Inter-Rater Reliability of the Strain Index

Abstract

The Strain Index is one of several tools available to evaluate exposure to musculoskeletal stressors in the workplace in order to predict whether workers are at an increased risk of developing distal upper extremity disorders. The purpose of this study was to determine the inter-rater reliability of the Strain Index. Fifteen raters initially estimated or measured data for five Strain Index task variables: (1) intensity of exertion, (2) duration of exertion, (3) efforts per minute, (4) hand/wrist posture, and (5) speed of work, using a computer to view 61 video segments of single task jobs. Video segments were organized to provide a balanced number for each of the five variables under study, where each segment was used to provide data for a single variable. Raters then assigned rating values to each task according to published procedures. For an additional 12 segments, raters performed complete Strain Index analyses, including calculation of Strain Index scores, which were used to determine hazard classifications. Raters assessed the tasks both as individuals and as members of five three-person teams. Inter-rater reliabilities of the procedures leading to and including the Strain Index scores were assessed using an intraclass correlation coefficient, (ICC(2,1)). Inter-rater reliability for the dichotomous hazard classification was assessed using Kuder-Richardson-20 (KR-20, an ICC for dichotomous data). For task variables and estimated data, ICC(2,1) varied between 0.66–0.84 for individuals and 0.48–0.93 for teams. The Strain Index score had an ICC(2,1) of 0.43 and 0.64 for individuals and teams, respectively. For the most important variable, hazard classification, KR-20 was 0.91 for the individuals and 0.89 for the teams.