Skip to main content


Reset A A Font size: Print

United States Court of Appeals,Federal Circuit.

Birdeye L. MIDDLETON, Claimant–Appellant, v. Eric K. SHINSEKI, Secretary of Veterans Affairs, Respondent–Appellee.

No. 2013–7014.

Decided: August 15, 2013

Before LOURIE, PLAGER, and TARANTO, Circuit Judges. Michael S. Sawyer, Covington & Burling LLP, of Washington, DC, argued for claimant-appellant. With him on the brief was Einar Stole. Of counsel on the brief were Barton F. Stichman and Katy Schuman Clemens, National Veterans Legal Services Program, of Washington, DC. Alex P. Hontos, Trial Attorney, Commercial Litigation Branch, Civil Division, United States Department of Justice, of Washington, DC, argued for respondentappellee. With him on the brief were Stuart F. Delery, Principal Deputy Assistant Attorney General, Jeanne E. Davidson, Director, and Scott D. Austin, Assistant Director. Of counsel on the brief were Michael J. Timinski, Deputy Assistant General Counsel, and Meghan Alphonso, Attorney, United States Department of Veterans Affairs, of Washington, DC.

Birdeye L. Middleton appeals from the decision of the United States Court of Appeals for Veterans Claims (the “Veterans Court”) affirming the decision of the Board of Veterans' Appeals (the “Board”) denying a disability rating in excess of 20% for his service-connected diabetes. See Middleton v. Shinseki, No. 10–4222, 2012 WL 2180580 (Vet.App. June 15, 2012) (unpublished). Because the Veterans Court did not err in interpreting the governing regulations and we lack jurisdiction to review the Veterans Court's application of the regulations to the facts, we affirm.


Middleton served on active duty from January 1964 to February 1990. He first sought compensation for his type II diabetes mellitus in October 2001. In July 2002, a Department of Veterans Affairs (“VA”) Regional Office (“RO”) granted service connection, assigning a disability rating of 20% pursuant to 38 C.F.R. § 4.119, Diagnostic Code (“DC”) 7913. See In re Middleton, No. 05–15 604, slip op. at 5 (Bd.Vet.App. Aug. 27, 2010). Middleton sought an increased rating in 2008, which the RO denied in March 2009 after a VA physical examination. Id. Middleton filed a timely Notice of Disagreement, and the RO issued a Statement of the Case (“SOC”) in December 2009. Id. Middleton then filed an appeal, and the RO issued a supplemental SOC in January 2010. Id.

In December 2009, the VA provided Middleton with a further physical examination that confirmed the diagnosis of type II diabetes mellitus. For that condition he was treated with three oral hypoglycemic agents and daily injections of the drug Byetta®. Id. at 6. Byetta® is a synthetic peptide that induces the body to secrete endogenous insulin. In August 2010, the Board again denied a rating increase despite Middleton's assertions that he met the criteria for a 40% rating on the grounds that his diet was restricted, his activities were regulated, and he used an oral hypoglycemic agent. Id.

The Board found that Middleton was only entitled to a 20% rating. It stated, “Though [Middleton] is on a restricted diet, has regulation of activities, and uses an oral hypoglycemic agent, he does not use insulin to regulate his diabetes.” In re Middleton, No. 05–15 604, slip op. at 4. The Board further found that neither Middleton's VA treatment records nor records from his private physician mentioned that he required insulin, and that treatment records from January through June of 2008 specifically referred to him as a non-insulin dependent diabetic. Id. at 6–7.

The Board ultimately found that, while Byetta® is a medication used to control diabetes, it is not insulin, and therefore the medical evidence of record showed that Middleton did not require insulin. Id. at 7. The Board emphasized that the “[u] se of insulin is a necessary element for the 40–percent rating; the fact that [Middleton] has not been required to use insulin thus precludes his being assigned this increased rating.” Id.

Middleton then appealed to the Veterans Court and again argued that he was entitled to a rating in excess of 20%. The court disagreed and affirmed the denial by the Board because medical evidence did not show that Middleton was prescribed insulin. Middleton, 2012 WL 2180580 at *2. Middleton argued that treatment with Byetta® injections was analogous to, yet admittedly not identical to, requiring insulin, but the court held that the plain language of DC 7913 recites “insulin” and does not include a supposed substitute. Id.

