AustJPhysiotherv55i4Szeto

7/30/2019 AustJPhysiotherv55i4Szeto

1/6Australian Journal of Physiotherapy 2009 Vol. 55 Australian Physiotherapy Association 2009 257

Szeto et al: Postural muscle activity in computing tasks

Introduction

Work-related musculoskeletal disorders are common amongcomputer users, especially affecting the neck and shoulderregion (Tittiranonda et al 1999, Gerr et al 2004). Typingand mousing are common ofce tasks, involving both nemotor control and associated postural activity of the musclesof the cervical spine. Physiotherapists treat patients withcomputer-related musculoskeletal disorders, and therefore anunderstanding of the mechanisms underlying impairmentsis important to guide management. Consistent differencesin patterns of muscle activation between symptomatic andasymptomatic people have been found during standardisedtyping tasks, suggesting differences in motor control maybe an underlying mechanism (Szeto et al 2005a to d).

Studies on patients with chronic neck pain have reportedincreased muscle activation compared to asymptomaticcontrols during tasks (Madeleine et al 1999, Nederhandet al 2003, Falla et al 2004). However, eld studies oncomputer workers have produced contradictory resultswith either small or no difference between symptomatic

and asymptomatic computer workers in muscle activityamplitudes or muscular rest (Hgg and strm 1997,Nordander et al 2000, Roe et al 2001, Westgaard et al2001, Blangsted et al 2003). The lack of control over tasksand environment in these eld studies may contribute tothe different ndings. Only two case-control laboratory

studies have examined differences in muscle activity duringtyping, mousing and type and mouse tasks. Sjgaard andcolleagues showed a trend for more upper trapezius muscleactivity in symptomatic workers (Sjgaard et al 2006, Thornet al 2007). The authors acknowledged that the lack ofsignicant differences may have been due to the brief taskduration (25 min) and poor EMG quality. A pilot studyby Kallenberg et al (2006) suggested several parametersof muscle activity may be different between symptomaticworkers and controls.

The classic method of examining surface EMG amplitudesin occupational studies has been through the computationof the amplitude probability distribution function (Jonsson1982). The 50th percentile values of the amplitudeprobability distribution function has been used widely as anindicator of the average muscle activity (commonly referredto as median amplitude) in studies of real or simulatedoccupational tasks (Hansson et al 2000, Nordander et al2000, Westgaard et al 2001, Blangsted et al 2003). For

example, median amplitudes of about 110% of maximumvoluntary contraction of the upper trapezius have beenreported during keyboarding and mousing tasks (Blangstedet al 2003, Dennerlein and Johnson 2006) in healthyadults.

During computing tasks symptomatic emale ofce workers

demonstrate a trend towards higher cervical postural

muscle load than asymptomatic ofce workers:

an experimental studyGrace PY Szeto1, Leon M Straker2 and Peter B OSullivan2

1The Hong Kong Polytechnic University, Hong Kong, 2Curtin University of Technology, Australia

Questions: Do symptomatic emale oce workers perorm computing tasks with higher cervical postural muscle loads (interms o higher amplitudes and less muscular rest) and more discomort compared with asymptomatic individuals? Are thesedierences in postural muscle loads consistent across bilateral (typing) and unilateral (mousing) conditions? Design: anexperimental case-control study. Participants: 18 symptomatic emale oce workers and 21 asymptomatic emale oceworkers. Intervention: Three conditions (typing, mousing, and type-and-mouse) were perormed in random order. Outcomemeasures: Muscle load was measured as median amplitude and gap requency using surace EMG o bilateral cervicalerector spinae and upper trapezius. Discomort was measured using a numerical rating scale. Results: The case groupdemonstrated 4.3% (95% CI 0.1 to 8.4) higher amplitude during typing and 3.5% (95% CI 0.1 to 6.9) higher amplitude duringtype-and-mouse in the right cervical erector spinaecompared with the control group. There was a similar dierence betweengroups in the let cervical erector spinaewhich also demonstrated a 1.2 gaps/min (95% CI 2.3 to 0.0) lower frequency duringtyping. The case group had signicantly higher discomort during all conditions compared with the control group. The casegroup demonstrated higher median amplitudes and lower gap requencies than the control group during bilateral conditions(typing and type-and-mouse) compared with unilateral conditions (mousing) or both muscle groups. Conclusion: There wasincreased amplitude and decreased muscular rest in the cervical erector spinae o oce workers perorming typing andmousing tasks. These ndings may represent a mechanism underlying computer-related musculoskeletal disorders. [SzetoGPY, Straker LM, OSullivan PB (2009) During computing tasks symptomatic female ofce workers demonstratea trend towards higher cervical postural muscle load than asymptomatic ofce workers: an experimental study.Australian Journal of Physiotherapy55: 257262]

