Manual materials handling in mining

ARTICLE IN PRESS

Applied Ergonomics 37 (2006) 709–718 www.elsevier.com/locate/apergo

Manual materials handling in mining: The effect of rod heights and foot positions when lifting ‘‘in-the-hole’’ drill rods Andre´ Plamondona,, Alain Delislea, Karin Trimbleb, Pierre Desjardinsa,c, Trevor Rickwoodd a

Institut de recherche Robert Sauve´ en sante´ et en sećurite´ du travail (IRSST), 505 Boul. De Maisonneuve Ouest, Montreál, Que´., Canada, H3A 3C2 b School of Human Kinetics, Laurentian University, Sudbury, Ont., Canada c ´ Ecole de reádaptation, Faculte´ de me´decine, Universite´ de Montreál, Montreál, Que´., Canada d Inco Limited Ontario Operations, Copper Cliff, Ont., Canada Received 19 July 2005; accepted 20 December 2005

Abstract There is a paucity of studies focusing on the lifting of rods or long awkward heavy objects. In-the-hole (ITH) drilling is a heavy repetitive mining task, which has been identified as having a relatively high incidence and severity rate of musculoskeletal injuries. The purpose of this study was to examine how the load experienced by ITH drill operators changed when lifting a vertical drilling rod (1.61 m, 35 kg) using two rod heights and four different foot positions. In addition, a symmetrical lift with a lifting index (LI) of 1.4 also served as a comparison to determine possible risk of low back injury. Eleven experienced ITH drill operators participated in the study. Each subject was required to lift a vertical drilling rod until the upper body was in an erect posture using four different foot positions (01 ¼ subject facing the rod, 451 ¼ subject oblique to the rod, 901 ¼ subject right side to the rod and freestyle). In addition, two rod height conditions were studied where the base of the vertical rod was supported either (1) at ground level (height of rod CG ¼ 0.83 m) or (2) on a 20 cm rack (height of rod CG ¼ 1.03 m). Finally, each subject lifted a 21.5 kg box in the sagittal plane, which corresponded to an LI of 1.4 in the NIOSH lifting equation. Reflective markers were placed on the subjects, and three video cameras and one force plate were used to record the forces and the motion of the subjects’ segments. Two surface electrodes were applied on the right and the left erector spinae (ES) at the level of L3. Back loading was defined by the level of the peak moments, the mechanical work and erector spinae muscle activity (EMG). It was found that the vertical height of the rod had the most significant impact on back loading, while the effect of the initial foot positioning relative to the rod was limited by the technique adopted by the drillers. Moreover, it was found that some of the subjects used techniques less strenuous for the back than others. Finally, the asymmetrical lifting component was found to be the most negative aspect of lifting an ITH drill rod compared to a standard symmetrical lift (NIOSH). r 2006 Elsevier Ltd. All rights reserved. Keywords: Lifting; Manual materials handling; Biomechanics

1. Introduction Manual handling of objects in an industrial setting has been a significant concern to occupational health professionals who attempt to prevent injury. Tasks that demand frequent and heavy lifting are associated with an increased risk of low back pain. To date, a majority of studies have focused on the lifting of rectangular shaped objects and to a limited extent on irregular shaped objects such as Corresponding author. Tel.: +1 514 288 1551; fax: +1 514 288 6097.

E-mail address: [email protected] (A. Plamondon). URL: http://www.irsst.qc.ca. 0003-6870/$ - see front matter r 2006 Elsevier Ltd. All rights reserved. doi:10.1016/j.apergo.2005.12.003

shopping bags or sacks. Although irregular shaped objects are often lifted, they are rarely studied. Lifting studies with cylindrical shaped objects are almost completely nonexistent to the authors’ knowledge, except for a couple of research studies (Gagnon et al., 2002; Jorgensen et al., 1985). Furthermore, research in underground mining has focused on different kinds of problems. For instance, Marras and Lavender (1991) investigated the effects of different models of scaling bars (a hand tool used in underground mining), mine roof height and the method of tool use on the activity of trunk muscles. Gallagher et al. (2002), Gallagher and Hamrick (1994), and Gallagher (2005), addressed work in unusual and restricted postures,

ARTICLE IN PRESS 710

A. Plamondon et al. / Applied Ergonomics 37 (2006) 709–718

which are associated with higher rates of musculoskeletal problems. There is, however, a paucity of studies focusing on the lifting of rods or long awkward heavy objects. In-the-hole (ITH) drilling is a heavy repetitive mining task, which has been identified as having a relatively high incidence and severity rate of musculoskeletal injuries. An ITH drill consists of a pneumatic hammer and bit that are connected to a rotating drill head by rods. The drill rods of interest in this study are 1.61 m long and 13 cm in diameter. They start at an approximate weight of 41 kg and end at 32 kg when decommissioned because of wear. The bit, hammer and rods are collectively called the drill string, which can include up to 40 rods. Rod handling tasks include drilling (lifting the rod from a storage point, carrying the rod to the drill and placing the rod on the drill and securing it) and pulling (removing the rod from the drill and carrying the rod to rod storage). Rods are added to the drill string as the hammer and bit advance down the hole one rod length at a time. Ground conditions will determine the drilling time per rod; however, it is typically between 3.5 to 5 min per rod. Pre-drilled holes sometimes require cleaning and this requires less time than regular drilling. Rod removal requires less than one minute per rod, as the drill string is ‘‘pulled’’ from the completed hole. Rods are usually stored upright either on the floor or on aluminum racks. Many aluminum racks have forkways in their bases to allow handling with a forklift. Racks with forkways place the rod about 20–25 cm above the ground. Rack orientation relative to the drill will determine the degree the worker has to turn with the load. Preliminary worksite observation showed that rod-lifting requires asymmetrical lifting with the center of mass supported almost entirely by one arm, coming into contact with the upper arm, forearm and hand. The opposite hand may come in contact with the rod below the center of mass providing additional lifting force. Once the rod is lifted, it is usually carried at the operator’s side or across the waist. The ground conditions vary from ideal, level firm and dry ground, to slippery, muddy and uneven conditions. The hazards identified were rod weight, rod condition (wet, dirty), occasional uneven, slippery and/or muddy footing, and occasional increased handling rates. The Mines and Aggregates Safety and Health Association (MASHA) in Ontario (Canada) reported 124 disabling injuries from 1991–2000 that were associated with ITH drilling; 22% were back injuries and 30% were due to overexertion while lifting or lowering. The immediate cause of the injury was ‘taking an improper position’ for the task (25%), ‘defective/hazardous tools, equipment or material’ (14%) and ‘improper lifting’ (13%). The risk of injury due to manual handling in ITH drilling is apparently high with the back being the most commonly injured body part. Because low back pain can be disabling, expensive and permanent, it is important to find ways to decrease the causes of injury. Examining how the load on the back (EMG and biomechanical modeling) changes with different lifting heights and foot

