VA OIG Estimates 38 Percent Error Rate in Painful Scar Rating Decisions

VA OIG Estimates 38 Percent Error Rate in Painful Scar Rating Decisions

The VA Office of Inspector General published Report No. 25-03303-266 on October 21, 2026, estimating that about 17,900 painful scar rating decisions completed over one year contained at least one error. The review examines how claims processors applied 38 C.F.R. § 4.118.

Information last verified on October 23, 2026. This is a developing story; we update it as the record changes.

Jurisdiction scope: This article addresses a federal oversight report on United States Department of Veterans Affairs disability compensation, specifically evaluations for painful scars under 38 C.F.R. § 4.118, diagnostic code 7804. It does not address state law, VA health care eligibility, or any other diagnostic code. For how VA arrives at a combined percentage once individual evaluations are assigned, see the arithmetic behind combined VA ratings.

What Happened

On October 21, 2026, the VA Office of Inspector General (OIG) published Report No. 25-03303-266, "Review of Accuracy of Claims Processors' Disability Evaluation Decisions for Painful Scars," prepared by its Office of Audits and Evaluations. The OIG conducted the work from August 2025 through August 2026. It reviewed a stratified random sample of 130 in-scope decisions completed from June 1, 2024, through May 31, 2025: 70 that granted service connection for painful scars and 60 that increased evaluations for scars already service connected. The sampling frame held 47,957 decisions, and the OIG estimated the actual in-scope population at about 47,300. According to VBA data cited in the report, roughly 538,000 veterans were receiving VA benefits for service-connected painful scars as of July 24, 2026.

Weighting the sample to the population, the OIG projected 17,856 decisions with at least one error, a 37.7 percent error rate, with a 90 percent confidence interval of 13,645 to 22,398 decisions. Split by stratum, the projections were 15,409 errors among grants of service connection (35.7 percent) and 2,447 among increased evaluations (58.3 percent). Seven of the sampled decisions involved an improper payment. From those seven, the OIG projected a total payment-error value of $12,524,919 and reported the one-sided 90 percent lower bound of $5,584,470 as its questioned cost. It reported a separate one-sided lower bound of 915 affected decisions against a point estimate of 2,129, stated in the narrative as a minimum of about 920. The report states the OIG "did not find any errors that resulted in underpayments."

After briefing VBA staff on preliminary findings in February 2026, the review team pulled 30 more decisions completed in January 2026 and found a 43 percent error rate, which the OIG said confirmed a similar rate was continuing as of that month. The team also conducted an in-person site visit at the VA regional office in Houston, Texas, in December 2025 and a virtual visit at the New Orleans, Louisiana, regional office in January 2026, interviewing 26 regional office managers and staff, including 20 claims processors.

What this report is, and what it is not. The review is an audit of VBA's process. It assesses whether claims processors followed the rating schedule and VBA's Adjudication Procedures Manual. It is not a rating reduction, a notice of indebtedness, or a collection action, and it imposes nothing on any veteran. The report's only instruction to anyone is directed at VBA: recommendation 1 asks the under secretary for benefits to "review and correct the processing errors on decisions the Office of Inspector General team identified" so they comply with 38 C.F.R. § 4.118. VBA concurred and set a target completion date of February 28, 2027. The report does not describe notice to affected veterans, does not say whether VBA will look beyond the decisions the OIG identified, and does not address how any individual overpayment would be handled.

VBA concurred with all three recommendations in a September 8, 2026, memorandum signed by J. Margarita Devlin, principal deputy under secretary for benefits, performing the delegable duties of the under secretary. The OIG considers recommendation 2 closed. Recommendation 2 asked VBA to revise the "Scars/Disfigurement" disability benefits questionnaire so it clearly elicits whether subjective reports of pain are supported by objective evidence; VBA reported the revision completed August 14, 2026, sent to its DBQ Change Control Group on August 25, 2026, with a tentative release date of January 10, 2027. Recommendation 3 asked VBA to complete additional actions so employees understand how to evaluate painful scars accurately; VBA said it will cover the topic on a Compensation Service Monthly Quality Assurance call, with a January 31, 2027, target.