Middleton also argued that his diabetes was more closely related to the criteria for a 40% rating and that 38 C.F.R. § 4.7, which provides that the higher of two evaluations will be assigned if the veteran's disability picture more nearly approximates the criteria required for that rating, was applicable to his claim. Id. at *3. The Veterans Court nevertheless held that § 4.7 did not apply because, following its own precedent in Camacho v. Nicholson, 21 Vet.App. 360 (2007), a veteran could not be rated at 40% for diabetes when he only satisfied two of the criteria for that rating, as did Middleton. Id.

This appeal followed.


Our jurisdiction to review decisions of the Veterans Court is limited by statute. 38 U.S.C. § 7292. We “have exclusive jurisdiction to review and decide any challenge to the validity of any statute or regulation or any interpretation thereof [by the Veterans Court] ․ and to interpret constitutional and statutory provisions, to the extent presented and necessary to a decision.” Id. § 7292(c). We may not, however, absent a constitutional challenge, “review (A) a challenge to a factual determination, or (B) a challenge to a law or regulation as applied to the facts of a particular case.” Id. § 7292(d)(2). We therefore generally lack jurisdiction to review challenges to the Board's factual determinations or to any application of law to fact. See, e.g., Johnson v. Derwinski, 949 F.2d 394, 395 (Fed.Cir.1991). But we do have jurisdiction here to determine the proper interpretation of a regulation such as DC 7913. See Amberman v. Shinseki, 570 F.3d 1377, 1381 (Fed.Cir.2009) (exercising jurisdiction over review of Veterans Court's interpretation of regulation with rating schedule); Forshey v. Principi, 284 F.3d 1335, 1338 (Fed.Cir.2002) (en banc).

Section 4.119 of the VA regulations sets forth a schedule of disability ratings for diseases of the endocrine system. See 38 C.F.R. § 4.119. Within that schedule, Diagnostic Code 7913 prescribes ratings for diabetes mellitus. See id., DC 7913. The code recognizes five levels of disability, expressed in terms of percentages, which “represent as far as can practicably be determined the average impairment in earning capacity resulting from” the corresponding descriptions of a veteran's condition. See 38 C.F.R. § 4.1. The code reads as follows:

§ 4.119, DC 7913.


Middleton maintains that the Veterans Court misinterpreted the “[r] equiring insulin” criterion of the 20% and 40% ratings in DC 7913 as requiring the direct administration of insulin. He asserts that the term refers more generally to a need for insulin, not a specific method of obtaining it, and that—even though he was not administered the substance insulin exogenously—he still “requires insulin” because he takes other medications such as Byetta® injections that cause his body to secrete insulin endogenously. Middleton argues that the court's interpretation limiting the regulation to require a prescription for a specific medication is inconsistent with the benefits scheme and regulatory history, which focus on the severity of the impairment and how well a veteran's diabetes is controlled. Moreover, he contends that interpretations of ratings that rely on specific medications rather than impairments become obsolete as new drugs are introduced; therefore, any ambiguity should be resolved in favor of referencing symptoms, e.g ., whether insulin is needed for control regardless whether it is directly injected or endogenously created after administering a medication such as Byetta®. The Secretary argues that the plain language of DC 7913 unambiguously recites “[r]equiring insulin,” which should be given its ordinary meaning and does not encompass using an insulin-inducing drug as analogous to using insulin.

We conclude that the Veterans Court did not err in interpreting DC 7913, as its plain language reciting the criterion “[r]equiring insulin” for each of the 20% and 40% ratings clearly requires that the veteran is administered insulin. See Lockheed Corp. v. Widnall, 113 F.3d 1225, 1227 (Fed.Cir.1997) (“To interpret a regulation we must look at its plain language and consider the terms in accordance with their common meaning.”).

In contrast to the position taken by Middleton, the code does not authorize a 40% rating premised on the administration of another medical compound or pharmaceutical agent than the substance insulin, reserving that rating only for those circumstances “[r]equiring insulin, restricted diet, and regulation of activities.” 38 C.F.R. § 4.119, DC 7913. The context of the code also demonstrates that, when the VA intended to specify treatment for diabetes with another substance, it identified such treatment directly. Specifically, a 20% rating provides for two possibilities: “[r] equiring insulin and restricted diet, or; oral hypoglycemic agent and restricted diet,” which expressly distinguishes between treatment via administration of insulin and treatment via administration of an oral hypoglycemic agent. If, as Middleton asserts, those regulatory provisions are obsolete, then it is not for us to rewrite them.

