Reliability of articular indices and function tests in a population study of rheumatic disorders

Reliability and method error were assessed for 25 clinical tests by test-retest on 31 subjects with rheumatic disorder (RD) and on 28 consecutive non-RD of a reference group (REF) from a female population sample. Low systemic differences were found for joint mobility tests in the RD group and correlation coefficients were generally above 0.7. A significant difference in the RD group was found only for climbing stairs, and in the REF group for shoulder external rotation and flexion, wrist extension and grip strength. The relative method error of grip strength was large (17%). A new assessment of activities of daily living could not be fully evaluated because the population studies had limited disability, but the error was acceptable. Articular indices (Lansbury, Ritchie, and American Rheumatism Association) showed large method errors (9-25%), acceptable test-retest correlations (greater than 0.75), and a systematic difference only in the Ritchie index. The analyses showed that a number of clinical tests are adequate for population studies of RD, but the metrical properties of these tests must be considered in the planning of clinical and population studies.