arm and trunk posture during work in relation to shoulder and neck pain and trapezius activity

7/29/2019 Arm and Trunk Posture During Work in Relation to Shoulder and Neck Pain and Trapezius Activity

1/10

ELSEVIER

Clinical Biomechanics Vol. 12, No. 1, 22-31,p. 1997Copyright fQ 1997 Elsevier Science Limited. All rights reservedPrinted in Great Britain0268-0033197 17.00 + 0.00

PII: s0268-0033(97)00048-4

Arm and trunk posture during work inrelation to shoulder and neck pain andtrapezius activity0 Vasseljen Jr PhD, R H Westgaard PhDDivision of Organization and Work Science, Norwegian University of Science andTechnology, Trondheim, Norway

AbstractObjective. To investigate work technique in relation to work-related shoulder and neck pain(SNP) and upper trapezius muscle activity.Design. A matched pair, case-control field study of female employees with and withoutSNP.Background It has proved difficult to distinguish subjects w ith SNP from those without byvocational electromyographic recordings from the upper trapezius muscle. Other potentialrisk indicators include psychosocial factors and work technique. This study focuses on thelatter.Methods. Manual (14 pairs) and office workers (24 pairs) were recorded during a 30-minwork period. Simultaneous recordings of upper trapezius activity by surfaceelectromyography and arm and upper back postures by inclinometers were analysed.Results. Cases and controls were not differentiated on the basis of arm elevation or of trunkposture in the sagittal plane. No significant correlations were found between variablesaveraging the muscle activity and the arm elevation over the recording period. Statisticallysignificant correlations were, however, found between these variables when analysingrecordings at high time resolution (0.2 s) and adjusting for the delay in arm elevationrelative to the upper trapezius muscle activity (r = 0.43, manual group; r = 0.32, officegroup).Conclusions. Factors other than arm elevation probably contribute more significantly to theload in the upper trapezius muscle, and to the development of work-related SNP in worksituations with moderate arm elevation.RelevanceThe study suggests that arm and trunk posture recordings are not a sensitive indicator ofrisk of shoulder and neck complaints in work with low to moderate biomechanicaldemands. @ 1997 Elsevier Science Ltd. All rights reserved.Key words: Postural angles, inclinometers, shoulder and neck, trapezius, EMG, case-control, workClin. Biomech. Vol. 12, No. 1, 22-31, 1997

IntroductionBoth electromyographic (EMG) recordings from thetrapezius and recordings of arm posture are used toindicate biomechanical exposure, i.e., forces generatedin the body to meet work demands, for the shoulderand neck muscles*2. laboratory studies have shownReceived: 15 February 1995; Accepted: 31 July 1996Correspondence and reprint requests to: Ottar Vasseljen Jr., Divisionof Organization and Work Science, Norwegian University of Scienceand Technology, Trondhe im, Norway

that there can be considerable deviations in the twomeasures3. t is not known whether this is the case alsoin vocational settings, and if guidelines for vocationalmuscle load and arm posture are interchangeable. Thisuncertainty affects commonly used guidelines foracceptable work load in the shoulder and neck region,e.g., average muscle load below 8% of maximal muscleforce4 and arm elevation below 1.5.Vocational studies of biomechanical exposure andrelated health effects have had problems showing adose-effect relationship between the level of muscle


2/10

Vasseljen and Westgaard: Arm and trunk posture during work 23

activity and complaints in the shoulder and neck regionat low biomechanical exposure.. Equivalent studies,using postural angles as an indicator of work load, arerare. In a cross-sectional study upper arm posture wasrelated to symptoms in the shoulder and neck. Thisrelationship weakened during 2-year and 3-year follow-up9. The results therefore need to be confirmed byother studies. Aaras and Westgaard showed that areduction in shoulder and neck complaints coincidedwith a reduction in vocational trapezius activity afterergonomic improvements, but there was no clearevidence of a concurrent reduction in postural angles ofthe arms. The study showed a positive correlationbetween arm flexion and trapezius EMG in standardizedwork tasks, an effect also demonstrated in laboratorystudies*,. This association was no longer presentwhen work tasks with more varied work contents wereincluded.