Key words: Computers, Word processing, Electromyography, Neck pain, Physiotherapy


2/6

Australian Journal of Physiotherapy 2009 Vol. 55 Australian Physiotherapy Association 2009258

Research

EMG gaps have also been reported in recent studies asa potential risk factor for musculoskeletal disorders. Thesegaps are dened as periods of rest with very low muscleactivity, ranging from 0.5% to 3% of maximal voluntarycontraction (Veiersted et al 1990). It has been proposedthat muscles with lower gap frequencies may have fewerrest periods in certain low-threshold motor units, and these

muscle bres may be exposed to overuse, thus contributing tomusculoskeletal disorders (Veiersted et al 1990, Hgg 1991).Past studies have shown differences in muscle activationduring non-computing tasks between symptomatic andnon-symptomatic individuals but there is no clear evidenceof these differences during computing tasks, except forour earlier reports of differences during typing (Szeto et al2005a and c). The present study aimed to compare whetherthere were differences in muscle activation, in terms ofmedian amplitude and gap frequency, between symptomaticand asymptomatic female ofce workers when performingtyping and mousing tasks. If differences in motor controlexist, it suggests that physiotherapists should directtheir management towards correcting the motor control

mechanisms instead of focusing on pain relief. Therefore,the specic research questions for this study were:

Do symptomatic female ofce workers perform1.computing tasks with higher cervical posturalmuscle loads (in terms of higher amplitudes and lessmuscular rest) and more discomfort compared withasymptomatic individuals?Are these differences in postural muscle loads2.consistent across bilateral (typing) and unilateral(mousing) conditions?

Method

Design

The study employed an experimental case-control design.Female ofce workers were recruited through conveniencesampling within the university and nearby ofces.They were divided into two groups: a case group and acontrol group. All participants were tested under threeconditions: 1) typing, 2) mousing, and 3) type-and-mousein a controlled environment. The order was randomisedby each participant drawing lots. Each condition lasted 20min with a 5 min rest in between. Cervical postural muscleload (median amplitude and gap frequency) and discomfortwere measured and compared between groups and betweenconditions.

ParticipantsTo be included, ofce workers had to be female, agedbetween 20 and 50 years, performing a minimum of 4hours of computer work daily, and carrying out mainly text-editing tasks. They were excluded if they had past traumaticinjuries or surgical interventions in their neck and upperlimb regions.

An interview questionnaire modied from the StandardisedNordic Questionnaire (Kuorinka et al 1987) was used tocollect information about musculoskeletal symptoms. Thosewith moderate to severe discomfort (> 4/10) in the neck andupper limb regions for at least 3 of the past 12 months were

assigned to the case group. Those with mild discomfort (


3/6

Australian Journal of Physiotherapy 2009 Vol. 55 Australian Physiotherapy Association 2009 259


Table 1. Characteristics o participants.

Characteristic Case group(n = 21)

Control group(n = 18)

Age (yr), mean (SD) 28 (9) 24 (2)Hand dominance, n right (%) 20 (95) 18 (100)Body mass index (kg/m2), mean (SD) 20.9 (2.7) 20.6 (1.2)

Height (m), mean (SD) 1.60 (0.04) 1.59 (0.04)Work experience (yr), mean (SD) 1.9 (2.2) 0.5 (1.2)Working duration (hr/wk), mean (SD) 45 (7) 40 (11)Computer use at work (hr/day), mean (SD) 7.2 (2.7) 6.0 (3.9)Keyboard use (hr/day), mean (SD) 4.5 (3.1) 4.7 (5.2)Mouse use (hr/day), mean (SD) 4.5 (3.9) 5.5 (4.2)Previous typing training, n yes (%) 7 (33) 1 (5)Typing method used, n (%)

Touch-typingTraditional typewriter styleCertain ngers only

13 (62)2 (10)6 (29)

8 (44)5 (28)5 (28)

Duration o discomort in past 12 months (days), mean (SD) 90 (165) 0.8 (5.7)Discomort in past 7 days, n yes (%) 19 (90) 2 (10)

Table 2. Mean (SD) o EMG or each group during the three conditions and mean dierence (95%CI) between groups.

