A field methodology for ergonomic analysis in occupational manual materials handling

Applied

00034870(95)ooo7&7

Ergonomics

Vol. 28, No. 3, PD. 203-208. 1997 0 1997 Ek&ier Science Lid All rights resetwd. Printed in Great Britain 0003-687Oi97 $17.00 + 0.00

ELSEVIER

A field methodology for ergonomic analysis in occupational manual materials handling P. Capodaglio, E.M. Capodaglio and G. Bazzini Ergonomics Unit, Rehabilitation Montescano (PV), Italy

Center

of Montescano,

Clinica

de1 Lavoro

Foundation,

27040

(Received 25 April 1995; in revised form September 1995)

A methodology is presented for the ‘on-site’ evaluation of work-related physical activities. In a first session, video recordings were made of six industrial plant workers during their routine occupational tasks. Heart rate (HR) and subjective perception of effort were monitored. The video recordings were then analysed with appropriate software to determine the musculoskeletal load (‘Vision 3000’, Promatek, Montreal) and to evaluate energy expenditure (‘Energy’, University of Michigan). The indirect estimates of energy expenditure were validated in a second session by monitoring the six subjects’ oxygen consumption ( VOz) during the same activities with a portable telemetric oxgen uptake analyser (Cosmed ‘K2’, *Rome). No statistically significant differences were found between direct measurements of V02 and the computerised estimates of energy expenditure. Biomechanical parameters obtained in the two sessions did not differ. Therefore, we conclude that the ‘Energy’ programme and the ‘Vision 3000’ program provide a fast and reliable profile of job requirements. @ 1997 Elsevier Science Ltd Keywords:

manual

materials

handling,

energy expenditure,

Introduction

biomechanical

analysis, musculoskeletal

injuries

loading from external loads. The methods available for measuring physical work loads can be categorised as follows: direct measurements, observations, interviews, diaries and questionnaires. Direct measurements include electromyographic recordings, posture and movement recordings with goniometers, inclinometers and accelerometers. These methods are quantitative and relatively accurate, but only a limited number of body regions can be assessed at one time. Moreover, carrying monitoring devices can hinder the worker and hereby influence work methods. The use of questionnaires, diaries and interview techniques provides an opportunity to study cumulative exposure over time, an important parameter which is not usually available with direct measurements. However, the relatively low reliability and validity of the questionnaires developed so far make their use debatable (Wiktorin et al, 1993). Recently, the validity of diaries and interviews was found to be considerably higher than that of questionnaires (Kilbom, 1994). Observation methods may strike a balance between the high cost of direct measurements and the low validity, and subjectivity of questionnaires, diaries and interviews. The most precise and accurate operational estimates of loading may be obtained by biomechanical modelling. Biomechanics provides calculations of compressive forces based on real time observations of joint position, external forces and anthropometry of the

Manual materials handling (lifting, carrying, pushing, pulling) represents an occupational risk factor that has to be confined within safe limits. Acute and chronic work-related injuries may be attributed to excessive force demanded by the task, inadequate osteoarticular structures, or insufficient general or local aerobic capacity. Energy expenditure measurements, subjective perceptions of general fatigue and biomechanical analyses of the specific tasks serve to develop a profile of the strain of work-related physical activities. A variety of methods of measuring energy expenditure have been reported. These include: direct measurement of oxygen consumption with an oxygen uptake analyser; indirect calculations based on normative values (Pezzagno and Capodaglio, 1991); estimates derived from the linear relationship between heart rate (HR) and oxygen consumption (VO,) under stead state conditions at medium-high intensities of effort ( K strand and Ryhing, 1954; Legge and Banister, 1986; Margaria et al, 1965; Roozbar et al, 1979); and predictive formulas of energy expenditure during elementar components of a complex work cycle (Garg et al, 1978). The indirect methods for energy expenditure estimates have the advantage of being more comfortable and less obstructive to manual handling than the direct methods. Work-related musculoskeletal disorders may be caused by prolonged static postures or biomechanical

203

Computer bused job analysis: P. Capodaglio

204

Methods

subject. The recently available dedicated software programs can be used to anaiyse the compression forces on the L5-Sl intervertebral disc and to estimate the biomechanicai loading at each joint. Computerised postural analysis represents an alternative to the traditional ‘observational’ techniques that have led to ‘macropostural’ [OWAS method] (Kahru et al, 1977; Heinsaimi, 1986) and ‘micropostural’ (Keyseriing, 1986; Kiibom et af, 1986; Armstrong et al, 1982; Genaidy et al, 1994) classifications. Winter (1990) has provided an excellent summar of the advantages and disadvantages of the imaging rnc asurement techniques. The aim of our paper is to present a case study of two computerised systems as a computer based field methodology for assessing the physical demands of either the repetitive phases or the most physically demanding phases of the work cycles.

