Carissa Thompson v. Pete Hegseth

Opinion

USCA4 Appeal: 24-2050 Doc: 73 Filed: 07/31/2026 Pg: 1 of 47

UNPUBLISHED

UNITED STATES COURT OF APPEALS FOR THE FOURTH CIRCUIT

No. 24-2050

CARISSA L. THOMPSON,

Plaintiff – Appellant,

v.

THE HONORABLE PETE HEGSETH, in his official capacity as Secretary of Defense; KENDALL FRANK, III; PHYSICAL DISABILITY BOARD OF REVIEW,

Defendants – Appellees.

-----------------------------

MILITARY-VETERANS ADVOCACY INC.,

Amicus Supporting Appellant.

Appeal from the United States District Court for the District of Maryland, at Greenbelt. Deborah K. Chasanow, Senior District Judge. (8:23-cv-02458-DKC)

Argued: January 28, 2026 Decided: July 31, 2026

Before QUATTLEBAUM, RUSHING, and BENJAMIN, Circuit Judges.

Affirmed by unpublished opinion. Judge Rushing wrote the majority opinion, in which Judge Quattlebaum joined. Judge Benjamin wrote a dissenting opinion. USCA4 Appeal: 24-2050 Doc: 73 Filed: 07/31/2026 Pg: 2 of 47

ARGUED: Brian A. Kulp, DECHERT, LLP, Philadelphia, Pennsylvania, for Appellant. Rebecca Ann Koch, OFFICE OF THE UNITED STATES ATTORNEY, Baltimore, Maryland, for Appellees. ON BRIEF: Steven A. Engel, Washington, D.C., Christopher J. Merken, Philadelphia, Pennsylvania, Michael A. Losco, Biaunca S. Morris, DECHERT LLP, New York, New York; Rochelle Bobroff, NATIONAL VETERANS LEGAL SERVICES PROGRAM, Arlington, Virginia, for Appellant. Erek L. Barron, United States Attorney, OFFICE OF THE UNITED STATES ATTORNEY, Baltimore, Maryland, for Appellees. John B. Wells, MILITARY-VETERANS ADVOCACY INC., Slidell, Louisiana; Melanie L. Bostwick, Anne W. Savin, Washington, D.C., Melanie R. Hallums, ORRICK, HERRINGTON & SUTCLIFFE LLP, New York, New York, for Amicus Curiae.

Unpublished opinions are not binding precedent in this circuit.

2 USCA4 Appeal: 24-2050 Doc: 73 Filed: 07/31/2026 Pg: 3 of 47

RUSHING, Circuit Judge:

United States Air Force veteran Carissa Thompson brought this lawsuit under the

Administrative Procedure Act against the Secretary of Defense, the Secretary of the Air

Force, and the Physical Disability Board of Review (PDBR). The PDBR, which no longer

exists, was a temporary entity created to review disability determinations for former

members of the armed forces who were separated from the forces due to a medical

condition during a certain time frame. After its review, the PDBR recommended no change

to Thompson’s disability rating, and she challenged that decision in this lawsuit. The

district court granted Defendants’ motion for summary judgment and denied Thompson’s

cross-motion for summary judgment, finding that Thompson had failed to demonstrate that

the decision not to modify her disability rating was unlawful. We affirm.

I.

A servicemember’s disability rating is relevant for determining the benefits she will

receive after leaving the armed forces. “A member of a regular component of the armed

forces entitled to basic pay,” 10 U.S.C. § 1201(c)(1), who is found “by the Secretary” of

the relevant military department to be “unfit to perform the duties of the member’s office,

grade, rank, or rating because of physical disability incurred while entitled to basic pay,”

may be “retire[d],” id. § 1201(a), or “separated,” depending on the degree of the member’s

disability, id. § 1203(a). A servicemember with fewer than 20 years of service may qualify

for retirement, with retirement pay and benefits, if the Secretary determines that the

servicemember’s “disability is at least 30 percent under the standard schedule of rating

disabilities in use by the Department of Veterans Affairs at the time of the determination.”

3 USCA4 Appeal: 24-2050 Doc: 73 Filed: 07/31/2026 Pg: 4 of 47

Id. § 1201(b)(3). But if the servicemember’s disability is determined to be less than 30

percent, she “may be separated from [her] armed force, with severance pay.” Id. § 1203(a).

The Air Force implements this statutory scheme through its Disability Evaluation

System. See Air Force Instruction (AFI) 36-3212 ¶ 1.1 (2006). Under that system,

attending physicians at medical treatment facilities conduct exams, “[p]repare the

documents required to identify medical defects or conditions that may disqualify the

member for continued active duty,” and then “[r]efer the case” to an Air Force Medical

Evaluation Board (MEB). Id. ¶ 2.2.1. Medical officers on the MEB then evaluate the

documentation, recommend the disposition of the case, and refer disability cases to an Air

Force Physical Evaluation Board (PEB). Id. ¶¶ 2.1, 2.2.2, 2.3.1. A PEB “is a fact-finding

body that investigates the nature, origin, degree of impairment, and probable permanence

of the physical . . . condition of any member whose case it evaluates.” Id. ¶ 3.1. If the PEB

“finds a member unfit, it recommends appropriate disposition based on the degree of

impairment caused by the disabling condition, the date incurred, and the member’s line of

duty status.” Id. The PEB assigns disability percentage ratings using the Department of

Veterans Affairs Schedule for Rating Disabilities (VASRD) and Department of Defense

Instruction 1332.39. Id. ¶ 3.18.

In 2008, as part of the Dignified Treatment of Wounded Warriors Act, Congress

created the PDBR “to review the disability determinations of covered individuals by

[PEBs].” 10 U.S.C. § 1554a(a)(1). Covered individuals are “members and former

members of the armed forces who, during the period beginning on September 11, 2001,

and ending on December 31, 2009 . . . are separated from the armed forces due to unfitness

4 USCA4 Appeal: 24-2050 Doc: 73 Filed: 07/31/2026 Pg: 5 of 47

for duty due to a medical condition with a disability rating of 20 percent disabled or less”

and “are found to be not eligible for retirement.” Id. § 1554a(b). Based on its review, the

PDBR could recommend to the Secretary “issuance of a new disability rating,”

“modification of the disability rating previously assigned,” “recharacterization of the

separation . . . to retirement,” or no change. Id. § 1554a(d). The Secretary could then

“correct the military records . . . in accordance with a recommendation made by the

[PDBR].” Id. § 1554a(e)(1). Action taken by the Secretary based on the PDBR’s

recommendation not to correct is final. Id. § 1554a(e)(3). The PDBR ceased operating in

October 2024. 1 See Memorandum, Office of the Under Secretary of Defense, Sunset of

Physical Disability Board of Review (July 9, 2024).

II.

Thompson entered the United States Air Force in November 2002. While she was

on authorized leave in July 2003, Thompson sustained an injury to her back when an all-

terrain vehicle she was riding overturned and landed on her. She was diagnosed with an

L1 vertebra burst fracture, placed in a back brace, and prescribed narcotics for pain. She

returned to military duty and received further treatment.

Around July 26, 2004, Thompson underwent an MEB examination. In February

2005, the MEB referred Thompson’s case to the PEB. The PEB determined that Thompson

was unfit for duty under diagnostic code 5235 (vertebral fracture or dislocation), assigned

1 Any requests for PDBR review by an Air Force veteran after that date must be referred to the Secretary of the Air Force for referral to the Air Force Board for Correction of Military Records. See Memorandum, Office of the Under Secretary of Defense, Sunset of Physical Disability Board of Review (July 9, 2024). 5 USCA4 Appeal: 24-2050 Doc: 73 Filed: 07/31/2026 Pg: 6 of 47

her a disability rating of 10%, and recommended discharge with severance pay. On May

23, 2005, Thompson was separated from the Air Force.

In August 2005, Thompson underwent Compensation and Pension (C&P) exams by

the Department of Veterans Affairs (VA). The VA assigned her a disability rating of 40%

for the L1 fracture.

In 2013, Thompson applied for PDBR review of her 10% disability rating from the

PEB. After review, the PDBR issued a written decision recommending no change in the

PEB adjudication and therefore no re-characterization of Thompson’s disability or

separation determination. The Air Force accepted the PDBR’s recommendation.

Thompson subsequently filed this lawsuit under the Administrative Procedure Act

(APA), challenging the Air Force’s decision, based on the PDBR’s recommendation, not

to change her disability rating so that she could be awarded a medical retirement instead of

severance pay. The district court granted Defendants’ motion for summary judgment.

Thompson v. Austin, No. 8:23-cv-02458, 2024 WL 4215726 (D. Md. Sept. 17, 2024).

Thompson timely appealed, and we have jurisdiction under 28 U.S.C. § 1291.

III.

We review the district court’s decision de novo, independently applying the same

standards as that court was required to apply. See Ren v. U.S. Citizenship & Immigr. Servs.,

60 F.4th 89, 93 (4th Cir. 2023); Nat’l Audubon Soc’y v. U.S. Army Corps of Eng’rs, 991

F.3d 577, 583 (4th Cir. 2021).

The APA instructs federal courts to “hold unlawful and set aside agency action,

findings, and conclusions found to be . . . arbitrary, capricious, an abuse of discretion, . . .

6 USCA4 Appeal: 24-2050 Doc: 73 Filed: 07/31/2026 Pg: 7 of 47

otherwise not in accordance with law,” or “unsupported by substantial evidence.” 5 U.S.C.

§ 706(2)(A), (E). Substantial evidence “is more than a mere scintilla” but “means only . . .

such relevant evidence as a reasonable mind might accept as adequate to support a

conclusion.” Biestek v. Berryhill, 587 U.S. 97, 103 (2019) (internal quotation marks

omitted).