As currently specified, “[r]equiring insulin” means being administered insulin. To read that criterion otherwise would be to ignore the plain language in the code that specifies alternative treatments: if requiring insulin does not mean administering insulin, then that criterion could arguably be satisfied by the alternative of administering an oral hypoglycemic agent or any other diabetes medication, and there would have been no reason for the Secretary to have expressly provided for such an alternative possibility. Accordingly, as a matter of interpretation, the Veterans Court did not err in holding that the “[r] equiring insulin” criterion of the 40% rating contemplated by DC 7913 means that the veteran must be administered insulin.


Middleton also argues that the Veterans Court's holding that satisfaction of the requiring insulin criterion is a necessary finding for a 40% rating conflicts with 38 C.F.R. § 4.21, which provides that “it is not expected ․ that all cases will show all the findings specified.” Appellant Br. 5, 30–33. He therefore maintains that the Veterans Court erred in holding that 38 C.F.R. § 4.7, concerning application of the higher of two evaluations, does not apply to his entitlement claim for a rating in excess of 20% for his service-connected diabetes.

Middleton asserts that his disability status more nearly approximates the criteria required for the 40% than the 20% rating of DC 7913 because control of his diabetes requires regulation of activities, which is only associated with ratings equal to or exceeding 40%. Middleton analogizes his situation to that of the claimant in Tatum v. Shinseki, 23 Vet.App. 152 (2009), which concerned an evaluation of the appropriate disability rating level for hypothyroidism under § 4.119, DC 7903. In that case, the Board initially determined that § 4.7 did not apply to the veteran's claim for entitlement to a 30% rating because it found that, although she had two of the three listed symptoms (fatigability and mental sluggishness), she did not suffer from the third requirement (constipation). Tatum, 23 Vet.App. at 154. The Veterans Court, however, held that § 4.7 was “necessarily ․ implicated,” set aside the Board's decision, and remanded for further consideration whether a 30% rating was more appropriate than a 10% rating, which required only fatigability or control via continuous medication. Id. at 156.

The Secretary responds that DC 7913 is a successive and cumulative rating schedule, necessitating that to warrant a 40% rating, a veteran must satisfy all of the criteria for that rating. The Secretary argues that the plain language use of the conjunctive “and” means that the three elements associated with the 40% rating are mandatory, consistent with our holding in Boyle v. Nicholson, 233 F. App'x 984 (Fed.Cir.2007), and the Veterans Court's holding in Camacho v. Nicholson, 21 Vet.App. 360 (2007). The Secretary contends that, because control of Middleton's diabetes does not require insulin—one of the three mandatory elements associated with the 40% rating—the Veterans Court did not err in concluding that § 4.7 did not apply.

The regulation at issue provides as follows:

Where there is a question as to which of two evaluations shall be applied, the higher evaluation will be assigned if the [veteran's] disability picture more nearly approximates the criteria for that rating. Otherwise, the lower rating will be assigned.

38 C.F.R. § 4.7.

At the outset, we note that determining whether Middleton's disability status more nearly approximates the 40% rating rather than the 20% rating requires an application of law to fact that is beyond our jurisdiction, as there is no constitutional issue presented. 38 U.S.C. § 7292(d)(2); Jackson v. Shinseki, 587 F.3d 1106, 1109 (Fed.Cir.2009).

As a matter of regulatory interpretation, however, the plain language of § 4.7 provides that the higher of two evaluations will be assigned only “[w]here there is a question as to which of two evaluations shall be applied.” But there is no question as to which evaluation shall be applied when a veteran does not satisfy all of the required criteria of the higher rating but does satisfy all of the criteria of the lower rating. We thus conclude that the Veterans Court did not err in its analysis of the unavailability of § 4.7 as a matter of law in this case because Middleton did not meet the “[r] equiring insulin” criterion of the 40% rating, given its plain meaning of “being administered insulin.”

