Current recommendations for assessing the reliability of a measurement tool: A SURVIVAL GUIDE FOR ORTHOPAEDIC SURGEONS

    loading  Checking for direct PDF access through Ovid

Abstract

Previous standards for assessing the reliability of a measurement tool have lacked consistency. We reviewed the most current American Society for Testing and Materials and International Organisation for Standardisation (ISO) recommendations, and propose an algorithm for orthopaedic surgeons. When assessing a measurement tool, conditions of the experimental set-up and clear formulae used to compile the results should be strictly reported. According to these recent guidelines, accuracy is a confusing word with an overly broad meaning and should therefore be abandoned. Depending on the experimental conditions, one should be referring to bias (when the study protocol involves accepted reference values), and repeatability (sr, r) or reproducibility (SR, R). In the absence of accepted reference values, only repeatability (sr, r) or reproducibility (SR, R) should be provided.

Take home message: Assessing the reliability of a measurement tool involves reporting bias, repeatability and/or reproducibility depending on the defined conditions, instead of precision or accuracy.

Related Topics

    loading  Loading Related Articles