positions would help to identify particular risks to the lower back. Comparing the recommended NIOSH lift to the lifting of ITH drill rods (35 kg) may help to characterize the risk of lifting heavy awkward objects. The purpose of the study was to examine how the load experienced by ITH drill operators changed when lifting a vertical drilling rod (1.61 m, 35 kg) using two rod heights and four different foot positions. In addition, a symmetrical lift with a lifting index (LI) of 1.4 also served as a comparison to determine possible risk of low back injury. In the present study, low back load was defined by the level of the peak moments, the mechanical work and the erector spinae muscle activity (EMG). 2. Methods 2.1. Subjects Eleven experienced ITH drill operators participated in the study. Their mean age was 37.5 yr (range ¼ 26–52 yr), mean weight was 90.1 kg (range ¼ 84.3–94.7 kg), mean height was 1.77 m (range ¼ 1.68–1.82 m), and mean manual material handling (MMH) experience was 12.5 yr (range ¼ 2–35 yr). All subjects were right-handed and had no history of serious back injury or any recent discomfort. They signed a written consent form that was approved by the ethics committee after they had been informed about the experimental protocol. All testing was conducted in the Biomechanics Laboratory at Laurentian University. 2.2. Experimental task ITH drilling is a heavy manual handling industrial mining task. Observation from video analysis revealed that the operators used drastically different foot positions ranging from 0o to 90o when lifting. A rod storage rack was sometimes used, but not often. To test these different conditions, the task of lifting an ITH drill rod from its storage point was simulated in the laboratory. Each subject was required to lift a vertical drilling rod (35 kg, 1.61 m in length, 13 cm diameter) until the upper body was in an erect posture and to take a step off the force plate. The independent variables consisted of using four different foot placements and two rod heights. The four different foot placements (Fig. 1) were: (1) 01 ¼ subject facing the rod (symmetrical), (2) 451 ¼ subject oblique to the rod (asymmetrical), Symmetrical

Asymmetrical

Asymmetrical

Rod

Rod

Rod

0o

45o

Fig. 1. Initial foot positions relative to the rod.

90o

ARTICLE IN PRESS A. Plamondon et al. / Applied Ergonomics 37 (2006) 709–718

711

and two behind the subject. The system collected the bidimensional position of 44 reflective markers on the subject at 60 Hz. Prior to data acquisition, 40 control points established a reference frame of spatial coordinates in a volume of 2.6 m3 (1.8 m 1.2 m 1.2 m). The 3D loci of the anatomical markers were obtained with a DLT algorithm and smoothed with quintic splines. EMG activity of the back muscles was measured with two active bipolar silver/silver chloride surface electrodes (electrode diameter: 0.84 cm; interelectrode distance: 2.12 cm; Therapeutic Unlimited, Iowa City). Raw EMG signals were preamplified at the electrode site, then amplified with a differential amplifier (Gain range: 1000 to 100,000; CMRR: 87 db at 60 Hz; input impedance:425 MO at dc; noise:o2.0 mV RMS) and stored on a hard disk with a sampling frequency of 1500 Hz. Video, force plate data, and EMG data were synchronized with the Peak Motus measurement system. Fig. 2. Two rod height conditions: (A) ground level (left); and (B) Rack level (right).

(3) 901 ¼ subject right side to the rod (asymmetrical), (4) Freestyle ¼ subject free to choose foot placements. Two rod height conditions (Fig. 2) were studied in which the base of the vertical rod was supported either (1) at the ground level (height of rod CG ¼ 0.83 m) or (2) on a 20 cm rack (height of rod CG ¼ 1.03 m). Two trials were performed with the second trial used for analysis unless a data collection error occurred; otherwise, the first trial would be analyzed. All lifts were performed toward the right side of the subject and in a randomized order. The subjects were told to lift at a rate that was comfortable and as close to the speed that they use when performing the task at work. A symmetrical lifting task, the NIOSH lift, was added for comparison with the ITH rod-handling task. In this task, the subject lifted a box (34 cm 34 cm 18 cm) weighting 21.5 kg with good coupling (handles), vertical location of 58 cm, horizontal location of 25 cm and travel distance less than 25 cm. The NIOSH lift was supposed to result in a lifting index (LI) of 1.0, but the real index calculated reached 1.4 because of a larger horizontal location (36 cm instead of 25) and a shorter vertical distance from the ground (58 cm instead of 75 cm). Two trials of this NIOSH defined task were performed with the second trial used for further analysis. 2.3. Materials and apparatus The lifting task was performed on a force plate (Kistler: 40 cm 60 cm; model 9865B) and within the dimensions of a three-dimensional calibration object (1.8 m 1.2 m 1.2 m). A Peak Motus System (Peak Performance Technologies Inc., Englewood, Colorado) with three video cameras (Panasonic WV CL350) were used to record the subjects’ anatomical markers during the task, one facing the subject

