OPINION
In this military pay case, plaintiff seeks review of a decision of the Board for Correction of Naval Records (“BCNR”) denying an increase in plaintiff’s disability rating from 20% to 50%. The underlying issue is whether the decision of the BCNR is arbitrary, capricious or unsupported by substantial evidence. For the reasons stated below, plaintiff’s motion for summary judgment is denied, defendant’s cross-motion for summary judgment is granted and plaintiff’s complaint is to be dismissed.
FACTS
Plaintiff, formerly a Lieutenant Commander in the U.S. Navy, began active duty in June 1973. In May 1977, he was diagnosed by Navy physicians as having diabetes mellitus and placed on insulin maintenance therapy. On September 22, 1977, a Navy Medical Board concluded that plaintiff suffered from “diabetes mellitus and partially resolved right pereoneal diabetic neuropathy”, and referred his case to a physical evaluation board.
On October 14, 1977, the Central Physical Disability Evaluation Board (“CPEB”) found that plaintiff was unfit for duty, rated his disability at 20%, and recommended his separation with severance pay. Plaintiff appealed these findings and requested a formal hearing with a Regional Physical Evaluation Board (“RPEB”). On November 9, 1977, the RPEB concluded: “In view of the abrupt onset of this party’s disease with neuropathy, the initial dosage of insulin in excess of 40 units, and the party’s youth, he is afforded the benefit of doubt and rated 40% at this time.”
The RPEB’s recommendations and hearing record were then transmitted to the Physical Review Counsel (“PRC”). As the final level of Navy disability evaluation, the PRC reviews findings of both the CPEB and RPEB and enters final disposition on behalf of the Secretary of the Navy. The PRC concurred in the RPEB findings and, on December 2, 1977, directed that plaintiff be placed on the Temporary Disability Retired List (“TDRL”) so that his disability could be medically evaluated over a period of time, prior to ultimate disposition of his case.
Plaintiff underwent physical examinations by Navy physicians in June 19791 and December 1980. On the basis of the latter examination, the physician concluded that plaintiff’s condition was stable, permanent, and normal, except for his insulin dependent diabetes. On April 17, 1981, the CPEB found plaintiff unfit for duty due to diabetes mellitus, rated his disability at 20%, and recommended his separation from the Navy with severance pay.
*319Plaintiff sought a formal hearing before a RPEB maintaining that he should be assigned a 40% rating. On July 22, 1981, the RPEB held: “The Board was of the opinion that the member is able to maintain excellent stability with little difficulty; although he uses over 40 units of insulin, he is ratable at 20%.” Plaintiff submitted a letter in rebuttal to this finding and appealed to the PRC. The PRC concurred in the findings of the RPEB and directed that plaintiff be permanently separated from the Navy with severance pay. Plaintiff was so discharged on October 15, 1981.
Plaintiff filed this action on July 22, 1982. On December 21, 1982, the action was stayed after referral of the case to the BCNR. Before making its decision, the BCNR solicited comment from the Naval Council of Personnel Boards, who in turn referred it to the Physical Disability Review Board (“PDRB”). The PDRB considered the entire administrative record, including several items which had not previously been presented to any board. These items included: (1) a November 27, 1981 neurological evaluation of plaintiff; (2) a physical examination conducted by the Veterans Administration (“VA”) on December 31, 1981, reporting an insulin dependence of 55 units/day; (3) a 50% combined disability rating by the VA dated April 20, 1982 (40% for diabetes and 10% for neuropathy); and (4) an affidavit by plaintiff dated January 15, 1983, describing the problems he exhibited as a result of his disability.
On March 9, 1983, the PDRB found that plaintiff was properly ratable at 20% and concurred with the finding of the PRC. After receiving plaintiffs response to this decision, the PDRB reconsidered its previous holding and, on May 18, 1983, reaffirmed its recommendations of March 9, 1983. Upon reviewing all of the evidence, including the advisory opinions of the PDRB, the BCNR, on June 23,1983, denied plaintiff’s petition to correct his military records.