Outcome Muscles Groups Dierence between groupsCase group

(n = 21)Control group

(n = 18)Case minus control

50th% muscle activity (%MVC)

Typing RCES 14.1 (7.3) 9.8 (5.2) 4.3 (0.1 to 8.4)LCES 14.7 (6.8) 10.5 (5.1) 4.2 (0.2 to 8.2)RUT 11.8 (10.1) 9.0 (11.4) 2.8 (4.2 to 9.7)LUT 11.3 (11.3) 5.8 (5.7) 5.5 (0.5 to 11.4)

Mousing RCES 11.2 (4.5) 8.2 (4.3) 3.1 (0.2 to 5.9)LCES 11.6 (4.9) 8.6 (4.9) 3.0 (0.2 to 6.2)RUT 8.5 (8.7) 6.6 (8.2) 1.9 (3.6 to 7.4)LUT 6.7 (7.2) 3.1 (2.3) 3.6 (0.1 to 7.2)

Type-and-mouse RCES 13.1 (5.6) 9.6 (4.8) 3.5 (0.1 to 6.9)LCES 13.7 (5.7) 11.0 (9.3) 3.6 (0.0 to 7.2)RUT 11.0 (9.3) 8.9 (9.7) 2.1 (4.0 to 8.3)LUT 8.2 (7.4) 5.2 (5.1) 3.1 (1.2 to 7.3)

Gap requencies (gaps/min)

Typing RCES 0.8 (3.7) 1.8 (4.1) 1.0 (3.5 to 1.5)LCES 0.0 (0.1) 1.2 (2.3) 1.2 (2.3 to 0.0)

RUT 7.5 (17.6) 8.5 (10.1) 2.7 (12.4 to 7.1)LUT 6.9 (15.7) 15.6 (21.7) 8.6 (20.8 to 3.5)

Mousing RCES 1.0 (4.4) 6.2 (13.3) 5.3 (12.1 to 1.6)LCES 0.3 (0.9) 6.8 (14.1) 6.6 (13.6 to 0.5)

RUT 5.6 (9.2) 3.2 (4.9) 2.6 (2.4 to 7.7)

LUT 10.4 (16.2) 14.0 (22.3) 3.6 (16.1 to 9.0)

Type-and-mouse RCES 0.9 (2.6) 1.9 (4.8) 1.1 (3.5 to 1.4)LCES 1.1 (3.4) 3.3 (7.5) 2.2 (5.9 to 1.5)

RUT 8.2 (14.3) 12.9 (12.9) 4.8 (13.7 to 4.1)

LUT 7.2 (12.7) 17.4 (20.7) 10.2 (21.7 to 1.3)

RCES = right cervical erector spinae, LCES = let cervical erector spinae, RUT = right upper trapezius, LUT = let upper trapezius, MVC =maximum voluntary contraction


4/6


Research

between groups. Repeated measures analysis of variancewas performed to compare the median amplitude and gapfrequency between groups and conditions. Signicance

level was set atp < 0.05.

Results

Flow of participants through the study

Thirty-nine female ofce workers were recruited over a one-year period. All the participants were initially screened fortheir eligibility through telephone calls. One participant wastransferred from the control group to the case group whena history was taken in detail, and one control participantwas unable to attend as scheduled. Therefore, the casegroup consisted of 21 participants and the control group of18. Baseline characteristics of participants are presented in

Table 1.Cervical postural muscle loads and discomfortacross groups

Group data for amplitude and gap frequency are presentedin Table 2. The case group had signicantly higher cervicalerector spinae amplitudes than the control group in all threeconditions bilaterally (except left side during mousing).The case group demonstrated a trend towards higher uppertrapezius amplitudes but this was only signicant forthe left upper trapezius during mousing. The case groupdemonstrated a general trend for a lower gap frequencythan the control group during the three conditions for bothmuscle groups, although this was only signicant for leftcervical erector spinae for typing (mean difference 1.2,95% CI 2.3 to 0.0) and marginal for mousing (meandifference 6.6, 95% CI 13.6 to 0.5).

Group data for discomfort across time are presented inTable 3. The case group had signicantly more discomfortat all times during the three conditions compared with thecontrol group.

