PTSD VA claims: what the Board relied on in grants and denials

What VA requires

Under 38 CFR 3.304(f), PTSD service connection needs three things: a diagnosis that meets 38 CFR 4.125(a), medical evidence linking the symptoms to a stressor in service, and credible supporting evidence that the stressor happened. The rule relaxes the third for combat, fear of hostile military or terrorist activity, prisoners of war and in-service personal assault, where the veteran's own account, or evidence from outside the service records, can be enough.

Most claims are sent back first

Of 20,930 veterans whose PTSD claim the Board decided from 2020 to 2025, it granted 24.7%, denied 20.9% and remanded 54.4%, usually for a new examination or more records. The comparison below sets the 5,163 grants against the 4,383 denials.

In grants (5,163 veterans) In denials (4,383 veterans)

The event in service

• Stressor verified or conceded

  83%

  12%

• Combat service

  37%

  17%

• Military sexual trauma

  19%

  7%

• Service records discussed

  83%

  90%

Medical evidence

• Positive nexus opinion

  61%

  16%

• Favorable VA exam

  43%

  13%

• Private medical opinion

  50%

  30%

• Treating doctor's opinion

  19%

  6%

• Unfavorable VA exam

  65%

  81%

• Negative nexus opinion

  41%

  50%

Statements and credibility

• Veteran's own statement

  92%

  94%

• Buddy statement

  14%

  5%

• Board found the veteran credible

  65%

  12%

• Board doubted the veteran's credibility

  5%

  24%

How the Board weighed it

• Benefit of the doubt applied

  77%

  14%

Share of granted (green) and denied (gray) veterans whose decision contained each item, as extracted by AI from the decision text. Remanded claims are left out of this comparison. This shows what the decisions contained, not what caused the result.

The stressor is the dividing line

A verified or conceded stressor was in 83.2% of grants and only 12.5% of denials. Combat service (36.8% against 17.2%) and military sexual trauma (19.5% against 7.1%) were also more common in grants, since the rule eases the proof for both.

The stressor and the medical opinion work together:

Positive nexus opinion

No positive opinion

Stressor verified

85%

granted · 2,792 of 3,293

35%

granted · 1,503 of 4,307

Stressor not verified

19%

granted · 352 of 1,831

4%

granted · 516 of 11,499

One count per veteran; a remand counts as not granted.

With both, the Board granted 2,792 of 3,293 claims (84.8%). A verified stressor without a positive opinion was granted 34.9%, and a positive opinion without a verified stressor 19.2%. With neither, 4.5%.

Credibility, not just a statement

The veteran's own statement was in 91.6% of grants and 93.9% of denials, so its presence alone tells little. What differed was how the Board weighed it: it found the veteran credible in 65.0% of grants and 11.6% of denials, and doubted the veteran's credibility in 24.4% of denials.

Buddy statements were rare, but more common in grants (13.7%) than in denials (5.3%). A statement from someone who saw the event, or saw the change in the veteran after it, is the kind of supporting evidence the rule describes.

Medical evidence is weighed, not counted

A positive nexus opinion was in 60.9% of grants and 16.1% of denials. An unfavorable VA examination was common in both, including 65.1% of grants, because the Board weighs the opinions against each other. A private opinion was in 50.3% of grants.

What denials said was missing

For 3,689 denials, the decision summary noted what would have been needed to grant. It named a diagnosis in 59.0% of them, the stressor in 51.4% and a medical link in 44.2%; many named more than one. A common pattern: VA examiners diagnosed depression or anxiety, but not PTSD under the diagnostic criteria.

Read the decisions

Recent decisions, with the Board's reasoning as summarized by AI:

• A25111373, December 2025: granted; military sexual trauma corroborated enough to resolve reasonable doubt, with a favorable VA nexus opinion.
• A25110290, December 2025: denied; examiners diagnosed PTSD and linked it to service, but the Board found the stressors not corroborated and the veteran's statements alone not enough.
• A25109877, December 2025: denied; the stressor was not verified by military records or fellow service members.
• A25110021, December 2025: denied; examiners diagnosed generalized anxiety disorder and another trauma-related disorder, but not PTSD.

What the data cannot tell you

These are what the decisions contained, not what caused the result. Some items describe the Board's conclusion as much as the evidence: a credibility finding, a verified stressor or the benefit of the doubt are part of how the Board explains a grant. The evidence items were extracted by AI from each decision and apply to the whole decision, not one claim. Historical outcomes, not predictions.

For representatives

• Start with the diagnosis: make sure an examiner diagnosed PTSD under 38 CFR 4.125(a), not only depression or anxiety.
• Corroborate the stressor: unit records, buddy statements, or for personal assault, the outside sources and behavior changes 38 CFR 3.304(f)(5) lists.
• Pair the stressor with a medical opinion that addresses it by name; the combination is where grants concentrate.
• Expect a remand when the record is thin, and use it to fill the gap the Board names.