cy (secon d s) frequency (secon d s) H I POSTURE SHIFT ^ HEAD MOVEMENT Figure 4. Postural change Assessment of work posture: Video was used to examine the frequency of postural change, amount of time spent in different postures and the relationship between work posture and physical symptoms. Figure 4 shows the frequency of shifts in posture and head movements. It is evident how the task performed determines the work posture. The speed of data- entry typing and emphasis on visual information retrieval lead to a comparative immobility o f this group. Shifts in posture occur only once every 102 seconds and head movements every 13 seconds.. Postures were coded according to the posture of the head, trunk and arms. The relative fixity of immobility of posture may be derived by comparing the percentage of time spent in the most frequent position for the various groups. Figure 5 reveals the trend of postural constraint. The data- typists’ typical position was with the head bent down and twisted to the left, reading from the source documents and typing with the right hand with the trunk against the backrest. Typists turn their heads from the screen to the source docu­ ment in quick succession. Researchers using VDTs are fairly constrained but do not use the terminals constantly. The Programmers’ work layout was unusual with a VDT occu­ pying the far left corner of the desk. Terminal work thus required twisting the neck. Hunched shoulders were also observed in 6 of the subjects as a result of high desk height. These factors are thought to be largely responsible for the neck problems reported by this group in the clinical examin­ ation. freq u en cy % □ □ HEAD E H TRUNK ARMS - B~ TREND Figure 5. Duration of postures (group median body position) To endeavour to eliminate some of the bias that must inevitably creep into observations made by one person it was arranged for the consultant physiotherapist to view the video films and give an independent judgement on the sites of probable symptoms according to the work postures of sub­ jects. The physiotherapist watched 22 video films selected at random. Common instances of bad posture were sitting witht back support, desks being too high or chairs to low, where the head was rotated, and other cramped positions. The identi­ fied sites of probable symptoms were later compared with the results of the clinical examination. In 16 of the 22 cases the correspondence between the two was exact. That is, where a probable site was identified as low back, then this symptom was revealed in the examination. Clearly the use of the “expert eye” of the physiotherapist may be a potential substitute for lengthy video analysis. Relationships between measures: The previous para­ graph is a good example of what the study set out to achieve. That is, to verify the accuracy of the different methods being Bladsy 16 Fisioterapie, Mei 1990, dee! 46 no 2 R ep ro du ce d by S ab in et G at ew ay u nd er li ce nc e gr an te d by th e P ub lis he r (d at ed 2 01 3. ) employed to collect data. In such case studies where the sample is small it is necessary to insure against methodologi­ cal inadequacies by using several techniques in order to get cross-validation. Tests o f significance should be treated with caution since the chance of introducing spurious errors is relatively high. However, a brief summary of how some meth­ ods compared may be of interest, whilst bearing in mind the need for caution in their interpretation. There was a strong association revealed between the general comfort ratings and the body part discomfort ratings at each time level using the Chi-square test (p0,001) - it appears these ratings are measur­ ing the same thing in the same subjects ie. discomfort. Both these subjective ratings were found to correlate with postural constraint (percentage of time in one position). Spearman’s rho rank coefficients were rs = 0,36 (p,05) for postural con­ straint and general comfort and rs = 0,41 (p,05) for postural constraint and body part discomfort. Again, this result is tenuous in view of the fact that the postural analysis is a rather gross measure and the technique is in its infancy. DISCUSSION The findings here may be described as interesting but not substantive. The sample was too small to permit generali­ sations to be made about occupational differences in work posture. Nevertheless, the findings are broadly in line with the results of overseas research. The most important result was that as much as 68% o f reported symptoms verified clinically had no known cause. It is possible that long forgotten events such as motor vehicle accidents, whilst not producing any pain at the time, could be the precursor of later problems. Mini­ traumas over long periods is also a possibility. When symp­ toms appear it is