Cervical postural muscle loads across groups(case versus control) and conditions (bilateralversus unilateral)

The case group demonstrated higher amplitudes than the

control group during bilateral conditions (typing and type-and-mouse) compared with unilateral conditions (mousing)for cervical erector spinae (p < 0.001) and upper trapezius(p < 0.001).

The case group demonstrated lower gap frequencies than the

control group during bilateral conditions (typing and type-and-mouse) compared with unilateral conditions (mousing)for both muscle groups (p = 0.04).

Discussion

Symptomatic female ofce workers performed computingtasks with some higher cervical postural muscle loads(in terms of higher amplitudes and less muscular rest)compared to asymptomatic individuals and these were likelyto be clinically signicant. Although the between-groupdifferences were mainly signicant for the amplitude forcervical erector spinae, the consistent pattern of increasedamplitude and reduced gap frequency across both musclegroups lends support to an overall clinically meaningfulnding despite individual statistical non-signicance(Hopkins et al 2009). The lack of statistically signicantndings may relate to large variation within groups, acommon problem in surface EMG studies (Mathiassenet al 2002). Post-hoc sample size calculation using gapfrequency data for right upper trapezius suggests that over100 individuals would be needed to nd a signicant effectsize of 0.8 between typing and mousing conditions (PASS2005).

Where present, higher amplitude and lower gap frequencyin the case group may indicate that their muscles wereworking harder and with less rest. Sjgaard et al (2006)were able to detect increased upper trapezius muscleactivity in symptomatic females only during a stress test,

but not during normal typing, mousing, and type-and-mouse tasks. Data presented by Kallenberg et al (2006)suggested dominant upper trapezius activity was greaterin symptomatic individuals during mousing but not typingor typing-and-mousing, though no statistical analysis wasreported. Our earlier studies showed signicantly increasedtrapezius muscle during typing tasks (Szeto et al 2005aand c). The current ndings thus provide the best availableevidence that the cervical muscles of symptomatic computerworkers work harder for the same physical task. This may bethe mechanism underlying the development of symptoms.

The present study found higher muscle activity in typingcompared to mousing, in line with previous reports by

Cooper and Straker (1998) and Dennerlein and Johnson(2006). We examined differences between symptomaticand non-symptomatic individuals for a type-and-mousecondition for the rst time. This condition was a combinationof bilateral and unilateral hand actions and muscle loadswere similar to that of typing. This suggests that when

Table 3. Mean (SD) discomort scores or each group and mean dierence (95% CI) between groups.

Discomort (0 to 100) Time(min)

Groups Dierence between groups

Case group Control group Case minus control

Typing 0 9 (10) 0 (0) 9 (5 to 14)10 12 (9) 0 (0) 12 (8 to 16)20 13 (11) 1 (2) 13 (8 to 18)

Mousing 0 13 (12) 0 (0) 13 (7 to 18)10 11 (10) 0 (0) 11 (6 to 15)20 12 (10) 0 (1) 11 (7 to 16)

Type-and-mouse 0 7 (8) 0 (0) 7 (4 to 11)10 10 (9) 0 (0) 9 (5 to 14)20 11 (11) 0 (0) 11 (6 to 16)


5/6

Australian Journal of Physiotherapy 2009 Vol. 55 Australian Physiotherapy Association 2009 261


ofce workers perform combined typing and mousing,their cervical postural muscles may be working at highlevels of activation, especially in symptomatic persons. Inaddition to type of computing task, physiotherapists shouldalso consider factors such as workstation setting, individualpostural habits, work duration and demand and mentalstress when examining computer users (Gerr et al 2004,

Johnston et al 2008).The case group also showed high activity in left (contralateralside) upper trapezius during mousing, suggesting a globalmotor response to the task. This may reect a lack of efcientmotor control of the head neck region. We have previouslyfound that symptomatic individuals demonstrate higheramplitudes in the 10th% and 90th% amplitude probabilitydistribution function indicating greater variation in motorcontrol (Szeto et al 2009).

As this study was cross-sectional, a causal relation betweenmuscle activity and discomfort cannot be concluded.However increased levels of neck and shoulder muscle

activity may result in increased loading of sensitisedstructures which may be pain provocative. Prospectivestudies could be carried out to nd evidence of cause andeffect.

