THE CLASSIFICATION OF ANATOMIC-BASED AND SYMPTOM-BASED LOW-BACK DISORDERS USING MOTION MEASURE MODELS

Study Design. This study observed the trunk angular motion features of healthy subjects and those experiencing chronic low back disorders as they flexed and extended their trunks in five symmetric and asymmetri c planes of motion. Trunk angular position, velocity, and acceleration were evaluated during several cycles of motion. Objective. The trunk angular motion features of the low back disorder group were normalized relative to the healthy subjects and used to 1) evaluate the repeatab ility and reliability of trunk motion as a measure of trunk musculoske letal status, 2) quantify the extent of the disorder, 3) determine the extent to which trunk motion measures might be used as quantifiable m eans to help classify low back disorders. Summary of Background Data. Given the magnitude of the low back disorder problem, it is problemati c that there are few quantitative methods for objectively documenting the extent of a disorder. Impairment rating of low back disorders can vary by as much as 70% using current systems. Diagnoses and classifica tion schemes are rarely based upon quantitative indicators and we are unable to easily assess and diagnose low back disorders. It is importa nt to qunatitatively evaluate low back disorders so that proper treatm ent can be administered and the risk of exacerbating the problem can b e minimized. Methods. Three-hundred-thirty-nine and women between 20 a nd 70 years old who had not experienced significant back pain were rec ruited as the healthy subjects in this study. One hundred-seventy-one patients with various chronic low back disorders also were recruited a nd compared with the healthy group of subjects. All subjects wore a tr iaxial goniometer on their trunks that documented the angular position , velocity, and acceleration of the trunk as the subjects flexed and e xtended their trunks in each of five planes of motion. Trunk motion fe atures first were normalized for subject gender and age. Several two-s tage eight-variable models that account for trunk motion interactions were developed to classify the 510 healthy and low back-injured subjec ts into one of 10 anatomic and symptom-based low back disorder classif ication categories. Results. Using conservative cross-validation measu res, it was found that the stage one eight-variable model could correc tly classify more than 94% of the subjects as either healthy or having a low back disorder. One of the stage eight-variable models was able to reasonably classify the patients with low back disorders into one o f 10 back disorder classification groups. Conclusion. The motion-relat ed parameters may relate to biomechanical or learned sensitivities to spinal loading. This study suggests that higher-order trunk motion cha racteristics hold great promise as a quantitative indicator of the tru nk's musculoskeletal status and may be used as a measure of the extent of a disorder and as a measure of rehabilitative progress. Furthermor e, once the interactive nature of these trunk motion characteristics i s considered, the model could help diagnose low back disorders. Howeve r, independent data sets are needed to validate these findings.