Two disputed items remain on the record. VBA declined to accept two of the errors the OIG identified. In one, the OIG treated a claim as invalid because the veteran had asked for a decision on an already submitted claim rather than an increase, pointing to manual guidance that VBA should avoid assuming a veteran's intent. In the other, VBA had implemented a Board of Veterans' Appeals decision directing evaluations under multiple diagnostic codes 7804; the OIG pointed to 38 C.F.R. § 20.105, which requires the Board to follow VA regulations, and to the rebuttal routes in 38 C.F.R. § 20.801 and § 20.1403. As of July 17, 2026, the report states, VBA had not responded to the OIG's additional evidence on those two.

What the Law Actually Says

Painful scars are rated under 38 C.F.R. § 4.118, the schedule of ratings for the skin, at diagnostic code 7804. The criteria count scars:

7804: Scar(s), unstable or painful

30 percent: Five or more scars that are unstable or painful

20 percent: Three or four scars that are unstable or painful

10 percent: One or two scars that are unstable or painful

Source: 38 C.F.R. § 4.118, diagnostic code 7804, as reproduced in Table 1 of VA OIG Report No. 25-03303-266 and verified against the eCFR text of title 38 at its October 21, 2026 issue date.

Three notes follow. Note (1) defines an unstable scar as "one where, for any reason, there is frequent loss of covering of skin over the scar." Note (2) provides that "if one or more scars are both unstable and painful, add 10 percent to the evaluation that is based on the total number of unstable or painful scars." Note (3) provides that "scars evaluated under diagnostic codes 7800, 7801, 7802, or 7805 may also receive an evaluation under this diagnostic code, when applicable."

Two points about that text matter for reading the report. First, the regulation's own words for diagnostic code 7804 set a counting rule and three notes. The OIG grounds the objective-evidence requirement in both sources of authority: its introduction states that under 38 C.F.R. § 4.118 the medical evidence, "such as documentation from a physical examination of the scar, must show objective evidence of tenderness to the touch," and its finding is that claims processors did not always ensure an evaluation was supported by objective evidence of a painful scar "in compliance with the rating schedule or the Adjudication Procedures Manual." The explicit instruction that a veteran's reported history is not enough on its own appears in the manual topic. The report cites M21-1 topic V.iii.10.3.b, "Objective Evidence of a Painful Scar," updated August 13, 2018, for the proposition that a claims processor should not assign a compensable evaluation of at least 10 percent without objective evidence supporting the medical history of subjective pain, and that a veteran's reported history alone is not enough. In practice, the report says, that means the examiner must document whether the scar is tender to touch.

Second, the zone-based rule that lets some scars earn separate evaluations does not reach 7804. Note (2) under diagnostic codes 7801 and 7802 permits a separate evaluation "for each affected zone of the body"; Note (1) defines the six zones as each extremity, the anterior trunk, and the posterior trunk. The OIG's report states that this guidance "does not apply to painful scars under diagnostic code 7804." All of a veteran's painful scars are counted once, under one 7804.

That is where pyramiding comes in. 38 C.F.R. § 4.14 is titled "Avoidance of pyramiding" and reads in part:

The evaluation of the same disability under various diagnoses is to be avoided. [...] Both the use of manifestations not resulting from service-connected disease or injury in establishing the service-connected evaluation, and the evaluation of the same manifestation under different diagnoses are to be avoided.

38 C.F.R. § 4.14

The OIG's three common error categories map directly onto those rules. The report attributes the remaining errors to other issues, such as not clarifying scar examinations before deciding, and notes that some decisions contained errors in more than one category, which is why the three category projections do not sum to 17,856. The largest, projected at 7,409 decisions, was a compensable evaluation assigned without objective evidence of a painful scar. In the report's first example, a disability benefits questionnaire recorded aching pain from a keloid removal scar on the right ear in the medical history section, but the physical examination section did not record tenderness to touch; a 10 percent evaluation issued under 7804 anyway, which the report describes as an ongoing overpayment totaling about $6,900 as of March 2026.

