Premium
Within‐Session and Between‐Session Reliability, Construct Validity, and Comparison Between Individuals With and Without Neck Pain of Four Neck Muscle Tests
Author(s) -
Martins Filipa,
Bento André,
Silva Anabela G.
Publication year - 2018
Publication title -
pmandr
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.617
H-Index - 66
eISSN - 1934-1563
pISSN - 1934-1482
DOI - 10.1016/j.pmrj.2017.06.024
Subject(s) - medicine , intraclass correlation , physical therapy , construct validity , neck pain , physical medicine and rehabilitation , reliability (semiconductor) , standard error , population , psychometrics , clinical psychology , power (physics) , statistics , physics , alternative medicine , mathematics , environmental health , pathology , quantum mechanics
Background A variety of tests have been proposed to measure the performance of neck flexor muscles, but head‐to‐head comparisons hardly have been documented. Objective To compare reliability indexes, construct validity, and ability to discriminate between individuals with and without neck pain of 4 muscle tests (deep neck flexors endurance test [DNFET]; 2 variations of the craniocervical flexion test [CCFT1 and CCFT2]; and dynamometry). Design Reliability and validity study. Setting General community. Participants A total of 66 participants, 33 with chronic idiopathic neck pain (mean ± standard deviation pain intensity: 3.2 ± 1.9) and 33 without neck pain, from the general population. Methods/Main Outcome Neck muscle functioning was assessed with the CCFT1, the CCFT2, the DNFET, and dynamometry on 2 separate sessions. Participants with neck pain also were assessed for pain intensity, disability, pain catastrophizing, and fear of movement. Results Relative reliability of all tests was at least moderate (intraclass correlation coefficient ≥ 0.62), whereas measurement error was high, particularly for the DNFET (95% minimum detectable change ≥ 23.00 seconds). All tests showed moderate correlation (r ≥ 0.3) with at least 2 pain‐related measures and moderate‐to‐strong correlations with each other. Principal component analysis retained 2 factors explaining 68%‐73% of the variance of the 4 muscle tests. Significant differences between groups were found for the DNFET and dynamometry ( P < .05). Conclusion The reliability indexes suggest that the DNFET and the CCFT may be more appropriate for group comparisons than for individual comparisons. The 4 tests seem to have construct validity, but they also seem to measure slightly different constructs. Level of Evidence III