The present investigation is part of a stratified case-control study of -female office and manual workers.Workers with shoulder and neck complaints andsymptom-free controls were individually matched onwork tasks, age and employment time. Biomechanicalexposure was measured by upper trapezius EMGrecordings and recordings of arm and trunk posture.Psychosocial and personality factors were assessed nstructured interviews. In previous reports it was shownthat EMG measurements, particularly muscle activityat rest, differentiated cases and controls in the manualgroup, while psychosocial and personality variableswere important differentiating variables for the officegroup4.1. The present study reports the results of thepostural recordings, and how these related to therecording of muscle activity in the upper trapeziusmuscle.The aim of this paper was twofold: (1) to investigatewhether cases and controls in the two workgroupscould be differentiated on the basis of arm and trunkposture, and (2) to examine to what extent armelevation and trapezius EMG recordings are equivalentmeasures of vocational biomechanical exposure in theshoulder and neck region.MethodsA case-control design with matching on work-relatedbiomechanical exposure factors was used. The subjectswere matched n age and on exposure factors such asworking hours, ength and type of employment, both

0:ith respect to c rrent and historical exposure factors;see Vasseljen ar d Westgaard14 for details on the5atching procedu e. EMG and postural angle record-ings were performed during a 30-min work period. Itwas ensured that the one or two most commonlyperformed work tasks were chosen for the workrecording, in which the same type of work was per-formed by both the case and the control of each pair.The study was approved by the regional medical

ethics committee. All participants gave signed in-formed consent.SubjectsBased on interviews of 297 employees in both manualand office work, 76 females were included in the study,giving a total of 38 pairs. The main selection criterionfor caseswas presence of continuous shoulder and neckpain for at least 2 weeks during the previous year, withpain at 3 or higher on a scale from 0 to 61h. Subjectswith no shoulder and neck pain during the same periodwere included as controls. Those with shoulder andneck pain due to injuries or systemic disease wereexcluded, as were pregnant women, those withdiagnosed fibromyalgia, and those with symptoms ofdegenerative joint disease of the cervical spine. It was arequirement that individuals in manual and office workhad been employed for a minimum of 1 and 2 yearsrespectively. The manual work group included 14 pairs(median age 26.5 (95%CI 25-31) years, medianemployment time 5.8 (95%CI 4.5-7) years), whereasthe office group included 24 pairs (median age 37.5(95%CI 34-43) years, median employment time 14.8(95%CI 10-19) years). The manual workers weresignificantly younger (P


3/10

24 Clin. Biomech. Vol. 12, No. 1, 1997

Ag) electrodes, 6 mm in diameter and with a 20-mmintercentre distance, were used. The centre of theintegrated electrode was placed at a point 2/3 thedistance from the spine of the seventh cervical vertebra(C,) to the lateral edge of the acromion. The EMGsignals were band-passed filtered at lo-1250 Hz andstored on a digitizing recorder(Earth Data 128). Thesignals were later full-wave rectified and averaged at0.2-s resolution. Detailed information on the EMGrecording technique is found in Vasseljen andWestgaard14.EMG calibrations were carried out before and afterthe vocational recordings. Two maximal contractionswere performed both in shoulder elevation (shrugs)and in 90 arm abduction. The contraction thatproduced the highest EMG amplitude was used as thebasis for the EMG calibration procedure (%EMG,,,).The vocational muscle activity was described by vari-ables averaging the EMG over the measurement period(static, median and peak levelsi) and by variablesaiming to quantify aspects of the tempora l EMGpattern or pauses in the muscle activity (EMG gaps,long gaps, gap timelY).

Postural angle recordingsPostural angles of the upper arms and upper back wererecorded continuously during the 30-min work period,in parallel with the bilateral EMG recordings from theupper trapezius. Postural angles were measured byelectrolytic liquid level sensors (Physiometer PHY-400,Premed A/S, Norway). Two sensors were placed atright-angles in a cylinder (weight 60 g, length 60 mm;diameter 30 mm), allowing measurements of posturalangles in two planes. The accuracy of the posturalmeasurements have been evaluated by mounting thepotentiometer on an isokinetic device and measuringdynamic movements at different speeds (3O*s- to903- ) .2 The difference between the predeterminedposition and the measured position varied by no morethan 3.6. The sensors are thus considered acceptablefor the purpose of ergonomic studies. The posturalsignals were recorded and stored on the digital taperecorder simultaneously with the EMG signal. Theywere later sampled in the laboratory by a computer andtime-averaged at a resolution of 0.2 s. Sensors wereattached to the lateral aspect of both upper arms at themid-point between the shoulder and elbow joint. Onthe upper back the sensor was placed at the upperthoracic spine, attached to a back-pack worn by thesubject. Postural angles were measured in terms ofdeviations from the reference body position; a relaxedstanding position with the arms hanging along the sideof the body and the eyes fixed at a distant eye-levelpoint. The back recording was analysed for movementin the sagittal plane only; i.e. trunk flexion andextension. For the upper arms, the sensors werelocated to nominally record movements in the sagittal