An agency action “is arbitrary or capricious if ‘the agency relied on factors that

Congress has not intended it to consider, entirely failed to consider an important aspect of

the problem, offered an explanation for its decision that runs counter to the evidence before

the agency, or is so implausible that it could not be ascribed to a difference in view or the

product of agency expertise.’” Nat’l Aubudon, 991 F.3d at 583 (quoting Hughes River

Watershed Conservancy v. Johnson, 165 F.3d 283, 287–288 (4th Cir. 1999)). “‘[S]o long

as the agency provides an explanation of its decision that includes a rational connection

between the facts found and the choice made, its decision should be sustained.’” Id.

(quoting Am. Whitewater v. Tidwell, 770 F.3d 1108, 1115 (4th Cir. 2014)).

Our review under this standard “is ‘narrow’ and ‘highly deferential.’” Ren, 60 F.4th

at 93 (quoting Ohio Valley Env’t Coal. v. Aracoma Coal Co., 556 F.3d 177, 192 (4th Cir.

2009)). We may not substitute our judgment for that of the agency, FCC v. Fox Television

Stations, Inc., 556 U.S. 502, 513 (2009), but rather “‘look only to see if [the agency has

committed] a clear error of judgment,’” Ren, 60 F.4th at 93 (additional internal quotation

marks omitted) (quoting Hughes River Watershed, 165 F.3d at 287). “[W]e need only

‘ensure that the agency has acted within a zone of reasonableness and, in particular, has

reasonably considered the relevant issues and reasonably explained the decision.’” Id.

7 USCA4 Appeal: 24-2050 Doc: 73 Filed: 07/31/2026 Pg: 8 of 47

(brackets omitted) (quoting FCC v. Prometheus Radio Project, 141 S. Ct. 1150, 1158

(2021)). Moreover, we may “‘uphold a decision of less than ideal clarity if the agency’s

path may reasonably be discerned.’” Fox, 556 U.S. at 513–514 (quoting Bowman Transp.,

Inc. v. Ark.-Best Freight Sys., Inc., 419 U.S. 281, 286 (1974)). Although our inquiry into

the administrative record “is to be searching and careful, the ultimate standard of review is

a narrow one.” Citizens to Pres. Overton Park, Inc. v. Volpe, 401 U.S. 402, 416 (1971).

IV.

With this deferential standard of review firmly in mind, we consider Thompson’s

challenge to the PDBR’s recommendation to maintain her 10% disability rating.

Thompson “contends that the PDBR: (1) relied on examinations that were insufficient to

be considered under the VASRD; (2) wrongfully discounted the VA’s C&P exam;

(3) wrongfully disregarded exams containing range of motion observations that

contradicted the PDBR’s assessment; and (4) failed to resolve reasonable doubt in her

favor.” Thompson, 2024 WL 4215726, at *5. We address each argument in turn.

A.

First, Thompson argues that the PDBR improperly relied on exam reports from June

11, 2004, July 26, 2004, and February 1, 2005, that were inadequate under the VASRD.

Specifically, she contends each report failed to record her range of motion, failed to record

the degree at which her pain began during range-of-motion testing, or failed to adequately

evaluate her functional loss due to pain. We conclude that the PDBR did not act arbitrarily,

capriciously, or contrary to law in considering these exam reports.

8 USCA4 Appeal: 24-2050 Doc: 73 Filed: 07/31/2026 Pg: 9 of 47

The VASRD defines “[d]isability of the musculoskeletal system [as] primarily the

inability, due to damage . . . in parts of the system, to perform the normal working

movements of the body with normal excursion, strength, speed, coordination and

endurance.” 38 C.F.R. § 4.40 (2005). 2 “It is essential that the examination on which

ratings are based adequately portray the anatomical damage, and the functional loss, with

respect to all these elements.” Id. The functional loss “may be due to pain, supported by

adequate pathology and evidenced by the visible behavior of the claimant undertaking the

motion.” Id. “Weakness is as important as limitation of motion, and a part which becomes

painful on use must be regarded as seriously disabled.” Id.

An exam relied on for rating evaluation purposes must “consider[] and portray[]”

the “factors cited in § 4.40 . . . as to functional loss.” DeLuca v. Brown, 8 Vet. App. 202,

206 (Ct. Vet. App. 1995). In the context of diagnostic codes based upon a loss of range of

motion, where the veteran experiences pain on use, including during flare-ups, the medical

examiner must “express an opinion on whether pain could significantly limit functional

ability during flare-ups” or following repetitive use. Id. “[T]hese determinations should,

if feasible, be ‘portray[ed]’ in terms of the degree of additional range-of-motion loss due

to pain.” Id. (citation omitted) (quoting 38 C.F.R. § 4.40).

2 The PDBR was required to apply the VASRD “in effect at the time the covered individual’s disability rating was assigned[,] . . . along with all applicable statutes, and any directives in effect at the time of the contested separation (to the extent they do not conflict with the VASRD in effect at the time of the contested separation).” Dep’t of Def. Instruction (DoDI) 6040.44, encl. 3, ¶ 3(e) (2015). Disability determinations must be made using the VASRD and “any applicable interpretation of the schedule by the United States Court of Appeals for Veterans Claims.” 10 U.S.C. § 1216a(a)(A). 9 USCA4 Appeal: 24-2050 Doc: 73 Filed: 07/31/2026 Pg: 10 of 47

If an examination report “does not contain sufficient detail, it is incumbent upon the

rating board to return the report as inadequate for evaluation purposes,” 38 C.F.R. § 4.2

(2005), or “explain[] why such action [is] not necessary,” Mitchell v. Shinseki, 25 Vet. App.

32, 44 (Vet. App. 2011). At the same time, however, the PDBR must interpret examination

reports “in the light of the whole recorded history, reconciling the various reports into a

consistent picture so that the current rating may accurately reflect the elements of disability

present.” 38 C.F.R. § 4.2 (2005); see also 10 U.S.C. § 1554a(c)(2) (review by the PDBR

“shall be based on the records of the armed force concerned and such other evidence as

may be presented to the [PDBR]”); DoDI 6040.44, encl. 3, ¶ 3(d) (2015) (same).

1.

The June 2004 and July 2004 MEB exam reports were adequate for rating purposes.

Thompson’s diagnostic code is based on loss of range of motion, 38 C.F.R. § 4.71a (2005), 3

and she complained of pain on use during the challenged exams but did not report flare-

ups. Accordingly, to be adequate for rating purposes under DeLuca, the exams must

“express an opinion on whether pain could significantly limit functional ability,” and such

determinations should be portrayed “in terms of the degree of additional range-of-motion

loss due to pain on use” if feasible. DeLuca, 8 Vet. App. at 206.

3 In relevant part, under diagnostic code 5235, a 40% disability rating is warranted where “forward flexion of the thoracolumbar spine [is] 30 degrees or less.” 38 C.F.R. § 4.71a (2005). A 20% disability rating is warranted where forward flexion is greater than 30 degrees but not greater than 60 degrees. Id. And a 10% disability rating is warranted where forward flexion is greater than 60 degrees but not greater than 85 degrees; the individual experiences “muscle spasm, guarding, or localized tenderness not resulting in abnormal gait or abnormal spinal contour”; or there is a “vertebral body fracture with loss of 50 percent or more of the height.” Id. 10 USCA4 Appeal: 24-2050 Doc: 73 Filed: 07/31/2026 Pg: 11 of 47

Both the June and July exam reports noted that Thompson “has full range of motion

of the spine, being able to flex forward and touch her toes to the floor, going to full

extension with some minor tightness at the level of her injury” and “is also able to fully

lateral side bend bilaterally, but with focal midline pain at the level of her injury.” J.A.

205, 357. By stating that Thompson has “full range of motion of the spine,” the reports

implicitly opined that pain did not significantly limit her functional ability. And because

she had full range of motion, she did not have any degree of “range-of-motion loss due to

pain.” DeLuca, 8 Vet. App. at 206. In other words, the reports indicated that Thompson

had pain, but the pain did not cause her any range-of-motion loss.

Relying on Mitchell, Thompson contends that “to adequately portray functional loss

as required by the VASRD, examiners must note where pain begins during range-of-motion

testing,” Opening Br. 24, and thus, the June and July 2004 reports were inadequate because

they “failed to record where pain began upon movement,” id. at 26.

We disagree. “[P]ain itself does not constitute functional loss.” Mitchell, 25 Vet.

App. at 37. Pain “may result in functional loss, but only if it limits the ability ‘to perform

the normal working movements of the body with normal excursion, strength, speed,

coordination[, or] endurance.’” Id. at 38 (quoting 38 C.F.R. § 4.40). In the context of

diagnostic codes “based upon limitation of motion,” id. at 41, “[p]ainful motion alone is

not limited motion,” id. at 38. Mitchell requires that an exam make clear “whether and at

what point” the veteran experiences “any limitation of motion . . . specifically attributable

to pain.” Id. at 44 (emphasis added). An exam does so by recording a “finding as to the

degree of range-of-motion loss due to pain on use.” Id. (emphasis added). In other words,

11 USCA4 Appeal: 24-2050 Doc: 73 Filed: 07/31/2026 Pg: 12 of 47

Mitchell requires “examiners [to] note where pain begins during range-of-motion testing,”

Opening Br. 24, when pain limits motion, see Mitchell, 25 Vet. App. at 44 (noting that “[i]t

is important for the medical examiner to note this information so that the VA rating official

can have a clear picture of the nature of the veteran’s disability and the extent to which pain

is disabling,” which “will allow the Board to ensure that the disabling effects of pain are

properly considered when evaluating any functional loss due to pain that is attributable to

the veteran’s disability” (emphasis added)). Here, the June and July 2004 reports stated

that Thompson experienced pain and tightness upon motion but had the full range of

motion in her spine, making clear that she did not experience any limitation of motion

specifically attributable to pain.

2.

The February 2005 exam was inadequate for rating purposes. The exam report

noted that Thompson “continues to have pain that is worse with activity,” but it expressed

no opinion about whether pain could significantly limit her functional ability in terms of

the degree of range-of-motion loss due to pain, and it did not explain why this detail could

not feasibly be provided. J.A. 232; see DeLuca, 8 Vet. App. at 206.

Nevertheless, the PDBR did not err by considering the February 2005 report.

