Secondary claim guide

Nexus Letter for Headaches Secondary to a Service-Connected Neck Condition

Medically reviewed by the Patriot Path Medical Team

Licensed MD reviewers • Last updated: November 7, 2026

Patriot Path Medical Team → • Our review process →

✓ Medically Reviewed

VA can service connect headaches secondary to a service-connected neck condition under 38 CFR 3.310 when medical evidence links them, and the Board has granted it under several headache diagnoses. Check your rating decision first: VA may already count the headaches. If the neck is not service connected yet, claim both now. A grant can pay 0 percent.

Four more things belong up front. VA must give you an examination or opinion when your record lacks enough medical evidence to decide the claim but shows a current disability and an indication that it may be associated with a service-connected disability (38 CFR 3.159(c)(4)), and in the 11 Board grants we read (found by searching, so not a rate), VA's own evidence or a VA favorable finding decided the headache link in 7. A new rating can leave your combined rating where it is. Doctors disagree about how to diagnose a headache that comes from the neck, and a neck MRI cannot settle it. And each Board decision cited here is "binding only with regard to the specific case decided" under 38 CFR 20.1303.

An illustrative example

A veteran has been service connected for a neck condition at 20 percent since 2016, and that is his only rating. His headaches start at the base of his skull two or three times a month, and someone has told him he needs a nexus letter. Before buying one, he reads his rating decision and finds his neck described as degenerative disc disease with headaches. The decision assigns no separate rating for the headaches.

On those facts the link may already be on his record, as it was in 18150820 and 1442004, and the open question is the rating. He files a claim for an increase and keeps a headache journal. If his records show characteristic prostrating attacks averaging once a month over the last several months, the 30 percent level of DC 8100 fits by analogy, and with the neck at 20 percent Table I of 38 CFR 4.25 gives 44, which rounds to 40. If the attacks are not prostrating, the headaches stay at 0 percent and his combined rating stays at 20.

We put this example together from the rating rules and two Board decisions to show the order of the checks. It is not a client and not a case result, and it predicts nothing for any claim.

Start Here: VA May Already Count Your Headaches

Before anyone writes about the link, read what VA has already decided. A veteran can be service connected for headaches without knowing it, when VA has folded the headaches into the neck grant instead of listing them as a separate rating.

What one 1985 rating decision did

In 18150820 (2018) the Board found that a June 1985 rating decision had granted service connection for "residuals, injury, cervical spine with associated headaches," and that it said: "Headaches are considered part and parcel of the spinal injury and will be included in the veteran's diagnosis." The Board held that "VA expressly service connected the headaches, albeit without an explicit separate noncompensable rating", effective October 12, 1984.

Then VA lost track of its own grant. When the veteran claimed "compensation benefits for headaches" in 2014, VA, in the Board's words, "incorrectly construed this statement as a service connection claim instead of as an increased rating claim", and denied it. The Board dismissed the service-connection appeal, because there was nothing left to grant, and ordered VA's records updated to show the 1984 date, saying the correction "confers no new benefit". It sent back for a VA examination the question VA should have decided: "Entitlement to a compensable rating for a headache disorder" (non-precedential, 38 CFR 20.1303).

The wording to look for varies

Do not search only for the phrase "associated headaches". In 1442004 (2014) the neck disability on appeal had been identified as "degenerative disc disease, C3-4 and C4-5 with headaches." Inside that neck appeal the Board found the headaches to be "a separate neurological manifestation of the service-connected cervical spine disorder" and ordered: "A separate rating for headaches as secondary to a cervical spine disorder is granted" (non-precedential, 38 CFR 20.1303).

So read every description of a service-connected disability on your codesheet and in the rating narrative for any headache wording, whatever form it takes. Then read the decision notice.

The list of favorable findings

Every decision notice issued since February 19, 2019 must include "A listing of any findings made by the adjudicator that are favorable to the claimant under § 3.104(c)" (38 CFR 3.103(f)(4)). A favorable finding "is binding on all subsequent agency of original jurisdiction and Board of Veterans' Appeals adjudicators, unless rebutted by evidence that identifies a clear and unmistakable error in the favorable finding" (38 CFR 3.104(c)).

That rule carried the headaches in A21018795 (2021). The Board wrote: "In the October 2019 rating decision, the AOJ made a favorable finding that the Veteran's headaches are secondary to his neck condition." The neck itself was not service connected at the time. Once the Board granted the neck, on a private opinion about the neck, it granted the headaches "Based on this favorable finding", with no separate headache opinion discussed (non-precedential, 38 CFR 20.1303).

If VA already counts them, you are not done

If your headaches are already service connected, on their own or as part of the neck, you do not need a nexus letter for the link. You can still ask VA to rate them separately through a claim for an increase (38 CFR 3.1(p)(1)(ii)), the question the Board sent back in 18150820 and the separate rating it granted in 1442004. If the decision that folded them into the neck was issued less than a year ago, you can instead ask for a review of it: a supplemental claim with new and relevant evidence, a higher-level review, or a Board appeal. An issue pursued that way without a break can keep the date of the claim that decision decided (38 CFR 3.2500(h)(1)), where a claim for an increase can reach back up to a year before the claim (38 CFR 3.400(o)(2)).