(flexion/extension) and frontal (abduction/adduction)planes. However, no recording was made of rotation inthe glenohumeral joint. For this reason, and since armmovement in both the sagittal and the frontal planes isconsidered to contribute to the risk of musculoskeletalcomplaints, the recordings were transformed into polarcoordinates according to the equations:arcsin2X = sin2a + sin2b * cos2a (1)arctan Y = sin b/sin a (2)where a and b are the two measured angles, X is theelevation angle, and Y indicates horizontal orientationangle relative to the sagittalifrontal planes. In thispaper only elevation angle is reported.The data reduction of the continuous recording o fpostural angles was performed by a cumulativeprobability function analysis over the entire recordingperiod, giving static (10% probability level), median(50%) and peak (90%) levels18*21,and by variablesattempting to quantify aspects of the temporal posturalangle pattern (time and number o f movements above30 and below 10 arm elevation).Statistical analysisAn ANOVA model with repeated measures was used forcomparisons of casesand controls in both work groups.Variables showing significant differences were retainedfor post hoc analyses. In compliance with the distri-bution of the observations, either unpaired t test orMann-Whitney U test was used to test differencesbetween work groups22. Comparisons were performedtwo-tailed and differences were considered significantat the PcO.05 level. Pearson product moment corre-lation for normally distributed data and Spearman rankorder correlation in case of non-normal distributionswere used to test correlationS22. Correlation coeffi-cients (r) were calculated and PcO.05 taken to indicatestatistical significance.It was anticipated that the response measured by thepostural angle is slightly delayed relative to the EMGresponse of muscles that contribute to the movement.Cross-correlation functions23 between trapezius EMGand arm elevation were therefore constructed bycalculating the correlation coefficient between the twochannels with the signals time-shifted in 0.2 s intervalsfrom zero to a maximal time shift of 30 s. The delay ofthe mechanical response relative to the muscle activitywas determined by the time shift of the central peak inthe cross-correlation function. The half-width of thecentral peak was used to indicate the sharpness ofassociation between the two variables.ResultsPostural angles vs. musculoskeletal complain tsThe side with the highest median arm elevation level


4/10


was considered most at risk for developing shoulderand neck myalgia, and thus used in the furtherpresentation of the group data. The static, median andpeak arm elevation levels are given for cases andcontrols in the manual group (Figure la) and the officegroup (Figure lb,.I There was no difference betweencases and controls in either work group. In fac t a highconsistency between cases and controls was found forall variables in both work groups. The considerablevariation in job characteristics o f the manual and officegroups was not to any appreciable extent reflected inthe static level (7.1 v,s 6.1, P=O.O4), but the manualgroup showed a marginally higher median (19.5 vs15.6, P=O.O03) and a clearly higher peak (40.6 vs26.6, P


5/10


Manual Office

0Manual Office

o-Manual Office

Figure 2. Temporal arm elevation pattern obtained from vocational recordings for both work groups, as illustrated with the number of arm movemantsper minute above 30 (A) and below 10 (C) arm elevation, and time per minute (s) with the arm located above 30 (B) and below 10 (0) arm elevation,Median values with 95% confiden ce intervals are given.


6/10


The variation in muscle activity at different intervalsof arm elevation (O-15, 16-30, 31-45 and >45) wasinvestigated after adjusting for the delay in the armelevation. A considerable variation in the muscleactivity was found for all intervals of arm elevation,exemplified in Figure 5 for a manual worker. Slightlylower variations were seen for the office workers.

