Psychometric properties of the Neck OutcOme Score, Neck Disability Index, and Short Form-36 were evaluated in patients with neck pain

Abstract

Objective: To assess reliability, construct validity, responsiveness, and interpretability for Neck OutcOme Score (NOOS), Neck Disability Index (NDI), and Short Form-36 (SF-36) in neck pain patients.

Study design and setting: Internal consistency was assessed by Cronbach alpha. Test-retest reliability was evaluated by intraclass correlation coefficient (ICC), and measurement error was estimated from the standard error of measurement. Responsiveness was assessed as standardized response mean (SRM) and interpretability from the minimal important difference (MID). Construct validity was tested correlating subscale scores from NOOS and SF-36 and NDI items.

Results: At baseline, 196 neck pain patients were included. Cronbach α was adequate for most NOOS subscales, NDI, and SF-36 with few exceptions. Good to excellent reliability was found for NOOS subscales (ICC 0.88-0.95), for NDI, and for SF-36 with few exceptions. For NOOS, minimal detectable changes varied between 1.1 and 1.9, and construct validity was supported. SRMs were higher for NOOS subscales (0.19-0.42), compared to SF-36 and NDI. MID values varied between 15.0 and 24.1 for NOOS subscales.

Conclusions: In conclusion, the NOOS is a reliable, valid, and responsive measure of self-reported disability in neck pain patients, performing at least as well or better than the commonly used SF-36 and NDI.

Keywords: Construct validity; Interpretability; Measurement instrument; Neck pain; Patient-reported outcome; Questionnaire; Reliability; Responsiveness.