Back strength cannot be predicted accurately from anthropometric measures in subjects with and without chronic low back pain

• Published in: Clinical biomechanics (Bristol) Vol. 18; no. 6; pp. 473 - 479
• Main Authors: Larivière, Christian, Gravel, Denis, Gagnon, Denis, Bertrand Arsenault, A., Loisel, Patrick, Lepage, Yves
• Format: Journal Article
• Language: English
• Published: England Elsevier Ltd 01.07.2003
• Subjects: Adult; Anthropometry; Back - physiopathology; Back muscle; Body Mass Index; Chronic Disease; Dynamometry; Fat free mass; Humans; L5/S1 extension torque; Low back pain; Low Back Pain - physiopathology; Male; Middle Aged; Motor learning; Muscle, Skeletal - physiopathology; Regression Analysis; Strength; Fat free mass; Anthropometry; Back muscle; Low back pain; Dynamometry; L5/S1 extension torque; Motor learning; Strength
• ISSN: 0268-0033; 1879-1271
• DOI: 10.1016/S0268-0033(03)00026-3

Objectives. (1) To develop a multiple regression equation using anthropometric measurements to predict back strength and (2) to estimate the effect of practice on the back strength results and back strength predictions. Design. Comparative study with repeated measures performed on three days. Background. The assessment of back muscle relative endurance (% maximal strength) requires the measurement of maximal back strength which is problematic with low back pain patients. Methods. The back strength (L5/S1 static extension moment), age and 26 anthropometric parameters were obtained from 83 male volunteers [42 healthy subjects and 41 chronic low back pain patients] aged between 20 and 60 years. A subsample of 20 healthy subjects and 20 patients were assessed through three days of testing to evaluate the variations of back strength with practice. Results. The final regression model ( n=42 healthy subjects) explained 39% of the variance in back strength. Back strength increased with practice ( Day 1 < Day 2 & 3) for both healthy and patients groups. However, the error of prediction of back strength derived from the regression model showed a significant improvement with practice for the patients only. Conclusions. Back strength cannot be predicted from simple anthropometric measures without important errors. The decrease of the error of prediction obtained for the patient group is indicative of a decreasing influence of psychological factors that are independent of motor learning because the effect of motor learning would have occurred on both groups if present. Relevance The prediction of back strength would be useful to determine (1) the relative force level of a given exercise or (2) the preinjury back strength. Unfortunately, anthropometric measurements are not enough good predictors of back strength.