Detecting relevant changes and responsiveness of Neck Pain and Disability Scale and Neck Disability Index.

Abstract

To investigate relevant change on the Neck Pain and Disability Scale (NPAD) and Neck Disability Index (NDI) and which questionnaire is the most responsive in patients with non-specific chronic neck pain (CNP). Seventy-six patients with non-specific CNP in an outpatient tertiary rehabilitation setting were dichotomized into "improved" and "stable" based on global perceived effect (GPE) scores. To investigate relevant change minimal detectable change (MDC) and minimal important change (MIC) with the receiver operator characteristic (ROC) cut-off point were assessed. Comparison of responsiveness was performed using areas under the ROC curve (AUC) and correlations between change scores of NPAD and NDI, and GPE. MDC and MIC on NPAD (scale 0-100) were 31.7 and 11.5 points, respectively. MDC and MIC on NDI (scale 0-50) were 8.4 and 3.5 points, respectively. Changes should exceed this MDC or MIC cut-off to be interpreted as relevant. AUC was 0.75 for both NPAD and NDI. Correlations between change scores of NPAD and NDI, and GPE were, respectively, 0.48 (95 % CI 0.29-0.64) and 0.49 (95 % CI 0.30-0.64). Relevant change on both NPAD and NDI assessed with MDC and MIC resulted in different cut-offs and consequently with different amounts of certainty that the patient is improved. Responsiveness of NPAD and NDI was similar.