Methods: A prospective observational study of a fixed cohort consisting of 357 military conscripts was completed. Data were collected at baseline and after 3 months of military service. The outcome variables were self-reported LBP and leg pain at any time during those 3 months. The main independent (predictor) variable was coping. Covariables were biological (age, height, weight, fitness, strength, and a history of previous problems) and social (type of work, years of education, and social support). The clinical usefulness was shown for the best model, with each outcome variable in relation to sensitivity/specificity, positive/negative predictive value, numbers correctly classified, and the area under the receiver operating characteristics curve.
Results: The SOC-13 questionnaire could successfully predict leg pain (odds ratio [OR], 3.3), but only 1 of its 13 items predicted LBP (OR 2.0). For leg pain, the strongest predictor was age (OR, 4.5), and for LBP, the strongest predictor was previous LBP (OR, 6.7). The receiver operating characteristics curves of the final models were 0.85 and 0.79, respectively.
Conclusions In these young Danish conscripts, coping or elements of coping (as defined with the help of the SOC-13 questionnaire) could predict LBP and leg pain occurring during a 3-month period. Biological and psychological variables dominated the final models, but none of the social variables could significantly predict LBP or leg pain.
This abstract is reproduced with the permission of the publisher. Full text is available by subscription. DOI Link