Measuring a job’s cumulative load on the lumbar spine – a concept seen as having great value relative to the prediction of low back injury – can vary greatly, according to Fischer et al., depending on:
1) the model (tool) used to calculate low back forces
2) work time standardization (a consistently defined period of exposure; i.e., including or not including planned/unplanned rest in the cumulative load calculation)
In comparing four compressive force models applied to one job, the authors found the cumulative compression force calculation varied 53%, 17% and 18% relative to the lowest calculation. When unplanned rest was included as a contributor to cumulative lumbar load, one model’s cumulative force value increased by 25.5% while a second model’s cumulative force was raised by 28.3%.
The authors conclude that in order to measure cumulative lumbar spinal loading, a new model is needed that can be easily applied and flexible to consider the variety of forces and moments in the workplace. Further, standardization of the exposure needs to be defined to allow for comparison of study findings in the literature.
Study Design
Subjects and Tasks
Twenty-eight automotive parts manufacturing plant workers (15 male, 13 female) were recruited from nine job areas. The tasks associated with the jobs were performed through a variety of symmetrical/asymmetrical lifts, pushes and pulls. Lifting loads ranged from 0.5-3 kg while push/pull maneuvers required forces from 15-25 kg.
Calculation of Spinal Compression
Tasks were divided into subtasks for data collection of posture (captured on video with a minimum of three cycles per worker recorded) and forces (measured with a Chatillon force gauge).
All tasks were analyzed even when the demands for posture or force were low.
Posture matching was performed frame by frame using the 3DMatch program. Based on elbow, shoulder, neck and trunk angles, the mass moment arm relative to L4/L5 was calculated (symmetrical and asymmetrical positioning was considered).
Lumbar compression forces were calculated by following four formulas:
1) Single muscle equivalents (SME) with a 6 cm moment arm (McGill et al. 1996).
2) EMG-based third order polynomial (McGill et al. 1996).
3) EMG-based third order polynomial with the intercept reduced to 53.6% of each participant’s individual body mass (body mass proportion above L4/L5 from Winter 2005).
4) Custom-designed hybrid approach to select the model output from either the SME or polynomial depending on the moments present.
The hybrid approach allowed for use of the SME for low load model instances while the polynomial was used when high loads occurred. An instantaneous force was multiplied by 0.33 (sample time duration) to produce a finite load value. All finite values were used to determine an average task cumulative load (mean value created from analysis of the three trials).
The average task cumulative load was extrapolated relative to the total number of times each task was completed during a work shift. Over the course of a shift, a worker performed one to seven tasks. Extrapolation was applied relative to each task to yield a shift cumulative spine compression exposure.
Spinal compression exposure during unplanned rest (i.e., time spent socializing, pausing to control pace) was also calculated. Unplanned rest time duration was determined by starting with the 7 hour work day and subtracting the work time (average cycle time for each task times the number of cycles completed during a shift). The unplanned rest time duration was multiplied by standing spinal compression (the postural position usually assumed during unplanned rest).
A time standardized cumulative spine load for each worker was produced by adding the working spinal compression value with the unplanned rest spinal compression value.
Evaluation of Findings
Cumulative spinal loading values as calculated from the four modeling methods were compared through two-way ANOVA design. Further, cumulative spinal load values including and excluding unplanned rest were compared. There were statistically significant differences between:
· each model calculation
· values that did and did not use unplanned rest within each model calculation
The cumulative compression predicted by the modified EMG-based third order polynomial yielded the lowest cumulative value: 53%, 17% and 18% less than the EMG-based third order polynomial, single muscle equivalents, and custom-designed hybrid approach, respectively.
Including unplanned rest produced a cumulative lumbar compression estimate 25.5% higher for the SME model and 28.3% higher when using the EMG-based third order polynomial.
Other Findings
The authors reviewed the limitations and real-life application shortcomings of spinal compression models including:
1) The SME assumes all spinal forces occur through one vector (one tendon) and occur in one plane of motion.
2) EMG models require extensive equipment and are difficult to use in the work environment.
3) Polynomial equations have assumptions not always consistent with tasks being assessed.
Article Title: Methodological considerations for the calculation of cumulative compression exposure of the lumbar spine: A sensitivity analysis on joint model and time standardization approaches
Publication: Ergonomics 50:9 1365-1376, 2007
Authors: S L Fishcer, W J Albert, A J McClellan and J P Callaghan
This article originally appeared in The Ergonomics Report™ on 2007-08-28.