Accuracy of motor tests related to musculoskeletal fitness as predictors of low back pain in adolescents: a pilot study

Abstract

Background: The prevalence of low back pain in adolescents is close to that found in the adult population. In view of the relationship between low back pain and the resistance of the spine stabilizing muscles, studies have sought to identify the ability of specific motor tests to predict this condition. Objectives: Our objective was to verify the predictive ability of three motor tests related to musculoskeletal fitness to identify adolescents with low back pain. Methods: The sample consisted of 150 adolescents, of both sexes, aged between 15 and 19 years. The Nordic questionnaire validated for Brazilian adolescents was applied to identify those with low back pain. Subsequently, three motor tests (one-minute sit-up test, Sorensen test and lateral plank test) were applied. Statistical analysis involved the ROC curve, to identify the Area Under the Curve (AUC), assuming a 95% confidence interval. Results: It was observed in males that all the tests had a low AUC (between 0.56 and 0.57), without statistical significance (p > 0.05). For females, AUC ranged from 0.62 to 0.66, with statistical significance (p < 0.05) for the cut-off points identified in the one-minute sit-up test (≤ 24 repetitions) and Sorensen test (≤ 28 seconds), however, without significance (p > 0.05) for the lateral plank test. Conclusion: Motor tests of abdominal and paravertebral muscle resistance were not predictors of low back pain in male adolescents. Despite the low accuracy, the cut-off points identified in the one-minute sit-up test and Sorensen test, can be used with some caution to predict low back pain in female adolescents.