A nexus letter is the document that decides whether the VA treats your condition as connected to your service or not. At the Board, the gap between a strong nexus and a weak one runs from grant rates around 93.7% down to 4.9%. That's not a margin. That's the entire claim.

Most veterans I talk to don't really understand what a nexus letter is until they've already been denied without one. The first denial letter says something like "the evidence does not show a medical link between your current condition and your military service," and a lot of people read that and think it means "I need more treatment records." It almost never means that. It means the file is missing the one specific document that ties the dots together for the rater, and unless that document shows up, the rest of the file can be a hundred pages thick and still lose.

This post is the definitional version. What a nexus letter is. Who writes one. What the legal standard is. How it differs from related documents like an IMO or a DBQ. When it's required. What it costs. What to watch out for. The companion post on what makes a nexus letter persuasive goes deeper on the language patterns the Board treats as strong versus weak. This one is the orientation.

TL;DR

  • A nexus letter is a medical opinion from a licensed clinician stating that a current condition is "at least as likely as not" related to military service. It's the third leg of the service connection stool under 38 CFR § 3.303 and § 3.310.
  • Any licensed medical professional can write one. The credentials matter less than the reasoning. In the Board dataset, opinions written by treating physicians were granted in 78.5 percent of 2,613 condition records, and opinions from private independent examiners in 73.8 percent of 6,827.
  • The legal standard is the 50/50 "at least as likely as not" threshold. The opinion doesn't have to say service caused the condition. Only that service and the condition are at least as likely related as not.
  • In the BVA dataset I've looked at, strong nexus opinions correlate with grant rates around 89.5% to 93.7%, and weak or missing ones correlate with grant rates between 1.5% and 4.9%. The middle barely exists. It's a cliff.
  • Private nexus letters typically cost between $300 and $2,500. Some VSO-affiliated clinics offer them at reduced cost or free.
  • "Nexus mills," providers who churn out template letters for low prices, show up repeatedly in BVA decisions as the source of opinions the Board called conclusory and dismissed.
What evidence wins a VA claimFree

Strong medical opinions won 89.7% of the time in our analysis. Weak ones won just 3.7%.

  • Know what makes a medical opinion harder to dismiss.
  • Use your own statement where it can carry real weight.
  • Catch common C&P exam problems before they hurt your claim.
Get the free guide

What a Nexus Letter Actually Is

A nexus letter is a written medical opinion that connects a current diagnosed condition to an event, injury, or exposure that happened during military service. The word "nexus" just means "connection."

The reason the document exists is that 38 CFR § 3.303, the regulation governing direct service connection, requires three things for VA to grant service connection:

  • A current diagnosis of the claimed condition
  • An in-service event, injury, or illness
  • A medical link between the first two

That third leg is the nexus. The nexus letter is the document that supplies it. Without something in the file that ties the in-service event to the current diagnosis, the first two pieces don't matter. You can have a textbook in-service injury and a textbook current diagnosis, and if there's no medical opinion connecting them, the claim still fails.

For secondary service connection under 38 CFR § 3.310, the structure is the same with one substitution. Instead of an in-service event, the trigger is an already service-connected primary condition. PTSD aggravating hypertension is a common example. The hypertension claim needs a nexus opinion saying the PTSD caused or aggravated it.

The regulation doesn't specify who has to write the opinion, how long it has to be, or what credentials the writer needs. It just says the link has to be established by competent medical evidence. The form that evidence takes is what we call a nexus letter.

Who Can Write One

Any licensed medical professional can write a nexus letter. A medical doctor. A doctor of osteopathic medicine. A nurse practitioner. A physician assistant. A psychologist with a doctoral-level license can write one for mental health conditions. An audiologist can write one for hearing loss and tinnitus.

