Observer variation in functional examination of the temporomandibular joint

Agreement between observers classifying TMJ sounds from data given in records of nonpatient adolescents was almost perfect in this interobserver study. Intraobserver and interobserver agreement in classifying all specific TMJ sounds at palpation and auscultation was acceptable to moderate (kappa value = 0.49 to 0.74). The agreement was considerably more reliable when classifying only one specific TMJ sound. Measurements of linear jaw opening showed small interobserver differences (coefficient of variation = 2.4 to 3.8). The significant difference found in calculating the angular mandibular opening may be the result of difficulties in maintaining maximum passive opening.