Whether a report is adequate for rating evaluation purposes and whether a report may be

considered by the PDBR are different questions. An examination report may be inadequate

for evaluation purposes because it lacks details required by Sections 4.40, 4.45, and 4.59.

See 38 C.F.R. § 4.2 (2005); Correia v. McDonald, 28 Vet. App. 158, 169 (Vet. App. 2016).

Yet the VASRD directs the PDBR to interpret examination reports “in the light of the

12 USCA4 Appeal: 24-2050 Doc: 73 Filed: 07/31/2026 Pg: 13 of 47

whole recorded history, reconciling the various reports into a consistent picture so that the

current rating may accurately reflect the elements of disability present.” 38 C.F.R. § 4.2

(2005); see also 10 U.S.C. § 1554a(c)(2) (review by the PDBR “shall be based on the

records of the armed force concerned and such other evidence as may be presented to the

[PDBR]”); DoDI 6040.44, encl. 3, ¶ 3(d) (2015) (same). The “whole recorded history”

will sometimes include exams that lack sufficient detail and thus, are inadequate for rating

purposes standing alone. Section 4.2 of the VASRD, however, does not prohibit the PDBR

from considering those exams along with other exams in the record that are adequate for

evaluation purposes. 4

Here, the PDBR interpreted the June and July 2004 exam reports “in the light of”

the February 2005 exam, the 2005 VA C&P exams, and other pertinent information from

Thompson’s medical history. 38 C.F.R. § 4.2 (2005). For example, the PDBR noted that

the June 2004 “MEB [Narrative Summary] examination indicated full back [range of

motion] with focal midline pain and ‘minor tightness’ of muscles at the level of the

[fracture].” J.A. 55. The PDBR “interpreted this description of [Thompson’s] [range of

motion] as reflecting [thoracolumbar range of motion] within functional limits.” J.A. 55

(emphasis added). The PDBR explained that “[t]his contention is supported by the second

4 Contrary to Thompson’s argument, Correia does not require that “exams without the required [range-of-motion] tests . . . be disregarded.” Reply Br. 17; see Correia, 28 Vet. App. at 170. Further, Thompson does not assert that “the range of motion testing described in the final sentence of [38 C.F.R.] § 4.59” “can be conducted on [her spine]”; therefore, she does not appear to be arguing that any of the exam reports she challenges are additionally inadequate because they failed to record the specific range-of-motion testing described in Section 4.59. Correia, 28 Vet. App. at 170. 13 USCA4 Appeal: 24-2050 Doc: 73 Filed: 07/31/2026 Pg: 14 of 47

orthopedic opinion in February 2005.” J.A. 55 (emphasis added). “No [range of motion]

of the [thoracolumbar] spine was documented at that visit, but the examiner documented

thorough muscle strength evaluation, including of the iliopsoas muscle, a hip flexor that

has its origins from the upper lumbar spine, with normal strength of all muscles noted and

no particular difficulty with pain or muscle spasm during strength testing was

documented.” J.A. 55. We conclude that the PDBR did not err by considering the February

2005 exam as it did.

B.

Second, Thompson contends that the PDBR unlawfully failed to “give particular

consideration” to her August 3, 2005 VA C&P exam. Opening Br. 33. That exam

diagnosed Thompson with an “L1 fracture . . . in 2003 with chronic low back pain and

decreased range of motion wearing a large, lumbosacral brace.” J.A. 118. The exam report

stated:

It was most difficult . . . to examine her back since she had this large brace on, but she was very limited in all ranges of motion. Forward flexion goes from 0 to 30 degrees with pain in the entire lumbar spine at 30 degrees, minus 60 degrees secondary to pain. Extension was 0 to 10 degrees with pain in the entire spine at 10 degrees, minus 20 degrees secondary to pain. Left and right lateral flexion was from 0 to 10 degrees, minus 20 degrees secondary to pain. Left and right lateral rotation is 0 to 25 degrees with pain in the entire spine, lumbar at 25 degrees minus 20 degrees secondary to pain. Active range of motion did not produce any weakness, fatigue, or incoordination.

J.A. 117.

The PDBR must “[c]ompare any VA disability rating for the specifically military-

unfitting condition(s) with the PEB combined disability rating” and “[c]onsider any

14 USCA4 Appeal: 24-2050 Doc: 73 Filed: 07/31/2026 Pg: 15 of 47

variance in its deliberations and any impact on the final PEB combined disability rating,

particularly if the VA rating was awarded within 12 months of the former Service

member’s separation.” DoDI 6040.44, encl. 3, ¶ 4(a)(5) (2015).

Here, the PDBR satisfied that standard. The PDBR noted that “[a]t the post-

separation C&P examination[, Thompson’s] [range of motion] was recorded as much

worse, but [she] had a rigid back brace on and it remained in place during the

examina[ti]on.” J.A. 55. The PDBR found that the thoracolumbar range of motion

“obtained while in a brace designed to restrict [range of motion] was not useful for the

purpose of providing a rating according to the VASRD” and accordingly “placed greater

probative value on the orthopedic exams” in 2004 and 2005 for its rating recommendation.

J.A. 55. The PDBR reasoned that “[t]he evidence in [the] record indicated that [Thompson]

was able to ambulate normally with or without the brace and was noted to have full [range

of motion], with aggravation of back pain by activity.” J.A. 55. Although it acknowledged

that the VA had assigned Thompson a 40% disability rating, the PDBR concluded that a

rating higher than 10% was not warranted based on range-of-motion criteria. J.A. 52, 55.

We agree with the district court that, by “providing an explanation for the [range-of-

motion] variance recorded in the C&P examination, compared to prior examinations, the

PDBR satisfied” its requirement to weigh the VA’s determination and explain why it

reached a different conclusion. Thompson, 2024 WL 4215726, at *7.

Thompson argues that the PDBR’s assessment “was arbitrary and capricious and

unsupported by evidence” because (1) the range of motion measurement in the VA C&P

exam “was based on [her] pain threshold, not the effects of the back brace,” (2) the PDBR

15 USCA4 Appeal: 24-2050 Doc: 73 Filed: 07/31/2026 Pg: 16 of 47

“wrongly asserted that the back brace was the cause of the pain that limited [her] range of

motion,” Opening Br. 34, and (3) “it was appropriate for [her] to wear the brace during the

C&P exam” because “[e]valuative ratings under the VASRD must be based on the

veteran’s ‘function under the ordinary conditions of daily life including employment,’” id.

at 37 (quoting 38 C.F.R. § 4.10).

Thompson’s arguments do not persuade. To begin, it is not entirely clear from the

C&P exam report whether Thompson’s range of motion was restricted by pain alone or by

both pain and her back brace. See J.A. 118 (C&P exam report noting “chronic low back

pain and decreased range of motion wearing a large, lumbosacral brace”), 68 (VA rating

decision noting “diagnosis was L1 fracture in service with chronic low back pain and

decreased range of motion while wearing a large lumbosacral brace”).

Next, the PDBR never asserted that the back brace caused Thompson pain that

limited her range of motion. Rather, the PDBR reasoned that the thoracolumbar range of

motion “obtained while in a brace designed to restrict [range of motion] was not useful”

for rating purposes. J.A. 55. Thompson does not dispute that her brace is designed to

restrict range of motion. And it is rational to conclude that a brace intended to restrict

range of motion may render range-of-motion measurements less helpful when worn during

an examination. Accordingly, it was not arbitrary or capricious for the PDBR to conclude

that range-of-motion measurements obtained while Thompson was wearing a brace

designed to reduce range of motion were not useful for rating purposes. See Am.

Whitewater, 770 F.3d at 1115 (holding that “so long as the agency provides an explanation

16 USCA4 Appeal: 24-2050 Doc: 73 Filed: 07/31/2026 Pg: 17 of 47

of its decision that includes a rational connection between the facts found and the choice

made, its decision should be sustained” (internal quotation marks and brackets omitted)).

That conclusion is not altered by the fact that, at that time, Thompson wore her back

brace every day. The VASRD provides that “[t]he basis of disability evaluations is the

ability of the body . . . to function under the ordinary conditions of daily life.” 38 C.F.R.

§ 4.10 (2005). “This imposes upon the medical examiner the responsibility of furnishing,

in addition to the etiological, anatomical, pathological, laboratory and prognostic data

required for ordinary medical classification, full description of the effects of disability upon

the person’s ordinary activity.” Id. (emphasis added). The medical examiner’s obligations

under Section 4.10 to provide the data for ordinary medical classification plus a description

of the disabling effects upon ordinary activity does not undermine the PDBR’s explanation

about why it found the VA range-of-motion measurements less useful for understanding

Thompson’s medical condition.

Thompson also asserts that “even absent the [range-of-motion] measurement, the

C&P exam report provided ample evidence to support the VA’s 40% disability rating.”

Opening Br. 37. But she makes no effort to show how her complaints of “chronic low back

pain” and difficulty sleeping, her morphine usage, or the statement that her condition

“definitely interferes with her daily activities in that she can hardly do anything” entitles

her to a higher disability rating under the VASRD’s highly specific criteria. 5 Id. (internal

5 Under diagnostic code 5235, a 40% disability rating is warranted where one of the following three criteria is met: (1) “[u]nfavorable ankylosis of the entire cervical spine”; (2) “forward flexion of the thoracolumbar spine [is] 30 degrees or less”; or (3) “favorable

17 USCA4 Appeal: 24-2050 Doc: 73 Filed: 07/31/2026 Pg: 18 of 47

quotation marks and emphasis omitted); cf. 38 C.F.R. § 4.71a (2005). By contrast, the

PDBR considered her “reported eight incapacitating episodes” and “radiating pain and

some numbness and tingling” and concluded that they did not warrant a higher disability

rating based on the criteria of VASRD Section 4.71a. 6 J.A. 55.

C.

Third, Thompson contends that the PDBR “arbitrar[ily] disregard[ed]” four other

exams in the record and so failed “to ‘consider all of the relevant evidence.’” Opening Br.

27–28 (quoting Valles-Prieto v. United States, 159 Fed. Cl. 611, 618 (Fed. Cl. 2022)).