Generally, the EMG activity at arm elevation above45 showed approximately twice the variation of theEMG activity at arm elevation below 15.Static, median and peak EMG activity wereinvestigated for the same postural intervals in eightmanual and eight office workers. Four cases and fourcontrols were selected at random from each workgroup. Large interindividual variation in the EMGvariables was seen for all arm elevation intervals inboth work groups (Figure 6). The static EMG activitylevel was similar in the two work groups, and littleaffected by arm elevation at angles up to 45. Both

A 2 14-c3g 12-g I0.$ a-rn.2 6-2mo 4-f;s 2-5f O

A0.70.6

0 IO 20 30 40Arm elevation

B 0.20.1 -0.0 IAL-0.1 I, I I I I I I

-30 -20 -10 0 10Seconds

20 30O* 0dc4 0I ' I I I

0 5 10 15 20 25 30B

0.7 ,Number of movements below IO a rm elevation (mine)

c -l0.6 1

0

i II I I I I0 10 20 30 40 50 60Time below IO arm elevation (wmin-)

Figurr 3. Relationships between simultaneously recorded arm elevationand upper trapezius EMG activity averaged over a IO-min work recordingfor manual (filled symbols, n = 28). and office workers (open symbols,n = 48). Median traperiu:; activity (%EMG,.,) against median armelevation (A), number of EMG gaps against number of movements below10 arm elevation (B), and time below 0.5 %EMG,, against t ime below10 arm elevation (C) are shown. Values are given for right arm andtrapezius.

-30 -20 -10 0 10 20 30Seconds

Figure 4. Cross-correlation functions between upper trapezius activityand arm elevation. One manual worker (A) and one office worker(B) areshown. Dots represent correlation coefficients for various time shiftintervals of 0.2 s from zero to + 30s. Zero co rresponds to correlationcoefficient without time adjustment, indicated by the arrows. Highestcorrelation occurred at a latency of 0.6 s for the manual worker and 0.4 sfor the office worker.


7/10


groups showed a clear increase in the static level at armelevations above 45.

DiscussionThis study has shown that -a number of variables,derived from the continuous recording of arm elevationangle, did not distinguish cases with complaints in theshoulder and neck region from healthy controls in thetwo study groups. Nor was there any significant effectof pain status on trunk posture in the sagittal plane forthe manual workers. Arm elevation and trapeziusEMG appeared to be largely independent representa-tions of biomechanical exposure in the shoulder andneck region for these work situations, where armelevation in excess of 45 rarely happened.

MethodsContinuous measurements of arm posture during work,for assessment of biomechanical exposure relating toshoulder and neck complaints, have been carried outonly to a limited extent. Such techniques have morefrequently been used to analyse back movements24725.In studies comparing observational and instrumenta-tion-based techniques it is argued that observationaltechniques are insufficient in field trials. Percentage ofworktime with flexed trunk (>20) is reported to varyby more than 20% when direct observation (the OWASmethod) was compared with simultaneous measure-ment of back postural angle by inclinometers25. Directobservation, self-reported postures and instrumenta-tion-based techniques for recording of posture gavesignificantly different results in another study2(j. Theinaccuracy of observational methods is to some extentacknowledged by their quantification systems, operatingonly with three or four categories of postural anglequantifications2.The introduction of apparatus that continuouslyrecords body postures allows quantification of posturalangles at high time and spatial resolution. The higheraccuracy relative to observational methods presumablyensures better reliability and sensitivity, but notnecessarily improved validity when used as a riskindicator for shoulder and neck pain. Several factorsmodify the relationship between arm posture andbiomechanical forces in the body, including armsupport, hand loads, speed and accuracy of move-ments, eccentric or concentric work, and shoulderelevation. Due to the pairwise matching of cases andcontrols on physical exposure factors at work, weassume that these factors do not seriously bias theinferences made from the results of this study.Only trunk and upper arm postures were recorded toavoid cumbersome instrumentation. The upper armposture was only quantified in terms of verticaldisplacement from the reference position and not

15 -O-16" A

10 -

5-

O- I,5 0 10 20 30 40 50

I 1630" B105

0 IL ( -, ,,5 0 10 20 30 40 50

10 131-46" C

5

0 IAl&+, ,,5 0 10 20 30 40 50

I >46" D10 I

5

0 L4hb40 10 20 30 40 50

Upper trapezius activity (%EMG,,)Figure 5. Distribution of the vocational EMG recorded trapezius muscleactivity at differe nt arm elevation intervals for a manual w orker. Theintervals O-15 (A), 16-30 (6). 31-45 (CL and >45 (D) are shown.Within each arm elevation interval the histogram columns are scaled toshow relative distr ibution of EMG activity levels.


