Reliability and stability of the Roland Morris Disability Questionnaire: intra class correlation and limits of agreement

Abstract

To analyse test-retest reliability and stability of the Dutch language version of the Roland Morris Disability Questionnaire (RMDQ) in a sample of patients (n = 30) suffering from Chronic Low Back Pain (CLBP). Patients filled out the Dutch language version of the RMDQ questionnaire twice, before starting the rehabilitation programme, with a 2-week interval. Intra Class Correlations (ICC), (one way random) was used as a measure for reliability and the limits of agreement were calculated for quantifying the stability of the RMDQ. An ICC of 0.75 or more was considered as an acceptable reliability. No criteria for limits of agreement were available. However, smaller limits of agreement indicate more stability because it indicates that the natural variation is small. The Dutch RMDQ showed good reliability, with an ICC of 0.91. Calculating limits of agreement to quantify the stability, a large amount of natural variation ( +/- 5.4) was found relative to the total scoring range of 0 to 24. The Dutch RMDQ proves to be a reliable instrument to measure functional status in CLBP patients. However, the natural variation should be taken into account when using it clinically.