Headaches That Start in the Neck: Cervicogenic Headache Basics

What it feels like

The description that opens this page — steady, non-throbbing, base of the skull — comes from a physiotherapy clinic's write-up of cervicogenic headache, and it's the most useful one-sentence portrait available 1 (Back In Motion, 2022). Note what it rules out as much as what it rules in: throbbing suggests a different headache family; pain centered behind the eyes with nausea suggests another. The cervicogenic pattern is duller, steadier, and anchored at the back of the head where it meets the neck.

Two features do most of the work in telling cervicogenic headache apart from its lookalikes. First, it's usually side-locked: the same side, episode after episode, rather than wandering. Second, the neck is involved — stiffness, reduced range of motion, or pain that worsens when the neck is pressed or moved in certain ways. The International Headache Society's criteria list exactly these as the features that tend to distinguish cervicogenic headache from tension-type headache: side-locked pain, provocation by neck pressure or head movement, and pain radiating from the back of the head forward — while honestly noting these features are "not unique to it" 2 (ICHD-3, 2018).

That "not unique to it" is doing important work, and it's why this page keeps insisting on clinicians. Migraine can also be one-sided. Tension-type headache can also involve the neck. The features point; they don't prove. And here's the complication the criteria openly admit: migrainous features — nausea, light sensitivity — may be present in cervicogenic headache too. So the presence of migraine-like symptoms doesn't rule the neck out, and their absence doesn't rule migraine out. The pattern is a clue, not a verdict — which is exactly why the diagnostic criteria demand the second half: proven causation, not just a matching feeling.

You know the feeling. Next: the table that separates cervicogenic headache from migraine. →

How it differs from migraine

This is the comparison the opener promised, and it pays off the open loop: how to tell cervicogenic headache from migraine. Read it as tendencies with honest overlap, because the sources insist on the overlap. The ICHD-3 criteria state outright that migrainous features may be present in cervicogenic headache — so "I feel nauseous, therefore it's migraine" is not a valid deduction (ICHD-3, 2018).

Cervicogenic headache vs migraine: the comparison table

The row that matters most is the last one. Migraine is diagnosed by its clinical pattern — the constellation of throbbing pain, nausea, sensitivity, aura in some. Cervicogenic headache is diagnosed by a two-part test: a cervical disorder that is known to be capable of causing headache, plus demonstrated causation. That asymmetry is the whole story of this page: migraine is a pattern you recognize; cervicogenic headache is a claim you prove. If your self-diagnosis stopped at "my neck hurts and so does my head," you've done the recognizing half and skipped the proving half.

One more honest complication. Neck pain and stiffness are common during migraine too — the neck gets blamed for a lot of migraines, and migraine gets blamed for a lot of neck-driven headaches. When both the neck and the head hurt, the question isn't "which one is real" — both are. The question is directional: is the neck driving the head pain, or is the head pain referring into the neck? The criteria's causation tests (temporal onset, parallel improvement, provocation, blockade) exist precisely to answer that directional question. Everything else is informed guessing.

Comparison made. Next: how clinicians approach cervicogenic headache — the 2-of-4 test. →

How clinicians approach it

Here's the 2-of-4 test from the meta title, unpacked. The International Headache Society's ICHD-3 criteria (2018) require two things for cervicogenic headache, and both must hold. First: clinical or imaging evidence of a cervical disorder or lesion known to be capable of causing headache — arthritis in the upper neck joints, a disc problem, or similar. Not "my neck feels stiff" — a diagnosed disorder. Second: causation demonstrated by at least two of four findings 2 (ICHD-3, 2018):

1. Temporal onset: the headache began in close time relationship to the onset or worsening of the cervical disorder.
2. Parallel improvement: the headache improves as the cervical disorder improves — treat the neck, watch the head.
3. Provocation: reduced neck range of motion, and pressing on the neck or moving the head in certain ways worsens the headache.
4. Blockade: the headache is abolished by a diagnostic nerve block of the suspected cervical source.

Notice what this demands of the patient story. "My headaches started around when my neck got bad" is criterion one. "When my neck feels better, the headaches ease" is criterion two. "Pressing here makes it worse" is criterion three. Any two clear the bar — but zero plus a stiff neck does not. This is why the page keeps saying most self-diagnoses skip the second half: feeling the neck-headache connection is common; demonstrating it is clinical work.

A physiotherapy clinic's take adds the management side: it describes physiotherapy — manual therapy plus exercise — as the usual first line for cervicogenic headache (Back In Motion, 2022). That's a clinic blog's framing (secondary source), not a guideline — but it matches the criteria's logic: if the neck is driving the headache, treating the neck is the intervention, and the headache's response is itself diagnostic. The "parallel improvement" criterion turns treatment into a test.

Text equivalent: headaches starting at the base of the skull trace toward a neck source; temple starts toward migraine-type patterns; jaw or front-of-ear starts toward the jaw joint. The start point suggests; only clinical testing confirms.

For readers whose headaches arrive with the morning rather than the neck, the companion pages sort the adjacent questions: waking up with a headache: what are the common causes? for the morning pattern, and can your pillow cause morning headaches? for the pillow question specifically. If the jaw is the start point, TMJ pain at night: sleep position and pillow setup is the right next read. And when neck stiffness is the constant companion, why do I wake up with a stiff neck every morning? covers that side.

The clinical bar is set. Next: what the prevalence numbers actually say — including a conflict. →

The numbers: what prevalence actually says

This section exists because honesty about numbers is the whole point of this site. The American Academy of Physical Medicine and Rehabilitation's cervicogenic headache reference gives a one-year prevalence of 0.17–2.2% — meaning in any given year, fewer than 3 in 100 people have it — while also noting it may account for up to 53% of headaches after whiplash 3 (AAPMR). Then the same page claims 15–20% of headaches are cervicogenic. Those figures are in internal tension: a condition with ~1% yearly prevalence doesn't straightforwardly make up a fifth of all headaches.

