Manual therapy for tension-type and cervicogenic headache: what the evidence shows

Why this question is worth asking carefully

If you get frequent headaches, someone has probably suggested hands-on treatment — a massage, a "neck adjustment," or a course of physiotherapy. It is a fair thing to try, and a fair thing to be curious about. Does manual therapy actually change how often these headaches come, how bad they are, and how long they last? And does the honest answer differ depending on which kind of headache you have?

This summary gathers the better-quality research — systematic reviews and meta-analyses, which pool many trials rather than resting on a single study — and reports what it found, how confident the researchers were, and where the gaps are. The short version is that there is some evidence of short-term benefit, that its certainty is generally low, that the effect often shrinks in the higher-quality trials and over longer follow-up, and that the picture is not identical for the two headache types. We are curating this evidence, not endorsing a treatment. The aim is to help you and a regulated professional have a better-informed conversation.

The two headaches, briefly

Tension-type headache is the most common primary headache — "primary" meaning the headache is the problem itself, not a symptom of another disease. It typically feels like a dull, pressing, band-like tightness on both sides of the head, mild to moderate, and it usually does not worsen with ordinary activity like walking [1]. It is described by how often it occurs, from infrequent to chronic (15 or more headache days a month) [1].

Cervicogenic headache is different in kind: it is a secondary headache, meaning the pain is referred to the head from a disorder in the neck's joints, muscles, or nerves [2]. That distinction matters for treatment research, because the formal criteria require evidence that the neck disorder is actually causing the headache — not merely that neck wear-and-tear is present, since such changes are common in people without headaches too [2]. Cervicogenic headache is also genuinely hard to diagnose: a systematic review of diagnostic studies concluded that no single history or examination finding reliably confirms it [3]. So when you read "manual therapy for cervicogenic headache," remember that even sorting who has the condition is contested — which quietly limits every efficacy trial built on top of that diagnosis.

What the evidence shows for tension-type headache

For tension-type headache, the pooled evidence is best described as suggestive but thin.

A 2021 systematic review and meta-analysis in The Journal of Headache and Pain examined several non-drug approaches, including manual joint mobilisation [4]. It found manual mobilisation had a potential positive effect on headache frequency (standardised mean difference −1.31) and quality of life (mean difference 0.84) measured at follow-up — but both estimates had confidence intervals that crossed the point of no effect (95% CI −2.70 to 0.07 for frequency; −0.09 to 2.67 for quality of life), meaning the result is imprecise rather than firmly established [4]. The authors graded the certainty of this evidence as very low, downgrading for risk of bias, inconsistency, and imprecision, and framed manual therapy as a weak, supplementary option rather than a proven treatment [4].

A 2023 systematic review in the International Journal of Environmental Research and Public Health pooled 15 randomised trials of physiotherapy approaches for tension-type headache — mobilisation with posture work, myofascial and trigger-point techniques, suboccipital inhibition, and upper-cervical manipulation among them [5]. Several showed reductions in headache intensity and frequency in the short to medium term. But the authors were candid about the limits: the methods varied widely, there is "no consensus on a single physical therapy protocol," only one trial followed patients out to the longer term (26 weeks), and overall certainty was again rated very low [5].

A third review, published in Cureus in 2021, looked at manual therapy and acupuncture together and reported that manual therapy's effect on tension-type headache was, in the authors' description, in the range of preventive medication such as tricyclic antidepressants [6]. That sounds encouraging, and it may be — but the same authors flagged low study quality, a lack of standardised sham comparisons that makes rigorous blinding difficult, and limited long-term data [6]. It is a reason to keep an open mind, not a basis for a strong claim.

Read together, the tension-type evidence points in a modestly positive direction for short-term relief, with consistently low certainty and a near-total absence of good long-term data.

What the evidence shows for cervicogenic headache

The cervicogenic evidence is a little more developed, and it illustrates a recurring pattern worth understanding: the better the trial, the smaller the effect tends to look.

A 2024 network meta-analysis in Physical Therapy pooled 28 randomised trials of physiotherapist-delivered interventions [7]. In the short term (within about 12 weeks), several combinations reduced headache intensity more than control — for example, manipulation plus dry needling (mean difference −4.87, 95% CI −8.51 to −1.24), muscle-energy technique plus exercise (−4.37, 95% CI −8.01 to −0.74), and soft-tissue techniques plus exercise (−3.01, 95% CI −5.1 to −0.92) [7]. Manipulation plus dry needling also reduced headache frequency by about 3 days per week (−3.09, 95% CI −4.93 to −1.25) [7]. The review did rank these combinations, but it rated all of them as low-certainty evidence, reported very high statistical heterogeneity between studies (I² ranging from roughly 78% to 94%), and found that by the longer term the effects were smaller and no longer clinically meaningful [7]. Its own conclusion was that the top-ranked interventions "can be considered," but that "no conclusive recommendation can be made due to the low certainty of evidence" [7]. Those rankings are a snapshot of noisy data, not a verdict on which approach is best.

