Comparison of expert and nonexpert physicians in the assessment of vertebral fractures using the semiquantitative method in Japan.

Assessment of vertebral fracture is critically important for the diagnosis and treatment of osteoporosis. This study aimed to clarify the effectiveness of the semiquantitative (SQ) method in the assessment of vertebral fractures in Japanese clinical practice. Forty-four physicians (seven experts and 37 nonexperts) assessed the spinal radiographs of 40 patients participating in the Adequate Treatment of Osteoporosis (A-TOP) Japanese Osteoporosis Intervention Trial (JOINT)-02 at the baseline, 12 months, and 24 months using the SQ method. The proportion of diagnosed fracture cases per spine was higher in the nonexpert group than in the expert group at each time point, and was especially high in the upper thoracic spine (T4-T6). The least mean squares spinal fracture index was significantly higher in the nonexpert group than in the expert group for all time points. The kappa statistics were also higher in the expert group than in the nonexpert group for all vertebral levels at all time points. Assessment of vertebral fractures using the SQ method tended to be overestimated by nonexpert physicians compared with the experts, with poor nonexpert interobserver reliability and well-matched expert interobserver reliability. Conscious efforts to avoid overestimation and to obtain higher reliability with the SQ method should be made to achieve more precise diagnoses and treatment of osteoporosis in Japanese clinical practice.