8/10


for horizontal movements (i.e. horizontal arm a simultaneous reduction in the forward flexion of theadduction/abduction), since ergonomic guidelines for head towards a neutral posture, trapezius EMG wasarm posture almost exclusively state risk of musculo- positively correlated with arm flexion and negativelyskeletal complaints in terms of arm flexion or abduction. correlated with head inclination. This indicates thatThe trapezius activity is 20-50% higher in armabduction relative to the same elevation in flexion3T*.

arm posture, rather than head posture, is the moreimportant postural variable to influence the trapeziusHorizontal arm position may thus partially explain theobserved dissociation between arm elevation andtrapezius EMG activity. Variation in head inclinationmay likewise contribute to the dissociation. However,in an occupational task with increasing arm flexion and

EMG level.Trapezius EMG vs. arm elevationIn the present study there was a concurrent increase in

:..: .2 1 .A * :

I I I I

.

.. .

E40 -

gp 30-Wsg 20-4%I IO-d

O-

:.

.;> . .1 l l..I I I I

F

25 1

O-1 I I I I

Figure 6. Vocational static, median, and peak upper trap&us EMG levels at different arm elevation intervals for eight manual (A,C,E; filled symbols) andeight office workers (B,D,F; open symbols). Lines represent median levels for the subjects shown.


9/10


arm elevation and median and peak EMG muscleactivity, but not in static EMG level at arm elevationangles below 45. Analysis of the complete material of76 subjects may have shown a statistically significantassociation also for static EMG level. However, thepractical result of a low association between staticEMG level and arm elevation at low to moderateelevation angles is likely to remain; thus there appearsto be a potential for short periods of rest for thetrapezius muscle at moderate arm elevation. This resultprobably rests on (1) the trapezius being a stabilizer ofthe shoulder girdle and not a primary mover o f the arm,and (2) dynamic movement rather than static posturewas examined.There were no correlations between EMG and armposture when both types of variables were averagedover the recording period. Correlations were, however,observed when analysing recordings at high timeresolution and adjusting for the time delay between theactivity in the upper trapezius and the mechanicalresponse of the arm. These correlations were strongerfor the manual workers, presumably due to the morerapid arm movements in this group.Considerable interindividual variation in the EMGactivity was observed at all arm elevation intervals andfor both work groups. Differences in the weighthandled have probably contributed to the inter-individual variation in case of the manual workers.Some of the postal workers were using dynamicthrowing movements, which may cause the hand loadto change according to arm elevation, thereby maskinga presumed relationship between EMG and armposture; However, the interindividual differences inthe handling of external loads were small or non-existent in case of the office workers. The dissociationbetween trapezius EMG levels and arm elevationangles is thus not easily explained by applying morecomplete biomechanical models including, e.g. handloads. Trapezius EMG and arm elevation may there-fore be considered independent indicators of posturalload in the shoulder and neck region for work situationswith low biomechanical exposure. Shoulder elevationand the level of agonist-antagonist cocontraction in thestabilization of the shoulder girdle may contribute tothe dissociation.Arm elevation as a risk indicator for complain ts in theneck and shoulderWork at constantly high arm elevation angles has beenfound to cause more pain and stiffness in the neck andshoulder than similar work tasks with variable workingheight .8 This is also an assumption of observationalmethods quantifying postural stress n the shoulder andneck region. Such methods have defined various lowrisk limits for developing shoulder and neck complaintsat arm flexion or abduction of O-45o29,3o,O-3031 or0- 15 (neutral)/45 (mild elevation)32. Thus mostsubjects of our study groups were in the low-risk