DISCUSSION
The standard of review in this type of case is well-settled. A determination by the Secretary of a military department that an individual is entitled to a particular disability rating will not be overturned unless plaintiff can show by clear and convincing evidence that the decision was arbitrary, capricious, unsupported by substantia] evidence, or contrary to applicable statutes and regulations. See Finn v. United States, 212 Ct.Cl. 353, 356, 548 F.2d 340, 342 (1977); Craft v. United States, 210 Ct.Cl. 170, 544 F.2d 468 (1976). Thus, to prevail, plaintiff must demonstrate “that the personnel involved ignored relevant and competent evidence, that they unreasonably construed the significant body of medical documents before them, or that in [some] other manner they failed to discharge their designated duties.” Stephens v. United States, 174 Ct.Cl. 365, 373, 358 F.2d 951, 955 (1966), quoted in Rutherford v. United States, 216 Ct.Cl. 163, 169, 573 F.2d 1224, 1226-1227 (1978).
Initially, plaintiff claims that the decision of the BCNR is arbitrary and capricious because the BCNR did not supply sufficient reasoning in writing to support its conclusion. Defendant rejoins that the BCNR implicitly adopted the well-reasoned opinion of the PDRB in denying plaintiff’s application. Defendant states that although “the Board’s decision fails to explicitly discuss the factual basis for its decision, the record supports that decision and allows meaningful review.”2
Ordinarily, agency decisions cannot be upheld under the “arbitrary and capricious” standard if the grounds therefor are not set forth in writing. See Istivan v. United States, 231 Ct.Cl. 671, 676, 689 F.2d 1034, 1038 (1982). In a military pay case, the Court of Claims noted that:
A naked conclusion and mere recitation that the opinion is based upon all of the evidence without an analysis of the evidence in writing ... is inimical to a rational system of administrative determination and ultimately inadequate ...
*320“In these circumstances [summary and sketchy findings and reasoning by the administrative Board] we cannot give as much deference to the Board’s determination as if it had given detailed findings to support, and fuller explanations of the reason for, its conclusion ...” [Beckham v. United States, 183 Ct.Cl. 628, 636, 392 F.2d 619 (1968) (Citations omitted).]
In the instant case, plaintiff points to the one-page letter from the BCNR as being wholly inadequate under this standard. Indeed, the Board’s June 23, 1983 letter, informing plaintiff of its decision, merely itemizes the evidence it considered, without engaging in any analytical discussion of the evidence. It is clear that the BCNR’s decision, standing alone, does not provide the required foundation for its decision. See Istivan, 231 Ct.Cl. 671, 689 F.2d 1034 (1982).
However, such an omission, by itself, will not support the conclusion that the BCNR’s decision was arbitrary and capricious. Rather, if the court can discern from the record that the Secretary had sufficient evidence before him to reach the conclusion he did, it must hold, based upon its limited scope of review, that the Secretary did not act arbitrarily or capriciously. See Selman v. United States, 723 F.2d 877, 881 (Fed.Cir.1983). This approach comports with the reasons underlying the requirement that written explanations be given for discretionary agency action, i.e., to inform the aggrieved person of the basic grounds for an adverse decision, and to enable a court to properly review the agency decision, if it is challenged. See Craft v. United States, 210 Ct.Cl. at 181, 544 F.2d 468; Dzialo v. United States, 5 Cl.Ct. 554 at 562-563 (1984).