understandable that people will associate present events with their occurrence. Given such complications it is not that easy to identify the cause of symptoms. However, the analysis of work pos­ ture, corroborated by the other evidence does suggest that in many instances work design is a major source of postural stress which appears to aggravate, if not actually cause, some symptoms. The reduction in the frequency of symptoms o f the programmers has occurred simultaneously with a change in the workstation layout. Neck complaints were caused by hun­ ched shoulders and twisted necks. New workstations with dedicated terminal desks were purchased, which no longer require these constrained postures. It therefore seems that interventions to change the pos­ tural demands of work by redesigning the layout of worksta­ tions will often be successful, especially if the problem is fairly obvious as in the case of the programmers. Indeed the main emphasis in the interventionist approach is that aetiology is frequently too obscure or complex and that if jobs posing risks can be identified then it makes more sense to reduce postural stress by redesigning the work. It is not so easy to assess the contribution o f work exposure to a chronic disorder first appearing as a result o f trauma. In several cases symptoms were reported to occur only when the individual was working or when engaged in another activity at home. Gardening for lo w back su ffer ers was a com m on r e sp o n se w h ilst neck/shoulder problems could be triggered by a range o f activities. Data-entry typing was an activity which led to prob­ lems and two aspects could be responsible for this the repe­ titive nature of the task and the static loading. Interventions overseas have concentrated on work reorganisation in this instance introducing more frequent rest pauses and other tasks to increase the amount o f postural variety. The conse­ quence is that postural fatigue is not permitted to build up and there is reduced discomfort4. CONCLUSION Further research should employ much larger samples but intervention programmes must be run longitudinally be­ fore the effects can be properly assessed. This presents a problem as far as continuous monitoring of subjective discom­ fort is concerned because o f the time involved in administer­ ing questionnaires. Diary recording has met with mixed success, the main problem being that subjects need to be highly motivated. If there is an appropriate health profes­ sional available such as an occupational health nurse then a daily monitoring programme may well be feasible. The most likely possibility would seem to be the administration o f a questionnaire on a regular basis over a number of years to monitor the progress of intervention efforts. It appears that the techniques used in the measurement of postural discomfort have considerable potential for ident­ ifying problems in work environments other than the office, where the hazards associated with musculoskeletal disorders are greater. A t present there is little involvem ent of physiotherapists in occupational settings in South Africa, possibly because the need for their presence has not been realised. It is argued that if the contribution o f occupational exposure to disorders is high 5then prevention efforts would indeed be more effective if initiated at this point. Physiother­ apists are uniquely qualified to identify the problems associ­ ated with poor posture and the field o f occupational physiotherapy offers an interesting challenge for those con­ cerned with the prevention o f symptoms rather than their cure. References 1. SHACKEL B, CHIDSEY KD, & SHIPLEY P. The assessment of chair comfort. Ergonomics 1969; 12(2):269-306. 2. CORLETT EN & BISHOP RP. A technique for assessing postural discomfort. Applied Ergonomics 1976; 19(2):175-182. 3. ROLAND M & MORRIS R. A study of the natural history of back pain. Part 1: Development of a reliable and sensitive measure o f disability in low-back pain. Spine 1983; 8(2):141-144. 4. ONG CN. V D T work place design and physical fatigue. A case study in Singapore. In Ergonomics and health in modem offices, ed by E Grand- jean. London: Taylor and Francis, 1984. 5. HAGBERG M & WEGMAN DH. Prevalence rales and odds ratios of shoulder-neck diseases in different occupational groups. British Journal o f Occupational Medicine 1987; 44:602-610. ACKNOWLEDGEMENTS Thanks to: Lynne Thompson (consultant physiotherapist) and her students Meryl Edley and Andrea Illsley whose contributions to this study were invaluable. Dr KF Mauer, Vice-President of the HSRC. All HSRC staff who participated and assisted in the study. ♦ Physiotherapy, May 1990, vol 46 no 2 Page 17 R ep ro du ce d by S ab in et G at ew ay u nd er li ce nc e gr an te d by th e P ub lis he r (d at ed 2 01 3. )