Trunk-head coordination in chronic neck pain, reliability, influence of visual feedback, and relationship to patient-reported outcomes.

Nineteen people with no a priori power calculation; the authors call the study exploratory and underpowered. Wide confidence intervals on most estimates, especially the inter-session ICCs and the correlations, so the true values could be quite different. Most participants had mild neck disability (NDI mean 23%) and mild dizziness (DHI mean 16%), so the findings may not apply to more severely affected patients. Thirteen of the 19 were women, reflecting the higher prevalence of neck pain in women but limiting generalisability. Only one trunk rotation per direction was used, to limit fatigue; previous studies used three, and more repetitions might improve reliability. In the 'visually constrained' condition the patient was not blindfolded and could potentially use the wall texture to orient the head. No criterion-validity study of the accelerometer-based THCT in a neck-pain population has been published, so the measurements are preliminary. Multiple analyses were performed, increasing the chance that at least one finding is a false positive. The inter-session smallest detectable changes approach the magnitude of the group medians, so individual-level tracking between visits is unreliable.

No conflict of interest declared. The Movesense sensors were funded by Physio Barillette Sarl, a private physiotherapy practice owned by the lead author (François Tharin), and by Satakunta University of Applied Sciences. The manuscript was reviewed by ChatGPT for linguistic quality.

Do not use this test to track one patient's improvement between visits. The smallest detectable change between sessions for the total path-length measure (SAE) is 38.78 m/s², while the group median is 55.27 m/s², so a real change in a single person is hard to separate from measurement noise. The good within-session reliability does not carry over to the between-session comparison.