Diagnostic accuracy and discriminative validity of clinician-administered physiotherapy examination tests in adults with nonspecific neck pain: a systematic review

Abstract

To synthesize evidence on the diagnostic/classification accuracy and discriminative validity of clinician-administered physiotherapy examination tests used in adults with nonspecific neck pain, distinguishing individual-level classification of prespecified clinical or impairment states from known-groups discrimination between nonspecific-neck-pain and asymptomatic samples. PubMed, Scopus, and Web of Science were searched. Eligible studies evaluated routine physiotherapy examination tests in adults with nonspecific neck pain. The review independently screened studies, extracted data, and assessed risk of bias using QUADAS-2 domains for studies contributing threshold-based classification or construct-aligned validity estimates against explicit comparators, and a design-matched case-control appraisal for known-groups discriminative-validity studies. Narrative synthesis was performed. Eighteen studies were included. No study evaluated direct clinical diagnostic accuracy in a sample reflecting diagnostic uncertainty. Most evidence came from known-groups comparisons between adults with nonspecific neck pain and asymptomatic controls. Two studies reported threshold-based classification or construct-aligned validity estimates, but these were derived from separate symptomatic and asymptomatic group design or construct-aligned designs rather than clinically uncertain symptomatic samples. A manual examination cluster showed the most favorable threshold-based classification estimate, but this finding reflected symptomatic-versus-asymptomatic group classification rather than direct diagnosis of nonspecific neck pain. Current evidence suggests that clinician-administered physiotherapy examination tests may help characterize impairment patterns in adults with nonspecific neck pain, particularly within sensorimotor, muscle-performance, and selected movement-control domains. However, direct clinical diagnostic-accuracy evidence is lacking, and most available evidence comes from studies comparing people with neck pain against asymptomatic controls. These designs help identify impairment differences between groups but provide limited information about whether a test can classify patients in routine clinical practice. The review did not evaluate differential diagnosis of nonspecific neck pain from specific cervical disorders. Rather, it examined classification and impairment-discrimination evidence within the nonspecific-neck-pain literature. These tests should therefore be interpreted as adjuncts within a multimodal clinical assessment rather than as standalone diagnostic tools. Registration: OSF osf.io/7twf2.

The authors' abstract, as published at the source. BMC Musculoskeletal Disorders, 2026 · DOI ↗