Reliability, construct validity, and responsiveness of the neck disability index and numeric pain rating scale in patients with mechanical neck pain without upper extremity symptoms

<b><i>Objective</i></b>: The purpose of this study was to examine the psychometric properties of the neck disability index (NDI) and numeric pain rating scale (NPRS) in patients with neck pain (NP) without concomitant upper extremity (UE) symptoms. <b><i>Design</i></b>: A secondary psychometric analysis of 107 patients with NP without UE symptoms. Test-retest reliability, construct validity, area under the curve (AUC), minimum detectable change (MDC), and minimum clinically important difference (MCID) were calculated.<b><i>Results</i></b>: The NDI exhibited excellent reliability (ICC = 0.88; [0.63 to 0.95]), while the NPRS exhibited moderate reliability (ICC = 0.67; [0.27 to 0.84]). The AUC for both the NDI (0.86; [0.79 to 0.93]) and NPRS (0.81 [0.73 to 0.90]) was acceptable. The MDC for the NDI was 6.9, and the MCID for the NDI was 5.5(Sn = 0.83; Sp = 0.79). For the NPRS, the MDC was 2.6, and the MCID was 1.5(Sn = 0.93; Sp = 0.64). <b><i>Conclusion</i></b>: The threshold for MCID for the NDI and NPRS in patients without UE symptoms is lower (NDI = 5.5; NPRS = 1.5) than that of patients with UE/radicular symptoms (NDI = 8.5 points; NPRS = 2.2). Knowledge of these cut-scores in each presentation of NP is needed for successful research and clinical treatment. Additional outcomes may be warranted for patients with UE symptoms.