Specifically, she identifies exams conducted on September 25, 2003, October 21, 2003,

December 13, 2004, and March 17, 2008.

When making a disability benefits determination, the Air Force focuses on the

veteran’s disability at the time of separation. Coleman v. Kendall, 74 F.4th 610, 613 (4th

ankylosis of the entire thoracolumbar spine.” 38 C.F.R. § 4.71a (2005). By contrast, a 10% rating is warranted where one of the following six criteria is met: (1) “[f]orward flexion of the thoracolumbar spine [is] greater than 60 degrees but not greater than 85 degrees”; (2) “forward flexion of the cervical spine [is] greater than 30 degrees but not greater than 40 degrees”; (3) “combined range of motion of the thoracolumbar spine [is] greater than 120 degrees but not greater than 235 degrees”; (4) “combined range of motion of the cervical spine [is] greater than 170 degrees but not greater than 335 degrees”; (5) “muscle spasm, guarding, or localized tenderness not resulting in abnormal gait or abnormal spinal contour”; or (6) “vertebral body fracture with loss of 50 percent or more of the height.” Id. 6 Under the alternative “[f]ormula for [r]ating [i]ntervertebral [d]isc [s]yndrome [b]ased on [i]ncapacitating [e]pisodes,” a 40% disability rating is warranted based on “incapacitating episodes having a total duration of at least 4 weeks but less than 6 weeks during the past 12 months,” whereas a 10% rating is warranted based on “incapacitating episodes having a total duration of at least one week but less than 2 weeks during the past 12 months.” 38 C.F.R. § 4.71a (2005). 18 USCA4 Appeal: 24-2050 Doc: 73 Filed: 07/31/2026 Pg: 19 of 47

Cir. 2023); see Ward v. United States, 133 Fed. Cl. 418, 431 (Fed. Cl. 2017) (holding that

“the relevant time for a determination of whether [a veteran] is entitled to military disability

benefits is when [she] was separated from the service”). As the district court observed,

“[t]he September 2003 and October 2003 examinations took place more than 12 months

before [Thompson’s] May 23, 2005, separation. Similarly, the March 2008 examination

occurred more than 2 years after [her] separation.” Thompson, 2024 WL 4215726, at *8.

We agree with the district court that “[t]he length of time between [her] separation and

these exams makes them less probative than the examinations done closer to [her]

separation date.” Id. Accordingly, the PDBR did not act arbitrarily or capriciously by

failing to account for the range-of-motion measurements in these exams.

As for the December 2004 exam, the PDBR discussed it in detail. As the PDBR

described, the exam occurred during a pain medicine clinic appointment (which resulted in

increasing her dosage) and recorded “flexion to 20” degrees. J.A. 231. For Thompson’s

diagnostic code, a 40% disability rating is warranted where “forward flexion of the

thoracolumbar spine [is] 30 degrees or less.” 38 C.F.R. § 4.71a (2005). In finding that a

disability rating higher than 10% was not warranted based on range-of-motion criteria, the

PDBR “placed greater probative value on the orthopedic exam[]” reports from June and

July 2004 and February 2005. J.A. 55. The PDBR observed that the June and July 2004

MEB reports, both performed by an “orthopedic specialist,” reflected thoracolumbar range

of motion “within functional limits.” J.A. 55. That contention, the PDBR found, was

“supported by the second orthopedic opinion in February 2005,” which “documented

thorough muscle strength evaluation, including of the iliopsoas muscle, a hip flexor that

19 USCA4 Appeal: 24-2050 Doc: 73 Filed: 07/31/2026 Pg: 20 of 47

has its origins from the upper lumbar spine, with normal strength of all muscles noted and

no particular difficulty with pain or muscle spasm during strength testing was

documented.” J.A. 55. That evidence, the PDBR concluded, “indicated that [Thompson]

was able to ambulate normally with or without the brace and was noted to have full [range

of motion], with aggravation of back pain by activity.” J.A. 55.

The PDBR’s failure to specifically distinguish the 20-degree flexion measurement

from the December 2004 pain medicine clinic appointment when discussing the range-of-

motion measurements in the June and July 2004 MEB orthopedic reports and the

subsequent February 2005 orthopedic evaluation does not render its decision arbitrary or

capricious. By regulation, the PDBR had a specific obligation to consider and compare

variances in proximate VA ratings, see DoDI 6040.44, encl. 3, ¶ 4(a)(5) (2015), but there

was no similar obligation to specifically compare and address measurements from other

exams. Rather, the PDBR was required to “reconcil[e] the various reports into a consistent

picture so that the current rating may accurately reflect the elements of disability present,”

38 C.F.R. § 4.2 (2005), and it satisfied that obligation. The PDBR reviewed and discussed

numerous exams, including pain clinic visits, with varying evaluations over the course of

Thompson’s medical history, but it emphasized the June and July 2004 MEB reports and

the February 2005 report, which it stressed were based on orthopedic exams, performed

during orthopedic clinic appointments, by orthopedic surgeons. Ultimately, the PDBR

“examined the relevant data and provided an explanation” that demonstrates “‘a rational

connection between the facts found and the choice made,’” thereby satisfying the APA’s

deferential standard of review. Ohio Valley, 556 F.3d at 192 (additional internal quotation

20 USCA4 Appeal: 24-2050 Doc: 73 Filed: 07/31/2026 Pg: 21 of 47

marks omitted) (quoting Motor Vehicle Mfrs. Ass’n v. State Farm Mut. Auto Ins. Co., 463

U.S. 29, 43 (1983)); see Fox, 556 U.S. at 513–514 (We may “uphold a decision of less than

ideal clarity if the agency’s path may reasonably be discerned.” (internal quotation marks

omitted)); Coleman, 74 F.4th at 620 (holding that PDBR decision was not arbitrary or

capricious for “placing more weight on” one evaluation rather than on a different

evaluation “that was closer to [the veteran’s] retroactive discharge date”).

D.

Finally, Thompson contends that the PDBR did not resolve reasonable doubt in her

favor, contrary to Sections 4.3 and 4.7 of the VASRD.

Section 4.3 instructs that “[w]hen after careful consideration of all procurable and

assembled data, a reasonable doubt arises regarding the degree of disability such doubt will

be resolved in favor of the claimant.” 38 C.F.R. § 4.3 (2005). “Reasonable doubt” exists

when there is “an approximate balance of positive and negative evidence which does not

satisfactorily prove or disprove the claim.” 38 C.F.R. § 3.102 (2005). “It is a substantial

doubt” and “not a means of reconciling actual conflict or a contradiction in the evidence.”

Id. Where there is no reasonable doubt, Section 4.3 does not apply. See Gilbert v.

Derwinski, 1 Vet. App. 49, 55–56 (Ct. Vet. App. 1990) (“[I]f a fair preponderance of the

evidence is against a veteran’s claim, it will be denied and the ‘benefit of the doubt’ rule

has no application . . . .”); Peterson v. United States, 104 Fed. Cl. 196, 209 (Fed. Cl. 2012)

(“Because of the lack of ‘an approximate balance of positive and negative evidence which

d[id] not satisfactorily prove or disprove’ plaintiff’s claim, 38 C.F.R. § 3.102, the ‘benefit

of the doubt’ doctrine is not applicable in this case[.]” (citing Gilbert, 1 Vet. App. at 56));

21 USCA4 Appeal: 24-2050 Doc: 73 Filed: 07/31/2026 Pg: 22 of 47

Mayhue v. Shinseki, 24 Vet. App. 273, 282 (Vet. App. 2011) (“[W]here, as here, the Board

concludes that the evidence is not in equipoise, specific consideration of § 4.3 is not

warranted.”).

Section 4.7 requires that “[w]here there is a question as to which of two evaluations

shall be applied, the higher evaluation will be assigned if the disability picture more nearly

approximates the criteria required for that rating. Otherwise, the lower rating will be

assigned.” 38 C.F.R. § 4.7 (2005).

The PDBR concluded that Thompson’s case did not give rise to a reasonable doubt

about the degree of disability. And assuming without deciding that Section 4.7 is triggered

here, the PDBR did not find that Thompson’s disability picture more nearly approximated

the criteria for a higher rating. The PDBR “agreed that a 10% rating of the back condition

was supported by the criteria of ‘muscle spasm, guarding, or localized tenderness not

resulting in abnormal gait or abnormal spinal contour; or, vertebral body fracture with loss

of 50 percent or more of the height.’” J.A. 55 (quoting 38 C.F.R. § 4.71a). Relying on the

June and July 2004 and February 2005 exam reports, the PDBR concluded that

Thompson’s range of motion was “within functional limits.” J.A. 55. It acknowledged

that her range of motion “was recorded as much worse” during the VA C&P exam, but

placed “greater probative value on the orthopedic exams in June 2004 and February 2005

for its rating recommendation” because the range-of-motion measurement obtained during

the C&P exam while Thompson was wearing a back brace was not useful for rating

purposes. J.A. 55. And the PDBR “concluded that a higher rating than 10% was not

warranted” “based on [range-of-motion] criteria” because “[t]he evidence in [the] record

22 USCA4 Appeal: 24-2050 Doc: 73 Filed: 07/31/2026 Pg: 23 of 47

indicated that [Thompson] was able to ambulate normally with or without the brace and

was noted to have full [range of motion], with aggravation of back pain by activity.” J.A.

55. In other words, the PDBR found that the evidence “satisfactorily prove[d]” a 10%

disability rating. 38 C.F.R. § 3.102 (2005). Therefore, Section 4.3 did not apply, and

Section 4.7 did not require a higher rating.

Thompson argues that “[t]he evidence strongly supported the 40% disability rating

and plainly established a reasonable doubt concerning the PEB’s 10% rating.” Opening

Br. 41. She highlights exams recording a range of motion of less than 30 degrees and notes

in her VA C&P exam report stating that she has chronic low back pain and “can hardly do

anything.” Id. (internal quotation marks omitted). However, the PDBR was entitled to

weigh the evidence and find that it satisfactorily proved a 10% disability rating. “Equal

weight is not accorded to each piece of material contained in a record; every item of

evidence does not have the same probative value.” Gilbert, 1 Vet. App. at 57. And the

PDBR’s finding is supported by substantial evidence. See Biestek, 587 U.S. at 103

(substantial evidence is “such relevant evidence as a reasonable mind might accept as

adequate to support a conclusion”).