What the Board cares about isn't the credential. It's the basis and the reasoning. In Nieves-Rodriguez v. Peake (2008), the Court of Appeals for Veterans Claims explicitly held that "the probative value of a medical opinion comes from its reasoning." A board-certified specialist who writes a conclusory opinion can be outweighed by a treating clinician with lesser credentials who actually explains how they got to the conclusion.

That doesn't mean credentials are worthless. Specialists relevant to the condition tend to be more persuasive than general practitioners, particularly when there's a competing opinion in the file. But a general practitioner with a strong, fully articulated opinion can outweigh a specialist who writes a one-paragraph letter.

The most common sources of nexus opinions I've seen:

Treating physicians

The veteran's regular doctor. They've seen the veteran across multiple visits and have clinical judgment grounded in real care. The Board tends to give treating physician opinions extra weight when they reflect longitudinal observation. The numbers bear that out: in Claim Raven's Board dataset, treating physician opinions were granted in 78.5 percent of 2,613 condition records, private independent examiner opinions in 73.8 percent of 6,827, and VA examiner opinions in 36.9 percent of 35,160. Your own doctor's reasoned opinion carried at least as much weight as a paid one.

Specialists

A condition-relevant specialist who's evaluated the veteran for the nexus question.

Independent Medical Examiners (IMEs)

A clinician hired specifically to review the file and render an opinion. No treating relationship.

VA C&P examiners

The clinicians who conduct VA-ordered Compensation and Pension exams. They can render nexus opinions on a DBQ. Their opinions tend to be shorter and less articulated than private opinions on average.

Any of these sources can produce a strong nexus letter. Any of them can produce a weak one. Who wrote it matters less than how it was written.

This phrase shows up in every nexus letter that's done correctly, and it's worth pausing on what it actually means.

VA service connection cases are governed by a 50/50 evidentiary standard. The veteran doesn't have to prove the condition is service-connected. The veteran only has to put forward evidence that makes service connection at least as likely as the alternative. If the probability is 50/50 or better, the benefit of the doubt under 38 CFR § 3.102 goes to the veteran. That's a lower bar than the standards used in criminal cases ("beyond a reasonable doubt") or most civil cases ("preponderance of the evidence," more than 50%).

In medical opinion language, that translates to a specific phrase: "at least as likely as not." A clinician who writes "it is at least as likely as not that the Veteran's [condition] is related to his/her military service" is hitting the regulatory standard exactly.

The three phrases that matter:

  • "At least as likely as not" = 50/50 or better. Favorable. Meets the standard.
  • "More likely than not" = above 50%. Strongly favorable.
  • "Less likely than not" = below 50%. Negative. Weighs against the claim.

Anything weaker than those, "possibly related," "could be related to," "cannot be ruled out as related to," falls short. The Board reads softer language as the doctor hedging, and a hedged opinion doesn't clear the 50/50 bar.

The exact wording matters because the regulation and the case law track those phrases. Doctors who aren't familiar with VA work sometimes write opinions that say something like "I believe Mr. Smith's condition is related to his service" without using the magic words. Ambiguous language often gets read as falling short.

Nexus Letter vs. IMO vs. DBQ

These three documents get confused often enough that it's worth separating them.

A nexus letter is a written medical opinion focused on the connection between a condition and service. It can be short or long, but its purpose is narrow. It answers the question "is this condition related to military service?" with reasoning attached.

An Independent Medical Opinion (IMO) is broader. An IMO is a more comprehensive review performed by a clinician who isn't otherwise involved in the veteran's care, often hired specifically for the case. It usually includes a full file review, an examination of the veteran, and a written report covering diagnosis, history, and opinion. Most IMOs include a nexus opinion as part of the report. A nexus letter from a treating physician is often a one or two page document. An IMO is often eight to twenty pages and costs more.

A Disability Benefits Questionnaire (DBQ) is a different animal. A DBQ is a standardized form VA uses to capture the clinical findings needed to rate the disability. It's not primarily about service connection. It's about how severe the condition is. Most DBQs include a section for a medical opinion on nexus, but the form's main purpose is rating, not connection. (I wrote a separate post on DBQs that goes deeper.)