The differences in EMG found in the present study couldbe secondary to peripheral and/or central pain mechanisms,psychological stress, repetitive activities or a combinationof these, and therefore contribute to an increase in tissueload, muscle fatigue, and tissue sensitisation (Visser andvan Dieen 2006). Furthermore, these ndings may alsobe associated with altered motor programming secondaryto pain, as seen in patients with acute whiplash injury(Nederhand et al 2003) or chronic neck pain (Falla et al

2004).The present ndings suggest that symptomatic individualshave a tendency to perform computing tasks with highermuscle activity and less muscular rest in the cervical posturalmuscles. Although typing and mousing tasks were different(being bilateral and unilateral tasks), the results weremostly consistent across the tasks. The results lend furthersupport to the nding that altered patterns of motor activity,characterised by increased activity of postural muscles, areassociated with work-related musculoskeletal symptoms.These ndings may have important implications for theassessment and management of these disorders. A focuson training relaxed and efcient motor control strategiesduring typing and mousing for patients with neck and armpain may be appropriate in the physiotherapy managementof these disorders. n

Footnotes: aNoraxon USA Inc, USA, bNational Instruments,Austin USA.

Ethics:The Human Research Ethics Committee of HongKong Polytechnic University approved this study. Writteninformed consent was gained from all participants beforedata collection began.

Competing interests: None declared.Support: Departmental research grant, Department ofRehabilitation Sciences, The Hong Kong PolytechnicUniversity, Hong Kong.

Acknowledgements: The authors acknowledge Mr PaulDavey who produced the software to analyse the EMGvariables, and Mr Man Cheung for technical support in theuse of the EMG equipment.

Correspondence: Dr Grace Szeto, Department ofRehabilitation Sciences, The Hong Kong PolytechnicUniversity, Hung Hom, Kowloon, Hong Kong. Email:[email protected]

References

Blangsted AK, Hansen K, Jensen C (2003) Muscle activityduring computer-based oce work in relation to sel-reportedjob demands and gender. European Journal of AppliedPhysiology89: 352358.

Cooper A, Straker L (1998) Mouse versus keyboard use: acomparison o shoulder muscle load. International Journal ofIndustrial Ergonomics22: 351357.

Dennerlein JT, Johnson PW (2006). Dierent computer tasksaect the exposure o the upper extremity to biomechanicalrisk actors. Ergonomics49: 4561.

Falla D, Bilenkij G, Jull G (2004) Patients with chronic neckpain demonstrate altered patterns o muscle activationduring perormance o a unctional upper limb task. Spine29: 14361440.

Gerr F, Marcus M, Monteilh C (2004) Epidemiology omusculoskeletal disorders among computer users: lessonlearned rom the role o posture and keyboard use. Journalof Electromyography and Kinesiology14: 2531.

Hgg GM (1991) Static work load and occupational myalgia: anew explanation model. In: Anderson P, Hobart D, DainoJ (eds) Electromyographical Kinesiology, Amsterdam:Elsevier, pp 141144.

Hgg GM, strm A (1997) Load pattern and pressure painthreshold in the upper trapezius muscle and psychosocialfactors in medical secretaries with and without shoulder/neck disorders. International Archives of Occupational andEnvironmental Health69: 423432.

Hansson G-, Nordander C, Asterland P, Ohlsson K,Stromberg U, Skerving S et al (2000) Sensitivity o trapeziuselectromyography to dierences between work tasks infuence o gap denition and normalization methods.Journal of Electromyography and Kinesiology10: 103115.

Hopkins WG, Marshall SW, Batterham AM, Hanin J (2009)Progressive statistics or studies in sports medicine andexercise science. Medicine and Science in Sports andExercise41(1): 312.

Johnston V, Jull G, Darnell R, Jimmieson NL, Souvlis T(2008) Alterations in cervical muscle activity in unctional

and stressul tasks in emale oce workers with neck pain.European Journal of Applied Physiology103: 253264.

Jonsson B (1982) Measurement and evaluation o localmuscular strain in the shoulder during constrained work.Journal of Human Ergology11: 7388.

Kallenberg LAC, Hermans HJ, Vollenbroek-Hutten MMR (2006)Distinction between computer workers with and withoutwork-related neck-shoulder complaints based on multiplesurace EMG parameters. International Journal of IndustrialErgonomics36: 921929.

Kuorinka I, Jonsson B, Kilbom , Vinterberg H, Biering-Srenson F, Andersson A, Jorgensen K (1987) StandardisedNordic Questionnaires or the analysis o musculoskeletalsymptoms. Applied Ergonomics18: 233237.