Nor does the evidence Thompson cites compel a finding that her disability picture

more nearly approximates the criteria required for a 40% rating. Her reports of chronic

low back pain and that she can hardly do anything do not approximate the VASRD criteria

that “forward flexion of the thoracolumbar spine [be] 30 degrees or less.” 38 C.F.R.

§ 4.71a (2005). As discussed above, three of the five range-of-motion measurements she

references are too remote from her separation to be probative, a fourth was not useful

23 USCA4 Appeal: 24-2050 Doc: 73 Filed: 07/31/2026 Pg: 24 of 47

because it was conducted while Thompson wore a rigid back brace, and the PDBR “placed

greater probative value on the orthopedic exam[]” reports from June and July 2004 and

February 2005. J.A. 55. The PDBR concluded that the various reports, reconciled and

interpreted in light of Thompson’s whole recorded history, supported a 10% disability

rating at the time of her separation, and that conclusion was not arbitrary or capricious. 7

See 38 C.F.R. § 4.2 (2005).

For the foregoing reasons, the district court’s judgment is

AFFIRMED.

7 Our colleague in dissent finds it “inconceivable” at “a commonsense level” that someone who was in an all-terrain vehicle accident and suffered a L1 vertebra burst fracture could “only be 10% disabled” approximately 21 months later. Diss. Op. 47. Of course, it was the PDBR’s role to reassess that medical determination applying the specific criteria of the VASRD. Our role is to ensure that the PDBR did its duty—which it did—not to substitute our own medical judgment for that of the agency. 24 USCA4 Appeal: 24-2050 Doc: 73 Filed: 07/31/2026 Pg: 25 of 47

DEANDREA GIST BENJAMIN, Circuit Judge, dissenting:

The majority decision upheld the Physical Disability Board of Review’s (PDBR)

conclusion that Carissa Thompson was properly assigned a disability rating of 10%.

Because I think the PDBR’s decision was arbitrary and capricious, I respectfully dissent.

I. Facts and Relevant Background 1

A. Thompson’s Service and Life-Threatening Injury

Carissa Thompson served as an Airman First Class in the United States Air Force.

Thompson enlisted in the Air Force after graduating from high school in 2002. From

November 25, 2002, to May 23, 2005, she worked as a Signals Intelligence Analyst in the

Air Force. Thompson was recognized for her service, as she was awarded the Global War

on Terrorism Service Medal and an Air Force Organizational Award.

On July 31, 2003—just under nine months into her service as a Signals Intelligence

Analyst and shortly after her nineteenth birthday—Thompson sustained a life-threatening

injury. That day, Thompson was thrown off an all-terrain vehicle. And after she was

thrown off, the vehicle rolled over on top of her. Thompson suffered a burst fracture 2 of

1 I largely agree with the majority’s recitation of the facts. However, for purposes of context, I lay out key facts and medical examination reports that bear directly on the issues addressed in my analysis. 2 “A burst fracture is an injury in which the vertebra, the primary bone of the spine, breaks multiple directions . . . . A burst fracture usually results from significant trauma that compresses the bone, such as a motor vehicle accident or a severe fall.” Columbia University Irving Medical Center, BURST FRACTURE,

25 USCA4 Appeal: 24-2050 Doc: 73 Filed: 07/31/2026 Pg: 26 of 47

her L1 vertebra and was hospitalized for weeks. Upon release, Thompson returned to

military duty, even though she remained in a large back brace and was prescribed narcotics

to manage her pain. Her duties were limited, as she was “restricted from running, jumping,

climbing, crawling, repeated bending, repeated twisting, engaging in ergometry testing,

and engaging in sports.” J.A. 20. 3 Further, she was restricted from “lifting anything greater

than 15 [pounds], and standing more than 15 minutes per hour.” J.A. 21. Thompson

continued to serve for nearly two years. But because of her enduring pain and continuing

back problems, Thompson’s service ended. She was officially separated from the Air

Force on May 23, 2005.

B. Veteran’s Disability Benefits

A servicemember who is found to be unfit for continued military service due to

physical disability may be entitled to some type of benefit or pay. To be entitled to some

benefit or pay, the Physical Evaluation Board (“PEB”) must first decide whether the

servicemember is unfit for duty. See Air Force Instruction (AFI) 36-3212 ¶ 3.1 (2006).

Once that determination is made, the PEB assigns a disability rating (from 0% to 100%)

based on the Veterans Affairs Schedule for Rating Disabilities (“VASRD”), 38 C.F.R.

§§ 4.1, et seq., Department of Defense Instruction, and any applicable interpretations of

the VASRD by the U.S. Court of Appeals for Veterans Claims. AFI 36-3212 ¶ 3.18.

https://www.neurosurgery.columbia.edu/patient-care/conditions/burst-fracture [https://perma.cc/YT8R-ZBEN]. 3 Citations to “J.A.” refer to the joint appendix filed by the parties. The J.A. contains the record on appeal from the district court. Page numbers for citations to the J.A. utilize the “J.A. or J.A. #” numbering at the bottom of the page on each document. 26 USCA4 Appeal: 24-2050 Doc: 73 Filed: 07/31/2026 Pg: 27 of 47

The disability rating’s accuracy is crucial. That is because a servicemember

assigned a disability rating of 30% or higher would be considered a “military retiree,” 10

U.S.C. § 1201, which would entitle her to extensive benefits, including healthcare for

herself and her family, monthly military pension payments, and access to military bases’

medical facilities and commissaries. See, e.g., id. §§ 1401 (monthly retired pay), 2481

(commissaries, exchanges, and morale, welfare, and recreation benefits). By contrast, a

servicemember assigned a disability rating lower than 30% would be classified instead as

“medically separated,” and thus only awarded a one-time lump-sum severance payment

and without disability retirement or military retirement benefits. Id. § 1203(a)–(b); see

also id. § 1212(a).

What does it mean to be given a 30% disability rating? And what is the practical

difference between a 30% disability rating, and a 20% disability rating? The VASRD

provides a “General Rating Formula for Diseases and Injuries of the Spine,” which sets

forth the benchmarks for assigning a disability rating for spinal injuries. See 38 C.F.R.

§ 4.71a. Pertinently, a 30% disability rating means that a servicemember has “[f]orward

flexion of the cervical spine [that is] 15 degrees or less; or, favorable ankylosis of the entire

cervical spine.” Id. And a 20% disability rating means that a servicemember has

“[f]orward flexion of the thoracolumbar spine greater than 30 degrees but not greater than

60 degrees; or, forward flexion of the cervical spine greater than 15 degrees but not greater

than 30 degrees.” Id.

The VASRD’s rating formula places significant emphasis on a servicemember’s

range of motion. That emphasis raises another critical question: how does a rating board

27 USCA4 Appeal: 24-2050 Doc: 73 Filed: 07/31/2026 Pg: 28 of 47

determine exactly what a servicemember’s range of motion is? It must review a

servicemember’s medical reports and examinations. A rating board must “interpret reports

of examination in the light of the whole recorded history, reconciling the various reports

into a consistent picture so that the current rating may accurately reflect the elements of

disability present.” 38 C.F.R. § 4.2. And importantly, “[i]f a diagnosis is not supported by

the findings on the examination report or if the report does not contain sufficient detail, it

is incumbent upon the rating board to return the report as inadequate for evaluation

purposes.” Id.

C. Thompson’s Medical Reports and Examinations

On September 25, 2003, Thompson attended a physical therapy consultation with

staff physical therapist Michael Curtin. J.A. 212. During Curtin’s examination, he

recorded that Thompson’s lumbosacral spine range-of-motion for forward flexion was 15

degrees, which would equate to a 40% disability rating. Id.

On October 21, 2003, Thompson attended an orthopedic spine appointment. Dr.

David Christensen recorded that her flexion tolerated 30 degrees before she experienced

pain. See J.A. 197. Christensen noted that Thompson wore her back brace “most of the

time,” with low back pain and fatigue occurring approximately 15 minutes after removing

the brace. Id. Thompson added that “when she flexes forward she has [an] increase in low

back pain.” Id.

Thompson’s health progression was not linear. On March 26, 2004, Christensen

examined Thompson and reported that a chiropractic treatment she received shortly before

the examination “made things worse and increased her pain in her back.” J.A. 208.

28 USCA4 Appeal: 24-2050 Doc: 73 Filed: 07/31/2026 Pg: 29 of 47

Christensen further reported that Thompson had “quite a bit of axial back pain,” and that

his examination of Thompson showed that she had “pain with forward flexion.” Id.

Christensen did not note Thompson’s range of motion. Id.

Christensen continued treating Thompson, seeing her again on June 11 and July 26,

2004. J.A. 205, 357. There were no noted differences between the two summer

examinations. During both examinations, Thompson stated that she “had persistent pain

since coming out of the brace, primarily centered at the area of injury.” J.A 205, 357.

Further, while Thompson was “able to walk approximately ten minutes, limited by back

pain,” she felt “pain and discomfort within 30 to 60 minutes of awakening and getting up

in the mornings.” J.A 205, 357. And her “pain persist[ed] throughout the day as long as

she is upright.” J.A 205, 357. Upon physical examination, Christensen noted that she had

“full range of motion of the spine, being able to flex forward and touch her toes to the floor

and going to full extension with some minor tightness at the level of her injury.” J.A 205,

357. In both examination reports, however, Christensen failed to make any range of motion

measurements in degrees.

Thompson’s condition worsened. On December 13, 2004, she visited a pain clinic

where staff anesthesiologist Dr. Christopher Frandrup noted that Thompson could only

tolerate 20 degrees of forward flexion. J.A. 231. Frandrup further reported that

Thompson’s pain was exacerbated by activity, including walking and standing, and that

her pain had worsened over the past six months. Id. Thompson also stated that she suffered

from a tingling sensation in her low back, radiating outward. Id.