A strong evidentiary package often includes all three: a DBQ documenting severity, a nexus letter or IMO establishing the connection, and supporting treatment records. Each does a different job. Conflating them is one of the most common mistakes I see. A veteran with a strong DBQ but no nexus letter often gets denied for "lack of medical nexus" even when the DBQ documents a severe condition. The DBQ doesn't replace the nexus letter.

When a Nexus Letter Is Required, And When It Isn't

Not every claim needs a nexus letter. There are categories of claims where the regulation either presumes service connection or treats certain conditions as inherently connected to specific service.

Presumptive conditions. Under 38 CFR § 3.309, § 3.317, § 3.320, and the PACT Act, certain conditions are presumed service-connected if the veteran served in a qualifying time and place. Vietnam-era veterans with Agent Orange-linked cancers. Gulf War veterans with qualifying chronic multisymptom illnesses. PACT Act burn pit exposures linked to certain respiratory and oncological conditions. If the condition is on the presumptive list and the service is qualifying, the nexus is presumed.

Chronic diseases manifesting within a presumptive period. Under § 3.309(a), certain chronic diseases (arthritis, diabetes, hypertension, others) are presumed service-connected if they manifested to a compensable degree within one year of separation.

Continuity of symptomatology. Under § 3.303(b), for certain chronic diseases, evidence of continuous symptoms from service to the present can establish service connection without a formal nexus opinion. The Federal Circuit narrowed this in Walker v. Shinseki (2013) to only those conditions specifically listed in § 3.309(a).

Combat presumption. Under 38 U.S.C. § 1154(b), combat veterans get a presumption that injuries consistent with the circumstances of combat occurred in service. This doesn't substitute for the medical nexus to the current condition, but it lowers the bar for the in-service event piece.

For everything outside those categories, a nexus letter is functionally required to win. Direct service connection without any of the presumptions essentially requires a medical opinion linking the current condition to service. Lay evidence alone, no matter how detailed, doesn't establish a medical nexus for conditions that require medical expertise.

Secondary service connection under § 3.310 almost always requires a nexus opinion, because the connection between two medical conditions is by definition a medical question.

The practical takeaway: most non-presumptive claims need a nexus letter to win. If there's any uncertainty about whether a claim falls under a presumption, the safer move is to have a nexus opinion in the file.

The Three Components Every Nexus Letter Needs

A nexus letter that does its job covers three core elements. The companion post goes deeper on what makes each one persuasive, but at the definitional level:

  • A current diagnosis. The letter has to identify what the current condition is. Not "back pain." Specifically "L4-L5 disc herniation with radiculopathy" or "degenerative disc disease of the lumbar spine." The diagnosis has to be specific enough to map to the rating schedule.
  • A clear identification of the in-service event, injury, or exposure. The letter has to name the specific in-service trigger. A specific injury date, a documented MOS-related exposure, a specific event in the service treatment records. Vague references like "exposures during service" don't carry the weight that specific events do.
  • The causal link with reasoning. This is the actual nexus. The opinion has to explain not just that the conditions are related, but why. What's the mechanism? How does the in-service event cause the current condition? An opinion that states a conclusion without explaining the link is what the Board calls conclusory, and per El-Amin v. Shinseki (2013), conclusory opinions have "little, if any, probative value."

The third element is where most nexus letters fail. Veterans get a letter that has the diagnosis right and identifies the in-service event correctly, but the connection between them is a single sentence with no explanation. The Board calls that conclusory and discounts it.

A nexus letter that explicitly walks through the mechanism, even briefly, lands differently. "The Veteran's documented in-service [event] caused acute [injury], which is a known precursor to the chronic [condition] currently diagnosed. The Veteran's continuous post-service symptoms, documented in [records], are consistent with this progression." That's a sentence or two longer than the conclusory version, but it does the work the Board needs the opinion to do.