category with respect to biomechanical exposure. Inthe manual group workers exceeded 45 arm elevation5-10% of the observation time, but there was noindication that this contributed to an increased risk ofshoulder and neck complaints, nor would it be con-sidered a significant risk situation in terms of the bio-mechanical guidelines.A potentially interesting extension of traditional bio-mechanical analysis is to consider time in low elevationpositions, i.e. below 10 of arm elevation, as an analogyto the EMG gap analysis. However, this variableshowed no sign of differentiating cases and controlseither in the manual group, who previously were differ-entiated on the basis of the EMG gap analysis14,or inthe office group. The result is consistent with thefinding that the static trapezius activity in both theoffice and the manual groups were not much affectedby arm elevation at elevation angles below 45.The present study does not rule out the possibilitythat those with pain had modified their earlier paineliciting posture and movements to those less loading,e.g. by cases changing work technique after paindevelopment. Longitudinal studies of the relationshipbetween postural variables and health effects areneeded to confirm the results of the present study.Cases and controls were, however, distinguished onbasis of variables not easily transformed into posturalvariables measured during occupational activity.Furthermore, a recent longitudinal study did not findevidence of a change in EMG variables following achange in pain status6,33.A masking of an underlyingbiomechanical exposure-health effect relationship bypain-related changes in work technique thereforeappears less likely.These results put a perspective on biomechanicalguidelines for arm posture. While near-neutral workpostures have been recommended, there is little evid-ence in this study to suggest that median arm elevationis a sensitive indicator of risk of shoulder complaints atelevations below 45. This conclusion is qualified by thefollowing points: (1) choosing cases with complaintslocated to the glenohumeral joint or rotator cu ff musclesmay have yielded a different result; (2) postural vari-ables not monitored (e.g. head inclination) may con-tribute as discriminating risk factors, either by them-selves or interacting with arm elevation; (3) the handload is an important contributor to biomechanical load,but was in this study hopefully eliminated by the pair-wise matching procedure; and (4) median arm eleva-tion may be a risk factor in work situations with morestatic arm postures than for the present workgroups.However, shoulder elevation, rest ing EMG activity(manual workers in this study and6), perceived generaltension (both groups of this study and34), and psycho-logical and psychosocial risk factors (office workersin this study and35) are probably more importantrisk factors in the development of shoulder and neckcomplaints at low biomechanical exposure levels.


10/10


AcknowledgementsThis study was supported by a grant from theNorwegian Research Council.References

1. Westgaard, R. H. Measurement and evaluation ofpostural load in occupational work situations. Eur JApplPhysiol, 1988,57. 291-304.2. Kilbom, A. Assessment of physical exposure in relationto work-related musculoskeletal disorders-whatinformation can be obtained from systematicobservations? Stand J Work Env iron Health, 1994,20[Special issue], 30-45.3. Mathiassen, S. E. and Winkel, J. Electromyographicactivity in the shoulder-neck region according to armposition and glenohumeral torque. Eur J Appl Physiol,1990,61,370-379.4. Grandjean, E. Fiiting the Task to the Man. A Textbook ofOccupational Ergonomics. 4th edn, Taylor & Francis,London, 1988,22-27.5. Chaffin, D. B. anld Andersson, G. B. J. OccupationalBiomechanics, 2nd edn, John Wiley, New York, 1991,357.6. Veiersted, K. B.. Westgaard, R. H. and Andersen, P.Electromyographic evaluation of muscular work patternas a predictor of trapezius myalgia. Stand J WorkEnviron Health, 1993, 19,284-290.7. Jensen, C., Nilsen K.. Hansen, K. and Westgaard, R. H.Trapezius muscle load as a risk indicator for occupationalshoulder-neck complaints. IntArch Occup EnvironHealth, 1993,64,415-423.8. Kilbom A., Persson, J. and Jonsson, B. G. Disorders ofthe cervicobrachial region among female workers in theelectronics industiy. Int J Ind Erg, 1986, !, 37-47.9. Jonsson, B. G., Persson.J. and Kilbom A. Disorders ofthe cervicobrachial region among female workers in theelectronics industry. A two-year follow up. Int J Ind Erg,1988.3, l- 12.10. Aaras, A. and Westgaard, R. H. Further studies onpostural load and musculoskeletal injuries of workers atan electro-mechanical assembly plant. Appl Ergon. 1987,l&211-219.11. Aaras, A., Westgaard, R. I-I. and Stranden, E. Posturalangles as an indicator of postural load and muscularinjury in occupational work situations. Ergonomics,1988,31,915-933.12. Sigholm, G.. Herberts, P ., Almstrom. C. and Kadefors,R. Electromyographic analysis of shoulder muscle load.J Orthop Res, 1984, 1,379-386.13. Schiildt, K., Ekholm, J.. Harms-Ringdahl, K., Nemeth,G. and Arboreliur,, U. P. Effects of changes n sittingwork posture on sratic neck and shoulder muscle activity.Ergonomics, 1986.29,1525-1537.14. Vasseljen, O., Jr. and Westgaard, R. H. A case-controlstudy of trapezius muscle activity in office and manualworkers with shoulder and neck pain and symptom-freecontrols. Int Arch Occup Environ Health, 1995.67,11-18.15. Vasseljen, O., Jr.. Westgaard, R. H. and Larsen, S. Acase-control study of psychological and psychosocial riskfactors for shoulder and neck pain at the work place. ZntArch Occup Environ Health, 1995,64,375-382.16. Westgaard, R. I-I. and Jansen, T. Individual and workrelated factors associated with symptoms ofmusculoskeletal complaints. I. A quantitativeregistration system. Br J Ind Med, 1992,49, 147- 153.17. Jensen, C., Vasseljen O., Jr. and Westgaard, R. H. Theinfluence of electrode position on bipolar surface EMG

18.