The BCNR followed accepted procedures in seeking advisory opinions from the PDRB. On March 9, 1983, the PDRB issued a detailed analysis of plaintiff’s case and clearly laid a foundation for its conclusion that previous boards had rated plaintiff properly at 20%. Plaintiff was given an opportunity to respond and did so by letter of April 10, 1983. The PDRB then reconsidered its opinion and, on May 18, 1983, reaffirmed its earlier conclusion. Plaintiff was again given an opportunity to respond and did so by letter of June 20, 1983. This letter, together with the remainder of the administrative record then before the PDRB, was considered by the BCNR. It is reasonable to conclude that, in reaching its decision, the BCNR concurred with the reasoning of the PDRB. Thus, by virtue of the congruous decisions of the PDRB and the other administrative boards, plaintiff was clearly apprised of the basic grounds of the adverse decision, even though the BCNR did not articulate the reasons for its decision in its letter to plaintiff. Therefore, assuming the record provides a sufficient basis for the Secretary’s decision, the BCNR’s failure to articulate its underlying reasoning constitutes harmless error.
In arguing that the record does not provide a sufficient basis for the Secretary’s decision, plaintiff contends that the BCNR, as well as the other military boards which reviewed his case, unreasonably construed the significant body of evidence before it in assigning plaintiff a 20% disability rating. We do not agree. Congress established the Veteran’s Administration Schedule for Rating Disabilities (VA schedule) as the standard under which percentage determinations, pursuant to Title IV of the Career Compensation Act of 1949, are to be made. In addition, Department of Defense (“DOD”) Directive 1332.18 (September 9, 1968) sets forth the standards relating to separation from military service by reason of physical disability. The pertinent portion of the Department’s general rating policies regarding application of the VA schedule is set forth below:
2. Essentials of Evaluative Rating. The VA 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. The percentage ratings represent, as far as can practicably be determined, the average impair- *321 merit in earning capacity resulting from such diseases and, injuries and their residual conditions in civil occupations.
3. Higher of Two Evaluations. In view of the number of atypical instances, it is not expected especially with the more fully described grades, that all cases will show all the findings specified in the VA Schedule. However, findings sufficiently characteristic to identify the disease and the disability therefrom and, above all, coordination of rating with impairment of function is required in all instances. It is not the intent of the VASRD that there be a rigid requirement for the presence of all enumerated manifestations of a given disability. Those manifestations sufficiently and significantly representative of the entity and the severity of limitations imposed on the member are the only requirements. Where there is a question as to which of two percentage evaluations shall be applied, the higher evaluation will be assigned if the disability picture more nearly approximates the criteria for that rating. Otherwise the lower rating will be assigned. When after careful consideration of all reasonably procurable and assembled data, there remains a reasonable doubt as to which rating shall be applied, such doubt will be resolved in favor of the member. [DOD Directive 1332.18 (end. 3) (Emphasis supplied).]
It is clear from the Department’s enunciated policy that the VA schedules are not to be strictly construed, but are to be used as a guide in determining the individual’s impairment in earning capacity in civil occupations resulting from his disease(s). Plaintiff contends that the military boards which reviewed his case misinterpreted the VA disability rating schedule and assigned him an improper rating. The applicable rating schedule for diabetes mellitus is as follows:
Moderately severe; requiring large insulin dosage, restricted diet, and careful regulation of activities, i.e., avoidance of strenuous occupational and recreational activities — 40%
Moderate; with moderate insulin or oral hypoglycemic agent dosage and restricted (maintenance) diet; without impairment of health or vigor or limitations of activity — 20%
In addition, DOD Directive 1332.18 provides further guidance in assigning a rating for diabetes mellitus:
a. The severity of each case is to be individualized, taking into consideration complications, age of the member, and ease or difficulty in the control of blood sugar levels. By established practice, “large” insulin dosage has come to be regarded as “more than 40 units daily.” This may be used as a general guide, but not as the determining factor in assigning percentage ratings. It is quite possible for a member whose average insulin dosage is 30 or 35 units, but with unstable control requiring frequent hospital observation to be more disabled in fact than one on 45 units with steady blood sugar levels on a regimen of normal activity. [Emphasis supplied.]