How do we handle a source that contradicts itself? By saying so, plainly, and not laundering either number into a confident claim. The 0.17–2.2% is labeled as one-year prevalence; the 15–20% is labeled as share of headaches; the whiplash figure (up to 53%) is a specific subpopulation where neck injury makes cervicogenic headache far more likely. They may come from different studies with different methods — or one may simply be wrong. The evidence pack flags the conflict explicitly, and this page carries the flag rather than resolving it by editorial fiat. If a page quotes you "15–20% of headaches" without mentioning the 0.17–2.2%, it's choosing the dramatic number. Now you know both.

What the numbers do agree on: cervicogenic headache is a secondary headache — meaning it's attributed to another condition, not a primary disorder like migraine — and it's uncommon in the general population but common after neck trauma. That "secondary" label is itself informative: it means the diagnostic work always points back at the neck. No neck disorder, no cervicogenic headache — the first half of the 2-of-4 test again.

Numbers weighed, conflict disclosed. Next: the myth this whole page exists to retire. →

Myth: all headaches start in the head

The myth persists because it's half true: most headaches are primary — migraine, tension-type — where the neck is a fellow traveler, not the driver. The error is the "all." The ICHD-3's entire section 11 exists because some headaches are secondary to neck disorders, and cervicogenic headache has the strictest proof requirements of the bunch. The myth says the neck never matters; the criteria say the neck sometimes matters and here's the test. The truth is the test.

There's a mirror-image myth worth retiring in the same breath: that every headache with neck pain is cervicogenic. It isn't — most aren't, per the prevalence numbers above. Neck pain during a headache is common across headache types; what makes it cervicogenic is the diagnosed cervical disorder plus the causation criteria. Both myths fail the same way: they skip the second half of the definition. The feeling is never enough; the proof is the point.

If neck symptoms are your constant companion beyond headaches, the warning-signs checklist is worth a read: neck pain warning signs: when to stop self-managing sorts which neck symptoms need prompt care from which ones don't.

Myths retired. Next: why neck trauma changes the odds. →

Why whiplash makes cervicogenic headache more likely

Remember the prevalence conflict from the last section — 0.17–2.2% in the general population, but up to 53% of headaches after whiplash 3 (AAPMR). That gap isn't a contradiction; it's the criteria working visibly. Cervicogenic headache needs a cervical disorder first, and whiplash is a cervical disorder with a timestamp. When the neck injury is documented, dated, and followed by headache, the first half of the diagnosis is already established — which is exactly why the post-whiplash population shows the highest rates.

Walk through the 2-of-4 test in a whiplash case and you can see why. Temporal onset: the headache began with or shortly after the neck injury — criterion one, nearly automatic. Parallel improvement: as the neck heals, the headaches track it — criterion two, observable over weeks. Provocation: the injured neck is tender and stiff, and pressing or moving it worsens the head pain — criterion three, testable in the exam room. Three of four criteria can be satisfied almost by the nature of the injury, which is why post-whiplash headache gets evaluated for a cervical source as a matter of course.

What this means in practice: if your headaches began after a car accident, a fall, or any neck trauma, say so in the first minute of any headache appointment. "Headache after whiplash" is one of the highest-yield sentences in the cervicogenic workup — it changes what the clinician looks for and which criteria they test first. It doesn't guarantee the diagnosis; plenty of post-whiplash headaches are other types. But it puts the neck at the front of the line, where the criteria can do their work.

And the boundary, as always: this is about evaluation, not self-treatment. A neck injured badly enough to cause ongoing headaches deserves proper assessment — the same red-flag rules from the warning box apply with extra force after trauma. The whiplash link explains why cervicogenic headache is common in this group; it doesn't make it a home diagnosis.

The trauma connection is mapped. Next: the sleep-position question everyone asks. →

Sleep position and the neck-headache link

The question behind "neck headache sleeping position" is reasonable: if the neck drives the headache, does the night's neck position matter? A physiotherapy clinic's cervicogenic headache write-up includes sleep advice — head neutral, on a relatively firm non-feather pillow when side sleeping 1 (Back In Motion, 2022). Read it for what it is: a clinic's practical suggestion for not aggravating the neck overnight, from a 2022 blog post (secondary source), not a trial result. The mechanism is the same one behind every pillow-height recommendation on this site — a neutral neck kinks less, and a less-kinked neck is less provocation for a neck-driven headache.

The boundary matters more than the advice. Nothing about a pillow treats cervicogenic headache — the criteria's causation tests run through diagnosed disorders and clinical interventions, not bedding. What sleep position can do is subtract one nightly aggravation: eight hours with the neck cranked up or sagging down is eight hours of provocation for a structure that's already implicated. Neutral alignment is harm reduction, not therapy. Anyone selling you a pillow as a cervicogenic headache treatment is selling past the evidence — including, to be clear, anyone on this site's commercial side.

Practically: side sleepers need enough loft to fill the shoulder-to-head gap so the neck sits level; back sleepers need less. That's the same alignment check the neck-pain pages teach — lie in your usual position, sight along the neck, level is the target. If morning headaches are the pattern rather than neck-driven ones, the morning-headache pages linked above are the better next read. And "headache from neck pain" as a search is exactly this page's topic in plain words: the head pain is real, the neck is the suspect, and the 2-of-4 test is how the suspicion gets confirmed or cleared.

The full picture is assembled. Below: the short version, then your exact questions answered. →