Name the earlier decision by date and quote its wording in your claim, because VA has misread its own grant before. Bring severity evidence too, because a separate rating pays only when the attacks meet the rating criteria (see what a headache rating can pay, below). That evidence comes from your records, a headache journal and the people who see the attacks, not from a nexus letter.

If Your Neck Is Not Service Connected Yet, Claim Both Now

A headache claimed as secondary to the neck needs a service-connected neck. That is a reason to claim the neck now, not a reason to wait for it.

Why the neck has to be part of the claim

In 18160361 (2018) VA treating physicians had attributed the veteran's headaches to his neck, one recording an impression of chronic headaches thought to be cervicogenic. The Board still declined to reopen the headache claim for lack of new and material evidence, holding that those opinions did not raise a reasonable possibility of substantiating it because the neck was not service connected and "service connection cannot be granted secondary to a nonservice-connected disability", citing Allen v. Brown, 7 Vet. App. 439 (1995) (en banc), and 38 CFR 3.310(a). The neck's own request to reopen failed in the same decision. That was a failed request to reopen a finally denied claim, not a ruling on the medical link (non-precedential, 38 CFR 20.1303).

Why waiting costs money

VA's manual dates a secondary grant from the claim it was awarded on: "The effective date of secondary SC must be based on the date that the application upon which SC was eventually awarded was filed." It adds: "The effective date assigned for a secondary SC disability cannot be earlier than the effective date assigned for the causal or primary SC disability" (M21-1 Part V, Subpart ii, Chapter 4, Section A, Blocks 3.f and 3.g). A headache claim filed a year after the neck claim cannot reach back to the neck's date on that rule.

If you are not ready to file a complete claim, file an intent to file now. Under 38 CFR 3.155(b), if VA receives a complete application "within 1 year of receipt of the intent to file a claim, VA will consider the complete claim filed as of the date the intent to file a claim was received."

The Board has granted both in one decision

Five decisions we read granted the neck and the headaches in the same decision: A21018795 (2021), 21072900 (2021), 18160149 (2018), 18157013 (2018) and 18146722 (2018). In 21072900 the Board found the requirements met "because the Veteran's cervical spine condition is now service-connected as discussed above". Each is non-precedential under 38 CFR 20.1303.

Name both conditions on the claim. VA's manual says a pending claim can reach an unclaimed condition that caused the claimed one, if that condition is shown to be related to service ("the pending claim reasonably encompasses a claim for benefits for the causal disability", M21-1 Part V, Subpart ii, Chapter 3, Section A, Block 2.d), but naming both removes the question.

If a neck claim or appeal is already pending

Raise the headaches in it. VA's manual dates a disability found within the scope of a claim from that claim: the effective date "is governed by the date of receipt of the expressly claimed issue which resulted in sympathetic reading and/or finding a disability within scope" (M21-1 Part V, Subpart ii, Chapter 4, Section A, Block 1.g). It also says "A specific claim is not required to award a within-scope complication/residual" and lists "neurological disabilities related to the spine" among the examples (Part V, Subpart ii, Chapter 3, Section A, Block 2.c). It does not say whether a headache is one.

The Board has taken headaches up in pending appeals: in 1442004 it granted a separate headache rating inside an appeal for a higher neck rating, and in A25007432 it recorded that a 2022 Board decision, in a legacy appeal whose subject the 2025 decision does not state, had found a separate headache rating, secondary to the cervical spine, "reasonably raised", after which VA granted service connection for the headaches (both non-precedential, 38 CFR 20.1303).

If your headaches began in service

Say so, with the dates. Direct service connection does not depend on the neck at all. In 18146722 (2018) the Board found as fact that "The Veteran's headaches are related to service" and granted the neck and the headaches on a direct basis. In 18157013 (2018) service records showed that the veteran "reported neck pain and headaches following these MVAs", but the Board granted the headaches on a secondary basis, "as either secondary to the Veteran's service connected hepatitis C or as secondary to her neck disability" (both non-precedential, 38 CFR 20.1303).

The timeline cuts both ways, so give it as it happened. In 1314955 (2013) a VA examiner found the veteran's neck pains "could not have caused his headaches as the symptoms of both began at the same time", and the Board denied the claim (non-precedential, 38 CFR 20.1303). An opinion has to answer the theory your actual history supports, and no one should shape the history to fit a theory.

Inside the Neck Rating or Secondary to It: Two Routes With Different Dates

VA could treat a neck-linked headache as part of the neck disability or as a separate secondary condition. The two routes carry different effective-date rules, and nothing we found says which one VA uses for headaches. Ask VA to consider both.

