METHODS: Two observers independently and in random order assessed the cervical range-of-motion in 30 subjects with a dysfunction in the neck and shoulder region (symptomatic subjects) and 30 subjects without known pathology (asymptomatic subjects). Measurements included rotation in neutral position, in flexed position and in extended position, flexion-extension, and lateral bending (all active and passive). Reliability was analyzed by intraclass correlation coefficient (ICC) and agreement by limits of agreement and percentage of paired observations within 5 degrees, 10 degrees, and 15 degrees.
RESULTS: For asymptomatic subjects, the ICC varied from 0.57 to 0.85, and the limits of agreement varied between 14.5 degrees and 27.0 degrees. The percentage of paired observations within 5 degrees, 10 degrees, and 15 degrees ranged from 31% to 57%, 58% to 90%, and 78% to 93%, respectively. For symptomatic subjects, the ICC varied from 0.36 to 0.91, and the limits of agreement varied between 9.6 degrees and 37.8 degrees. The percentage of paired observations within 5 degrees, 10 degrees, and 15 degrees ranged from 17% to 60%, 33% to 93%, and 50% to 97%, respectively.
CONCLUSIONS: Despite the use of a standardized protocol and a sophisticated measurement system, the interobserver reliability of neck mobility was variable in quality, with reliability being good in rotation in neutral position, flexion-extension, and lateral bending.
Click on the above link for the PubMed record for this article; full text by subscription. The abstract is reproduced here with the permission of the publisher.