A 2022 systematic review in Chiropractic & Manual Therapies makes the point about study quality most clearly [8]. Pooling 20 trials (1,439 patients), it found manual and exercise therapy produced large short-term effects on headache intensity and moderate-to-large effects on frequency, shrinking to small effects in the longer term [8]. Crucially, when the authors restricted the analysis to the two trials at low risk of bias — both testing spinal manipulation — the effects became small across intensity, frequency, and disability at both time points [8]. Overall certainty was rated very low in the short term and low in the longer term, with the low-bias subset rated moderate [8]. The honest reading is that the headline "large effect" is partly an artefact of weaker studies.

A 2025 network meta-analysis in Frontiers in Neurology (14 trials, 1,297 patients) compared manipulation, mobilisation, sustained natural apophyseal glides (a mobilisation technique), and massage [9]. It produced favourable rankings for some hands-on techniques on pain and neck disability — but the authors stressed wide variability in protocols, small samples, and possible publication bias, and were explicit that they did not systematically extract adverse-event data [9]. On safety they reported a low incidence of adverse effects when techniques were delivered by experienced clinicians, while noting that cervical manipulation carries a rare but serious risk (such as vertebral artery injury) that gentler mobilisation and massage largely do not [9].

How this is studied, and why certainty stays low

Several features of this literature keep confidence low, and they are worth naming plainly.

• Blinding is genuinely hard. There is no accepted "sham" hands-on treatment that convincingly mimics real manual therapy while doing nothing, and the practitioner always knows what they delivered [6]. That opens a well-recognised door to bias, and it is a structural limit, not a fixable oversight.
• The trials are small and heterogeneous. "Manual therapy" is an umbrella over manipulation, mobilisation, soft-tissue work, dry needling, and their combinations, delivered at different doses to different populations. The cervicogenic network meta-analysis reported heterogeneity high enough (I² up to ~94%) to make pooled averages slippery [7].
• Higher-quality trials show smaller effects. The clearest single lesson from the cervicogenic data is that restricting to low-risk-of-bias studies shrank the benefit to "small" [8]. Impressive numbers from weaker trials should be read with that in mind.
• Durability is largely unknown. Most benefit is measured in the short term; where longer follow-up exists, effects fade toward the not-clinically-meaningful [5][7]. Very few trials look past a few months at all.
• Safety data are under-reported. Even the reviews that comment on safety note that adverse events were not consistently captured, so the harm side of the ledger is thinner than the benefit side [9].

What guidelines and reviews conclude

The syntheses that translate this evidence for clinicians land in a consistent, measured place: manual therapy is a reasonable option, usually best combined with exercise, and not a stand-alone answer.

A 2026 clinical practice guideline on chiropractic management of these headaches — developed by a panel including chiropractors and neurologists — recommended spinal manipulation as an option for cervicogenic headache, and for tension-type headache only within multimodal (combined) care rather than on its own [10]. The same guideline emphasised careful history-taking, red-flag screening, and appropriate referral, and acknowledged that limitations in the underlying evidence precluded strong recommendations for several modalities used in isolation [10]. Across the reviews above, the through-line is the same: short-term benefit is plausible, certainty is low, combined and active care is favoured over passive treatment alone, and no single technique is established as superior [4][7][8][10].

What this means for you

If you are weighing hands-on treatment for frequent headaches, here is a fair summary. For both tension-type and cervicogenic headache, manual therapy shows some short-term benefit in the pooled research; that evidence is low-certainty; the effect tends to be smaller in the best-designed trials and to fade over time; serious harms appear rare — though under-studied — with gentler techniques carrying less risk than cervical manipulation. None of that is a promise, and none of it can tell you what will happen for your headaches.

The most useful next step is an assessment by a regulated health professional — a physician, physiotherapist, chiropractor, or registered massage therapist — who can take your history, examine you, distinguish tension-type from cervicogenic (or something else), rule out the uncommon serious causes that need different care, and help you weigh manual therapy alongside options like exercise. They can also tell you which specific technique they would use and its likely benefits and risks for you. Those are questions the research, on its own, cannot answer.