Six industrial plant workers volunteered to participate in the study. Their anthropometric characteristics are reported in Tab/e I. Ail subjects gave written informed consent. Two field sessions were held on different working days to verify the consistency of the parameters measured. In the first session subjects were observed while performing three consecutive work cycles. Dimensional variables of the six work cycles were recorded. Details of the work cycles are reported in Table 2. Subjects (Table I) and work cycles (Table 2) were numbered correspondingly. Subjects were asked to wear a black body suit and reflective body markers were positioned on one shoulder, elbow, wrist, hip, knee, ankle and on the L5-Sl intervertebral space of each subject investigated (American Academy of Orthopaedic Surgeons, 1965). Video recordings with a Sony Video8 TM CCD-F501 camera were made of each subject during the work cycles. The camera was set perpendicular to the plane of motion. The lens iongitudinai axis height was set at 1.2 m from the floor. Heart rate (HR) was monitored every 30 set with a cardiotachometer (Polar Sport Tester, Finland) and subjective perception of effort was estimated with a IO-point scale (Borg, 1982). Borg’s IO-point scale is a simple category scale for differential use that has positive attributes of a general-ratio scale. Numbers are anchored by readily understood verbal expressions describing peripheral effort sensations such as aches and pain. Tbe video recordings were then anaiysed using the ‘Energy’ program (from the University of Michigan) and the ‘Vision 3000’ analysis system [from Promatek Ltd, Montreal] (Blench, 1993).

Table I Physical characteristics of the subjects. Maximal HR values were calculated from the formulas: HRmax = 220 - age (y) for male and female subjects under 30 years, HRmax = 200 - age Q for female subjects over 30 years Subject (n”)

Sex

1 2 3 4 5 6 (mean)

f f m m m m

Age (years) 45 47 42 36 48

Body mass (kg)

Height (cm)

HR max

69 52 75 71 83 65

159 165 172 171 181 160

155 153 178 184 177 176

(69)

(168)

Table 2 Description of the six work cycles and relative elementary components maintained during each work cycle are specified

Work

Wription

cycle (n”) 1

et al

Duration (mia)

investigated.

The duration,

number of frames anaiysed and posture

FiWlleS analysed

Elementary components

Posture (% work cycle)

Preparation of sacks (25 or 50 kg) at the frequency of l-1 .Ymin; automatic filling and manual fastening. Carrying (1.5-2 m) and placing on the footboard (0.3-l m)

1.6

288

Walking Li~in~lowering Carrying Arm work

Erect (98%) Bent back (2%)

Handling small sacks; small sacks (0.5 kg) are taken with one hand at the frequency of 291min from a trolley and handed at a height of 0.7 m to another worker nearby

1

180

Lifting/lowering Arm work

Erect (70%) Bent back (30%)

Emptying barrels containing solids (50 kg); the worker performs arm work

6

1080

Walking Lifting/lowering Carrying Arm work

Erect (100%)

Emptying barrels containing dust; the barrels (50 kg) are taken from a bench (height I .5 m) and emptied into mixer (height 0.9 m) at the frequency of 4/min

0.8

144

Li~in~lowering Arm work

Erect (100%)

Placing packagings on a footboard; the boxes (26.5 kg) are taken from a conveyor belt (height 0.9 m) at the frequency of 3/min, carried for 2.5 m and placed on a footboard (height between 0.4 m and 1.15 m)

1

180

Walking Lifting/lowering Carrying Arm work

Erect (80%) Bent back 20%

1.1

198

Walking Lifting/lowering Carrying Arm work

Erect (100%)

Emptying sacks; the sacks (20-50 kg) are taken at the frequency of ilmin at variable height (1.7-0.3 m). carried for 4 m and emptied down a shute (height 0.7 m)

Computer

based job analysis:

Table 3 Elementary components, postures and related dimensional variables of the work cycles according to the ‘Energy’ program Elementary components Posture