Two-column diagram with no people, titled two routes for a neck-linked headache, and what each one does to the date. The two columns are drawn alike. The left column is headed inside the neck rating, 38 CFR 4.71a, Note (1), and reads: rated separately if it qualifies, and VA's manual treats new spinal complications as a claim for increase; written for objective neurological abnormalities, and it does not name headaches; effective date under 38 CFR 3.400(o), which can reach back up to a year if an increase was factually ascertainable, for a neck already service connected. The right column is headed secondary service connection, 38 CFR 3.310, and reads: rated separately where the symptoms do not overlap the neck rating (Esteban; 38 CFR 4.14); effective date based on the claim it was awarded on, and not earlier than the neck's. A bar across the bottom reads: no rule we found says which route VA uses for headaches, so ask VA to consider both. If the neck is not service connected or claimed yet, claim the neck and the headaches now. — The two routes, the date rule each one carries, and what is not settled.

Route one: a complication counted with the neck

The spine formula in 38 CFR 4.71a rates the neck "With or without symptoms such as pain (whether or not it radiates), stiffness, or aching in the area of the spine affected by residuals of injury or disease". Its Note (1) adds: "Evaluate any associated objective neurologic abnormalities, including, but not limited to, bowel or bladder impairment, separately, under an appropriate diagnostic code."

VA's manual treats those abnormalities as a worsening of the spine rather than a new disability: "a claim asserting new complications of spinal disease is considered a claim for increase rather than a claim for secondary SC" (M21-1 Part V, Subpart iii, Chapter 1, Section B, Block 3.d). For the date, it says objective neurological abnormalities "establish a worsening of the spinal disability" and that 38 CFR 3.400(o) applies (Part V, Subpart ii, Chapter 4, Section A, Block 5.e, Example 2). Under 3.400(o)(2) the date can be the "Earliest date as of which it is factually ascertainable based on all evidence of record that an increase in disability had occurred if a complete claim or intent to file a claim is received within 1 year from such date". That can reach back up to a year before the claim, for a neck that is already service connected. Where the neck and a neurological complication are granted on one claim, the manual's example dates both from that claim (M21-1 Part V, Subpart iii, Chapter 1, Section B, Block 3.e, Example 2). It is a possibility, not a promise.

Route two: secondary service connection

38 CFR 3.310(a) service connects a disability "which is proximately due to or the result of a service-connected disease or injury". VA's manual says "A secondarily SC disability cannot be treated as an increased evaluation under 38 CFR 3.400(o) for effective date purposes", and the date follows the claim the grant was awarded on (same section, Block 3.f). The general rule in 38 CFR 3.400 is "the date of receipt of the claim or the date entitlement arose, whichever is later."

The Board has used both framings for the same kind of headache. It decided neck-linked headaches as secondary claims in 1507123, 1500511, 21067969 and 21072900, and it granted a separate headache rating under Note (1) inside a neck appeal in 1442004, whose order reads "as secondary to a cervical spine disorder" (all non-precedential, 38 CFR 20.1303). The word secondary on a decision does not settle the route or the date: VA's manual says its rating system requires that complications rated separately from a service-connected disability "be coded as secondary", and that this convention "does not impact assignment of an effective date" (M21-1 Part V, Subpart ii, Chapter 4, Section A, Block 5.e).

Why neither route is settled for headaches

The word headache does not appear in VA's manual section on the spine, and in 38 CFR 4.71a it appears once, in the fibromyalgia code. Nothing we read says whether a headache is one of the "associated objective neurologic abnormalities" in Note (1). One signal points away from it: a note under the codes for organic diseases of the central nervous system (DCs 8000 to 8025) in 38 CFR 4.124a describes "residuals not capable of objective verification, i.e., headaches, dizziness, fatigability". That note governs a different set of codes and is not a rule about the neck.

So plead both. Ask VA in writing to consider a separate rating for the headaches as part of the neck disability, secondary service connection under 38 CFR 3.310, and direct service connection if the headaches began in service. We have not verified how VA's application form handles more than one theory; whatever form you use, name each theory in writing.

If the neck made an existing headache disorder worse

If you had migraines or another headache disorder before the neck made it worse, the route for that worsening is aggravation under 38 CFR 3.310(b), and the regulation sets a gate: VA will not concede aggravation "unless the baseline level of severity of the nonservice-connected disease or injury is established by medical evidence created before the onset of aggravation or by the earliest medical evidence created at any time between the onset of aggravation and the receipt of medical evidence establishing the current level of severity of the nonservice-connected disease or injury."

In 23005147 (2023) an examiner who agreed the neck could "serve as triggers" for the veteran's migraines, yet found no aggravation without an adequate explanation, gave an inadequate opinion (non-precedential, 38 CFR 20.1303).