2.4. Procedure Upon arrival at the Biomechanics Laboratory, the subjects signed a written consent and changed from their clothes into black spandex shorts. Anthropometric measurements of each subject were recorded, including their height and weight. Reflective markers were then placed on 44 locations distributed over 15 segments to be used to estimate the joint center position of the segments. The locations of these skin markers were based on the anthropometric data provided by Webb and Associates (1978) and reported by Chaffin et al. (1999). EMG electrodes were placed over the right and left sides of the lumbar erector spinae (ES) approximately 3 cm lateral to the L3 spinous process (McGill, 1991). Prior to electrode placement, the skin was shaved, abraded and cleaned with alcohol. An electrolytic paste was used on the surface electrodes. The ground electrode was applied to the anterior bony part of the tibia. Each subject was then placed in a structure that allowed maximal isometric exertions to be performed in extension. The subject stood facing a wall and exerted effort against a load cell (Intertechnology, Don Mills, Ontario) in a cable that connected the harness around the subject’s upper trunk. Each subject performed two maximal voluntary isometric contractions (MVIC) in extension measured with a tensor dynamometer and the maximum of these two MVICs was kept for analysis. A picture was taken with a digital camera to determine the net torque exerted during the MVIC. Before the lifting tasks were performed, the subjects were positioned in different calibration postures to help recover missing markers during the 3D reconstruction. Lifting tasks were then performed by the subjects. 2.5. Data processing Two dynamic 3D multi-segment models able to estimate the net reaction moment at L5/S1 were defined: (1) a lower


712

body model which included seven segments, the feet, shanks, thighs and pelvis; (2) an upper body model which included nine segments, the hands, upper arms, arms, head, trunk and the rod. Details of the models are described in Desjardins et al. (1998) and Plamondon et al. (1996). The L5/S1 moments were computed in majority from the lower body model, except for two subjects for which technical problems with one of the channels of the force platform imposed the use of the data from the upper body model. The L5/S1 moments were expressed relative to a 3D anatomical coordinate system of the pelvis. The flexion/ extension moment (symmetrical moment) was about the transverse axis of the pelvis. The longitudinal and sagittal moments were reduced to one component, an asymmetrical moment, defined as the square root of the sum of the squares about the longitudinal and sagittal moments (Gagnon, 2003). The symmetrical and asymmetrical moments were also combined to give a net resultant moment. The mechanical work done on the rod was estimated during the lifting phase by integrating power as a function of time, with power being the dot product of the force exerted on the rod and its center-of-gravity velocity (Gagnon, 2003). Because there was no deposit phase, only the vertical work to lift the rod was considered. Three kinetic variables were selected to compare the different experimental conditions: the peak resultant moment, the peak asymmetrical moment during lifting and the vertical mechanical work done on the load. Kinematic variables included peak trunk angle from the vertical and peak rod velocity during lifting (positive and negative). EMG signals were band-pass filtered at 20–500 Hz; (Datapac 2000 software, Run Technologies, Laguna Hills, CA) and transformed using a moving root mean squares (RMS) processing method. A time window of 200 ms was used for the MVIC, and 100 ms for the lifting task. The peak RMS values (left and right) computed during the MVIC represented the maximum voluntary EMG (MVE),

and the peak RMS values found during the lifting task (EMGPEAK) were used to reflect peak muscle activation for the entire task. To express the level of muscle activation (% MVE), the RMS EMGPEAK values during the lifting tasks were normalized relative to the maximum voluntary EMG (MVE) observed. The differences between the NIOSH and ITH various rod-handling tasks were compared since the NIOSH task (1991 NIOSH lifting equation) is a recommended standard for preventing injury. The ITH rod-handling conditions that were highest and lowest in the peak resultant moment were compared with the NIOSH task. 2.6. Statistical analysis The calculated dependent variables were—peak resultant moment, peak asymmetrical moment, vertical mechanical work (positive and negative), peak rod velocity (positive and negative), maximal trunk angle and the normalized erector spinae EMG signal (EMGPEAK) for both the left (ESL) and right side (ESR). The independent variables were the vertical height of the rod (ground and rack) and foot position (01, 451, 901 and freestyle). To examine the influence of the independent variables, repeated measures analysis of variance (ANOVA) were conducted. To avoid compounding alpha error by conducting t-test with all ITH rod-lifting conditions, only the best and the worst ITH rodlifting conditions were compared with the NIOSH lift. An alpha level of .05 was used for all statistical tests, and Scheffe post-hoc analyses were subsequently used when needed. 3. Results The results will be presented in four parts that consider the effect of the vertical height of the rod, the effect of the different foot positions, the comparison with the NIOSH

Table 1 Summary table of the results Variable

Rack

a

Peak resultant moment (Nm) Symmetrical moment (Nm)a

Asymmetrical moment (Nm)a Peak asymmetrical moment (Nm)b Resultant moment (Nm)b Symmetrical moment (Nm)b a

Ground

01

451

901

Free

01

451

901

Free

01

M SD M SD M SD

190 52 185 53 39 17

212 53 202 59 59 20

217 54 206 57 61 29

195 57 186 64 43 24

251 63 246 65 44 15

243 55 237 56 48 29

273 54 261 58 61 46

259 85 256 85 35 21

226 36 225 36 18 7

M SD M SD M SD

77 24 139 47 112 50

74 21 177 50 159 54

81 26 193 52 171 58

74 23 150 56 127 60

69 13 170 77 152 83

72 19 179 78 157 89

86 34 226 71 203 83

70 16 160 76 138 85

22 7 198 46 197 45

Symmetrical or asymmetrical moments during the peak resultant moment. Resultant moment or symmetrical moment during the peak asymmetrical moment.

b

NIOSH


lift and other effects. Tables 1 and 2 present a summary of all the results, and Table 3 the main statistical results.

