Baseline Cervical Joint Position Error and Headache Intensity

Cervicogenic headache (CGH) is a secondary headache disorder associated with cervical musculoskeletal dysfunction. Clinical features may include unilateral headache, headache provoked by neck movement or sustained cervical postures, restricted cervical range of motion, neck or shoulder pain, and evidence of cervical musculoskeletal dysfunction. Physiotherapy is commonly used in the management of CGH; however, outcomes following physiotherapy may vary between individuals.

Cervical joint position error (JPE) is a measure of cervical sensorimotor accuracy. It assesses the ability of an individual to accurately return the head to a previously established starting position following active cervical movement. Increased JPE may reflect impaired cervical proprioception or sensorimotor control. Although elevated cervical JPE has been reported in individuals with neck pain and CGH, its potential prognostic relevance for subsequent headache outcomes following physiotherapy remains uncertain.

This retrospective cohort study will examine whether baseline cervical JPE is associated with headache intensity at the observed treatment-completion assessment among adults clinically classified with CGH who received routine clinician-directed physiotherapy. The study will use routinely collected clinical records from the outpatient physiotherapy department of Saveetha College of Physiotherapy, Saveetha Institute of Medical and Technical Sciences, Chennai, Tamil Nadu, India. Clinical records from January 2023 through December 2025 will be reviewed.

The study population consists of adults clinically classified with CGH according to a retrospective operational classification rule derived from the Cervicogenic Headache International Study Group criteria. The retrospective classification required documentation of at least three of five clinical features: unilateral headache without sideshift, headache provoked by neck movement or sustained cervical postures, restricted cervical range of motion, ipsilateral neck/shoulder/arm pain, and evidence of cervical musculoskeletal dysfunction on clinical examination.

The primary exposure is baseline mean cervical JPE. JPE was assessed before physiotherapy using a CROM device. Patients were seated with the back supported and feet flat on the floor, with the eyes closed. A head-mounted laser pointer attached to the CROM device projected onto a wall-mounted target. After memorizing the starting position, the patient actively moved the cervical spine into right rotation, left rotation, flexion, and extension and then attempted to return to the starting position. Angular displacement from the starting position was recorded in degrees. Three trials were performed for each direction, and the mean absolute repositioning error across the four directions was used as the primary JPE measure.

The primary outcome is headache intensity recorded at the observed treatment-completion assessment using the 11-point Numeric Pain Rating Scale (NPRS). The treatment-completion assessment was recorded at the final treatment session or within one week of treatment completion. Because treatment duration was not standardized, the observed treatment-completion assessment represents the outcome at the time the patient's routine physiotherapy episode was completed rather than a standardized follow-up time point.

The primary analysis will evaluate the adjusted association between baseline cervical JPE and treatment-completion NPRS using multivariable linear regression within an ANCOVA framework. Baseline NPRS, age, sex, and baseline Neck Disability Index will be included as prespecified covariates based on clinical relevance and the conceptual framework for the study. The analysis will estimate the difference in treatment-completion headache intensity associated with each one-degree increase in baseline mean cervical JPE while holding the prespecified covariates constant.

The study distinguishes a prognostic factor from a treatment-effect modifier. A prognostic factor is a baseline characteristic associated with an outcome among patients receiving treatment, whereas a treatment-effect modifier would require evidence that the baseline characteristic changes the relative benefit of one treatment compared with another treatment or control condition. Because this study is retrospective and does not randomly assign or compare treatment interventions, it is designed to evaluate prognostic association and cannot establish whether baseline JPE modifies the effect of physiotherapy.

Secondary and exploratory analyses will examine whether baseline JPE is associated with achievement of a prespecified reduction of at least 2 points in NPRS during the observed treatment period. Additional exploratory analyses include receiver operating characteristic analysis, assessment of discriminatory ability, bootstrap internal validation, and direction-specific analyses of JPE. These analyses are considered exploratory because of the structural dependence of the responder definition on baseline and treatment-completion NPRS and because the study does not include an independently defined prospective prediction endpoint.

The study will also describe changes in headache intensity and, where available, cervical JPE between baseline and treatment completion. These analyses are descriptive and will not be interpreted as evidence that physiotherapy caused the observed changes because the study has no untreated comparison group and treatment was not standardized or assigned by the investigators.

Patients in the retrospective cohort received routine clinician-directed multimodal physiotherapy according to clinical practice. Documented treatment approaches included combined manual therapy and exercise, exercise alone, and manual therapy alone. Interventions commonly included cervical joint mobilization, deep cervical flexor training, cervical stabilization exercises, scapular strengthening, postural correction, cervical proprioceptive retraining, and prescribed home exercises. Treatment exposure was not determined by the study investigators.

The study uses routinely collected clinical information and does not alter patient treatment. Individual informed consent was waived because of the retrospective design and use of de-identified clinical data. The study was approved by the Institutional Review Board of Saveetha Institute of Medical and Technical Sciences (IHEC/SDC/PHYSIO/2025-0142) and was conducted in accordance with the Declaration of Helsinki. Reporting follows the STROBE guidance for observational cohort studies.

The principal analytical objective is to determine whether baseline cervical JPE provides information about subsequent headache intensity beyond baseline headache severity, disability, age, and sex. The study does not seek to establish a causal effect of JPE on headache outcomes, determine the effectiveness of a specific physiotherapy intervention, or recommend treatment selection based on JPE.

The retrospective nature of the study introduces potential limitations, including residual confounding, selection related to treatment engagement, non-standardized treatment duration, variation in treatment exposure, and non-standardized timing of treatment-completion assessment. The study therefore provides preliminary evidence regarding group-level prognostic association. Prospective studies with standardized diagnostic criteria, protocolized treatment, blinded JPE assessment, fixed follow-up time points, and external validation would be required to determine the clinical utility of JPE as a prognostic factor.

Inclusion criteria

• Adults aged 18-65 years.
• Clinically classified with cervicogenic headache (CGH) based on the retrospective classification rule derived from the Cervicogenic Headache
• International Study Group criteria.
• At least 3 of the 5 documented CGH features were present.
• Baseline cervical joint position error (JPE) was measured before physiotherapy.
• Baseline headache intensity (NPRS) was documented.
• Completed at least 4 physiotherapy sessions.
• Had at least one treatment-completion outcome assessment.
• Clinical records contained sufficiently complete baseline and outcome information.

Exclusion criteria

• History of cervical spine surgery.
• Cervical trauma or whiplash within the previous 6 months.
• Cervical fracture, instability, or severe structural pathology.
• Rheumatoid or other inflammatory cervical disease.
• Neurological disorders affecting cervical sensorimotor function.
• Vestibular disorders affecting JPE testing.
• Migraine or another primary headache disorder as the predominant diagnosis.
• Systemic malignancy affecting the cervical region.
• Incomplete baseline JPE or treatment-completion outcome data.