The second category, projected at 5,490 decisions, was the use of more than one diagnostic code 7804 on the same rating. The report's example involved scars on the right hand and the left posterior flank, both documented as tender to touch, each given its own 10 percent evaluation under 7804. Under § 4.118, the OIG wrote, the correct result was a single 10 percent evaluation under one 7804 covering both scars, since two painful scars fall in the "one or two scars" band. The report attributes about $5,200 in payments as of March 2026 to that error. Claims processors interviewed by the team split on recognizing the pattern: shown a sample codesheet with duplicate 7804 entries, 14 of 20 identified the error and 6 did not. The VBMS-Rating application flagged pyramiding for other diagnostic codes but not for 7804. VBA told the OIG on July 9, 2026, that a pyramiding conflict notification for multiple 7804 codes had been added to VBMS-Rating on April 19, 2026, and demonstrated it on July 13, 2026. Because VBA had already acted, the OIG made no recommendation on that point.

The third category, projected at 1,328 decisions, ran the other way. Note (3) lets a scar already rated under another code pick up a separate 7804 evaluation once it becomes painful. The report's example describes back scars rated 0 percent under diagnostic code 7805, which covers other effects of scars, where the 7805 evaluation was deleted and replaced with a 10 percent evaluation under 7804. The OIG said the veteran remained entitled to the 0 percent 7805 evaluation alongside the separate 7804, and while removing it caused no underpayment then, it "could affect future disability evaluations if scars in this category were to become compensable." How separate percentages interact once assigned is a question of 38 C.F.R. § 4.25, the combined ratings table; readers who want to see that arithmetic can work it through with our VA disability calculator.

Two other bodies of regulation sit behind the report without being its subject. When VA reduces an evaluation that would lower compensation currently being paid, 38 C.F.R. § 3.105(e) requires a rating proposing the reduction "setting forth all material facts and reasons," notice to the beneficiary at the latest address of record, and 60 days "for the presentation of additional evidence to show that compensation payments should be continued at their present level." By a different route, § 3.105(a)(1)(ii) sits inside § 3.105(a)(1), the provision on revising a final decision for clear and unmistakable error, and provides that "[e]xcept as provided in paragraphs (d) and (e) of this section, where an award is reduced or discontinued because of administrative error or error in judgment, the provisions of § 3.500(b)(2) will apply." Paragraph (e) is the reduction-in-evaluation route described just above, so it is carved out of that rule rather than paired with it. Section 3.500(b)(2), itself subject to exceptions in paragraph (r) of that section and in § 3.501(e) and (g), sets the "date of last payment on an erroneous award based solely on administrative error or error in judgment." Separately, 38 C.F.R. § 1.963(a), issued under 38 U.S.C. § 5302, provides that recovery of an overpayment "shall be waived" where there is no indication of fraud, misrepresentation, or bad faith and recovery "would be against equity and good conscience," with requests generally due within one year of a notice of indebtedness. Those are the general rules in the regulations. The OIG report does not apply them, and it does not address how any particular overpayment would be handled.

The published review options after a rating decision are set out in 38 C.F.R. § 3.2500. Within one year of notice of an agency of original jurisdiction decision, a claimant may elect either a higher-level review under § 3.2601 or an appeal to the Board under § 20.202, and a supplemental claim under § 3.2501 may be filed at any time after notice of a decision. Section 3.2500(b) prohibits electing two of those options concurrently on the same issue. A Notice of Disagreement under § 20.202 must be filed on the form prescribed by the Secretary and must indicate whether the claimant wants direct review, a Board hearing, or an opportunity to submit additional evidence without a hearing. We cover those three lanes and their deadlines in more detail in our guide to the three VA review options after a rating decision.

Analysis: Why This Matters

The following is analysis from the Recording Law Editorial Team.

The most useful thing in this report is not the headline error rate. It is the diagnosis of why the errors happen, because both causes are largely structural.

Take the objective-evidence category. Diagnostic code 7804 turns on a single word, "painful," and a compensable rating depends entirely on whether the examination record crosses a line the regulation's own text never draws. The OIG attributes the requirement to the rating schedule and the Adjudication Procedures Manual together; the explicit requirement of documented tenderness to touch, and the instruction that a reported history of pain is not enough without it, sit in a manual topic last updated in August 2018. It is operationalized through a form that has a medical history section where a veteran's reported pain is recorded and a physical examination section where tenderness to touch is recorded, without labeling either as subjective or objective. The OIG reports that 5 of the 20 claims processors interviewed said they had difficulty understanding the questionnaire for painful scars, and 13 said revising it could help clarify what objective evidence means. One suggested moving the subjective section closer to the objective section. That is a form-design finding, and recommendation 2, the one the OIG has already closed, is a form-design fix.