One deduction is gone, and the baseline deduction remains

3.310(b) still recites a deduction for natural progress. VA's manual, following Spicer v. McDonough, 61 F.4th 1360 (Fed. Cir. 2023), says VA "will apply the broader but-for standard and will no longer consider natural progress of the NSC disability when deciding a claim based on aggravation of an NSC disability by an SC disability" (M21-1 Part V, Subpart ii, Chapter 2, Section D, Block 1.a, in a note). That is VA's stated practice in its manual, not new regulation text. The same block applies the but-for test to secondary service connection as well, for disabilities that "would not have occurred but for" a service-connected one, and to aggravation, where a non-service-connected disability "would have been less severe but for the SC disability, including where the SC disability has interfered with or impeded treatment for the NSC disability".

The baseline deduction survives: "The evaluation assigned will reflect the difference between the current and baseline levels of severity" (Block 1.g). The baseline can rest on medical evidence or on "any other credible evidence, including lay evidence, that supports establishment of the baseline", though lay evidence on its own may not be enough (Block 1.c). And the manual cuts against the veteran here: "Do not assume a baseline of 0 percent when a baseline cannot be established by the available evidence" (Block 1.f). The same block says an aggravation claim should be denied for lack of a baseline only after VA's duty to assist is fulfilled, and that the duty includes developing evidence of the baseline, which may include an examination or medical opinion when warranted.

If VA service connected a TBI

Check the TBI rating before claiming anything here. VA's manual evaluates subjective complaints of headaches "as part of the TBI evaluation under 38 CFR 4.124a, DC 8045 rather than under a separate DC", and rates a distinct headache disorder separately "as long as the manifestations do not overlap with those used to assign the evaluation of TBI" (M21-1 Part V, Subpart iii, Chapter 12, Section B, Block 2.d). The same headache cannot be paid twice (38 CFR 4.14). That manual rule is written for TBI, not for the neck.

For the TBI side of the question, see our guide to headaches secondary to TBI.

Can a Neck Condition Cause Headaches? What the Medicine Supports

The medical question in this claim is about now: whether your current neck disorder is the source of your current headache, or makes it worse. Doctors recognize headache that comes from the neck, they disagree about how to diagnose it, and the research does not answer the question for any one veteran. Each finding below is labelled for what it measured and in whom.

A recognized diagnosis with written criteria

The International Classification of Headache Disorders (ICHD-3, entry 11.2.1) recognizes cervicogenic headache as a headache caused by a disorder of the cervical spine, its bone, discs or soft tissues. Paraphrasing its criteria: there must be clinical or imaging evidence of a disorder or lesion in the cervical spine or neck soft tissues of a kind known to be able to cause headache; at least two of four signs of causation (the headache began around the time the neck disorder began; it significantly improved or resolved as the neck disorder improved or resolved; neck motion is reduced and provocative maneuvers make the headache significantly worse; or a diagnostic block of a neck structure or its nerve supply abolishes it); and no other ICHD-3 diagnosis that accounts for it better. ICHD-3 adds that cervical spondylosis may or may not qualify, depending on the individual case.

Doctors disagree about how to diagnose it

A 2009 review in The Lancet Neurology (Bogduk and Govind) noted that some clinicians remain skeptical of the diagnosis, that the clinical diagnostic criteria it assessed had not been shown to be valid, and that a neck source can be established with fluoroscopically guided, controlled diagnostic nerve blocks (PMID 19747657). In a 1999 study in which physicians examined 24 headache patients in person, agreement on the diagnosis was a kappa of 0.83 between two expert headache neurologists but 0.43 to 0.62 among other physicians; the authors concluded that, with Sjaastad's criteria applied strictly, cervicogenic headache was diagnosed about as reliably as migraine and tension-type headache under the International Headache Society's criteria (PMID 10595292). In a 2022 review of 11 diagnostic studies, one bedside test, the cervical flexion-rotation test, had a pooled sensitivity and specificity of 83.0 percent each across 4 studies and 182 participants in telling cervicogenic headache apart from migraine, other headaches or no headache, which shifts the probability modestly; the review supported taking a headache history and examining the neck to make the diagnosis, and judged the value of findings other than this test unclear (PMID 36088782). A 2021 meta-analysis found less rotation on that test (by 17.67 degrees) and weaker neck flexion in cervicogenic headache than in migraine, each pooled from 2 studies (75 and 80 patients), and its authors called for more high-quality studies and wrote that physical findings alone cannot settle the diagnosis between cervicogenic headache and migraine (PMID 34479514). ICHD-3 itself says the features that tend to set cervicogenic headache apart are not unique to it and do not necessarily establish causation.

Controlled nerve blocks are the reference test

In a 2022 study of 166 consecutive patients with probable cervicogenic headache, controlled diagnostic blocks found the painful joint in 75 percent of those whose main complaint was headache (the C2-3 joint in 62 percent) and in 67 percent of those whose neck pain was worse than their headache (PMID 33484154). In 100 consecutive patients with neck pain for more than three months after whiplash, double-blind controlled blocks of the third occipital nerve confirmed that nerve's headache in 27 percent (95 percent CI 18 to 36), and no feature of the history or examination settled the diagnosis before the blocks, although patients who had it were more likely to be tender over the C2-3 joint (PMID 7931379). Neither study was of veterans.