Madeleine P, Lundager B, Voigt M, Arendt-Nielsen L (1999)Shoulder muscle co-ordination during chronic and acuteexperimental neck-shoulder pain. European Journal ofApplied Physiology79: 127140.

Mathiassen SE, Burdor A, van der Beek AJ (2002) Statisticalpower and measurement allocation in ergonomic intervention


6/6


Research

studies assessing upper trapezius EMG amplitude. A casestudy o assembly work. Journal of Electromyography andKinesiology12: 4557.

Nederhand MJ, Hermens HJ, Ijzerman MI, Turk DC, Zilvold G(2003) Chronic neck pain disability due to an acute whiplashinjury. Pain102: 6371.

Nordander C, Hansson G-, Rylander L, Asterland P, BystromJU, Ohlsson K et al (2000) Muscular rest and gap requencyas EMG measures o physical exposure: the impact o worktasks and individual related actors. Ergonomics43: 19041919.

Power Analysis or Statistical Sotware (2005) NCSS and PASS.Number Cruncher Statistical Systems, Kaysville, Utah.

Roe C, Bjorklund RA, Knardahl S, Waersted M, VollestadNK (2001) Cognitive perormance and muscle activationin workers with chronic shoulder myalgia. Ergonomics 44:116.

Sjgaard G, Sgaard K, Hermens HJ, Sandsj L, Lubli T, ThornS, Vollenbrock-Hutten MMR et al (2006) Neuromuscularassessment in elderly workers with and without workrelated shoulder/neck trouble. European Journal of AppliedPhysiology96: 110121.

Szeto GPY, Straker LM, OSullivan PB (2005a) A comparisono symptomatic and asymptomatic oce workers perormingmonotonous keyboard work 1. Neck and shoulder musclerecruitment patterns. Manual Therapy10: 270280.

Szeto GPY, Straker LM, OSullivan PB (2005b) A comparisono symptomatic and asymptomatic oce workers perormingmonotonous keyboard work 2. Neck and shoulderkinematics. Manual Therapy10: 281291.

Szeto GPY, Straker LM, OSullivan PB (2005c) The eects ospeed and orce o keyboard operation on neckshouldermuscle activities in symptomatic and asymptomatic oceworkers. International Journal of Industrial Ergonomics35:429444.

Szeto GPY, Straker LM, OSullivan PB (2005d) The eectso typing speed and orce on motor control in symptomaticand asymptomatic oce workers. International Journal ofIndustrial Ergonomics35: 779795.

Szeto GPY, Straker LM, OSullivan PB (2009) Examining thelow, high and range measures o muscle activity amplitudes insymptomatic and asymptomatic computer users perormingtyping and mousing tasks. European Journal of Applied

Physiology106: 243251.Thorn S, Forsman M, Zhang Q, Taoda K (2002) Low-

threshold motor unit activity during a 1-h static contractionin the trapezius muscle. International Journal of IndustrialErgonomics30: 225236.

Thorn S, Sgaard K, Kallenberg LAC, Sandsj L, SjgaardG, Hermens HJ, Kadeors R, Forsman M (2007) Trapeziusmuscle rest time during standardized computer work acomparison o emale computer users with and withoutself-reported neck/shoulder complaints. Journal ofElectromyography and Kinesiology17: 420427.

Tittiranonda P, Burastero S, Rempel D (1999) Risk actorsor musculoskeletal disorders among computer users.Occupational Medicine: State of the Art Reviews14: 1738.

Veiersted KB, Westgaard RH, Andersen P (1990) Pattern omuscle activity during stereotyped work and its relation tomuscle pain. International Archives of Occupational andEnvironmental Health62: 3141.

Visser B, van Dieen JH (2006) Pathophysiology o upperextremity muscle disorders. Journal of Electromyographyand Kinesiology16: 116.

Westgaard RH, Vasseljen O, Holte KA (2001) Trapeziusmuscle activity as a risk indicator or shoulder and neck painin emale service workers with low biomechanical exposure.Ergonomics44: 33953.

Were changing our name!

From Vol 56 No 1 March 2010,Australian Journal of Physiotherapy becomes

Journal of PhysiotherapyPrint ISSN: 1836-9553 Online ISSN: 1836-9561

AustJPhysiotherv55i4Szeto

Documents

Transcript of AustJPhysiotherv55i4Szeto