Precision based guidelines for sub-maximal normalisation task selection for trunk extensor EMG.

AIM: The object of this study was to quantify the contribution of sub-maximal normalisation to the overall variance of exposure parameters describing erector spinae (ES) activity, and to provide guidelines for task selection which minimize methodological variance.

METHODS: ES EMG was measured from three locations (T9, L1 and L5 levels) on fifteen men performing a manual materials handling task in the laboratory on three separate days. Four repeats of each of eleven sub-maximal normalisation tasks (eight static, three dynamic) were collected, work data were normalised to each task and repeat, and exposure parameters calculated. The unique contribution of normalisation to the overall variance was determined for each task and exposure parameter using variance component analyses. Normalisation tasks were scored according to their relative contributions to the overall variance and coefficients of variation.

RESULTS: A prone task, similar to the Biering-Sørensen test posture, was the most repeatable for all electrode locations and across all exposure parameters. Thoracic level normalisation typically showed poorer repeatability than lumbar normalisation.

DISCUSSION: To maximize measurement precision, we recommend that future ES EMG studies employing sub-maximal normalisation utilise said prone task. An alternate normalisation task specific to thoracic level ES muscles may be warranted.