One block, or an occipital block, does not settle it

Against controlled blocks, a single uncontrolled neck-joint block gave a false-positive result for 16 of 60 joints that had responded to a first block, 27 percent (95 percent CI 15 to 38), in 55 patients with neck pain for more than three months after a motor vehicle accident (PMID 8358135). The same research group recommended placebo-controlled blocks where diagnostic certainty is critical, naming a medicolegal context (PMID 8535040). Greater occipital nerve injections also helped after 26 of 57 injections in migraine and 13 of 22 in cluster headache in a London clinic audit (PMID 16527404), so relief after an occipital nerve block does not by itself place the source in the neck.

An MRI cannot settle it

Cervical MRI showed disc bulging in 87.6 percent of 1,211 volunteers with no symptoms (PMID 25584950). In a small comparison, disc bulging appeared in 10 of 22 patients with cervicogenic headache (45.4 percent) and 9 of 20 controls (45.0 percent) (PMID 14510932). ICHD-3 treats upper-cervical imaging findings as a hint of causation, not firm proof, and its chapter introduction says degenerative neck changes appear in nearly everyone over 40 and are as widespread in people without headache as in people with it.

Neck pain comes with other headaches too

Pooled clinic studies found neck pain in 77.0 percent of people with migraine (95 percent CI 69.0 to 86.4) against 23.2 percent of people without headache (PMID 35166137). In a population sample of 797 people studied by researchers at the Danish Headache Center, past-year neck pain was reported by 76.2 percent with migraine alone and 88.4 percent with tension-type headache alone, against 56.7 percent without headache (PMID 24853166). ICHD-3 leans toward classing headache from neck muscle pain as tension-type headache.

After a neck injury, the long-term evidence runs both ways

In Lithuania, 210 people in police-recorded rear-end collisions, 28 percent of whom reported neck pain at first (lasting no more than 17 days), and 210 matched people without a collision showed no significant difference in neck pain or headache after one year (PMID 10084524). In another Lithuanian group of 202 people in police-recorded rear-end collisions, 53 percent reported headache 1 to 3 years later against 50 percent of controls, and 10 of the 12 with daily headache had it before the crash (PMID 8622449). A Swedish insurance cohort found more headache and other complaints 7 years later in 232 drivers who reported a whiplash injury than in 3,688 unexposed people (relative risks 1.6 to 3.7 across complaints), and no increase in exposed drivers without a reported injury (PMID 11470396). A Japanese study that re-examined 75 of its original 508 whiplash patients and 181 of its 497 comparison volunteers 20 years later found headache in 24.0 percent against 12.2 percent (odds ratio 2.39), while neck MRI findings were similar in both groups and did not track symptoms (PMID 33394988). None of these diagnosed cervicogenic headache by ICHD criteria or by blocks, and none studied service members.

Military figures are context, not causation

Among 985 service members evacuated from Iraq and Afghanistan for headache between 2004 and 2009, 33.6 percent returned to duty overall, against 29.7 percent of those diagnosed with cervicogenic headache and 23.1 percent with occipital neuralgia (PMID 21994113). Among 38,426 Iraq and Afghanistan veterans who first received VA care in 2008 and received care in each year from 2009 to 2011, a 2008 headache diagnosis went with a coded neck-pain diagnosis (adjusted odds ratio 2.44, 95 percent CI 2.14 to 2.77) and more strongly with TBI plus PTSD (10.16) (PMID 26688427). We found no study that followed a neck injury sustained in military service to a headache that began years later. Neither study above shows that a neck injury caused a later headache.

What this means for a claim

Put together: a neck disorder can be the source of a headache, a cervical source can be shown in selected patients with controlled blocks, and examination findings can help separate it from migraine. The research does not show that headache plus neck pain plus an imaging finding means the neck is the source, and it gives no rate for veterans.

That makes the question a fair one for a physician to weigh in one veteran's file, against the ICHD-3 criteria and the record, and not one that the science settles in general. An opinion that says the science settles it overstates the science, and a VA examiner can say so.

Occipital neuralgia is a different diagnosis

ICHD-3 lists occipital neuralgia separately (entry 13.4): shooting or stabbing pain at the back of the scalp in the territory of the occipital nerves, which, under its criteria, a local anesthetic block of the nerve relieves for a time, and which has to be told apart from pain referred from the upper neck joints. VA's rating schedule has no entry that names the occipital nerve. If your records name occipital neuralgia, the analysis on this page does not carry over directly; ask the clinician who made the diagnosis how it was confirmed.

It Does Not Have to Be Called Cervicogenic, and Every Symptom Counts

The Board has granted neck-linked headaches under several labels. Nothing in the rating rules asks for one name, and nothing on this page should push you toward a label or away from a symptom.