The pyramiding category is cleaner still. VBMS-Rating already alerted users to pyramiding for other diagnostic codes. It did not for 7804. All 20 claims processors interviewed agreed that adding the control could prevent the error, and VBA added it on April 19, 2026, during the review. That sequence is worth noting precisely because it shows the cheapest fix in the report was a configuration change in an existing system control, and it shipped before publication without a recommendation attached.

The statistical dispute in the report deserves to be read in full rather than summarized to one side. VBA's technical comment is that 130 decisions is roughly 0.27 percent of a 47,300-decision population, that 60 errors were found, and that only 7 of those 60 resulted in improper payments, which VBA characterized as about 5.4 percent of the sample. VBA's point is that more than 88 percent of the sampled errors were technical or procedural without a payment consequence. The OIG's response is that the share of a population a sample represents has very little bearing on statistical precision unless the population is very small, and that precision turns on absolute sample size and variability. Both statements can be true at once. The projection of about 17,900 errors is a defensible estimate of decisions containing at least one error in that year; it is not an estimate of 17,900 incorrect payments, and the report does not claim it is. The money figure carries its own hedge: because only 7 of the 60 sampled errors drove the monetary estimate, variability was high, so the OIG reported the one-sided 90 percent lower bound of $5,584,470 rather than its $12,524,919 point estimate. A reader who treats $5.6 million as the measured loss is reading a deliberately conservative floor as a central estimate.

One asymmetry in the findings is stated plainly in the report and should not be read past: the OIG found no errors that resulted in underpayments. That is a finding about this sample under this diagnostic code over this twelve-month window, and the report offers no theory for it. The third error category complicates the picture somewhat, since deleting a 0 percent evaluation under 7805 costs a veteran nothing today while narrowing what is on the codesheet later.

How This Affects You

Nothing in this report changes any veteran's rating, creates a debt, or requires anyone to do anything. The report is addressed to the under secretary for benefits, its recommendations run to VBA, and its target dates are VBA's. The one qualification the record carries is that the report asks VBA to review and correct the specific decisions the OIG identified by February 28, 2027, and it says nothing about how any resulting adjustment would be handled in an individual case.

Recommendation 1 asks VBA to review and correct the processing errors on the decisions the OIG team identified, with a February 28, 2027, target completion date. The decisions the OIG identified are the sampled decisions it found to contain errors, a list it gave VBA on November 25, 2025; 60 of the 130 in-scope sampled decisions contained at least one error. The report does not say whether VBA will review decisions outside that list, does not describe what notice affected veterans would receive, and does not address collection, waiver, or offset of any overpayment. Reporting beyond what the report says would be speculation, so we are not going to do it. Anyone who receives correspondence from VA about their own rating is reading a document this report does not generate and does not describe.

In general terms, the regulations set out what a reduction looks like when VA pursues one. Section 3.105(e) requires a proposed-reduction rating, notice, and a 60-day window to submit additional evidence before a reduction takes effect, and § 3.500(b)(2) sets the effective date at the date of last payment where an erroneous award rested solely on administrative error or error in judgment, although § 3.105(a)(1)(ii) excepts reductions made under paragraph (e) from that rule, so which provision governs depends on the facts. Waiver of an overpayment is governed by 38 C.F.R. § 1.963 and 38 U.S.C. § 5302. The general rules for when a corrected effective date changes what is owed are the subject of our explainer on effective dates and retroactive VA payments. Whether any of that applies to a particular decision is a question about that decision, not about this report.

The published review lanes in 38 C.F.R. § 3.2500 remain available to any claimant who disagrees with a rating decision on its own terms: a supplemental claim at any time, or, within one year of the decision notice, a higher-level review or an appeal to the Board. Those routes existed before this report and are unchanged by it.

The one concrete, dated change to watch is the questionnaire. VBA's revised "Scars/Disfigurement" disability benefits questionnaire is tentatively scheduled for release on January 10, 2027. If it ships, examinations for painful scars conducted after that date will be documented on a form VBA says it revised to meet the OIG's recommendation that the questionnaire clearly elicit whether reported scar pain or tenderness is supported by objective evidence.

