Pragmatic application of manipulation versus mobilization to the upper segments of the cervical spine plus exercise for treatment of cervicogenic headache: a randomized clinical trial.
Featured on Physle Daily · Thursday, 13 August 2026
Cervicogenic headache comes from problems in the upper neck, and both spinal manipulation (a quick thrust technique) and mobilization (slower, graded movement) are used to treat it, often alongside exercise. This trial randomized 45 patients with cervicogenic headache to receive either manipulation or mobilization of the upper cervical spine, both combined with exercise, and measured neck disability, pain, headache impact, and patient-rated change at baseline, 48 hours, discharge, and one-month follow-up. The statistical interaction between treatment group and time was not significant for neck disability, pain, or headache impact, and there were no significant between-group differences in patient-rated global change or symptom acceptability. This suggests manipulation and mobilization produced similar outcomes when applied pragmatically alongside exercise.
The main limitation is the small sample size of 45 participants, which limits the ability to detect smaller between-group differences and constrains how confidently these findings can be generalized.
If you've dealt with cervicogenic headache, a type of headache that starts from problems in the upper neck, you may know that physiotherapists sometimes use hands-on techniques like manipulation (a quick, targeted movement) or mobilization (slower, gentler movement) to treat the neck, usually paired with exercises. This study looked at whether one approach worked better than the other. Researchers randomly assigned 45 people with cervicogenic headache to receive either manipulation or mobilization of the upper neck, both combined with exercise.
They tracked neck disability, pain, and headache impact over about a month. The results showed no meaningful difference between the two treatments, both groups experienced similar outcomes over time.
This means the study didn't find evidence that one technique outperforms the other for this type of headache when exercise is included. Because the study only included 45 people, it's a fairly small trial, and this limits how much certainty we can place in the findings or how well they might apply to everyone with cervicogenic headache.
This is a summary of research, not medical advice. Talk to your own healthcare provider about anything affecting your care.
This pragmatic RCT compared upper cervical manipulation versus mobilization, both paired with exercise, in 45 patients with cervicogenic headache, using a mixed-model ANOVA with treatment group and time (baseline, 48 hours, discharge, 1-month follow-up) as factors. Primary outcomes were NDI and NPRS, with HIT-6, GRC, and PASS as secondary measures. The treatment-by-time interaction was not statistically significant for NDI (p=0.91), NPRS (p=0.81), or HIT-6 (p=0.89), and no significant between-group differences emerged for GRC or PASS.
These findings indicate no evidence of superiority of manipulation over mobilization within this pragmatic clinical selection framework, rather than evidence that the two techniques are equivalent in effect. The modest sample size of 45 participants limits statistical power to detect smaller between-group differences and constrains confidence in the null result. The pragmatic design, allowing clinicians to select technique based on clinical presentation, enhances applicability to real-world practice but complicates isolating the specific contribution of manipulation versus mobilization technique itself.
Longer-term outcomes beyond one month were not assessed, and adverse events were not reported in the abstract.
In 45 patients with cervicogenic headache, manipulation and mobilization of the upper cervical spine plus exercise produced statistically similar changes in neck disability, pain, and headache impact.
No significant between-group differences were found in patient-rated global change or symptom acceptability at any follow-up point.
The small sample size of 45 participants limits confidence in these null findings and restricts detection of smaller between-group differences.
Source
The Journal of manual & manipulative therapy
Lerner-Lentz A, O'Halloran B, Donaldson M et al. · 2021
doi: 10.1080/10669817.2020.1834322