If you have a service-connected knee, shoulder, hip, ankle, or wrist rated at 0 percent, there is now a federal audit that says there is roughly a coin flip’s chance the rating is wrong.
That is not advocacy language. It is the finding of the VA Office of Inspector General, published August 6, 2026, in a report titled Review of Assignment of Noncompensable Musculoskeletal Joint Disabilities (report number 25-02645-131). The OIG pulled a statistical sample of decisions in which VA assigned or confirmed a 0 percent rating for a joint, checked the work, and estimated that about 32,000 of 64,000 such decisions were decided incorrectly. Half.
About 12,000 of those errors cost veterans money right away, an estimated $45 million in underpayments at a minimum.
What a 0 percent rating actually is
VA calls it a noncompensable evaluation. It means VA agreed the condition came from service, put it on your rating sheet, and then assigned a percentage of zero. No monthly payment for that condition.
This is not a rare outcome. As of fiscal year 2024, veterans held 15.9 million service-connected musculoskeletal disabilities. Of those, 3.3 million were rated at 0 percent. The musculoskeletal system is the single largest category of service-connected disability in the entire VA system, and joints are most of it.
The rule the audit is about
There is a regulation, 38 C.F.R. § 4.59, that exists specifically to keep painful joints from falling into the zero bucket. Its intent is to recognize a joint that is actually painful, unstable, or malaligned as entitled to at least the minimum compensable rating for that joint. For most joints that minimum is 10 percent.
Two things about that rule matter for the audit.
First, it does not require the pain to be severe, or to limit your motion to a particular number of degrees. The rating schedule at § 4.71a sets out degree thresholds, and most veterans do not meet them. Section 4.59 is the separate path: the joint hurts when it moves, so it gets the minimum.
Second, the evidence that establishes painful motion does not have to be an objective finding by the examiner. The courts that review VA decisions held in 2015 that a veteran’s own report of painful motion, if the adjudicator finds it credible and competent, is sufficient to support a compensable rating. VA issued guidance after that decision saying so. The OIG confirmed the guidance exists.
Claims processors are still not applying it.
What the OIG actually found
The team reviewed a sample of 100 decisions completed between April 1, 2024, and March 31, 2025, then went back and reviewed another 20 decisions from October and November 2025 to see whether anything had changed. It had not.
Projected across the year, the numbers look like this:
- About 32,000 of 64,000 decisions were wrong, roughly 50 percent.
- Of those 32,000, an estimated 19,000 files contained both objective and subjective evidence of painful motion. Not one of them received a compensable rating.
- An estimated 12,000 errors had immediate monetary impact, totaling at least $45 million in underpayments.
- The other 20,000 errors did not change the veteran’s combined percentage or payment at the time. That is a function of how VA’s combined rating table rounds, and it does not mean the error is harmless. A 10 percent that should have been there can change a combined rating later, when another condition is added or increased.
The report gives one example. A veteran filed for an increase on a service-connected left knee. At the exam, the examiner wrote down in the medical history section that the veteran reported pain when bearing weight while walking. During testing, the examiner also noted objective pain with repeated use. The rating decision assigned 0 percent and never addressed painful motion at all. After the OIG flagged it, VA issued a corrected decision and the veteran received $29,130 in retroactive payment.
One file. One paragraph in an exam report that nobody weighed.
Why it is happening
This is the part of the report that should get attention, because the OIG did not find carelessness. It found that VA does not agree with itself about what the regulation means.
The team interviewed 51 claims processors at three regional offices. Some said pain alone is enough for a compensable rating. Others said pain has to be tied to motion. Others were not sure. Asked to give an example of subjective evidence of pain, several gave examples of objective evidence instead. One processor said flatly that if the pain appeared only in the medical history section of the exam and not in the objective findings, they would assign 0 percent. Another said they would send the exam back to the examiner as inconsistent. In both cases, the evidence in the file was already enough.
Then the OIG went up the chain to VA Central Office and asked the same question of Compensation Service leadership. The answers did not match there either. The chief of the procedures manual staff said motion has to be involved. The assistant director of training thought pain alone was probably enough, but was not certain. The acting assistant director for quality said motion is not required, then added that pain and motion should go together.
These are the offices that write the guidance.
The manual they write reflects the split. The section telling processors how to apply § 4.59 opens by saying a painful joint can support a compensable rating, then a few paragraphs later switches to painful motion. It calls the veteran’s own reports “subjective” evidence in one place and “non-objective” evidence in another.