Headache labels in Board orders on neck-linked headaches (non-precedential, 38 CFR 20.1303)

Headache labels in Board orders on neck-linked headaches (non-precedential, 38 CFR 20.1303)
Decision	Label in the Board's words
1500511 (2015)	chronic headaches
1507123 (2015)	occipital headaches secondary to cervical spine DDD
1524862 (2015)	secondary service connection for migraine headaches
18160149 (2018)	tension headaches, including as secondary to a neck disorder
A21018795 (2021)	cervicogenic headaches secondary to degenerative arthritis, cervical spine
21072900 (2021)	headaches, as secondary to a cervical spine condition
21067969 (2021)	headaches as secondary to a service-connected cervical spine disability

Each label is quoted from the decision's order granting service connection.

A migraine diagnosis did not stop a neck-secondary grant

In 1524862 (2015) the Board found as fact: "The Veteran's migraine headaches are the result of his service-connected cervical degenerative disc disease." A private physician's 2011 note said the neck pain was worst at the base of the skull and "sometimes associated with headaches", adding "I suspect that this comes from the C3-4 spondylosis", and VA treatment notes later tied the headaches to the neck pain (non-precedential, 38 CFR 20.1303).

A split diagnosis can connect one part and not another

In 1507123 (2015) the VA examiner "diagnosed migraine and tension headaches." Three VA opinions tied the veteran's occipital headaches to his cervical spine, and one found that "the headaches/pressure behind the eyes and in his temples, the photo/phonophobia, and visual disturbance is part of the migraine syndrome and is not related to his cervical spine DDD." The Board granted "occipital headaches secondary to cervical spine DDD" and not the migraine component (non-precedential, 38 CFR 20.1303).

When only one component is connected and the medical evidence separates its symptoms, the rating counts only that component: 38 CFR 4.14 bars "the use of manifestations not resulting from service-connected disease or injury in establishing the service-connected evaluation". If the symptoms cannot be separated, VA's manual says "all symptoms should be deemed related to the SC disability and rated accordingly", though the other condition does not become service connected, and it treats whether they can be separated as a medical determination (M21-1 Part V, Subpart ii, Chapter 3, Section D, Block 2.c). In 0919523 (2009), after a grant for cervicogenic headaches, later diagnoses of tension and migraine headaches led the Board to ask whether they were "separate and distinct entities unrelated to the cervicogenic headaches" (non-precedential, 38 CFR 20.1303).

Describe every symptom as it happens

Tell every clinician and examiner about every symptom as it happens, including nausea and sensitivity to light or sound. ICHD-3 notes that migraine-like features such as nausea, vomiting and light or sound sensitivity can occur with cervicogenic headache, generally to a lesser degree than in migraine. For migraine, VA's manual says DC 8100 requires VA to "consider all symptoms experienced as a result of migraine attacks" (M21-1 Part V, Subpart iii, Chapter 12, Section A, Block 3.a, citing Holmes v. Wilkie).

Neither the ICHD-3 features nor the rating levels are a script. A headache described as it is, and diagnosed by the clinician who examined you, is what an honest opinion can work from.

If the injury that hurt your neck also started the headaches

ICHD-3 classifies headache caused by neck trauma, including headache after whiplash, in its chapter on trauma rather than under cervicogenic headache, even though it grants that headache after whiplash may well come from pathology in the neck. Every type in that chapter requires the headache to begin within 7 days of the injury, of regaining consciousness, or of regaining the ability to sense and report pain; ICHD-3 itself calls the 7-day interval somewhat arbitrary but keeps it for lack of evidence to change it. Where the neck condition comes from an injury in service, the physician has to say which diagnosis fits and why. A headache that began with the injury in service points to direct service connection; one that began later, with a current neck disorder as its source, is the secondary question this page covers.

What a Headache Rating Can Pay, and When It Pays Nothing

The rating schedule has no code for cervicogenic headache. VA rates a neck-linked migraine under the migraine code, DC 8100, and other neck-linked headaches by analogy to that code; either way, the percentage turns on prostrating attacks, not on the link.

Rated by analogy to DC 8100

In 1525697 (2015) the Board wrote: "There is no diagnostic code corresponding to cervicogenic headache disability in VA's rating schedule." The regional office had "rated the Veteran's cervicogenic headaches by analogy under Diagnostic Code 8100", and the Board raised them from 30 to 50 percent. In 18155577 (2018) the headaches "have been rated in accordance with DC 8199-8100", and the Board granted 30 percent on the veteran's own testimony (both non-precedential, 38 CFR 20.1303).

The code number follows 38 CFR 4.27, under which for a condition the schedule does not list "the last 2 digits will be "99" for all unlisted conditions", so it reads 8199-8100. 38 CFR 4.20 allows the analogy and adds a caution: "Conjectural analogies will be avoided, as will the use of analogous ratings for conditions of doubtful diagnosis, or for those not fully supported by clinical and laboratory findings." That caution concerns the diagnosis in the individual case, which is one more reason documented findings matter. It did not stop the Board from rating neck-linked headaches by analogy in the decisions above. Our guide to headaches after a TBI walks through the same analogy in its section on headaches that are not migraines.