Lifting/ lowering

Modality

Variable

Sitting Standing Standing,

Time bent back

stoop Squat Semi-squat One-handed Arm

Walking

Carrying

Weight lifted Initial/final height

Distance Time Slope Extended arms on the sides At chest height

Load Distance Time Slope

Holding

Extended arms on the sides At chest height Extended arms, one hand

Load Time

Pushing/ pulling

At bench height At a height of 1.52 m

Force exerted Hand height Distance

Hand-work

One or two hands, light One or two hands, heavy

Time

Arm-work

Time Lateral (180”), both arms Lateral (1809, one arm Lateral (90”) standing Lateral @I”), sitting, both hands Lateral (!N”), sitting, one hand Horizontal, standing Horizontal, sitting One arm, light One arm, heavy Two arms, light Two arms, heavy

The ‘Energy’ program is a mathematical model for the indirect estimation of energy expenditure. The program is based on the formulae developed by Garg et al (1978) which consider the elementary components of a complex work cycle (Table 3). The observed dimensional parameters of the work cycles become the variables of the predictive equations of energy expenditure. According to the anthropometric parameters of the subject examined and the dimensional variables of the work cycle, the program performs the partial (a single component) and total (a whole work cycle) calculation of the energy expenditure. Data obtained can be reported as metabolic equivalent units (MET) [l MET = 3.5 ml Oz/kg min]. The ‘Vision 3000’ program is a two-dimensional (2-D) biomechanical analysis system based on the acquisition and elaboration by means of a 80486 MSDOS personal computer of frames recorded on video during the work cycle. A camera auto advance unit automatically advances the camera when capturing the video frames. The camera records 30 frames per sec. Reflective markers are placed on the joints to be investigated and a source of light is needed. The system recognises the spatial position of the marked joints during movement and represents the body-image as a ‘stick diagram’. The program provides the following

P. Capodaglio

et al

205

data: kinematic description of the work cycle investigated; range of motion performed by each marked joint and percentage of working time spent in each position; compressive forces on the L5-Sl segment; determination of the recomended weight limits for the lifting tasks investigated. The 2-D biomechanical model provides a means to analyse symmetrical lifts in the sagittal plane as a sequence of static postures, thus not considering the dynamic components of the compressive forces on the L5-Sl segment (The University of Michigan, College of Engineering, 1990). The program provides the revised NIOSH lifting equations (Waters et al, 1993; Waters et al, 1994) for complex tasks, including repetitive and varied lifting and twisting of the trunk associated with other activities such as carrying. These equations provide the following recommended weight limits: l

l

RWL (recommended weight limit) defined as the weight that nearly all healthy workers could handle during a whole working shift (8 hr) without an increased risk of musculoskeletal injury; LZ (lifting index) defined by the ratio weight actually lifted/RWL, provides a relative estimate of the level of physical stress associated with a specific lifting task

The revised NIOSH lifting equation is a specialised risk assessment tool designed to help prevent lift-related injuries. Its multiplicative model RWL=LC.HMeVM.DM.AM.FM.CM (in which LC = weight constant, M = multiplier, horizontal distance, V = vertical distance, H= origindestination distance, A = angle of D= assymmetry, F = frequency of lifting, C = coupling) provides a weighting for six task dimensional variables, thus identifying the risk factors to modify in order to improve safety during the work cycle. The software also determines the centroids of the body markers and calculates the respective angles based on the ‘Joint Method of Measuring and Recording’ Motion: (American Academy of Orthopaedic Surgeons, 1965) and the ‘Guides to the Evaluation of Permanent Impairment, 3rd edn’ (American Medical Association, 1991). In addition, postures may be analysed during the work cycle to provide data concerning the biomechanical loading at each joint. Additional risk factors, such as external loads, can be computed (Promatek, 1991). In the second session, each of the six *subjects performed the same working cycle while VOZ and ventilation (VE) were monitored with a ‘K2’ (Cosmed, Rome) portable telemetric oxygen-uptake analyser (Dal Monte et al, 1989). The last of three consecutive work cycles was considered for data analysis. The ‘K2’ consists of a transmitting unit, battery, face mask and receiving unit. The transmitting unit, battery pack and face mask, in total weighing approximately 850 g, are attached to the individual by way of a harness. This device does not contain a CO2 electrode and assumes a respiratory exchange ratio of one in order to calculate VOz. The activity was performed for 15 min to reach steady state conditions. Direct measurements of V02 (second session) and computerised estimates of energy consumption (first session) were compared using a ttest for independent samples. A p value of less than 0.05 was considered statistically significant.