And the software has the same problem. Claims processors are required to use a calculator built into VA’s claims system called the evaluation builder. The OIG found the tool is laid out differently for different joints: the knee has a painful motion box with separate flexion and extension options, the shoulder handles it another way, and nowhere does the tool distinguish between objective and subjective painful motion. If a processor does not know that both count, the tool will not tell them.
VA quality staff apparently knew something was wrong. On a national call in March 2025, they reminded processors to remember to check the painful motion boxes, and warned that if the evaluation builder returns a 0 percent for a joint, it is worth going back and verifying the inputs. When the OIG visited regional offices in June and July 2025, the confusion was still there.
Where it stands now
The OIG made four recommendations: get a binding interpretation of § 4.59 from VA’s Office of General Counsel, fix the manual’s terminology, make sure processors understand they must weigh all the evidence, and fix the evaluation builder.
VA concurred with the first two and concurred in principle with the other two. On April 9, 2026, VA asked its Office of General Counsel for an opinion on what § 4.59 means, including the specific question of whether pain without painful motion is enough. As of the report’s publication, that opinion had not issued. The manual revision, the training assessment, and the software fix are all queued behind it.
VA also reported that by February 2026 it had corrected every error the OIG identified. Read that carefully. The OIG identified errors in 120 sampled files. The 32,000 figure is a projection. The other roughly 31,900 veterans were not part of the sample, have not been contacted, and are not being reviewed.
Nothing in the report suggests VA plans a mass readjudication. If a decision in that group is going to be corrected, someone has to raise it.
If you have a 0 percent joint rating
Two documents tell you most of what you need to know: the rating decision, and the C&P exam it was based on. Both are in your VA file, and you can get them through VA Form 20-10206 (Freedom of Information Act or Privacy Act Request) or through your MyHealtheVet and VA.gov file access.
Read the exam first, and read all of it. The error pattern the OIG described lives in a specific place: pain that the examiner wrote down in the medical history section, where the veteran describes the condition, that never made it into the objective findings section. That is still evidence. It still counts. Processors have been treating it as if it does not.
Then read the rating decision and look for whether it says anything at all about painful motion. In the OIG’s example, the decision simply did not address it. Silence in the reasons and bases, where the exam plainly documented pain, is the signature of this error.
A few other things worth checking:
Did you report pain and it is not in the exam anywhere. That is a different problem, an inadequate exam rather than a misapplied regulation, and it has its own remedy.
Does § 4.59 cover more than pain in your case. The regulation also reaches joints that are unstable or malaligned. Instability in particular gets rated separately from limitation of motion, so it can be worth more than the minimum.
Is the joint’s diagnostic code one that provides a compensable level for limitation of motion. Section 4.59 gets you the minimum compensable rating for the joint, so which code VA used matters to what that minimum is.
Do you have several joints at 0 percent. Multiple 10 percent ratings that should have been assigned combine, and the difference between a combined 40 and a combined 50 is real money every month.
What you can do about it
If the decision is less than a year old, you have all three lanes open: a supplemental claim (VA Form 20-0995) if you have new evidence such as treatment records documenting pain, a higher-level review (VA Form 20-0996) if the evidence is already in the file and the error is in how it was weighed, or a Board appeal (VA Form 10182). A misapplication of § 4.59 on evidence already in the record is the classic higher-level review posture, because the whole argument is that a senior reviewer looking at the same file should reach a different result.
If the decision is older than a year, a supplemental claim with new and relevant evidence is the route. The effective date will generally run from the date of the supplemental claim rather than the original decision, which is why the timing matters and why waiting costs money.
There is also a narrower path worth knowing about. Where a rating decision applied the wrong regulation to evidence that was in front of it, and the outcome would have been different, that can raise a clear and unmistakable error claim, which reaches back to the original effective date. The standard is demanding and most 0 percent ratings will not meet it. But an audit finding that half these decisions misapplied a specific regulation is a useful piece of context to have when evaluating one.
The larger point
Most VA errors are hard to see from the outside. This one is not. VA studied itself, found a 50 percent error rate on a specific and very common decision, traced the cause to its own guidance and its own software, and published the whole thing.
The 0 percent rating on a joint is easy to accept and move past. It sounds like a decision that went your way, since the condition is service connected. It reads like a technicality.
For roughly half the veterans who got one in the last two years, it was a mistake worth money.
If you have a service-connected joint sitting at 0 percent and you remember telling the examiner it hurts, that decision is worth pulling out and reading again.
Blackmore Veterans Law, P.C. is a solo practice representing veterans in VA disability claims and appeals. This article is general information about how the VA claims process works. It is not legal advice about any individual claim, and reading it does not create an attorney-client relationship. Attorney advertising.