Reliability and measurement error of 3-dimensional regional lumbar motion measures: a systematic review

Abstract

Objective: The purpose of this study was to systematically review the literature on reproducibility (reliability and/or measurement error) of 3-dimensional (3D) regional lumbar motion measurement systems.

Methods: Electronic searches were performed in PubMed, Cumulative Index of the Nursing and Allied Health Literature, Embase, and Mantis databases. To be included, original studies had to report on the reproducibility of a 3D computerized regional lumbar spinal motion analysis system in human subjects. A detailed checklist was developed based on guidelines for reporting reliability and agreement studies, the standards for reporting of diagnostic accuracy, and quality assessment of diagnostic accuracy studies and used for data extraction and quality assessment. The checklist consisted of descriptive items divided into 4 domains: study population, testing circumstances, equipment, and data analysis and presentation. The descriptive items were used as foundation for the quality assessment reflecting the reporting level of the included articles.

Results: A total of 15 articles were included in this study. We found incomplete reporting in 1 or more domains in all articles. A varying amount of measurement error was reported in 8 of the 15 articles. Because of incomplete reporting, these reliability and measurement error estimates are difficult to interpret.

Conclusions: The current literature on the reliability and measurement error of measures created by regional 3D spinal instruments contains uncertainties especially in relevant clinical populations. There is uncertainty with respect to the degree that repeated measurements by 3D regional spinal motion instruments are reproducible. However, limited to the studies where reliability estimates were provided, most instruments used under standardized conditions may be considered reliable enough to be used for research purposes on the group level, but it is uncertain if they can be used on the individual patient level.