Changes in the amount and structure of motor variability during a deboning process are associated with work experience and neck–shoulder discomfort

Abstract

In this field study, the size and structure of kinematics variability were assessed in relation to experience and discomfort during a deboning task. Eighteen workers divided in groups with low/high experience and with/without neck–shoulder discomfort participated. Standard deviation and coefficient of variation (amount of variability), as well as approximate entropy and sample entropy (complexity) and, correlation dimension (dimensionality) were computed for head–shoulder, shoulder–hip and elbow–hip displacement in the vertical direction. A longer work experience was associated with shorter work cycle duration and decreased amount of variability while complexity increased for the head–shoulder displacement, P < 0.05. Shorter work cycle, lower amount of variability and, lower dimensionality for the head–shoulder displacement were found in relation to discomfort, P < 0.05. While the amount of variability, complexity and dimensionality increased for the elbow–hip displacement, P < 0.05. These findings suggest a functional role of experience via learning effects and discomfort through compensatory mechanisms on the size and structure of motor variability.

Introduction

Meat processing involves a considerable amount of manual operations and several studies have demonstrated that workers performing meat processing tasks are at high risk of contracting work-related musculoskeletal disorders (WMSD) (Magnusson et al., 1987, Marklin and Monroe, 1998, Punnett and Wegman, 2004). In the meat processing industry, WMSD most commonly affect the upper extremities including neck and shoulder (Viikari-Juntura et al., 1991, Frost et al., 1998) and the key physical risk factors include exerted force, repetitive movements, lack of recovery, and awkward postures (Sommerich et al., 1993). A general lack of quantitative field studies in meat industry has been expressed (Marklin and Monroe, 1998). Most knowledge about WMSD due to slaughterhouse operations is from experimental studies investigating the effect of cutting force (McGorry et al., 2004) and muscle activity (Grant and Habes, 1997, Madeleine et al., 1999) during specific limb movements, rather than during functional and in vivo work activities (Juul-Kristensen et al., 2002). Quantitative biomechanical analysis can be used to identify motor patterns during work.

Mathiassen et al. (2003) have suggested in a rather unspecific manner that differences in motor patterns and motor control could explain why some workers performing a constrained industrial work task develop WMSD while others, performing the same work task, do not. A recent laboratory study by our group (Madeleine et al., 2008b) demonstrated that the size of cycle-to-cycle variability was larger among experienced butchers compared with novices and suggested that motor patterns change with learning and experience. Moreover, discomfort or pain has also been reported to change the size of motor variability (Parakkat et al., 2007, Madeleine et al., 2008a, Madeleine et al., 2008b). Variability is a central characteristic of all human movement because of its role in motor learning and control (Latash and Anson, 2006). A few studies have suggested a possible beneficial effect of e.g. varying load and movement pattern for the prevention of WMSD (Kilbom and Persson, 1987, Madeleine et al., 2008a, Madeleine et al., 2008b). To understand the nature and complexity of the motor variability, a collection of different types of variability measures needs to be considered (Newell and Corcos, 1993, Stergiou, 2004). Thus, a combination of linear parameters along with nonlinear estimators such as approximate entropy or/and correlation dimension has been suggested (Stergiou, 2004).

Linear descriptors such as standard deviation and coefficient of variation are commonly used to characterize the amount of variability in movements (Stergiou, 2004) and to date, variability in ergonomics has only been measured by means of these linear descriptors (Moller et al., 2004, Madeleine et al., 2008a, Madeleine et al., 2008b). However, these approaches do not provide information about the true structure of motor variability (Slifkin and Newell, 1999, Buzzi et al., 2003). Nonlinear analysis derived from chaos theory have contributed the understanding the variations over time of biological signals through e.g. aging and disease (Vaillancourt and Newell, 2002). In this context, approximate and sample entropy as well as correlation dimension can be computed to characterize the complexity and dimensionality of the kinematics signals (Buzzi et al., 2003, Georgoulis et al., 2006). Thus, the idea of combining linear and nonlinear techniques is sound and may expand our knowledge on the amount and structure of variability in ergonomics, and thereby provide valuable information about motor strategies in relation to e.g. experience and discomfort.

This cross-sectional field study focused on motor techniques and variability among slaughterhouse workers performing deboning work. The purpose was to assess motor variability in relation to the subjects' work experience and reported discomfort in the neck–shoulder region using both linear and nonlinear techniques.