The levels, and the one that pays nothing

DC 8100 has four levels. The paying ones are 10 percent for characteristic prostrating attacks averaging one in 2 months, 30 percent for attacks averaging once a month, and 50 percent for very frequent completely prostrating and prolonged attacks productive of severe economic inadaptability, each measured over the last several months. The bottom row reads "With less frequent attacks" and pays 0 percent. The full table is on our migraines page, linked below.

VA's manual defines prostrating as "causing extreme exhaustion, powerlessness, debilitation or incapacitation with substantial inability to engage in ordinary activities" (M21-1 Part V, Subpart iii, Chapter 12, Section A, Block 3.b). The regulation does not define the 0 percent level; the manual reads the structure of the code to mean "Characteristic prostrating attacks, on average, are more than two months apart over the last several months" (Block 3.f).

Winning the link can pay nothing

In A25007432 (2025) VA granted service connection for cervicogenic headaches in October 2022 and "assigned a noncompensable disability rating, effective September 22, 2021, under diagnostic code 8199-8100." The examiner had described headaches about twice a week that came with neck pain, resolved in an hour or two and did not make the veteran leave work. The Board sent the rating back because the examiner had assumed each headache would be treated with medication, and because chiropractic records had not been requested (non-precedential, 38 CFR 20.1303). In 1801003 (2018) another veteran's headaches sat at 0 percent for about nine years before a later examination supported 50 percent; that decision does not state the basis of the grant (non-precedential, 38 CFR 20.1303).

A grant at 0 percent still puts the headaches on the record as service connected, and a favorable finding binds later decision makers unless clear and unmistakable error is shown (38 CFR 3.104(c)). It can matter in other ways, and we do not count any of them as a reason to buy a letter. The law directs VA to furnish the hospital care and medical services it finds needed for a service-connected disability, with no percentage threshold (38 U.S.C. 1710(a)(1)(A)), and hospital and outpatient care for a 0 percent service-connected disability is not subject to the copayments in 38 CFR 17.108 (paragraph (e)(1)). VA also charges no copayment for medication for a service-connected disability, at any percentage (38 CFR 17.110(c)(2)). If the headaches later worsen, a claim for an increase can be dated up to a year before the claim when the increase is factually ascertainable (38 CFR 3.400(o)(2)), where a new secondary claim is dated from the claim. A condition the headaches cause can itself be service connected (38 CFR 3.310(a)). And if none of your service-connected disabilities is rated above 0 percent, a single 10 percent rating may be assigned under 38 CFR 3.324 when two or more separate permanent ones clearly interfere with normal employability.

What proves the frequency

The percentage turns on how frequent and how prostrating the attacks are, and that evidence comes from you and the people around you. VA's manual, in guidance written for migraine under DC 8100, lists probative evidence of frequency: "medical progress notes", lay evidence of the symptoms "(as long as those symptoms have been competently identified as symptoms of migraine headaches)", "contemporaneous notes (a headache journal)", prescription refills and witness statements. It adds: "The absence of treatment reports is not necessarily probative on the question of headache frequency as a claimant may not seek treatment for headaches during every episode" (Block 3.g).

Journals "which routinely and relatively contemporaneously record headache episodes" may be accepted as credible lay testimony (Block 3.h), and "Headaches recorded on non-work days may be used to prove frequency and prostration. However, they will not generally be relevant to work availability, and performance or limitations, which are considerations in determining severe economic inadaptability." The manual also says "medical evidence is required to establish that the reported symptoms are due to the SC migraine headaches" (Block 3.c). Record every headache, the mild ones too, at the time it happens.

VA's examination guidance treats a headache examination report as insufficient "if it does not indicate the frequency of prostrating headaches and whether the headaches are migraine-type or non-migraine type" (M21-1 Part IV, Subpart i, Chapter 3, Section B, Block 1.b). In 18145348 (2018) an employer's report carried limited weight because it did not show a history of several months of prostrating attacks and "does not differentiate the Veteran's headache pain from his neck pain" (non-precedential, 38 CFR 20.1303). A nexus letter supplies none of this. It addresses the link, not the frequency of prostrating attacks.

Beside the neck rating, but not counted twice

A headache rating sits beside the neck rating only for what the neck rating does not already count. 38 CFR 4.14 says "the evaluation of the same manifestation under different diagnoses are to be avoided." VA's manual states the test in a note: "Separate evaluations are warranted where none of the symptomatology for any one condition duplicates or overlaps another" (M21-1 Part V, Subpart ii, Chapter 3, Section D, Block 2.b, citing Esteban v. Brown). Neck pain belongs to the neck rating; attacks of head pain rated on DC 8100's criteria are measured on different things. In the decisions we read, headache ratings stood beside neck ratings in 1442004, 1801003, 18145348 and 18155577.

