Neck Pain and Disability Scale and Neck Disability Index: validity of Dutch language versions

To investigate the validity of the Neck Pain and Disability Scale Dutch Language Version (NPAD-DLV) and the Neck Disability Index (NDI)-DLV. NPAD–DLV, NDI–DLV, Short-Form-36 Health Survey (SF-36)-DLV, visual analog scale (VAS)pain and VASdisability were administered to 112 patients with non-specific chronic neck pain in an outpatient tertiary rehabilitation setting. Twenty seven hypotheses were formulated regarding validity. NPAD–DLV and NDI–DLV were evaluated for content validity (normal distribution total scores, missing items, floor and ceiling effects), internal consistency (Cronbach’s alpha and Spearman Item–total correlations), construct validity (Pearson correlations with SF-36 domains, VASpain and VASdisability and Pearson correlation between total scores of NPAD–DLV and NDI–DLV). NPAD–DLV and NDI–DLV scores were distributed normally. Missing items were negligible. Floor and ceiling effects were absent in NPAD–DLV and in NDI–DLV two items had floor effects and one item had a ceiling effect. Cronbach’s alpha of NPAD–DLV was 0.93 and of NDI–DLV 0.83. Item–total correlations ranged for NPAD–DLV from 0.45 to 0.73 and for NDI–DLV from 0.40 to 0.64. The correlation between, respectively, NPAD–DLV and NDI–DLV and: SF-36 domains ranged from −0.36 to −0.70 and from −0.34 to −0.63; VASpain was 0.54 and 0.43; VASdisability was 0.57 and 0.52. The correlation between the total scores of NPAD–DLV and NDI–DLV was 0.77. Twenty six hypotheses were not rejected and one hypothesis was rejected. The NPAD–DLV and NDI–DLV are valid measures of self-reported neck-pain related disability.