713

moment (84 Nm), and the others having almost equal moments (73 Nm). Furthermore, the asymmetrical component of the peak resultant moment (Table 1) was not

3.1. Vertical height effect of the rod Lifting from a rack significantly reduced (p o.01) the peak resultant moment from 256 Nm (SD ¼ 64 Nm) at ground level to 204 Nm (SD ¼ 53 Nm) as illustrated in Fig. 3. However, the effect of the rack was not significant in the peak asymmetrical moment (p ¼ .50), from 77 Nm (SD ¼ 23 Nm) with a rack to 74 Nm (SD ¼ 23) at ground level (Fig. 3). The result with the peak resultant moment is consistent with the EMG results for the left (ESL) and right (ESR) erector spinae where the EMGPEAK was always smaller (po.01) with the use of a rack (61% MVE) compared to ground level (72% MVE). Moreover, maximal trunk forward bending with the vertical was significantly lower with the rack (p o.01). As expected, the positive vertical work was significantly less (po.01) when lifting with the rack, 81 J (SD ¼ 34 J) compared to the 136 J (SD ¼ 35 J) without it. Interestingly, there was no significant effect (p ¼ 0.47) in the downward vertical work in the order of 23 J, meaning that the downward motion of the rod was not affected by the initial height of the rod. Overall, when the rod was lifted from the rack, back loading was significantly less than for lifting from the ground. 3.2. Foot position effect There was a significant difference (po.05) in the peak resultant moment at L5/S1 between the different foot positions (Fig. 4). The magnitude ranged between 220 Nm (01 face) and 245 Nm (901 side). The difference in the peak asymmetrical moment was not significant (p ¼ .19) between the conditions, with the 901 having the highest

Table 3 Repeated analysis of variance (ANOVA) for the most important variables as a function of the rack height and foot positions Dependant variables

Rack (1)

Foot (2)

Interaction 12

Peak resultant moment

F df p

156.30 1,10 o0.010

3.53 3,30 0.027

3.05 3,30 0.043

Peak asymmetrical moment

F df p

0.48 1,10 0.503

1.68 3,30 0.193

0.73 3,30 0.543

Positive mechanical work

F df p

532,01 1,10 o0.010

1.15 3,30 0.345

0.58 3,30 0.630

Negative mechanical work

F df p

0,47 1,10 0,508

4.13 3,30 0.015

0.332 3,30 0.802

EMGa

F df p

10.71 1,10 o0.010

2,40 3,30 0.088

0.16 3,30 0.921

Trunk angle

F df p

35.52 1,10 o0.010

0.37 3,30 0.770

2.36 3,30 0.092

Rod peak positive F vertical velocity df (m/s) p

79.73 1,10 o0.010

2.06 3,30 0.130

1,71 3,30 0.180

3.15 1,10 0.110

3.46 3,30 0.028

0.55 3,30 0.650

Rod peak negative vertical velocity (m/s)

F df p

a

There was no significant difference between right and left side muscles (F1,10 ¼ 1.42, p ¼ 0.261) and no significant interaction.

Table 2 Summary of the other results Variable

Rod positive vertical work (J) Rod negative vertical work (J) EMGPEAK (%) left side EMGPEAK (%) right side Max flexion trunk angle(1)b Rod peak positive vertical velocity (m/s) Rod peak negative vertical velocity (m/s) a

NA—non applicable. Flexion angle from the vertical.

b

Rack

M SD M SD M SD M SD M SD M SD M SD

Ground

NIOSH

01

451

901

Free

01

451

901

Free

01

79 33 18 11 64 28 48 17 43 6 0.45 0.10 0.10 0.05

80 36 23 17 69 37 62 31 45 9 0.45 0.10 0.15 0.10

84 37 29 14 73 44 55 18 42 9 0.55 0.15 0.20 0.10

83 36 22 17 62 29 53 18 44 6 0.50 0.15 0.15 0.10

138 35 18 9 69 29 66 23 49 9 0.70 0.15 0.05 0.30

130 35 20 10 76 46 72 28 51 10 0.70 0.20 0.10 0.05

136 42 28 15 83 43 68 28 51 11 0.75 0.25 0.15 0.05

139 34 22 15 69 33 71 34 49 8 0.70 0.20 0.15 0.10

NAa NA 64 34 63 28 NA NA NA


714

350

Moment (Nm)

300 **

250

**

Peak Resultant Peak Asymmetrical

200 150 100 50 0 Rack

Ground

Fig. 3. Effect of rod height on the peak resultant moment and in the peak asymmetrical moment (n ¼ 11). **Significantly different at po.01.

350

Moment (Nm)

300

Peak Resultant Peak Asymmetrical

*

250

*

200 150 100 50 0 0°

45° 90° Feet position (°)

Free

Fig. 4. Effect of the four initial foot positions in the peak resultant moment and the peak asymmetrical moment (n ¼ 11). * Scheffe test— significantly different at po.05.