The disc-disease exception is used only when it gives the higher combined rating

If your neck is rated for intervertebral disc syndrome on incapacitating episodes, VA's manual says "a separate evaluation may not be assigned for LOM, radiculopathy, or any other associated objective neurological abnormality as doing so would constitute pyramiding under 38 CFR 4.14" (M21-1 Part V, Subpart iii, Chapter 1, Section B, Block 3.a). Whether that reaches a headache is unverified: neither the regulation nor the manual names headaches. And VA uses that method only when it gives more: disc syndrome is rated under the general formula or the incapacitating-episode formula, "whichever method results in the higher evaluation when all disabilities are combined under § 4.25" (38 CFR 4.71a).

How a headache rating combines with the neck

Ratings are not added. 38 CFR 4.25(a) combines them through Table I, largest first, and the combined value "will then be converted to the nearest number divisible by 10, and combined values ending in 5 will be adjusted upward." Under 4.25(b) that conversion "will be done only once per rating decision, will follow the combining of all disabilities, and will be the last procedure in determining the combined degree of disability."

Read the table below in both directions: the same 10 percent headache rating moves a lone 20 percent neck rating to 30, and leaves a file whose ratings combine to exactly 60 at 60. Each row is scoped to the file it names. With other ratings in your file, every rating is combined in order of severity using unrounded values, and these rows do not carry over.

Combined value under 38 CFR 4.25 Table I, for the file named in each row

Combined value under 38 CFR 4.25 Table I, for the file named in each row
Ratings already in the file	Headache rating	Table I value	Combined rating after rounding
Neck 20 percent, nothing else	0 percent	No Table I step	20 percent (no change)
Neck 20 percent, nothing else	10 percent	28	30 percent
Neck 20 percent, nothing else	30 percent	44	40 percent
Neck 20 percent, nothing else	50 percent	60	60 percent
Neck 20 percent and one other rating of 50 percent (Table I value 60)	10 percent	64	60 percent (no change)
Neck 20 percent and one other rating of 50 percent (Table I value 60)	30 percent	72	70 percent

Cells read from Table I in 38 CFR 4.25 as published in the eCFR issue of August 27, 2026. Three ratings are combined in order of severity: 50 and 20 give 60, and 60 and 10 give 64; 50 and 30 give 65, and 65 and 20 give 72.

Table diagram with no people, titled what a headache rating adds, 38 CFR 4.25 Table I, for the file named in each row. Columns: ratings already in the file, headache rating, Table I value, combined after rounding. Six rows: neck 20 alone and headaches 0 percent, no Table I step, combined 20, no change; neck 20 alone and headaches 10 percent, 28, combined 30; neck 20 alone and headaches 30 percent, 44, combined 40; neck 20 alone and headaches 50 percent, 60, combined 60; neck 20 and one other 50 (Table I 60) and headaches 10 percent, 64, combined 60, no change; neck 20 and one other 50 (Table I 60) and headaches 30 percent, 72, combined 70. A footer reads: rounding to the nearest 10 happens once, after every rating is combined; with other ratings in the file, these rows do not carry over. — The same examples, scoped to the file named in each row.

The full DC 8100 table and the definitions of its terms are on our migraines and VA disability page.

What Board Decisions on Neck-Linked Headaches Show

We read 25 Board decision files for this page, each at its order or remand block and the reasoning behind it. On the link, 11 granted service connection or a separate rating, 1 denied, 1 declined to reopen, 1 dismissed because VA had already connected the headaches, and 4 sent the question back. Seven more concern the percentage or a diagnosis that changed after the grant. These decisions came from searches, not a random draw, so the counts are not grant rates and predict nothing.

38 CFR 20.1303 says "previously issued Board decisions will be considered binding only with regard to the specific case decided", that "Prior decisions in other appeals may be considered in a case to the extent that they reasonably relate to the case", and that "each case presented to the Board will be decided on the basis of the individual facts of the case in light of applicable procedure and substantive law." We describe each decision for the issue it decided and for no more.

What carried the headache link in the 11 grants: VA's own evidence in 6 (1507123, 18157013, 18160149, 21067969, 21072900, 1442004); an earlier VA favorable finding, whose evidence the decision does not describe, in 1 (A21018795); VA evidence together with treating clinicians' notes in 2 (1500511, 1524862); a private chiropractor's opinion, with a June 2018 VA examiner reaching the same conclusion, in 1 (21024692); and an examiner whose status the decision does not state in 1 (18146722). In two of the grants a private opinion went to the neck, not the headaches (A21018795, and 18160149, which cites a June 2013 private opinion on the neck). Two private headache opinions failed: in 1314955 the Board found one "deserves little probative value", and in 23064169 it called one "speculative in nature".

Service connection or a separate rating granted

Citation Nr 1507123 (decided February 18, 2015)

Granted: "Service connection for occipital headaches secondary to cervical spine DDD is granted." The migraine component was not connected.

The Board relied on "the December 2009, June 2012, and August 2014 VA opinions", which tied the occipital part of the headaches to the cervical spine and found the migraine features unrelated to it.

Why it matters. VA's own examiners carried the grant, and a mixed headache picture was split.