The Neck Bournemouth Questionnaire cross-cultural adaptation into Dutch and evaluation of its psychometric properties in a population with subacute and chronic whiplash associated disorders

Abstract

Study design: Cross-cultural adaptation of an outcome questionnaire.

Objective: The aim of the study was to cross-culturally adapt the Neck Bournemouth Questionnaire (NBQ) for the Dutch language in a population of people with subacute and chronic whiplash associated disorders (WAD), and to assess its psychometric qualities.

Summary of background data: The NBQ covers the salient dimensions of the biopsychosocial model of pain, and has been shown to be reliable, valid, and responsive to clinically significant change in patients with non specific neck pain. However, no Dutch validated version was available for patients with WAD at the time our study was initiated.

Methods: The English version of the NBQ was translated into Dutch (NBQ-NL) and back-translated according to established guidelines. The internal consistency (with help of Cronbach's alpha), construct validity, and convergent validity were estimated in a different group of 92 patients. Agreement and correlation between the NBQ-NL scores and counterpart questionnaires (SF-36, Neck Disability Index, Hospital Anxiety and Depression Scale, the General Perceived Self-Efficacy and a numerical rating scale for perceived pain) were investigated, using Bland and Altman method and Spearman rank correlation coefficient. Subsequently, 34 patients with subacute WAD completed the questionnaire twice over 1 to 3 weeks, to assess its test-retest reliability.

Results: Ninety-two patients, with subacute and chronic WAD, completed the study. Their mean age was 41 years (SD = 11). There were relative high mean scores on the Neck Disability Index (25.5; SD = 8.2), and the NBQ (43.9; SD = 14.8). The NBQ-NL was granted face-validity. Spearman rank correlation coefficient was 0.51 to 0.82 (except for General Perceived Self-Efficacy (0.21). The limits of agreement of normalized scores were relative variable, from small to wide. Cronbach's alpha (internal consistency) for the NBQ-NL whole scale was 0.87. The Intraclass correlation coefficient for the test-retest reliability was excellent (0.92) and the SEM was relatively low (3.7).

Conclusion: Overall, the rank correlation level in general was good, whereas the agreement between questionnaires outcome was variable, most of which seems to be attributable to absolute scale differences. The NBQ-NL is a useable patient-orientated tool for assessing disability in clinical studies and clinical diagnosis in Dutch speaking patients with WAD.