significantly affected by foot positions (F3,30 ¼ 0.10; p ¼ 0.10) even if the 901 condition had a larger value (about 61 Nm) compared to the 01 condition (rack ¼ 39 Nm). In the freestyle condition, most of the subjects adopted either the 01 foot position or a position where the right foot was at 01 and the left foot at 451; therefore the difference between the 01, the 451 and the freestyle conditions was small and not significant. There was no significant difference in the positive vertical work (range from 105 to 111 J) and in the EMGPEAK level between the different foot positions (range from 62% to 70% MVE). However, the downward vertical work was affected significantly by the foot position from –18 J for 01 to 29 J for 901. One explanation is that the subjects changed their technique to lift the rod: at 01, most of the handlers supported the upper part of the rod with their right arms; this changed to the left arm at 901 for seven of the subjects (Fig. 5A). Instead of turning the back in torsion to face the rod, they simply let the rod fall in front of them and stopped the downward motion with their left arm. This is probably why the 901 compared to the 01 has a more pronounced downward motion and a greater peak downward velocity of the rod (901 ¼ 0.20 m/s; 01 ¼ 0.10 m/s) and a greater peak resultant moment. The technical modification was the most noticeable change between the 01 and the 901 foot positions. Overall, the foot positions had some significant effect on back loading but not as large as expected. The results also showed a significant interaction (po.05) between the height conditions (rack vs ground) and the foot positions. It was observed that all foot conditions produced a significant increase in the resulting moment during the ground condition as compared to the rack. However, the increase was smaller in the 451 foot position, as seen in Fig. 6. Video observations and data analysis did not help to find an explanation for this result.

Fig. 5. Three different rod handling techniques. (A) 901 asymmetrical lift; (B) 01 symmetrical with lower hand in supination; and (C) 01 symmetrical with lower hand in pronation.


200

380 Rack

*

Ground

330

160 280

* Work (J)

Peak resultant moment (Nm)

715

230 180 130 0°

45° 90° Foot positions (degrees)

Free

Fig. 6. Significant interaction (F3,30 ¼ 3.02; po.045) between the height conditions (rack vs. ground) and the foot positions.

120

80

40

0 > 200 Nm

< 200 Nm Resultant moment

3.3. Rod-lifting vs NIOSH lift The best and the worst case scenarios selected to be compared with the NIOSH task were respectively the rack at 01 with the lowest resultant moment, and the ground at 901 with the highest (Table 1). The difference between NIOSH and ITH rod-lifting was significant for both the peak resultant moment (F2,20 ¼ 18.65; po.01 ) and the peak asymmetrical moment (F2,20 ¼ 6.20; po.01). Surprisingly, the resultant peak moment for the NIOSH lift (lifting index of 1.4) was higher on average from the one obtained during rod-lifting from the rack at 01 (NIOSH ¼ 226 Nm; 01 ¼ 191 Nm) but was significantly lower (Scheffe test po.01) than the lifting from the ground at 901 (273 Nm). Comparison of the EMGPEAK level agreed with these results where it reached on average 56%, 76% and 63% MVE respectively for 01 with rack, 901 in the ground condition and the NIOSH lifting. The most negative aspect of the ITH rod-lifting was found in the peak asymmetrical moment—the NIOSH lift reached approximately 22 Nm compared to the 77 Nm and 86 Nm of the ITH rod-lifting for respectively the 01 lift with rack and the 901 lift from the ground (Table 1). 3.4. Variability in lifting strategies Fig. 7 shows that the subjects having a peak resultant moment less than 200 Nm in all the conditions (group of four subjects) were those with the lowest vertical positive work. In other words, four subjects used a lifting strategy that appeared to have decreased back loading by minimizing the vertical displacement of the rod as the rack did. Figs. 5(B) and (C) illustrates two techniques generally used by our subjects. The difference is mostly in the way the lower hand grasps the rod—eight of our subjects supported the rod in supination and the three others in pronation. This difference appeared to be important in the resultant moment and also in the asymmetrical moment. By switching the lower hand position from supination to pronation, three of our subjects significantly reduced the vertical work

Fig. 7. Mechanical work vs. peak resultant moment. Subjects were separated in two groups: (1) subject with the average of the eight trials (2 heights 4 feet) in the peak resultant moment higher than 200 Nm (n ¼ 7); (2) subject with the average of the eight trials in the peak resultant moment lower than 200 Nm (n ¼ 4). *Significantly different: F1,9 ¼ 5.76; p ¼ .04.

on the rod by grabbing it at a higher level and consequently decreasing the back loading. Interestingly, the partial correlation (removing the effect of weight and height) between the positive vertical work and the MVIC was generally greater than 0.70 (po.05), reaching even 0.89 (901 lift with rack). Then subjects with less back strength (or likely not to exert MVIC) appeared to have used a technique less strenuous for the back and more adapted to their physical capacity. 4. Discussion It is known from the literature that the following aspects of lifting are potentially hazardous—weight of load, horizontal and vertical location of the load, shape and size of load, frequency of lifting, stability of load, coupling, duration of lifting, workplace geometry, asymmetric lifting and environment (Kroemer and Grandjean, 1997). Several of these aspects were covered in this study where three main findings were found: the first one confirmed the use of a rack to reduce back loading; the second showed that inappropriate foot positions can increase back loading but less than expected; the third indicated that some workers were able to reduce the back load with the use of an efficient technique. In addition, the asymmetrical lifting component was found to be the most negative aspect of ITH rod-lifting compared to a standard symmetrical lift (NIOSH). 4.1. Vertical height effect of the rod Vertical height had a significant effect on the resultant moment, mechanical work and EMGPEAK. The higher the vertical position of the rod at the start of the lifting, the lower was the load on the back. It has been reported that