In applying these guidelines to plaintiff’s situation, the CPEB, RPEB, PRC, and PDRB all concurred that plaintiff suffers no significant impairment of his earning capacity as a result of his diabetes and assigned him a 20% rating. These decisions were based on the extensive administrative record which clearly outlines the undisputed facts of the effects of plaintiff’s disability. Noting that plaintiff was required to maintain prudent regulation of his activities and curtail certain recreational activities, the PDRB agreed with the conclusion that, despite plaintiff’s large insulin dosage (55 units/day according to the VA physical exam of December 31, 1981), his diabetic condition best approximated the criteria for a 20% rating.
In addition, upon reviewing the evidence before the other boards and the November 27, 1981, neurological examination submitted by plaintiff to the BCNR, the PDRB found that plaintiff was ratable at 0% for *322his peripheral neuropathy.3 Therefore, the PDRB determined that plaintiffs previously awarded disability rating of 20% best reflected his overall disability.
We find that there is adequate evidence in the record to support such a finding. This court does not sit as a finder of fact, but rather has a very limited scope of review with respect to decisions of the BCNR. Although the decision to rate plaintiff at 20% instead of 40% or 50% may appear to the court to be a close question, the underlying facts are not sufficient to uphold a finding that the decision of the BCNR was arbitrary or capricious. At best, the evidence of record might establish a reasonable basis for disagreement with the BCNR’s decision; but such evidence is insufficient to justify overturning that decision. See Snell v. United States, 168 Ct.Cl. 219, 227 (1964). It is clear that, pursuant to DOD Directive 1332.18, the CPEB, RPEB, PRC, PDRB and BCNR each examined plaintiff’s overall disability as it affected his earning capacity in civil occupations and found that there was no significant impairment of his ability to earn a living as a civilian. None of the boards indicated any doubt regarding the propriety of their decisions. In the absence of reasonable doubt, there is no mandate to apply the higher of two ratings considered. See DOD Directive 1332.18 (end. 3) 113. This court will not substitute its judgment for that of the numerous military boards, with specific expertise in the area of disability rating, which are in accord on the evaluation of plaintiff’s condition.
Similarly, the VA rating of 50% and the Navy’s TDRL rating of 40% do not affect the propriety of the BCNR’s disability determination. The TDRL is designed to “safeguard the member from being permanently retired with a condition which may ... develop into a more serious permanent disability ... [Mjembers whose disabilities have not stabilized to a degree where permanent disposition is warranted will be placed on the TDRL ...” DOD Directive 1332.18. Once permanent disposition of the member’s case is achievable, the member may be removed from the TDRL and his final rating assigned. Plaintiff was placed on the TDRL on January 1, 1978 so that his condition could be evaluated over a period of time and his disability assigned the proper rating. In light of the purpose of placing a member on the TDRL, the initial rating of 40% assigned to plaintiff is not controlling in any subsequent disability evaluation. See, e.g., Newman v. United States, 185 Ct.Cl. 269 (1968).
Moreover, the BCNR was not bound to follow the VA’s determination that plaintiff is entitled to a 50% rating. As the Court in Finn v. United States, 212 Ct.Cl. 353, 357, 548 F.2d 340 (1977) stated:
[Although the VA's determination of a plaintiff’s condition may be relevant evidence, it is in no way binding upon the court nor conclusive on the issue of disability retirement. [Citations omitted.]
Both the PDRB and the BCNR considered the VA rating, but found it to be unpersuasive. There was nothing improper in that action, especially in light of the fact that there is sufficient evidence in the record to warrant a rating of 20%.
CONCLUSION
Because the decision of the BCNR is supported by an extensive administrative record, and is in agreement with numerous other administrative board evaluations, we find that plaintiff has failed to meet his burden of proof. See Gerber v. United States, 2 Cl.Ct. 311 (1983). Consequently, the decision of the BCNR is neither arbitrary, capricious, nor unsupported by substantial evidence, and the failure of the BCNR to articulate in writing the reasons for its decision is deemed to be harmless error.
For the above-stated reasons, plaintiff’s motion for summary judgment is DENIED, defendant’s cross-motion for summary judgment is GRANTED, and the complaint is to be DISMISSED.