Translation of the Neck Disability Index and validation of the Greek version in a sample of neck pain patients

Marianna N Trouli¹^,2 , Howard T Vernon³ , Kyriakos N Kakavelakis⁴ , Maria D Antonopoulou¹ , Aristofanis N Paganas¹ and Christos D Lionis¹

¹Clinic of Social and Family Medicine, School of Medicine, University of Crete, Greece

²Department of Physiotherapy, University Hospital of Heraklion, Crete, Greece

³Canadian Memorial Chiropractic College, Toronto, Ontario, Canada

⁴Department of Orthopaedic Surgery and Traumatology, University of Crete, Greece

author email corresponding author email

BMC Musculoskeletal Disorders 2008, 9:106doi:10.1186/1471-2474-9-106


Published:	22 July 2008

Abstract

Background

Neck pain is a highly prevalent condition resulting in major disability. Standard scales for measuring disability in patients with neck pain have a pivotal role in research and clinical settings. The Neck Disability Index (NDI) is a valid and reliable tool, designed to measure disability in activities of daily living due to neck pain. The purpose of our study was the translation and validation of the NDI in a Greek primary care population with neck complaints.

Methods

The original version of the questionnaire was used. Based on international standards, the translation strategy comprised forward translations, reconciliation, backward translation and pre-testing steps. The validation procedure concerned the exploration of internal consistency (Cronbach alpha), test-retest reliability (Intraclass Correlation Coefficient, Bland and Altman method), construct validity (exploratory factor analysis) and responsiveness (Spearman correlation coefficient, Standard Error of Measurement and Minimal Detectable Change) of the questionnaire. Data quality was also assessed through completeness of data and floor/ceiling effects.

Results

The translation procedure resulted in the Greek modified version of the NDI. The latter was culturally adapted through the pre-testing phase. The validation procedure raised a large amount of missing data due to low applicability, which were assessed with two methods. Floor or ceiling effects were not observed. Cronbach alpha was calculated as 0.85, which was interpreted as good internal consistency. Intraclass correlation coefficient was found to be 0.93 (95% CI 0.84–0.97), which was considered as very good test-retest reliability. Factor analysis yielded one factor with Eigenvalue 4.48 explaining 44.77% of variance. The Spearman correlation coefficient (0.3; P = 0.02) revealed some relation between the change score in the NDI and Global Rating of Change (GROC). The SEM and MDC were calculated as 0.64 and 1.78 respectively.

Conclusion

The Greek version of the NDI measures disability in patients with neck pain in a reliable, valid and responsive manner. It is considered a useful tool for research and clinical settings in Greek Primary Health Care.