29 USCA4 Appeal: 24-2050 Doc: 73 Filed: 07/31/2026 Pg: 30 of 47

On February 1, 2005, Thompson met with spine specialist Dr. Steven Cyr. J.A. 232.

Cyr reported that Thompson “continues to have pain that is worse with activity,” and that

she was “on a Fentanyl patch for her pain control.” Id. Though Cyr did record that

Thompson’s “motor is 5/5 in bilateral lower extremities, in the iliopsoas, quadriceps,

hamstrings, tibialis anterior, extensor hallucis longus, peroneals, and gastrocnemius

solius,” id., Cyr did not measure the forward flexion of Thompson’s back or any range of

motion.

On August 3, 2005, the United States Department of Veterans’ Affairs (VA)

conducted a Compensation and Pension (“C&P”) examination. J.A. 112–20. The C&P

report stated that Thompson occasionally used a wheelchair and could take only “a few

steps for short distances, satisfying bathroom needs, etc., but she [had] considerable pain

and [felt] numbness in both legs.” J.A. 113. Thompson noted that she had “chronic low

back pain, constant, every day, 24 hours a day,” id., and that “pain [was] getting worse

instead of better,” J.A. 116. Physician Assistant Jim Kokel measured Thompson’s range

of motion, with her back brace in place during the examination. He noted that Thompson’s

“[f]orward flexion goes from 0 to 30 degrees with pain in the entire lumbar spine at 30

degrees, minus 60 degrees secondary to pain.” J.A. 117.

On March 17, 2008, the VA conducted a second C&P examination. J.A. 129–30.

Thompson “noted a tingling sensation in her anterior thighs and legs” which felt “like a

muscle tremor or ‘ants.’ ” J.A. 129. This time, Thompson’s forward flexion was measured

at “less than 10 degrees, and more in the range of 5 degrees,” with pain noted on “all

motions.” J.A. 130.

30 USCA4 Appeal: 24-2050 Doc: 73 Filed: 07/31/2026 Pg: 31 of 47

D. Thompson’s Disability Review Process and Administrative Appeal

On February 17, 2005, the Medical Examination Board (“MEB”) issued a report

referring Thompson’s case to a PEB. J.A. 23, 431. And on April 6, 2005, the PEB

determined that Thompson was unfit for duty due to the fracture of her L1 vertebra and

assigned her a disability rating of 10%. J.A. 347. Because Thompson was assigned a 10%

disability rating, she was “medically separated,” and did not receive any long-term benefits.

J.A. 431. Instead, she was given a one-time lump-sum payment of $6,190.80. Id.

On August 25, 2005, the VA made its own rating decision and assigned Thompson

a disability rating of 40% “for forward flexion of the thoracolumbar spine of 30 degrees or

less; or, favorable ankylosis of the entire thoracolumbar spine” under Code 5235. J.A. 68.

In rendering its decision, the VA considered “Service Medical Records from September

13, 2002 through February 17, 2005,” the August 2005 C&P examination, and an

additional August 11, 2005 examination. Id. And in Thompson’s second C&P

examination on March 17, 2008, the VA retained the 40% disability rating for her injury.

J.A. 129–30.

On November 19, 2013, Thompson requested that the PDBR review her 10%

disability rating in light of her new 40% rating by the VA. J.A. 26. It is within the PDBR’s

jurisdiction to review the applications of veterans who were medically separated between

September 11, 2002, and December 31, 2009, with a disability rating under 30%. 4 See 10

4 Congress enacted the PDBR to combat the prevalence of under-ratings that deprived veterans of their earned benefits and reduced the government’s corresponding liabilities. See National Defense Authorization Act for Fiscal Year 2008, Pub. L. No. 110-

31 USCA4 Appeal: 24-2050 Doc: 73 Filed: 07/31/2026 Pg: 32 of 47

U.S.C. § 1554a(a)–(b). On April 29, 2016, the PDBR concluded that Thompson’s 10%

disability rating was appropriate. J.A. 55. The Air Force Secretary affirmed the PDBR’s

decision, J.A. 51, making the agency decision final.

II. Procedural History

Thompson then brought this action in federal court, seeking judicial review under

the Administrative Procedure Act (“APA”). J.A. 488. Thompson specifically challenged

the PDBR’s decision declining to alter her disability rating. Id. Both parties filed summary

judgment motions. Id. The district court granted the Government’s summary judgment

motion, thereby dismissing Thompson’s case in full. Thompson v. Austin, No. 23-cv-2458,

2024 WL 4215726 (D. Md. Sept. 17, 2024).

The majority writes to affirm the district court’s order upholding the PDBR’s

recommendation on Thompson’s disability rating. For the reasons below, I disagree.

III. Standard of Review

In APA cases, “[w]e review the district court’s evaluation de novo, independently

assessing whether the agency action was unlawful.” Vanda Pharms., Inc. v. Ctrs. for

181, 122 Stat. 3, tit. XVI, §§ 1601–76. In 2007, the chairman of the Veterans’ Disability Benefits Commission testified that military departments frequently assigned lower disability ratings than the VA for the same servicemember, even though both the VA and the military departments purportedly relied on the same disability rating criteria (i.e., the VASRD). Hearing to Receive Testimony on the Department of Defense and Veterans Affairs Disability Rating Systems and the Transition of Servicemembers from the Department of Defense to the Department of Veterans Affairs, S. Hrg. 110–212, at 104 (Apr. 12, 2007). 32 USCA4 Appeal: 24-2050 Doc: 73 Filed: 07/31/2026 Pg: 33 of 47

Medicare & Medicaid Servs., 98 F.4th 483, 490 (4th Cir. 2024) (cleaned up). We set aside

an agency’s action if it is “arbitrary, capricious, an abuse of discretion, . . . otherwise not

in accordance with law,” or “unsupported by substantial evidence.” 5 U.S.C. § 706(2)(A),

(E).

“[A]n agency decision is arbitrary and capricious if ‘the agency has relied on factors

which Congress has not intended it to consider, entirely failed to consider an important

aspect of the problem, offered an explanation for its decision that runs counter to the

evidence before the agency, or is so implausible that it could not be ascribed to a difference

in view or the product of agency expertise.’ ” Sierra Club v. Dep’t of the Interior, 899 F.3d

260, 293 (4th Cir. 2018) (quoting Motor Vehicle Mfrs. Ass’n of the U.S., Inc. v. State Farm

Mut. Auto. Ins. Co., 463 U.S. 29, 43 (1983)). In other words, the “APA’s arbitrary-and-

capricious standard requires that agency action be reasonable and reasonably explained.”

F.C.C. v. Prometheus Radio Project, 592 U.S. 414, 423 (2021). While this standard of

review is deferential, it “is not meant to reduce judicial review to a ‘rubber-stamp’ of

agency action.” Ohio Valley Env’t Coal. v. Aracoma Coal Co., 556 F.3d 177, 192 (4th Cir.

2009) (quoting Ethyl Corp. v. EPA, 541 F.2d 1, 34 (D.C. Cir. 1976)). A “court must

nonetheless engage in a ‘searching and careful’ inquiry of the record.” Id. (quoting Citizens

to Preserve Overton Park v. Volpe, 401 U.S. 402, 416 (1971)).

After conducting such an inquiry, I think that the PDBR’s decision was arbitrary

and capricious and that the majority’s decision is a rubber stamp to the agency’s action.

33 USCA4 Appeal: 24-2050 Doc: 73 Filed: 07/31/2026 Pg: 34 of 47

IV. Applicable Law

As the APA instructs, an agency’s decision must comport with the applicable laws.

See U.S.C. § 706(2)(A). The PDBR was required to review the disability rating of a

“covered individual in accordance with the VASRD,” Department of Defense instruction,

and “any applicable interpretation of the [VASRD] by the United States Court of Appeals

for Veterans Claims,” DoDI 6040.44 Enc. 3 § 4(f)(1) (first quotation), 10 U.S.C.

§ 1216a(A) (second quotation). It did not do so here.

The VASRD defines disability as the inability “to perform the normal working

movements of the body with normal excursion, strength, speed, coordination, and

endurance” due to damage or infection in the musculoskeletal system. 38 C.F.R. § 4.40.

Beyond the VASRD’s rating formula, see id. § 4.71a, “[i]t is essential that the examination

on which ratings are based adequately portray the anatomical damage, and the functional

loss, with respect to all these elements.” Id. Such functional loss “may be due to pain,

supported by adequate pathology and evidenced by the visible behavior of the claimant

undertaking the motion.” Id.; see also DeLuca v. Brown, 8 Vet. App. 202, 206 (Vet. App.

1995) (requiring a military review board to consider whether pain could significantly limit

functional ability during flare-ups or with repeated use over a period of time). Painful

motion with joint or periarticular pathology is considered disabling. 38 C.F.R. § 4.59; see

also Burton v. Shinseki, 25 Vet. App. 1, 2–3 (Vet. App. 2011) (Section 4.59 applies to joint

conditions other than arthritis). Further, disability evaluations must be based on the

servicemember’s “function under the ordinary conditions of daily life including

employment.” 38 C.F.R. § 4.10. Again, “[i]f a diagnosis is not supported by the findings

34 USCA4 Appeal: 24-2050 Doc: 73 Filed: 07/31/2026 Pg: 35 of 47

on the examination report or if the report does not contain sufficient detail, it is incumbent

upon the rating board to return the report as inadequate for evaluation purposes.” Id. § 4.2.

The PDBR is also required to “[c]ompare any VA disability rating for the

specifically military-unfitting condition(s) with the PEB combined disability rating” and

“[c]onsider any variance in its deliberations and any impact on the final PEB combined

disability rating, particularly if the VA rating was awarded within 12 months of the former

Service member’s separation.” DoDI 6040.44, Enc. 3, § 4(a)(5) (2015) (emphasis added).