lower lifting heights increase the reaction moments and, therefore, the compressive and shear forces acting on the disc (Buseck et al., 1988; de Looze et al., 1993; Dolan et al., 1994; Freivalds et al., 1984; Kingma et al., 2004; Lavender et al., 2003; Leskinen et al., 1983; Schipplein et al., 1995; Tsuang et al., 1992). The importance of the initial vertical height in lifting was recognized and incorporated as a component in the NIOSH equation for the recommended limit (Waters et al., 1993). Low heights and greater horizontal distances increase the load on the spine for two principal reasons—first, the load moment increases, requiring greater internal forces and thus resulting in greater compression and shear spine loading. Second, a greater proportion of the weight of the body is at a greater distance from the spine, which increased the load moment (National Research Council, 2001; Waters et al., 1993). These results combined with those found in this study suggest that the use of a rack is an important element to reduce the load of the back during ITH drilling operations. 4.2. Foot position effect The 901 foot position compared to the three other conditions resulted in a significantly higher resultant peak moment but not in the peak asymmetrical moment and EMGPEAK. Therefore, foot position had some significant effect on back loading but not as great as expected. Past studies, similar to the 901 foot condition with the initial load to the side of the subjects, found a large asymmetrical component of the moments during lifting compared to a symmetric lifting (Gagnon et al., 1993; Kingma et al., 1998; Lavender et al., 1999; Plamondon et al., 1995). The subjects in these studies were forced to rotate their trunk to face the load in order to lift it because of the fixed position of their feet. In this study, instead of turning their trunks in torsion to face the rod, the subjects let it fall in front of them (Fig. 5A), thus avoiding large asymmetrical postures and, consequently, limiting the magnitude of the asymmetrical moment. This technical change was not planned, but was probably the consequence of their working experience, which gave them the ability to adapt to different constraint conditions. This ability is generally found in expert handlers, which may allow them to achieve objectives such as reducing stress to the joints, maintaining balance, controlling the load or reducing fatigue (Authier et al., 1995). The magnitude of the asymmetrical moment was elevated, around 68–86 Nm, but could be explained by the weight of the rod which imposes large asymmetrical loads on the spine when the lifting is predominantly over one side of the body as was the case in this study. In the freestyle condition, most of the subjects adopted either the 01 foot position or a position where the right foot was at 01 and the left foot at 451. These conditions did not appear to affect the magnitude of spine loading, but according to Gagnon (2003, 2005), foot mobility plays a role in asymmetry and expert workers seem to anticipate their displacement by orienting their feet towards deposit, always facing the load,

which has the potential of improving handling maneuvers and reducing back efforts. 4.3. Rod-lifting vs. NIOSH lift The 1991 NIOSH lifting equation gives the ‘recommended weight limit’ (RWL) for the given task. The maximum RWL was established at 23 kg (225 N) under ideal conditions. The lifting index is the ratio of the load lifted to the RWL and it may be used to estimate the percentage of the workforce that is at risk of developing lifting-related low back pain. It has been suggested that most of the working population should be able to perform jobs with LIs less than 1.0 without significant risk of LBP. A lifting index greater than ‘one’ poses an increased risk of lifting-related back pain for some of the workforce, and an LI greater than 2 indicates that a worker is at significantly greater risk of having LBP (Waters et al., 1993; Waters et al., 1998). The mean value of the peak resultant moment in the NIOSH lift reached 226 Nm, which is equivalent to the value found under similar conditions (Davis and Marras, 2000) or in simulated lifts (Leskinen and Haijanen, 1996; Potvin, 1997). In the current study, the NIOSH lift resulted in an LI of 1.4. ITH rod-lifting under ideal conditions (01 lifting with rack) and the NIOSH lifting reached a similar level of spine loading in the resultant moment. This was not expected because rod handling tasks have been identified as having a relatively high incidence and severity rate of musculoskeletal injuries. A possible explanation for this is that although the weight of the rod was greater (rod ¼ 35 kg vs. NIOSH ¼ 21.5 kg), the vertical location of the rod was higher than the NIOSH lift (CG rod ¼ 0.83 m, with rack 1.03 m vs. NIOSH ¼ 0.58 m) reducing the need for the trunk to bend forward. In addition, the rod was held very close to the body, limiting the horizontal distance and, consequently, the resultant moment. On the other hand, the level of asymmetrical loading in ITH rod-lifting was far more important than in the NIOSH lifting, increasing considerably the physical demand for this task. Asymmetrical loading is different from asymmetrical posture because you can be in a symmetrical posture but have high asymmetrical loading, e.g. when carrying a heavy suitcase at your side. This large asymmetrical component puts the ITH driller at greater risk of back injury because of the reduced back strength to lift the load (Gravel et al., 1997; Vink et al., 1992). During asymmetric exertions, dynamic trunk motion is under the control of muscles that are smaller in cross-sectional area and are consequently more likely to suffer injury to perform the task (Marras and Mirka, 1992). Another drawback of asymmetrical loading concerns the tissue stress distribution on the disc. For instance, the rise in nucleus pressure generated by a lateral bending moment appears to be greater than when the same moment is applied in forward bending (Adams et al., 2002; Schultz et al., 1979). Finally, NIOSH defines heavy physical work as muscular exertion greater than 70% of maximum


voluntary contraction (Waters et al., 1993). In this study, the average EMGPEAK level varied between 48% and 83% MVE, depending on the side of exertion, foot positioning or the use of a rack. Therefore, ITH rod-lifting could be considered a job at significantly greater risk of LBP than the NIOSH lift. 4.4. Variability in lifting strategies Our results showed that some of the subjects used techniques less strenuous for the back than others. At the beginning of the lift, these subjects were able to grasp the rod at a higher level than the others, using a different hand position, minimizing the vertical distance to travel. Again, these strategies are generally found in expert handlers. It has been found that expert handlers adopted very different strategies when compared to novices (Authier et al., 1995; Authier et al., 1996). Experts should be seen here as different from experienced handlers as they are recognized for their professional skills by teammates and management (Authier et al., 1996). Some differences between novices and expert lifters include foot positioning, the role of load tilts/hand and shoulders positioning (Gagnon, 2003; 2005). For instance, experienced workers positioned their supporting foot and pelvis toward the deposit site and also rotated the box in the direction of the deposit site. This could help reduce changes in direction during the lift, reduce the asymmetry of posture, and would also allow for a continuous motion (Authier et al., 1996; Gagnon, 2003; Delisle et al., 1999; Gagnon, 2005). The selection of our subjects was not based on their skills, but some of them showed more abilities than others in handling the rod. Recent publications by Gagnon (2003; 2005) and Lortie (2002) demonstrated the importance of emphasizing expert workers’ strategies to improve the training program with the purpose of reducing back injuries. The preventive strategies that are used to reduce the load on the back associated with lifting tasks involve engineering controls (redesigning the workplace) and administrative controls, such as training and instruction in lifting posture. In the case of ITH drill operators, the best engineering approach is to completely eliminate rod handling by using a mechanical-assist devices. For instance, Lavender and Marras (1990) recommended an articulated arm connected to a mining vehicle to eliminate the need for overly stressful exertions when manipulating a jackleg drill (weight of 52.2 kg). As suggested by Gallagher et al. (1992, 2002, 2005), efforts should be made to provide workers with mechanical assistance when performing demanding tasks in mines in order to reduce the stresses and facilitate the task. When this is not possible, it was found in this study that using engineering controls such as providing the workers with rack support is an efficient way to reduce back effort. An additional way to reduce back effort is to use strategies such as changing hand position to reduce the distance of the vertical lift during rod handling. These administrative control strategies combined with engineering controls are

