Psychometric Properties Study of the Oswestry Disability Index in a Spanish Population With Previous Lumbar Disc Surgery: Homogeneity and Validity.

STUDY DESIGN

Evaluation of the psychometric properties of a questionnaire.

OBJECTIVE

To assess the psychometric properties of the Spanish version of the Oswestry Disability Index (ODI) in a Spanish population with previous lumbar disc surgery.

SUMMARY OF BACKGROUND DATA

ODI is frequently used for measuring disability in spinal disorders. In 1995, ODI was translated and transculturally adapted into the Spanish context; its content and apparent validity, internal consistency and test-retest reliability were demonstrated for a Spanish population with lumbar pain. However, this score has not been tested in terms of discriminative capacity (floor and ceiling effects) and construct validity.

METHODS

Two hundred seventy-five patients who had previously undergone surgical treatment for disc lumbar herniation completed the ODI, Short-Form 36 (SF36), EuroQol-5D (EQ5D), and Numerical Rating Scale for Back Pain. Internal consistency, floor and ceiling effects and construct validity (convergent, divergent, and "known-groups" validities) were assessed.

RESULTS

Spanish ODI showed a very good internal consistency: Cronbach-α coefficient for ODI score was 0.928; Cronbach-α coefficient, if the item was deleted, did not increase by more than 0.1 for each item; and Item-total correlations ranged from strong to very strong. There was floor effect for both ODI score and for all individual items. There was no ceiling effect. Spanish ODI showed very good construct validity because 88% (7/8) of the hypotheses about convergent, divergent and "known-groups" validities were supported. Concerning convergent validity, ODI was strongly correlated with the Numerical Rating Scale for Back Pain (0.717; P < 0.001), SF36-Utility (rho = -0.786; P < 0.001), Physical Component Summary score SF36 (rho = -0.787; P < 0.001), Mental Component Summary score SF36 (rho = -0.605; P < 0.001), EQ5D-Utility (rho = -0.833; P < 0.001) and some EQ5D physical components.

CONCLUSION

Spanish ODI showed good psychometric properties (internal consistency and construct validity), similar to others validation studies. Spanish ODI is a valid instrument for the measurement of disability in Spanish patients with previous lumbar disc surgery, specifically in patients with moderate/intense disability.

LEVEL OF EVIDENCE