Lumbar movement control in non-specific chronic low back pain: Evaluation of a direction-specific battery of tests using item response theory

Question: What tests are most useful to identify poor lumbar movement control (LMC) and what are the dimensional and psychometric properties of these tests? Design: Multicenter, cross-sectional cohort study. Participants: Adults with non-specific chronic low back pain (NSCLBP). Methods: A literature review was conducted to identify LMC tests with good reliability (κ≥ 0.61) and their dimensionality was examined. Based on item response theory, psychometric properties of individual items and the entire battery of LMC tests were determined. Results: 277 participants with NSCLBP were included and tested by 21 physiotherapists in 19 clinics in Germany and Austria. 15 tests for LMC were assessed. The battery of LMC tests showed a direction-specific structure representing extension, flexion and rotation/lateral flexion control, for which unidimensionality and local independence were confirmed (eigenvalue >1; factor loading >0.4, Yen's Q3 <0.2). 4 items for flexion control, 4 items for extension control and 5 items for rotation/lateral flexion control were extracted. The flexion control items were the easiest items (item difficulty: 1.98 to −1.31). The rotation/lateral flexion control items were the most difficult (−1.3 to −0.08). More than 80% of all participants showed at least one incorrect direction of LMC. Conclusions: A battery of LMC tests is proposed as the most appropriate to examine individuals with precise to poor LMC. Each direction of LMC should be examined separately. Tests can be sorted according to test difficulty, so that only three tests are initially required to screen for poor LMC.