717

efficient ways to minimize back loading, asymmetries and energy requirements. 4.5. Limitations The task of lifting the drill rod from the storage point was the only one studied. Carrying tasks and handling tasks at the drill itself were not included. The lifting conditions reproduced in the lab were ideal. Important considerations of the ITH drill works were not included, such as uneven, slippery and/or muddy footing, rod surface contaminated with sand, water or oil or the use of gloves and other protective equipment. The foot positions and the placement of the rod were constrained. The placement of the rod was based on workers’ comments and on video observations from the work site, but the subjects were not permitted to step-off of the force platform to grab the rod. It would have been possible for the workers to get closer to the rod if they had been allowed to place their feet the way they wanted without any constraint. All these factors could have an impact on the level of spine loading during rod handling. On the other hand, these variables should not affect the main findings of this study, such as the importance of a rack and the importance of observing expert workers for improving the content of training programs. It is essential to understand that trying to find the ‘best’ technique is not realistic. A training program should, therefore, include several strategies so that the worker can choose the most appropriate one for himself and the specific context. 5. Conclusions The current study showed that the vertical height of the rod had the most significant impact on back loading while the effect of initial foot positioning relative to the rod was limited by the technique adopted by the drillers. Implementation of racks at least 20 cm in height raises the center of mass of the rod and will help to significantly reduce back loading when lifting ITH rods. On the other hand, asymmetrical loading during rod handling was important and constitutes an increased risk of back injuries. It was also found that some of the subjects used handling strategies less strenuous for the back than others. These strategies, such as changing hand position to reduce the distance of the vertical lift during rod handling, could help to decrease back loading. Training sessions based on the observation of experienced workers could be helpful to find ways to decrease back injuries. Acknowledgements This study was funded by the Institut de Recherche Robert Sauve´ en Sante´ et en Sećurite´ du travail du Que´bec (IRSST) and the Natural Sciences and Engineering Research Council of Canada (NSERC). The participation of ITH drill operators from Inco Limited Ontario Operations is gratefully acknowledged.



References Adams, M.A., Bogduk, N., Burton, A.K., Dolan, P., 2002. The Biomechanics of Back Pain. Churchill Livingstone, Toronto. Authier, M., Gagnon, M., Lortie, M., 1995. Handling techniques: The influence of weight and height for experts and novices. Int. J. Occup. Safety Ergonom. 1, 262–275. Authier, M., Lortie, M., Gagnon, M., 1996. Manual handling techniques: comparing novices and experts. Int. J. Ind. Ergonom. 17, 419–429. Buseck, M., Schipplein, O.D., Andersson, G.B.J., Andriacchi, T.P., 1988. Influence of dynamic factors and external loads on the moment at the lumbar spine in lifting. Spine 13, 918–921. Chaffin, D.B., Andersson, G.B., Martin, B.J., 1999. Occupational Biomechanics, third ed. Wiley, Toronto. Davis, K.G., Marras, W.S., 2000. Assessment of the relationship between box weight and trunk kinematics: does a reduction in box weight necessarily correspond to a decrease in spinal loading? Hum. Factor. 42, 195–208. de Looze, M.P., Toussaint, .M., van Dieen, J.H., Kemper, H.C.G., 1993. Joint moments and muscle activity in the lower extremities and lower back in lifting and lowering tasks. J. Biomech. 26, 1067–1076. Delisle, A., Gagnon, M., Desjardins, P., 1999. Kinematic analysis of footstep strategies in asymmetrical lifting and lowering tasks. Int. J. Ind. Ergonom. 23, 451–460. Desjardins, P., Plamondon, A., Gagnon, M., 1998. Sensitivity analysis of segment models to estimate the net reaction moments at the L5/S1 joint in lifting. Med. Eng. Phys. 20, 153–158. Dolan, P., Earley, M., Adams, M.A., 1994. Bending and compressive stresses acting on the lumbar spine during lifting activities. J. Biomech. 27, 1237–1248. Freivalds, A., Chaffin, D.B., Garg, A., Lee, K.S., 1984. A dynamic biomechanical evaluation of lifting maximum acceptable loads. J. Biomech. 17, 251–262. Gagnon, M., 2003. The efficacy of training for three manual handling strategies based on the observation of expert and novice workers. Clin. Biomech. 18, 601–611 (Bristol, Avon). Gagnon, M., 2005. Ergonomic identification and biomechanical evaluation of workers’ strategies and their validation in a training situation: summary of research. Clin. Biomech. 20, 569–580 (Bristol, Avon). Gagnon, M., Delisle, A., Desjardins, P., 2002. Biomechanical differences between best and worst performances in repeated free asymmetrical lifts. Int. J. Ind. Ergonom. 29, 73–83. Gagnon, M., Plamondon, A., Gravel, D., 1993. Pivoting with the load. An alternative for protecting the back in asymmetrical lifting. Spine 18, 1515–1524. Gallagher, S., 2005. Physical limitations and musculoskeletal complaints associated with work in unusual or restricted postures: a literature review. J Saf. Res. 36, 51–61. Gallagher, S., Hamrick, C.A., 1992. Acceptable workloads for three common mining materials. Ergonomics 35, 1013–1031. Gallagher, S., Hamrick, C.A., 1994. Dynamic biomechanical modelling of symmetric and asymmetric lifting tasks in restricted postures. Ergonomics 37, 1289–1310. Gallagher, S., Marras, W.S., Davis, K.G., Kovacs, K., 2002. Effects of posture on dynamic back loading during a cable lifting task. Ergonomics 45, 380–398. Gravel, D., Gagnon, M., Plamondon, A., Desjardins, P., 1997. Development and application of predictive equations of maximal static moments generated by the trunk musculature. Clin. Biomech. 12, 314–324 (Bristol, Avon). Jorgensen, K., Andersen, B., Horst, D., Jensen, S., Nielsen, A., 1985. The load on the back in different handling operations. Ergonomics 28, 183–196. Kingma, I., Bosch, T., Bruins, L., van Dieen, J.H., 2004. Foot positioning instruction, initial vertical load position and lifting technique: effects on low back loading. Ergonomics 47, 1365–1385.