Middleton's reliance on Tatum is misplaced. Aside from the fact that we are not bound by a decision of the Veterans Court, the Veterans Court itself distinguished Camacho in Tatum, recognizing that, in contrast to hypothyroidism ratings under DC 7903, diabetes ratings under DC 7913 involve successive criteria. Tatum, 23 Vet.App. at 155–56. We agree that the enumerated elements of DC 7913 required for a 40% rating are part of a structured scheme of specific, successive, cumulative criteria for establishing a disability rating: each higher rating includes the same criteria as the lower rating plus distinct new criteria. For example, a 10% rating is warranted when a veteran's diabetes is “[m] anageable by restricted diet only.” § 4.119, DC 7913. The restricted diet criterion is an element in each of the alternatives defining eligibility for the 20% rating, i.e., “[r]equiring insulin and restricted diet, or; oral hypoglycemic agent and restricted diet.” Id. And satisfaction of the in-the-alternative criterion for the 20% rating is required to obtain the 40% rating, to which is added the elements “[r] equiring insulin” and “regulation of activities.” Id.

As we held in Boyle, which we recognize was not precedential, use of the conjunctive “and” in the 40% rating of DC 7913 necessitates that there are three elements that a veteran must satisfy to warrant such a rating. Boyle, 233 F. App'x at 987 (citing Watson v. Dep't of the Navy, 262 F.3d 1292, 1299 (Fed.Cir.2001) (inclusion of conjunctive “and” in regulation indicated that all three enumerated criteria had to be demonstrated)). In contrast, the 20% rating uses the connector “or” to establish alternate factors. For the distinction between the ratings in this successive code to have any significance, we must give meaning to the “and” in the higher evaluation. Thus, because the 40% rating does not contemplate alternative considerations, a veteran must demonstrate all of the required elements in order to be entitled to that higher evaluation.

Accordingly, we discern no error by the Veterans Court with respect to the unavailability of § 4.7 to Middleton's claim for a rating in excess of 20% when it concluded that he could not be rated 40% disabled because there was no question that the higher evaluation did not apply when he only satisfied two of the required elements.


We have considered Middleton's remaining arguments and conclude that they are without merit. Because the Veterans Court did not err in interpreting the governing regulations, we affirm.



The medical experts who designed the VA disability rating system for veterans with service-connected disabilities had no difficulty in understanding that different people with the same illness do not always present the same symptoms, and that different people with the same illness do not always respond in the same way to the same drugs. They wisely built two foundational concepts into the rating schedule to provide for these understandings.

First, they stated at the outset that: “This rating schedule is primarily a guide in the evaluation of disability resulting from all types of diseases and injuries encountered as a result of or incident to military service.” 38 C.F.R. § 4.1 (emphasis added). In case anyone should miss the point that a guide is not a rigid diagnostic code or a treatment prescription, they added that the rating schedule should be interpreted broadly and in a manner that is veteran friendly. See 38 C.F.R. § 4.3 (quoted in full below).

Then, because even using their best efforts they could not anticipate the many ways that illnesses present, nor could they anticipate changes in treatment that new drugs might support, they added a second caveat to the “General Policy in Rating”:

Where there is a question as to which of two evaluations shall be applied, the higher evaluation will be assigned if the [veteran's] disability picture more nearly approximates the criteria for that rating. Otherwise, the lower rating will be assigned.

38 C.F.R. § 4.7.

In the case before us, the central issue is whether these foundational concepts apply to the rating schedule for diabetes mellitus. All agree that Mr. Middleton, who suffers from diabetes mellitus, has a compensable illness; the question is whether, under the rating guides in DC 7913, he is properly compensated at the 20% or 40% disability level. The record shows that his symptomology and his treatment regimen place him somewhere between the two descriptive guides for the two ratings; he does not fit squarely into either.

Mr. Middleton takes oral hypoglycemic agents, requires a restricted diet, regulates his activities, and receives daily injections of the drug Byetta®. He appears to meet all of the criteria for the 20% level, and all of the criteria for the 40% save one: insulin. Mr. Middleton argues that his Byetta injections are analogous to injections of insulin, thus putting him ‘nearly approximate’ to the 40% level. However that may be, and despite the majority's concern with all this, the facts of Mr. Middleton's particular case are irrelevant to the issue we must decide.