If “there is a question as to which of two evaluations shall be applied, the higher evaluation

will be assigned if the disability picture more nearly approximates the criteria required for

that rating.” 38 C.F.R. § 4.7. Any reasonable doubt regarding the degree of disability must

“be resolved in favor of [the veteran].” Id. § 4.3.

V. Analysis

Thompson asserts that the PDBR’s decision was unlawful because it (A) relied on

examinations that did not record her range of motion or degree at which her pain began

during the range-of-motion testing; (B) did not give particular consideration to the VA’s

40% disability rating that was closer in time to Thompson’s separation from the Air Force;

and (C) did not resolve any reasonable doubt in favor of Thompson. Opening Br. at 19–

20. I address each contention in turn below.

A. The February 2005 Examination Report Lacked Sufficient Detail.

Thompson’s first contention turns on a threshold question: whether the relied-upon

examination reports contained sufficient detail for evaluation purposes. If they did not, the

35 USCA4 Appeal: 24-2050 Doc: 73 Filed: 07/31/2026 Pg: 36 of 47

PEB should not have considered them, as it was required to return them as inadequate under

38 C.F.R. § 4.2. Thompson specifically challenges the use of the February 2005, and the

June and July 2004 examination reports. I agree with Thompson that the February 2005

examination report did not contain sufficient detail. Thus, the PDBR acted contrary to law

by not returning the report as inadequate for evaluation purposes.

As stated above, § 4.71a requires a rating board to assign a specific disability rating.

That disability rating is based on a servicemember’s forward flexion and range of motion

limitations. See id. § 4.71a. While range of motion testing is crucial for accurate disability

ratings, a rating board’s obligations under the VASRD do not end with § 4.71a. See

DeLuca, 8 Vet. App. at 205–06 (holding that § 4.71a “does not subsume” other sections of

the VASRD and that holding that a medical examination was insufficient because it

“merely recorded the veteran’s range of motion at that time,” without addressing functional

loss due to pain with motion). “It is essential that the examination on which ratings are

based adequately portray the anatomical damage, and the functional loss, with respect to”

the inability “to perform the normal working movements of the body with normal

excursion, strength, speed, coordination, and endurance.” 38 C.F.R. § 4.40. Functional

loss “may be due to pain, supported by adequate pathology and evidenced by the visible

behavior of the claimant undertaking the motion.” Id. “Weakness is as important as

limitation of motion, and a part which becomes painful on use must be regarded as seriously

disabled.” Id.

Section 4.59 further provides that the intent of the VASRD “is to recognize painful

motion with joint or periarticular pathology as productive of disability.” Thus, the “joints

36 USCA4 Appeal: 24-2050 Doc: 73 Filed: 07/31/2026 Pg: 37 of 47

involved should be tested for pain on both active and passive motion, in weight-bearing

and nonweight-bearing and, if possible, with the range of the opposite undamaged joint.

Id. § 4.59. And these examinations must be viewed in light of “the ordinary conditions of

daily life.” Id. § 4.10.

Considering the pertinent sections of the VASRD and accompanying interpretations

of the United States Court of Appeals for Veterans Claims, the PDBR should have

“return[ed] the [February 2005] report as inadequate for evaluation purposes.” Id. § 4.2.

The examination report does not measure Thompson’s range of motion or forward flexion

in degrees, let alone make any reference to range of motion at all. See J.A. 232. There is

also no notation of “test[ing] for pain on both active or passive motion.” 38 C.F.R. § 4.59.

It is hard to understand how a rating board could review a disability rating (that is supposed

to be based on exact range-of-motion measurements) when the relied-upon examination

contained no mention of range of motion.

The PDBR concluded that Cyr’s “muscle strength evaluation” was an adequate

substitute for the lack of any range of motion testing. J.A. 55. But that conclusion is

unfounded. As a reminder, the PDBR affirmed the PEB’s 10% disability rating, which

means that the rating board believed that Thompson’s “[f]orward flexion of the

thoracolumbar spine [was] greater than 60 degrees but not greater than 85 degrees; or [the]

forward flexion of the cervical spine [was] greater than 30 degrees but not greater than 40

degrees.” See 38 C.F.R. § 4.71a. Cyr did note that Thompson’s “motor is 5/5 in the

bilateral lower extremities” and “in the iliopsoas.” J.A. 232. But it is not apparent how

the evaluation of Thompson’s muscular strength informs any of the VASRD’s criteria for

37 USCA4 Appeal: 24-2050 Doc: 73 Filed: 07/31/2026 Pg: 38 of 47

a 10% disability rating. Further, Cyr noted that Thompson “continues to have pain that is

worse with activity.” J.A. 232. That point is significant because the “5/5” strength

evaluation does not indicate when pain occurred during movement, which is relevant under

§ 4.40. See id. § 4.40 (noting functional loss “may be due to pain . . . and [is] evidenced

by the visible behavior of the claimant undertaking the motion”); see also Mitchell v.

Shinseki, 35 Vet. App. 32, 43–44 (2011).

The majority and I agree that the “February 2005 exam was inadequate for rating

purposes.” Maj. Op. at 12. Our disagreement is whether the PDBR erred by considering

it. See id. at 12–13. The majority reasons that the PDBR did not err because the VASRD

directs the PDBR “to interpret examination reports ‘in the light of the whole recorded

history, reconciling the various reports into a consistent picture so that the current rating

may accurately reflect the elements of disability present.’ ” Id. (quoting 38 C.F.R. § 4.2).

But that directive does not relieve the VASRD’s independent directive to return an

inadequate medical report. Section 4.2’s plain text is clear: “if the report does not contain

sufficient detail, it is incumbent upon the rating board to return the report as inadequate for

evaluation purposes.” 38 C.F.R. § 4.2 (emphasis added). The VASRD’s use of the word

“the” denotes a specific report. See MERRIAM-WEBSTER DICTIONARY,

https://www.merriam-webster.com/dictionary/the (defining the word “the” “as a function

word to indicate that a following noun . . . is a unique or a particular member of its class”);

see also Freytag v. C.I.R., 501 U.S. 868, 901–02 (1991) (Scalia, J., concurring) (reasoning

that “the Courts of Law” does not generally refer to “Courts” or “Courts of Law”

generally). At bottom, if a report does not contain detail, it must be returned as inadequate

38 USCA4 Appeal: 24-2050 Doc: 73 Filed: 07/31/2026 Pg: 39 of 47

for evaluation purposes. Thus, the PDBR did not act in accordance with § 4.2 because it

did not return the insufficiently detailed February 2005 examination report.

B. The PDBR’s Conclusion Ran Counter to the Evidence Before it.

Even if each examination report was detailed enough to be relied on, the PDBR’s

affirmance of Thompson’s 10% disability rating still ran “counter to the evidence before”

it. Sierra Club, 899 F.3d at 293. The PDBR “ignore[d] evidence that undercut[] its

judgment.” Genuine Parts Co. v. EPA, 890 F.3d 304, 312 (D.C. Cir. 2018) (citing Morall

v. DEA, 412 F.3d 165, 179–80 (D.C. Cir. 2005)). And “it may not minimize such evidence

without adequate explanation.” Id.

A servicemember’s disability rating must serve as “a snapshot of the service

member’s condition at the time of separation from the service.” Ward v. United States,

133 Fed. Cl. 418, 431 (2017) (emphasis added) (quoting Stine v. United States, 92 Fed. Cl.

776, 795 (2010)). Additionally, the PDBR must “[c]ompare any VA disability rating for

the specifically military-unfitting condition(s) with the PEB combined disability rating”

and “[c]onsider any variance in its deliberations and any impact on the final PEB combined

disability rating, particularly if the VA rating was awarded within 12 months for the former

Service member’s separation.” DoDI 6040.44, Enc. 3, § 4(a)(5).

So, the closer a medical examination report is to a separation date, the more

probative it is. Thompson’s separation date was May 23, 2005. The examination reports

in order of proximity to Thompson’s separation date are as follows: August 2005, February

2005, December 2004, July 2004, June 2004, October 2003, September 2003, and March

2008.

39 USCA4 Appeal: 24-2050 Doc: 73 Filed: 07/31/2026 Pg: 40 of 47

At the outset, I agree with the majority that the PDBR’s limited reliance on the

October 2003, September 2003, and March 2008 examination reports makes sense. The

October and September 2003 examinations were conducted more than a year and a half

before Thompson’s separation date, and the March 2008 report came almost three years

after her separation. With that in mind, I discuss the five examinations closest to

Thompson’s separation date, from furthest to closest: June and July 2004, December 2004,

February 2005, and August 2005.

1. The June & July 2004 Examination Reports

In Thompson’s June and July 2004 examinations, it seemed like her range of motion

was improving from her October 2003 examination, where her forward flexion was limited

to 30 degrees. The report still noted significant issues, such as “persistent pain since

coming out of the [back] brace, primarily centered at area of her injury,” “pain and

discomfort within 30 to 60 minutes of awakening” which “persists throughout the day as

long as she is upright,” and “exacerbated [pain] by prolonged periods of standing or sitting

in one place.” J.A. 205. However, the report noted that Thompson had “full range of

motion of the spine . . . with some minor tightness at the level of her injury.” Id.

2. The December 2004 Examination Report

But Thompson’s back health did not continue to trend upward. In the December

2004 examination report, Frandrup noted that Thompson’s pain had “worsen[ed] over [the]

past six months,” and that pain was “exacerbated by activity,” including walking and

standing. J.A. 231. Thompson also reported that she suffered from a tingling sensation in

40 USCA4 Appeal: 24-2050 Doc: 73 Filed: 07/31/2026 Pg: 41 of 47

her low back, radiating outward. See id. The report stated that Thompson could now only

tolerate 20 degrees of forward flexion. Id.

3. The February 2005 Examination Report

While I have discussed why the PDBR should have returned the February 2005

examination as inadequate, even its findings support the fact that Thompson’s back pain

was worsening. The report indicated that Thompson “continue[d] to have pain that [was]

worse with activity.” Id. at 232. Because Thompson did not exhibit an improvement in

pain from December 2004 to February 2005, there is no reason to believe her range of

motion improved from December 2004 to February 2005.