This is general legal information, not legal advice. It addresses a federal United States oversight report on VA disability compensation under 38 C.F.R. part 4 and reflects sources verified on October 23, 2026. It does not address state law or any other diagnostic code, and it is not advice about any individual rating, claim, or debt. Laws, regulations, and VA procedures change, and this story is developing; consult a lawyer or an accredited representative about your specific situation.

Related articles

• VA disability benefits: ratings, pay rates, and how claims work
• The three VA review options after a rating decision
• How VA math works: combined ratings explained
• VA disability back pay and effective dates
• VA claim sharks: the fee rules and how to verify anyone

---

Last updated: 2026-10-23. This is a developing story; details verified as of 2026-10-23.

Frequently Asked Questions

Does the VA OIG report mean my disability rating is being reduced?

No. The report published October 21, 2026 is an audit of how Veterans Benefits Administration claims processors applied 38 C.F.R. § 4.118. It is not a rating action and it names no veteran. Its recommendations run to the under secretary for benefits, and recommendation 1 asks VBA to review and correct the specific decisions the OIG identified, so a decision in that group can be revisited.

Does the report say veterans will have to repay the $5.6 million in overpayments?

The report does not address how any individual overpayment would be handled. It identifies $5,584,470 as a questioned cost under the Inspector General Act and asks VBA to review and correct the processing errors on the decisions the OIG team identified, with a February 28, 2027 target date. It says nothing about collection, notice, waiver, or offset.

What are the rating criteria for painful scars under diagnostic code 7804?

Under 38 C.F.R. § 4.118, diagnostic code 7804 assigns 30 percent for five or more scars that are unstable or painful, 20 percent for three or four, and 10 percent for one or two. Note (2) adds 10 percent if one or more scars are both unstable and painful. Note (1) defines an unstable scar as one where there is frequent loss of covering of skin over the scar.

What is pyramiding, and why does it apply to painful scars?

38 C.F.R. § 4.14, titled Avoidance of pyramiding, provides that the evaluation of the same disability under various diagnoses is to be avoided and that evaluating the same manifestation under different diagnoses is to be avoided. The OIG found claims processors sometimes assigned more than one diagnostic code 7804 on the same rating. Because 7804 is counted by the total number of painful scars, all of them belong under a single 7804.

How many decisions did the OIG actually review?

The OIG reviewed a stratified random sample of 130 in-scope decisions completed between June 1, 2024, and May 31, 2025, consisting of 70 grants of service connection and 60 increased evaluations. It then projected those results to an estimated population of about 47,300 decisions. It separately reviewed 30 decisions from January 2026 and found a 43 percent error rate in that smaller follow-up sample.

Did VBA agree with the OIG's findings?

VBA concurred with all three recommendations in a September 8, 2026 memorandum and concurred with the draft report findings. It filed one technical comment arguing that the projections should be read against the underlying case review, noting that 7 of the 60 errors found in the sample resulted in improper payments. The OIG made no material change to its sampling or projections.

Has VA fixed anything yet?

Partly, according to the report. VBA added a pyramiding conflict notification for multiple diagnostic codes 7804 to the VBMS-Rating application on April 19, 2026, and demonstrated it to the OIG team on July 13, 2026, so the OIG made no recommendation on that point. The OIG closed recommendation 2 after VBA revised the Scars/Disfigurement questionnaire, which VBA tentatively scheduled for release on January 10, 2027. Recommendations 1 and 3 remain open with target dates of February 28, 2027 and January 31, 2027.

Were any veterans underpaid because of these errors?

The report states the OIG did not find any errors that resulted in underpayments in its sample. It did identify one error category, projected at about 1,328 decisions, where a 0 percent evaluation under diagnostic code 7805 was deleted rather than kept alongside a new 7804 evaluation. The OIG said that caused no underpayment but could affect future disability evaluations.

What are the published options for disagreeing with a VA rating decision?

Under 38 C.F.R. § 3.2500, a claimant may file a supplemental claim under § 3.2501 at any time after notice of a decision, or, within one year of that notice, elect a higher-level review under § 3.2601 or an appeal to the Board of Veterans' Appeals under § 20.202. Section 3.2500(b) prohibits electing two of those options concurrently on the same issue. This is a description of the regulation, not a recommendation about any individual claim.