Reproducibility of a battery of commonly used clinical tests to evaluate lumbopelvic motor control

Objective: To determine reproducibility of a battery of clinical tests for evaluating lumbopelvic motor control (LMC). Design: Test-retest design. Participants: Fifty healthy subjects. Outcome Measures: Two raters independently examined performance on 12 clinical tests for evaluating LMC. All tests were scored on a seven-point scale, based on qualitative and quantitative performance. Subjects were measured twice, with a two week interval between examinations. Intra- and inter-rater reproducibility of each test were determined using intraclass correlation coefficients (ICCs), standard error of measurement, smallest detectable change (SDC) and limits of agreement. Results: Reliability of the tests ranged from poor to excellent. Intra-rater ICCs ranged from 0.00 to 0.82, whereas inter-rater ICCs varied from 0.00 to 0.96. SDC values were smallest for supine leg raising, bent knee fall out, prone bridge and unilateral prone bridge (<2 points). Conclusion: This study shows limited reproducibility of a battery of 12 clinical tests for the evaluation of LMC in a healthy population. Supine leg raising, bent knee fall out, prone bridge, and unilateral prone bridge showed the smallest measurement errors. The other 8 tests were found to have large measurement errors. Based on these results, dichotomization of the rating method might be considered in order to improve reproducibility values.