Validity and reliability of a German version of the Neck Disability Index (NDI-G)

Abstract

The Neck Disability Index (NDI) is a widely used questionnaire in the assessment of disability of neck patients. The aim of this study was to translate the NDI according to established guidelines into German (NDI-G) and to test the psychometric properties. Patients with acute (ACU) and chronic neck pain (CHR) and a healthy control group (HCG) completed the NDI-G twice with a mean test-retest interval of 3 days. The total score of NDI-G showed high reliability (Intraclass correlation coefficient (ICC2,1) = 0.92) and a high Cronbach's alpha (α = 0.96). The minimal detectable change was 7 points. The Bland-Altman plot revealed a small positive systematic error of 1.02 points. The Kruskal-Wallis test showed significant differences in the NDI-G total score among the three groups (χ(2) = 29.77, p < 0.001). Mann-Whitney U tests showed significant differences in the total score between ACU and HCG (p < 0.001), and CHR and HCG (p < 0.001). The factor analysis of NDI-G yielded 2 factors that together explained 67% of the variance. Spearman's phi coefficients showed no correlation between the NDI-G total score and the visual analogue scale (VAS) in the ACU group (phi = 0.23, p = 0.40), and a moderate correlation in the CHR group (phi = 0.55, p = 0.03). The item analysis of the NDI-G revealed moderate to good reliability for all items. Only the item 'work' could differentiate between the ACU and CHR group. The NDI-G emerged from this study as a valid and reliable assessment. Its psychometric properties are comparable with the original version. Thus, the NDI-G is recommended for research and clinical settings in neck pain in German speaking countries.

Keywords: NDI; Neck; Pain; Reliability; Validity.