19.

20.

21.22.23.

24.

25.

26.

27.

28.

29.

30.

31.

32.

33.34.

35.

recordings of the upper trapezius muscle. Eur JApplPhysiol, 1993,67,266-273.Jonsson, B. Kinesiology - with special reference toelectromyographic kinesiology. In Contemporary ClinicalNeurophysiology, ed. W. A. Cobb and H. Van Duijn,EEG [Suppl34], 1978,417-428.Veiersted, K. B., Westgaard, R. H. and Andersen, P.Patterns of muscle activity during stereotyped work andits relation to muscle pain. Int Arch Occup EnvironHealth, 1990,62,31-41.Hagen, K. B., Sorhagen, 0. and Harms-Ringdahl. K.Influence of weight and frequency on thigh and lower-trunk motion during repetitve lifting employing stoopand squat techniques. Clinical Biomechanics, 1995. 10,122-127.Aaras, A. and Stranden, E. Measurement of posturalangles during work. Ergonomics, 1988,31,935-944.Altman, D. G. Practical Statisticsfor Medical Research.Chapman and Hall, London, 1991.Box, G. E. P. and Jenkins G. M. Time Series Analysis:Forecasting and Control. Holden-Day. San Francisco,1976,393-395.Nordin M., Grtengren R. and Andersson, G. B. J.Measurement of trunk movements during work. Spine,1984,9,465-469.Burdorf, A., Derksen, J., Naaktgeboren, B. and vanRiel. M. Measurement of trunk bending during work bydirect observation and continuous measurement. AppliedErgonomics, 1992,23,263-267.Baty. D., Buckle, P. W. and Stubbs, D. A. Posturerecordings by direct observation, questionnaireassessmentand instrumentation: a comparison based ona recent field study. In The Ergonomics of WorkingPosture, ed. N. Corlett, J. Wilson and I. Manenica.Taylor & Francis, London, 1986.283-292.Genaidy, A. M., Al-Shedi. A. A . and Karwowski. W.Postural stress analysis in industry. Applied Ergonomics,1994.25,77-87.Sakakibara, H., Miyao, M., Kondo. T. and Yamada, S.Relation between overhead work and complaints of pearand apple orchard workers. Ergonomics, 1987,30,805-815.Armstrong, T. J., Foulke, J. A., Joseph, B. S. andGoldstein, S. A. Investigation of cumulative traumadisorders in a poultry processing plant. Am Ind HygAssocJ. 1982,43,103-116.Keyserling, W. M. Postural analysis of the trunk andshoulders in simulated real time. Ergonomics. 1986,29,569-583.Kilbom, A., Persson, J. and Jonsson, B. R isk factors forwork-related disorders of the neck and shoulder-withspecial emphasis on working postures and movements. InThe Ergonomics of Working Postures. Models, Methodsand Cases, eds. N. Corlett, J. Wilson and I. Manenica.Taylor & Francis, London, 1986,44-53.Genaidy, A. M.. Guo, L., Eckart, R. and Tischbein, D.A postural stress analysis system for eva luating bodymovements and positions in industry. Proc ErgonomicsSot Conf, 1993,346-351 (Abstract).Veiersted. K. B. Sustained muscle tension as a risk factorfor trapezius myalgia. Znt JZnd Erg, 1994,14,333-339.Westgaard, R. H. and Jansen, T. Individual and workrelated factors associated with symptoms ofmusculoskeletal complain ts. II. Different risk factorsamong sewing machine operators. BrJ Ind Med. 1992,49,154-162.Westgaard, R. H.. Jensen, C. and Hansen, K. Individualand work-related risk factors associated with symptomsof musculoskeletal complaints. Znt Arch Occup EnvironHealth. 1993,64.405-413.

arm and trunk posture during work in relation to shoulder and neck pain and trapezius activity

Documents