As the majority correctly notes, our question is not to which of these ratings Mr. Middleton is entitled. For us to determine whether Mr. Middleton satisfies every element of the 20% level and ‘approximately’ meets every element of the 40% level, or not, involves application of the law to the facts, which is beyond our jurisdiction under the peculiar standard of review Congress gave us over decisions of the Court of Appeals for Veterans Claims. Rather, the question of interpretation of VA law—over which we do have jurisdiction—is whether the foundational concepts set out in the provisions of the rating schedule quoted above apply to DC 7913, structured as it is (and presumably other provisions structured like it1 ).

The Government argues that when there is a successive and cumulative rating schedule, as here, the veteran seeking the higher rating can only obtain it if the language of the higher rating does not have specific requirements for that rating; if it does, the veteran must meet the requirements in haec verba. But the standard in § 4.7 is “more nearly approximates the criteria for that rating,” clearly providing that something approximating the criteria—not the criteria itself—is what to look for. If verbal compliance with the words of the guideline is what is required, § 4.7 has no meaning.

Recognizing the weakness in this argument, the Government couples it with the classic “read the statute” first argument. And it is certainly true that we judges spend much of our time interpreting statutes, seeking understanding of what the Congressional verbiage means by parsing the verbs and the nouns of a statute as if they contain some secret code that only we can penetrate. My colleagues, putting their interpretive skills to use, find in the stated rating schedule a controlling difference between the “and” in the 40% rating and the “or” in the 20% rating. Maj. Op. at 10. “For the distinction between the ratings in this successive code to have any significance, we must give meaning to the ‘and’ in the higher evaluation.” Id. The fact that the syntax and punctuation surrounding the “or” in the lower evaluation guide makes little grammatical sense is of no moment—the truth is in the words, and in the “plain language” of the ratings guide.

With due respect, the verbal statements in this ratings schedule are, as the regulations themselves state, only guides; calling it “the code,” as the majority frequently does, cannot change that fact. These are guides, not for the display of interpretive technique, but guides to what a sensible application of the two foundational concepts addresses. In a veteran-friendly system, what outcome is called for when the symptoms and prescribed treatment fall somewhere between the ratings, for example because a different drug—Byetta—is injected to treat the veteran's particular version of the illness rather than the standard drug—insulin, the one mentioned in the guide? Does that really turn on the difference between an “and” and a confusingly mistyped “or”? Should not the outcome instead respond to a common-sense analysis reflecting the illness and its treatment, and the purpose of a disability program for veterans who have honorably served their country?

I do not mean to say that the VA can begin its analysis of an application for benefits by setting aside criteria in the rating schedule. But if the VA's analysis reveals that the veteran's disability falls between two ratings, § 4.7 directs the VA to determine whether the disability picture more nearly approximates the criteria for the higher rating. If it does, § 4.7 honors substance over form by awarding the veteran the higher rating. The medical experts who designed the system wanted it that way. Indeed, as the “General Policy in Rating” tells us, “[i]n view of the number of atypical instances it is not expected, especially with the more fully described grades of disabilities, that all cases will show all the findings specified.” 38 C.F.R. § 4.21 (emphasis added). The argument that an illness with a “successive and cumulative rating schedule” is exempt from the ameliorative purposes of § 4.7 cannot be right. Nothing in DC 7913 suggests that the court should exempt it, or others like it, from § 4.7 and the clear policy of the ratings schedule.

Simply put, the majority's interpretation of § 4.7 is incorrect. The interpretation that it does not apply to provisions like DC 7913 offends the general policies and procedures understood by the medical profession. It snatches away the flexibility that the VA needs to battle the epidemic of diabetes and hands them a clipboard with a checklist. We should not hamper the VA's efforts to carry out their stated policy:

It is the defined and consistently applied policy of the Department of Veterans Affairs to administer the law under a broad interpretation, consistent, however, with the facts shown in every case. When 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.

The majority opinion cites a nonprecedential opinion for its legal support.2 However mistaken its authority, there is little this dissent can do to correct the disservice of the decisional outcome in this veteran's case. Nevertheless, it is to be hoped that the majority's treatment of veteran's law generally will be given the same weight as their nonprecedential authority, and that it will not be followed in future cases as a correct understanding of the law applicable more broadly to other such cases.

I respectfully dissent.

LOURIE, Circuit Judge.

Dissenting opinion filed by Circuit Judge PLAGER.

Copied to clipboard