Computer based job analysis: P. Capodaglio et al

206

Results The times required to apply this field methodology were: 5-15 min for the observation of the work cycles, data collection and video recording; 30 min for the analysis with the ‘Energy’ program; 45-60 min for the biomechanical analysis with the ‘Vision 3000’ program. During all the work cycles investigated, HR values ranged from 60% to 70% of the maximal HR, as calculated by the formula: HRmax = 220 - age (y) (Figure I). All tasks, except #3, were given a mean rating of ‘moderate’ effort (CR-3) on Borg’s lo-point scale; work cycle #3 was perceived as ‘heavy’ (CR-5). The indirect energy expenditure estimates obtained with the ‘Energy’ program did not differ significantly (p > 0.05) from the direct measurements with the ‘K2’ oxygen-uptake analyser (Table 4). The r correlation coefficient between the two measurements was: r = 0.96 [MET (Energy) = 0.1 + 0.8449 MET & Table 5 summarises the results of the biomechanical analysis of each working cycle. The results did not significantly differ between the two sessions.

Discussion A combination of direct observations and video recordings allowed the determination of energy expenditure in the occupational setting without interfering with the workers’ activities. This approach is based on the recording of the dimensional parameters of the work cycles. In our study, the estimates of energy expenditure obtained with the ‘Energy’ program did not differ significantly from direct measurements obtained with a portable telemetric oxygen-uptake analyser. Crandall et al (1994) demonstrated the validity and accuracy of the portable telemetric oxygen-uptake analyser ‘K2’ when compared with a standardised metabolic system, but they concluded that its accuracy could be compromised by factors such as the duration of the test and the warm-up. Other studies have demonstrated that the ‘K2’ system underestimates oxygen consumption when

Table 4 Energy expenditure reported in MET (1 MET = 3.5 ml OJ kg.min) as calculated by the ‘Energy’ program in the first session and with a ‘K2’ apparatus in the second session. The ‘Energy’ program calculates the kcal for each task and for the entire work cycle (I kcal = 200 ml 0,). Differences were not statistically significant. Task #3 in the second session was not considered due to substantial modifications of the cycle Task

1

2

3

4

5

6

MET Energy MET K2

2.1 2.9

1.7 2

9 -

2 2.2

2.2 2.5

2.5 3

Table 5 Biomechanical analysis of the six occupational activities. *RWL and LI are reported at the origin and at the destination of the lifts. Risk factors are identified from the multiplicative model of the revised NIOSH equation (Waters et al, 1993). A postural analysis was conducted throughout each work cycle to identify the articular stress. Risk of musculoskeletal injuries has been expressed in accordance with the NIOSH limits (NIOSH, 1981) Task (n”)

Load (kg)

Amount of load handled daily (kg)

RWL* (kg)

LI*

1

1o-25

4ooC 10000

load accept

load accept

2

0.5-t

3375

load accept

3

50-80

4000-4800

4

50

5 6

Risk factors

Articular stress

Risk of musculoskeletal

None

Elbows

Nominal

load accept

None

-

Nominal

unpredict

unpredict

High load trunk flex

Shoulders L-S

Very high

3100

15-15

3.3-3.3

High load high frequency trunk torsion

L-S

High

20-26.5

15000

17-9

1.63

Low height destination

L-S

Moderate lowering

20-50

3300

17-17

3-3

Inadequate grip height of the origin

Shoulders L-S

High

of the

injury

risk during

Computer

based job analysis:

compared with a standardised metabolic analyser (Peel and Utsey, 1993; Kawakami et al, 1992). The reasons for these discrepancies are not clear but could be related to the limitations demonstrated by Crandall et al (1994) in testing the ‘K2’. One such limitation is the warm-up time required by the ‘K2’ prior to calibration to ensure that the temperature at the O2 electrode is stable. Furthermore, data should not be collected for durations greater than 15 min without recalibration (Crandall et al, 1994). The ‘K2’ equipment is supplied with a correction factor for respiratory exchange ratio (Vacumetrics, Vacumed Division, 1991), but correction would not be required to significantly improve the accuracy of the ‘K2’ to measure V02 (Crandall et al, 1994). Recently, Forkink and Frings-Dresen (1994) showed that the ‘K2’ is a useful instrument for measuring physiological work load in the field, but that in heavy to extremely heavy work situations (VO, > 1500 ml min-‘) the device underestimates the V02 values and a correction is needed. In tasks of a static nature, postural loads on muscles and joints can lead to muscular fatigue and pain. In strenuous tasks, biomechanical loading from external loads and from the muscular exertions themselves can increase the risk of injury (Andersson, 1985). Various techniques for recording posture parameters have been devised, providing the following basic parameters: location of major body joints, body joint angles, body part location relative to a standard posture (Corlett, 1976), classifications of posture types (Kahru et al, 1977; Heinsalmi, 1986; Keyserling, 1986; Kilbom et al, 1986; Armstrong et al, 1982; Genaidy et al, 1994), body outlines (photographs, video recordings, sketches). In our study, posture analysis during entire work cycles was performed by the ‘Vision 3000’ program; specifically, surface reference points approximated the location of major body joints in a 2-D reference system. This software provides a means to analyse the effects of the task demands on the different parts of the body and to identify the dimensional variables needing to be modified in order to decrease the load on each joint. The computerised approach is less time-consuming than the traditional ‘observational’ methods. One of the most attractive features of computerised postural analysis is that it allows the three main dimensions of exposure (level, repetitiveness, duration) to be considered simultaneously. A recent review of the literature on work-related shoulder-neck disorders showed that these dimensions are rarely investigated concurrently, thus leading to imprecise estimates of mechanical exposure (Winkel and Westgaard, 1992), Burdorf and Laan (1991) reported that questionnaires for assessing postural load are not viable tools with which to classify the postural load of workers. Computerised methods represent an opportunity to develop valid and practical techniques for assessing exposure to postural load. Furthermore, the ‘Vision 3000’ program allows the comparison with normative data regarding the lifting capacity of a control group of industrial workers (NIOSH, 1981). However, the ‘Vision 3000’ system represents only part of the comprehensive effort required to prevent work-related disability. Other causes established as risk factors include whole body vibration and direct trauma. For some exposure factors, like twisting of the head and trunk, and prono-

P. Capodaglio

et

al

207

supination of the forearm, pushing or pulling, there are still no suitable instruments available although methodological development has started (Kilbom, 1994). Conclusions This field methodology allows the rapid collection and analysis of a large body of data. The method entails a brief ‘on-site’ phase and a longer analysis phase in the laboratory. The total time expenditure is, however, within reasonable limits when compared with other techniques described in the literature. This combined approach with two software programs seems likely to make an important contribution to preventive medicine and ergonomics. ‘On-site’ job evaluation could include a simple and reliable estimate of the physical demands with the ‘Energy’ program and a list of eventual ergonomics interventions based on the multiplicative equation included in the ‘Vision 3000’ program. Acknowledgement The authors wish to thank editorial assistance.

MS Gillian

Jarvis for

References American Academy of Orthopaedic Surgeons 1965 Joint Motion: Method of Measuring and Recording American Academy of Orthopaedic

Surgeons,

Chicago

American Medical Association 1991 Guides to the Evaluation of Permanent Impairment. Third Edition American Medica? Association,

Chicago

Armstrong, T.J., Foulke, J.A., Joseph, B.S. and Goldstein, S.A. 1982 ‘Investigation of cumulative trauma disorders in a poultry processing plant’ Am Ind Hyg Assoc J 43(2), 103-16 distrand, P.O. and Ryhing, I. 1954 ‘A nomogram for calculation of aerobic capacity (physical fitness) from pulse maximal work’ J Appl Physiol7, 218-221

rate

during

sub-

Blench, M.A. 1993 Validation of Promatek Medical System 11~‘s Vision 3000 (c) Promatek Ltd, Ottawa Borg, G. 1982 ‘A category scale with ratio properties for intermodal and interindividual comparisons’ in Geisler, H.G. and Pet&d, P. (eds) Psychophysical judgement and the process of perception VEB Deutscher Verlag der Wissenschaft Publishers, Berlin, pp 25-34 Burdorf, A. and Laan, J. 1991 ‘Comparisons of methods for the assessment of postural load on the back’ Second J Work Envion Health 17, 425429 Crandall, C.G., Taylor, S.L. and Raven, P.B. 1994“Evaluation of the Cosmed K2 portable telemetric oxygen uptake analyzer’ Med Sci Sports Exert 26( 1) , 108-l 11 Dal Monte, A., Faina, M., Leonardi, L.M., Todaro, A., Guidi, G. and Petrelli, G. 1989 ‘Maximum oxygen consumption by telemetry’ Institute of Sport Science no. 15, Rome Forkink, A. and Frlngs-Dresen, M.H.W. 1994 ‘Measurement of oxygen consumption

with the Cosmed

K2: a comparative

study’