What a Strong Letter Costs

There's no clean public dataset on private nexus letter pricing. From what I've seen across veteran communities and provider websites:

Treating physician letters: $0 to $300. When a veteran's existing treating physician is willing to write one as part of the care relationship, it's often free or a nominal administrative fee.

Standalone nexus opinions from dedicated providers: $300 to $1,500. A clinician who specializes in VA work and offers nexus letters as a focused service typically charges in this range.

Full Independent Medical Opinions: $1,000 to $2,500, sometimes higher. Comprehensive review, examination, and detailed written report. Complex cases involving multiple conditions or large records can run $3,000 or above.

Specialist evaluations: highly variable. Specialty conditions (neurology, cardiology, psychiatry) often run higher. A specialist IMO for a complex case can run $2,500 to $5,000.

The cost is almost always paid by the veteran. The VA doesn't reimburse private nexus letters. Some VSO-affiliated clinics offer them at reduced cost. Attorneys on contingency sometimes front the cost in expectation of recovering it from back pay if the case wins.

For veterans without resources, the options narrow. The C&P exam provides a VA-ordered nexus opinion at no cost, but C&P opinions tend to be shorter and less articulated than private opinions on average. Some veterans groups, university veterans clinics, and pro bono medical-legal partnerships offer nexus opinions at reduced or no cost. Availability varies by location.

The honest assessment: a strong nexus letter is often the single most consequential dollar a veteran spends on a claim. A lot of denials for lack of medical nexus happen to veterans whose conditions are genuinely service-connected. The connection just never made it to paper.

Nexus Mills and What to Watch For

A category to be cautious about: what veterans communities sometimes call "nexus mills."

A nexus mill is a provider, clinic, or service that produces nexus letters at high volume, often for low prices, with minimal individualization. The pattern is recognizable. The letter uses the same paragraph structure, the same phrasing, sometimes the same word-for-word language across cases for unrelated conditions. The doctor's name appears on opinions for hundreds or thousands of unrelated veterans. It's a template with the diagnosis and name dropped in.

The Board has gotten familiar with these patterns. When a provider's name shows up repeatedly on conclusory opinions, the Board's language tends to be predictable. "Conclusory," "lacks rationale," "appears templated," "fails to address the specific medical history." A nexus letter from a known mill provider often gets less weight than a thoughtful letter from an unknown but careful clinician.

A few flags I've seen veterans use to assess a provider before paying:

  • Does the provider review the actual file, or just the diagnosis sheet?
  • Does the letter address the specific in-service event by date and detail, or just refer to "events during service"?
  • Does the letter explain the mechanism, or only state a conclusion?
  • Does the letter engage with any existing C&P opinion in the file, or ignore it?

Attorneys and experienced VSOs often know which providers' work tends to hold up at the Board and which doesn't. Asking before paying is reasonable.

The cheapest letter isn't always the worst. The most expensive isn't always the best. The variable that matters is whether the opinion does the work the regulation and case law require.

Service Connection vs. Increased Rating: When a Nexus Letter Isn't the Right Tool

One distinction worth flagging. A nexus letter is a tool for establishing service connection. It answers the question "is this condition related to military service?" It doesn't answer the question "how severe is this condition?"

For an increased rating claim, the condition is already service-connected. The question is whether the condition is more severe now than the current rating reflects. A nexus letter doesn't help. What helps is a strong DBQ or treatment record evidence documenting the current severity.

For a secondary service connection claim, even though the primary is already service-connected, the question is whether the secondary is connected to the primary. That's a connection question, so a nexus letter is the right tool.

For a CUE motion (clear and unmistakable error), the question is whether the original decision was wrong based on the evidence that existed at the time. A new nexus letter doesn't help, because the question is what the file looked like back then.

For a supplemental claim with new and relevant evidence, a new nexus letter can be the right move if the original denial was for lack of nexus.