It is safe to assume that her range of motion in February 2005 was closer to her

range of motion in the December 2004 report than the June and July 2004 reports. The

December 2004 report was just a few months removed from the February 2005 report and

indicated that Thompson’s pain had worsened over the prior six months, while the June

and July 2004 reports were much further in proximity from the February 2005 report and

were made before the downward trend in Thompson’s back health.

4. The August 2005 Examination Report

The last pertinent examination was in August 2005. It was the VA’s C&P

examination, and it was closest in time to Thompson’s May 23, 2005, separation date. The

report confirmed for the third time that Thompson’s “pain [was] getting worse instead of

better.” J.A. 113. She had “chronic low back pain, constant every day, 24 hours a day,”

making it difficult to sleep and ranging from a constant ache to a stabbing pain. Id. at 116.

Thompson also stated that she occasionally used a wheelchair and could take only “a few

41 USCA4 Appeal: 24-2050 Doc: 73 Filed: 07/31/2026 Pg: 42 of 47

steps for short distances,” and had “considerable pain and numbness in both legs.” Id. at

113. Kokel determined that her “[f]orward flexion goes from 0 to 30 degrees with pain in

the entire lumbar spine at 30 degrees.” Id. at 117. Although this forward flexion

measurement was measured while Thompson wore the back brace she used daily, nothing

in the report indicates that her range of motion has improved since the December 2004

examination.

* * *

To sum up, in June and July of 2004, Thompson’s range of motion seemed to fall

within healthy limits. See J.A. 205, 357. But in December 2004, Thompson’s forward

flexion decreased to 20 degrees, see id. at 231, and in August 2005, it was measured again

at 30 degrees, id. at 117. And the February 2005 examination that was between the

December 2004 and August 2005 examinations did not make any range of motion

measurements. See J.A. 232. So the most logical conclusion is that Thompson’s forward

flexion on her separation date was somewhere between 20 and 30 degrees, corresponding

to a 40% disability rating. Especially considering that the PDBR is supposed to give

particular consideration to VA ratings awarded within 12 months of a servicemember’s

separation. DoDI 6040.44, Enc. 3, § 4(a)(5).

The PDBR’s decision defies this logical reasoning, as it ignores the evidence in front

of it and fails to “articulate a satisfactory explanation for its action.” Motor Vehicle Mfgs.

Ass’n, 463 U.S. at 43. In concluding that the 10% disability rating was well-supported, the

PDBR reasoned that:

42 USCA4 Appeal: 24-2050 Doc: 73 Filed: 07/31/2026 Pg: 43 of 47

The MEB NARSUM examination was performed by an orthopedic surgeon and was performed on 11 June 2004 and the NARSUM was written in July 2004 by the same orthopedic specialist, with no change in the status of the back condition noted, approximately a year after the fx had occurred. The MEB NARSUM examination indicated full back ROM with focal midline pain and “minor tightness’ of muscles at the level of the fx. The Board interpreted this description of the CI’s ROM as reflecting TL ROM within functional limits. This contention is supported by the second orthopedic opinion in February 2005.

J.A. 55.

Yet noticeably missing is any mention of the December 2004 report. The PDBR

does not even attempt to articulate an explanation for Thompson’s worsening health and

range of motion as documented in the December 2004 report. Thus, its decision is arbitrary

and capricious because it “entirely failed to consider an important aspect of the problem.”

Sierra Club, 899 F.3d at 293. The PDBR likely ignored the December 2004 examination

because it “undercut[] its judgment.” Genuine Parts Co., 890 F.3d at 312. To affirm such

a judgment would amount to a rubberstamping of unlawful agency action.

The majority holds that the PDBR “satisfied its obligation,” Maj. Op. at 20, pointing

to the deferential standard of review that states that we may “uphold a decision of less than

ideal clarity if the agency’s path may reasonably be discerned,” id. at 21 (quoting FCC v.

Fox Television Stations, Inc., 556 U.S. 502, 513–514 (2009)). But I am unable to discern

the PDBR’s path as it completely omitted the December 2004 report from its reasoning.

See J.A. 55. It is not that the PDBR placed more weight on the June and July 2004 exams

rather than the December 2004 exam; it is that the PDBR failed to “grapple with contrary

evidence [i.e., the December 2004 exam]—disregarding entirely the need for reasoned

43 USCA4 Appeal: 24-2050 Doc: 73 Filed: 07/31/2026 Pg: 44 of 47

decisionmaking.” Crockwell v. Austin, 2024 U.S. Dist. LEXIS 55955, at *29–30 (D.D.C.

Mar. 28, 2024) (quoting Fred Meyer Stores, Inc. v. N.L.R.B., 865 F.3d 630, 638 (D.C. Cir.

2017)).

Setting aside the fact that the PDBR unlawfully erred by relying on the insufficient

February 2005 examination and failed to grapple with the December 2004 examination, I

turn specifically to the VA’s August 2005 C&P examination. I think that the PDBR also

erred by failing to give particular consideration to that examination. The PDBR stated:

At the post‐separation C&P examination the CI’s ROM was recorded as much worse, but the CI had a rigid back brace on and it remained in place during the examination[sic]. The Board consensus was that TL ROM obtained while in a brace designed to restrict ROM was not useful for the purpose of providing a rating according to the VASRD and, therefore Members placed greater probative value on the orthopedic exams in June 2004 and February 2005 for its rating recommendation.

This reasoning is again flawed because the PDBR continued to ignore the December

2004 examination, which recorded a range of motion that was even more severe than the

August 2005 examination. Further, the PDBR’s conclusion lacks rationale. It gave greater

probative value to the February 2005 examination than to the August 2005 examination,

even though the February 2005 exam recorded no range of motion at all and was farther

from Thompson’s separation date.

Thompson’s use of a back brace during her August 2005 range of motion test is

heavily disputed. The PDBR functionally disregarded the entire August 2005 examination

because of the back brace. See J.A. 55. But the language in the August 2005 examination

44 USCA4 Appeal: 24-2050 Doc: 73 Filed: 07/31/2026 Pg: 45 of 47

seems to contradict the PDBR’s decision to do so. Kokel wrote that it was “difficult” for

him “to examine her back since she had this large brace on, but she was very limited in all

ranges of motion.” J.A. 117 (emphasis added). Kokel’s statement, read plainly, shows that

he had difficultly examining her back, not difficulty finding that Thompson was limited in

all ranges of motion.

And as noted above, the VASRD requires examinations to look at functional loss

due to pain with motion, DeLuca, 8 Vet. App. at 206, in light of “the ordinary conditions

of daily life,” 38 C.F.R. § 4.10. The August 2005 examination report does exactly that.

Thompson’s use of a back brace was ordinary in her daily life, as she used it “every day.”

J.A. 117. And Kokel noted exact pain points: “[f]orward flexion goes from 0 to 30 degrees

with pain in the entire lumbar spine at 30 degrees.” Id.

And even if I were to agree that the August 2005 range of motion test warranted less

value, the remainder of the examination included ample evidence in line with the fact that

Thompson’s condition was worsening over time. As stated above, the report noted that

Thompson had “chronic low back pain, constant every day, 24 hours a day.” J.A. 116.

And that pain “definitely interferes with her daily activities in that she can hardly do

anything.” Id. at 117. The PDBR seemed to ignore this evidence entirely.

Thus, the PDBR’s failure to give particular consideration to the August 2005

examination’s range of motion testing and failure to grapple with the December 2004

examination amounted to arbitrary and capricious action.

45 USCA4 Appeal: 24-2050 Doc: 73 Filed: 07/31/2026 Pg: 46 of 47

C. The PDBR Failed to Resolve Reasonable Doubt Regarding the Degree of Disability in Favor of Thompson

Lastly, the PDBR had explicit instruction from the VASRD that when “reasonable

doubt arises regarding the degree of disability such doubt will be resolved in favor of the

claimant.” 38 C.F.R. § 4.3. The VASRD further provides that if “there is a question as to

which of two evaluations shall be applied, the higher evaluation will be assigned if the

disability picture more nearly approximates the criteria required for that rating.” Id. § 4.7.

The PDBR ignored both directives. In fact, the PDBR appeared to have resolved all

reasonable doubt against Thompson.

I will not recount Thompson’s medical examination timeline again, see Section I.C.,

but when “interpret[ing] reports of examination in light of the whole recorded history,” 38

C.F.R. § 4.2, it is clear that on Thompson’s separation date, her health and range of motion

was steadily declining. From December 2004 to March 2008, it was consistently recorded

that Thompson’s health was worsening. See J.A. 231 (noting that Thompson’s back

“worsen[ed] over [the] past 6 months”); J.A. 232 (noting that Thompson “continues to have

pain that is worse with activity”); J.A. 117 (noting that Thompson’s “low back pain has

continued to get worse”); J.A. 130 (noting that there was “pain with all motions”). The

evidentiary record and recorded range of motion tests more nearly approximate the VA’s

disability rating of 40%, rather than the PEB’s disability rating of 10%. And even if the

evidence and examination reports recounted above amount to some reasonable doubt, the

PDBR should have resolved it in Thompson’s favor. See 8 C.F.R. § 4.3. The PDBR’s

46 USCA4 Appeal: 24-2050 Doc: 73 Filed: 07/31/2026 Pg: 47 of 47

conclusion that Thompson had nearly a full range of motion defies logic, cannot be squared

with the evidence, and more importantly, flouts the VASRD.

VI. Conclusion

For the above reasons, I would have held that the PDBR’s decision was arbitrary

and capricious. Thompson’s disability rating of 10% ran counter to the evidence before

the PDBR. I would have reversed the district court’s summary judgment order and directed

it to remand Thompson’s case to the Air Force Board for Correction of Military Records

for proper re-evaluation of her disability rating. At a commonsense level, it is

inconceivable how a servicemember who was thrown from and run over by an all-terrain

vehicle—suffering a burst fracture in her L1 vertebra—could only be “10% disabled.”

With respect for my good colleagues, I dissent.