Applied Ergonomics 25(2), 95-100 Garg, A., ChafBn, D.B. and Herrin, G.D. 1978 ‘Prediction of metabolic rates for manual materials handling job’ AIHAJ 39(S), 661-674 Genaidy, A.M., Al-Shedi, A.A. and Karwowski, W. 1994 ‘Postural stress analysis in industry’ Applied Ergonomics 25(2), 77-87 Heinsalmi, P. 1986 ‘Method to measure working posture loads at working sites (OWAS)’ in Corlett, E.N., Wilson, J. and Manenica, I. (eds) The Ergonomics of Working Postures. Models, Methods and Cases Taylor & Francis, London, pp 100-104 Karhu, O., Kansi, P. and Kuorinka, I. 1977 ‘Correcting working postures in industry: a practical method for analysis’ Applied Ergonomics S(4), 199-201 Kawakami, Y., Nozaki, D., Matsuo, A. and Fukunaga, T. 1992

208

Computer based job analysis: P. Capodaglio et al

‘Reliability of measurement of oxygen telemetric system’ Eur J Appl Physiol65, Keyserling, W.M. 1986 ‘Postural analysis of in simulated real time’ Ergonomics 29(4),

uptake

by a portable

409-414

the trunk and shoulders 569-583 Kilbom, A., Persson, J. and Jonsson, B. 1986 ‘Risk factors for workrelated disorders of the neck and shoulder-with special emphasis on working postures and movements’ in Corlett, E.N., Wilson, J. and Maneuica, I. (eds) The Ergonomics of Working Postures. Models. Methods and Cases Tavlor & Francis, London, pp 44-53 Kilbom, h. 1994 ‘Assessment of physical exposure in relation to

work-related musculoskeletal disorders - what information can be obtained from systematic observations?’ Stand J Work Environ Health 20, 30-45 Legge, B.J. and Banister, E.W. 1986 ‘The Astrand-Rhyming nomogram revised’ J Appl Physiol61, 1203-1209 Margaria, R., Aghemo, P. and Rovelli, E. 1965 ‘Indirect determination of maximal 0, consumption in man’ J Appl Physiol20, 1070107 NIOSH 1981 Work Practices Guide for Manual Lifting. NIOSH Technical Report No. 81-122 U.S. Department of Health and

Human Services, National Institute for Occupational Safety and Health, Cincinnati Peel, C. and Utsey, C. 1993 ‘Oxygen consumption using the K2 telemetry system and a metabolic cart’ Med Sci Sports Exert 2.5(3), 396400

Pezzagno, G. and Capodaglio, E. 1991 Criteri di valutatione energetica delle attivitd fisiche La Goliardica Pavese Ed, Pavia Roozbar, A., Bosker, G.W. and Richerson, M.E. 1979 ‘A theoretical

model to estimate some ergonomic parameters from age, height and weight’ Ergonomics 22, 43-58 The University of Michigan, College of Enghwering, Center for Ergonomics 1990 User’s Manual for the Static Strength Prediction Program The University of Michigan, Ann Arbor Vacumetrics Inc, Vacumed Division 1991 Operator’s Manual for the K2 System Vacumetrics Inc, Vacumed Division, Ventura Waters, T.R., Putz-Anderson, V., Garg, A. and Fine, L.J. 1993

‘Revised NIOSH equation for the design and evaluation of manual lifting tasks’ Ergonomics 36(7), 749-776 Waters, T.R., Putz-Anderson, V. and Garg, A. 1994 Applications Manual for the Revised NIOSH Lifting Equation DHHS (NIOSH) Publication no. 94-I 10, Cincinnati Wiktorin, C., Karlqvist, L., Winkel, J. and Stockholm MUSIC I Study Group 1993 ‘Validity of self-reported exposures to work postures and manual material handling’ Stand .t Work Environ Health 19, 208-214 Winkel, J. and Westgaard, R. 1992 ‘Occupational and individual risk factors for shoulder-neck complaints: part II - The scientific basis (literature review) for the guide’ lnt .t Ind Ergon 10, 85-104 Winter, D.A. 1990 Biomechanics and Motor Control of Human Movement John Wiley, New York