The general principle: a nexus letter answers the connection question. If the connection question isn't on the table, it probably isn't the right tool.

Bottom Line

A nexus letter is the document that establishes the medical link between a current condition and military service, and outside of presumptive claims, most claims need one to win. The regulation requires three things for service connection: a current diagnosis, an in-service event, and a medical link. The nexus letter is the link.

Any licensed medical professional can write one, and the credentials matter less than the reasoning. The legal standard is "at least as likely as not," a 50/50 threshold under the benefit-of-the-doubt rule. The three elements every nexus letter needs are a specific diagnosis, a specific in-service event, and a causal explanation with rationale. Letters that hit those three elements with real reasoning tend to win at high rates. Letters that don't tend to get characterized as conclusory and dismissed.

A nexus letter isn't a DBQ and it isn't an IMO, though IMOs usually contain nexus opinions and a comprehensive evidentiary package usually contains all three. It's the right tool for service connection questions and the wrong tool for increased rating questions.

Costs run from free (treating physician) to $2,500 or more (comprehensive specialist IMO). Veterans almost always pay out of pocket. Nexus mills exist and are recognizable by their templated output and the Board's predictable language when discounting their work.

The marginal return on a strong nexus letter is enormous when the existing file doesn't have one. The marginal return is much smaller when the file already has a thorough opinion. Where the dollar goes depends on what's already there.

Methodology and Limitations

  • Regulatory and statutory references: 38 CFR § 3.102 (benefit of the doubt), § 3.303 (direct service connection), § 3.304 (combat presumption), § 3.309 (chronic diseases and presumptions), § 3.310 (secondary service connection), § 3.317 (Gulf War undiagnosed illness), § 3.320 (PACT Act presumptions), § 4 (rating schedule). 38 U.S.C. § 1154(b) (combat presumption). All public and verifiable.
  • Case law references: Nieves-Rodriguez v. Peake, 22 Vet. App. 295 (2008); El-Amin v. Shinseki, 26 Vet. App. 136 (2013); Walker v. Shinseki, 708 F.3d 1331 (Fed. Cir. 2013). All published and citable.
  • Grant rate context: Strong nexus opinion grant rate of 89.5% to 93.7% and weak/missing nexus opinion grant rate of 1.5% to 4.9% are drawn from Claim Raven's aggregate BVA analysis published in companion posts. Specific condition-level breakdowns are not in this post.
  • Cost figures: Reflect a range observed across veteran-focused provider websites, veteran community discussions, and IME service marketing. Not survey data. Actual costs vary by provider, condition complexity, geography, and case demands.
  • Limitations:
  • Most observations are based on patterns from BVA-level cases. Cases resolved at the regional office level without appeal aren't in the BVA dataset.
  • "Nexus mill" pattern recognition is qualitative. There's no published list of which providers fall in that category. The flags described are general indicators, not definitive markers.
  • Presumptive condition lists change as VA updates regulations and Congress passes legislation. The current list is on VA.gov.
  • The companion post on what makes a nexus letter persuasive goes deeper on language patterns. This post is the definitional orientation.

The aggregate picture is consistent across the dataset. In Claim Raven's analysis of 101,518 condition records drawn from 49,876 Board decisions, cases with a nexus opinion the Board characterized as strong granted at 89.5% (12,509 cases). Adequate granted at 68.9% (18,597 cases). Weak granted at 3.3% (18,260 cases). Missing granted at 1.5% (10,718 cases). The drop between adequate and weak is where most claims fail, and it's where the cost of a thoughtful nexus opinion shows up.

Disclaimer

I'm not accredited by VA, not a lawyer, not a VSO. This is data analysis, not claim advice. These are patterns from cases that made it to the BVA, they don't predict individual outcomes. If you need help with your claim, work with an accredited representative.

Where to go next

-Landon Founder, Claim Raven | U.S. Army Veteran