Kingma, I., van Dieen, J.H., de Looze, M., Toussaint, H.M., Dolan, P., Baten, C.T.M., 1998. Asymmetric low back loading in asymmetric lifting movements is not prevented by pelvic twist. J. Biomech. 31, 527–534. Kroemer, K.H.E., Grandjean, E., 1997. Fitting the Task to the Human. A Textbook of Occupational Ergonomics. Taylor & Francis, London. Lavender, S.A., Andersson, G.B.J., Schipplein, O.D., Fuentes, H.J., 2003. The effects of initial lifting height, load magnitude, and lifting speed on the peak dynamic L5/S1 moments. Int. J. Ind. Ergonom. 31, 51–59. Lavender, S.A., Li, Y.C., Andersson, G.B.J., Natarajan, R.N., 1999. The effects of lifting speed on the peak external forward bending, lateral bending, and twisting spine moments. Ergonomics 42, 111–125. Lavender, S.A., Marras, W.S., 1990. An electromyographic analysis of an ergonomic intervention with the jackleg drill. Appl. Ergonom. 21, 90–100. Leskinen, T., Haijanen, J., 1996. Torque on the low back and the weight limits recommended by NIOSH in simulated lifts. Grenoble, France, pp. 1–7. Leskinen, T.P.J., Stalhammar, H.R., Kuorinka, I.A.A., Troup, J.D.G., 1983. A dynamic analysis of spinal compression with different lifting techniques. Ergonomics 26, 595–604. Lortie, M., 2002. Manutention: Prise d’information et dećision d’action. Le Travail Humain 65, 193–216. Marras, W.S., Lavender, S.A., 1991. The effects of method of use, tool design, and roof height on trunk muscle activities during underground scaling bar use. Ergonomics 34, 221–232. Marras, W.S., Mirka, G.A., 1992. A comprehensive evaluation of trunk response to asymmetric trunk motion. Spine 17, 318–326. McGill, S.M., 1991. Electromyographic activity of the abdominal and low back musculature during the generation of isometric and dynamic axial trunk torque: implications for lumbar mechanics. J. Orthop. Res. 9, 91–103. National Research Council, 2001. Musculoskeletal disorders and the workplace: low back and upper extremities. National Research Council and Institute of Medicine. Plamondon, A., Gagnon, M., Desjardins, P., 1996. Validation of two 3-D segment models to calculate the net reaction forces and moments at the L5/S1 joint in lifting. Clin. Biomech. 11, 101–110 (Bristol, Avon). Plamondon, A., Gagnon, M., Gravel, D., 1995. Moments at the L5/S1 joint during asymmetrical lifting: effects of different load trajectories and initial load positions. Clin. Biomech. 10, 128–136 (Bristol, Avon). Potvin, J.R., 1997. Use of NIOSH equation inputs to calculate lumbosacral compression forces. Ergonomics 40, 691–707. Schipplein, O.D., Reinsel, T.E., Andersson, G.B.J., Lavender, S.A., 1995. The influence of initial horizontal weight placement on the loads at the lumbar spine while lifting. Spine 20, 1895–1898. Schultz, A.B., Warwick, D.N., Berkson, M.H., Nachemson, A.L., 1979. Mechanical properties of human lumbar spine motion segments—Part I: responses in flexion, extension, lateral bending, and torsion. J. Biomech. Eng. 101, 46–52. Tsuang, Y.H., Schipplein, O.D., Trafimow, J.H., Andersson, G.B.J., 1992. Influence of body segment dynamics on loads at the lumbar spine during lifting. Ergonomics 35, 437–444. Vink, P., Daanen, A.M., Meijst, W.J., Ligteringen, J., 1992. Decrease in back strength in asymmetric trunk postures. Ergonomics 35, 405–416. Waters, T.R., Putz-Anderson, V., Baron, S., 1998. Methods for assessing the physical demands of manual lifting: a review and case study from warehousing. Am. Ind. Hyg. Assoc. J. 59, 871–881. Waters, T.R., Putz-Anderson, V., Garg, A., Fine, L.J., 1993. Revised NIOSH equation for the design and evaluation of manual lifting tasks. Ergonomics 36, 749–776. Webb, Associates, 1978. Anthropometric Source Book, NASA 1024. National Aeronautics and Space Administration, Washington, DC.