Manipulation beat mobilisation for neck pain, by less than the patient would notice

Design

randomised mechanistic crossover trial, two sessions 72 hours apart

Population

89 individuals with chronic mechanical neck pain, recruited in Zurich and Trois-Rivières, 2023-2024

Primary outcome

immediate change in cervical range of motion and pain intensity, with electromyographic and pressure pain threshold measures

Effect

range of motion improved equally; pain fell 0.786 points after manipulation against 0.373 after mobilisation (β 0.413, 95% CI 0.160-0.665, P=0.002); larger electromyographic responses after manipulation

Eighty-nine people with chronic mechanical neck pain each received both cervical manipulation and cervical mobilisation, in random order, in two sessions 72 hours apart, at centres in Zurich and Trois-Rivières. Range of motion, pain intensity, grip strength and pressure pain thresholds were measured before and after each treatment, with surface electromyography recording muscle activity during it.

Range of motion improved after both treatments with no difference between them. Pain intensity showed a significant interaction favouring manipulation: a mean decrease of 0.786 points on the numeric rating scale against 0.373 for mobilisation (β 0.413, 95% CI 0.160-0.665, P=0.002). Electromyographic responses were larger after manipulation in both sternocleidomastoid and upper trapezius bilaterally. Pressure pain thresholds rose after both. Grip strength did not change consistently.

The authors then do the thing that makes this worth reporting: they state that results reaching statistical significance did not reach clinical significance. A difference of 0.41 points on an 11-point pain scale is well below the two-point change usually taken as the minimum a patient notices, and the crossover design measured immediate effects only, in a single session, with no follow-up. So the finding is that manipulation produces a larger immediate neuromechanical response than mobilisation and a pain difference too small to matter. For a clinician choosing between them, that argues on safety and patient preference rather than on efficacy - and cervical manipulation carries a rare but real vertebral artery risk that mobilisation does not.

• Do not choose cervical manipulation over mobilisation on the strength of a pain difference this size
• Where manual therapy is used for chronic neck pain, let safety profile and patient preference decide the technique
• Screen for vertebral artery and upper cervical instability risk factors before any cervical manipulation
• Keep manual therapy as an adjunct to active exercise, which is where the durable evidence in chronic neck pain sits
• Note this measured immediate effects in a single session - it says nothing about outcomes at weeks or months

Why it matters

It removes efficacy as the reason to prefer the technique that carries the rare catastrophic risk.

Don't overread it

Immediate post-treatment measurements in a single crossover session - no follow-up, and the authors state the differences were not clinically significant.

The statistics, in plain English

A between-treatment difference of 0.413 points on a 0-10 scale, with a confidence interval of 0.160 to 0.665, is statistically solid and clinically invisible: the whole interval sits below the roughly two-point change a patient can reliably detect. This is the textbook case of statistical significance without clinical significance, and it is worth noting that the authors say so themselves rather than leaving a reader to work it out. The electromyographic differences are larger in relative terms but are a mechanistic measurement, not an outcome.

Read the rest in the app

You have read your two free briefings this month. The app carries all 27 specialties, every morning, free — and this finding is waiting in it.

Get Daily Doctor — free

Scan to keep reading on your phone. No account needed to start.