Inter-rater reliability of a video-analysis method measuring low-back load in a field situation

Abstract

Valid and reliable low-back load assessment tools that can be used in field situations are needed for epidemiologic studies and for ergonomic practice. The aim of this study was to assess the inter-rater reliability of a low-back load video-analysis method in a field setting.

Five raters analyzed 50 work site manual material handling tasks of 14 workers. Peak and mean moments at the level of L5S1, and segment angles were obtained using the video-analysis method. Intra-class correlation coefficients (ICCs) and median standard deviations across raters were calculated.

ICCs revealed excellent inter-rater reliability (>0.9) for peak and mean moments, ICCs of segment angles were variable. Median standard deviations showed relatively small inter-rater variance for moments (standard deviation <10 Nm) and segment angle variation ranging from 0° to 20°. The proposed video-analysis method, provides a reliable tool for obtaining low-back loads from occupational field tasks.

Introduction

High low-back loads that may occur at work (e.g. during lifting, pushing and pulling of objects or working in awkward body positions) are associated with low-back pain (LBP; e.g., Marras et al., 2010; van Dieën et al., 1999). These associations have often been confirmed in epidemiological studies using self-reported exposures or field observations (da Costa and Vieira, 2010; Griffith et al., 2012; Lötters et al., 2003). However, other epidemiological studies did not find support for the association between high low-back loads and LBP, possibly as a result of the lack of appropriate measurement designs (Bakker et al., 2009). Therefore, valid and reliable low-back load assessment methods that can be applied in field settings are needed. Three types of measurement methods can be adopted: self-reports, observational techniques and direct measurement techniques (Burdorf, 2010; David, 2005). Although self-reports are highly efficient, they are assumed to be less reliable than observational techniques and direct measurements (Balogh et al., 2004; Hansson et al., 2001). On the other hand, direct measurement techniques (e.g., measuring muscle activity or body posture recordings using marker tracking or goniometry) are much more accurate but difficult to apply in large scale field studies. In field measurements of low-back load, there thus seems to be a trade-off between efficiency (in terms of time, money and resources) and accuracy. Besides, it can be argued that crude observational low-back exposure measures (e.g., the number of lifts, time spent in a flexed trunk position) provide less detailed information on low-back load than dose metrics (i.e., low-back moments), since different exposures (e.g., lifting and bending) affect the same dose. Therefore, dose-estimates can provide more insight into the etiology of LBP (Wells et al., 2004) and these metrics are more predictive of future LBP than postural exposure measures (Coenen et al., 2013).

Video-based methods using postural exposure data in biomechanical models to calculate low-back load dose estimates have been shown to be a promising category of observational techniques (e.g., Chang et al., 2010; Coenen et al., 2011; Norman et al., 1998; Potvin, 1997; Sutherland et al., 2008) in the assessment of low-back load metrics such as static (Neumann et al., 2001), cumulative (Sutherland et al., 2008) or peak low-back moments (Norman et al., 1998). Furthermore, these coding systems allow raters with minimal training and minor use of equipment to collect occupational low-back load data. High inter-rater agreement has been found when using these kinds of models to calculate cumulative low-back moments (Cann et al., 2008; Sullivan et al., 2002). However, testing of these models was only performed in laboratory situations or in mock-ups of field situations, whereas, applicability of these methods for epidemiological studies or in ergonomic practice can best be assessed when applied to actual field situations. The aim of the present study therefore was to test the inter-rater reliability of a low-back load video-analysis method in a field setting. The model that will be tested in our study has been validated against a lab-based reference method (Coenen et al., 2011) and inter-rater reliability has been assessed in a laboratory situation (Xu et al., 2011). Although these authors suggest that the method might be valid and reliable in field studies, reliability has not yet been assessed in field settings.