Bringing this action under ERISA, 29 U.S.C. § 1001, et seq., plaintiff-appellant Edward E. Lynd (Lynd) alleged in his complaint that the benefits he had been receiving pursuant to a long-term disability plan were wrongfully terminated. In his appeal of the district court’s rulings on the parties’ cross motions for summary judgment, Lynd presently contends that the district court reviewed the plan administrator’s decision to terminate these benefits under an inappropriate standard of review, and that the grant of summary judgment dismissing his suit was erroneous.
Facts and Proceedings Below
Lynd was employed by defendant-appellee Ford, Bacon & Davis, Inc. (FBD) on December 18, 1989. In September of 1990, Lynd became unable to work and began receiving short-term disability benefits under FBD’s Employee Welfare Benefit Plan (the plan). After six months, Lynd applied for and began receiving long-term disability benefits. The group policy associated with this long-term disability plan was issued by defendant-appellee Reliance Standard Life Insurance Company (Reliance).
Long-term disability payments were made to Lynd for twenty-four consecutive months. At the close of this two-year period, on March 9, 1993, the plan administrator terminated these payments to Lynd. The administrator made this decision to terminate benefits based on a limitation provision found in both the master policy and the certificate of insurance which stated that, “Monthly Benefits for Total Disability due to mental or nervous disorders will not be payable beyond twenty-four (24) months unless you are in a Hospital or Institution at the end of the twenty-four (24) month period.”
Following the termination of these benefits, Lynd filed a petition in the Fourth Judicial District Court of Louisiana alleging that his disability did not result from a “mental or nervous disorder[],” and that his benefits under the plan were therefore wrongly terminated by defendants-appellees. The action was removed to federal district court pursuant to 28 U.S.C. § 1331, and the parties thereafter filed cross motions for summary judgment. The district court denied Lynd’s motion and, in granting appellees’ motion, held that the plan administrator had not abused its discretion in deciding to terminate benefits.
On appeal, Lynd contends that the district court erred by reviewing the plan administrator’s decision under an abuse of discretion standard. Lynd argues that the district court should have reviewed the plan administrator’s decision de novo. Furthermore, Lynd maintains that, regardless of the standard of review employed, his long-term disability benefits were wrongfully terminated.
Discussion
Whether the district court employed the appropriate standard in reviewing an eli*981gibility determination made by an ERISA plan administrator is a question of law. See Chevron Chemical Co. v. Oil, Chemical and Atomic Workers Local Union 4-447, F.3d 139, 142 (5th Cir.1995). Therefore, we review the district court’s decision de novo.
In Firestone Tire and Rubber Co. v. Bruch, 489 U.S. 101, 113-17, 109 S.Ct. 948, 956-57, 103 L.Ed.2d 80 (1989), the Supreme Court established that a denial of ERISA benefits by a plan administrator should be reviewed de novo by the courts unless the plan gives the administrator “discretionary authority to determine eligibility for benefits or to construe the terms of the plan.” However, it remains unclear precisely what language must be employed in the plan to confer such discretionary authority upon the plan administrator. In Duhon v. Texaco, Inc., 15 F.3d 1302 (5th Cir.1994), this Court applied the analysis from Bruch to the language of an ERISA plan and held that de novo review was inappropriate because:
“[I]t is clear that the plan administrator has the discretionary authority to make a final and conclusive determination of the claim. This court has not imposed a linguistic template to satisfy this requirement ... but in this ease the plan’s plain language provides that the administrator may make an independent and final determination of eligibility.” Id. at 1305-06 (citations omitted).1
Additionally, we have observed that the requisite grant of discretionary authority cannot be inferred from the language of an ERISA plan. In Chevron Chemical Co., supra, in the course of holding abuse of discretion was the proper standard of review, we stated that:
“[T]he Supreme Court ‘surely did not suggest [in Bruch ] that ‘discretionary authority’ hinges on incantation of the word ‘discretion’ or any other ‘magic word.’ Rather, the Supreme Court directed lower courts to focus on the breadth of the administrators’ power — their ‘authority to determine eligibility for benefits or to construe the terms of the plan’.... ’ On the other hand, discretionary authority cannot be implied ... ‘an administrator has no discretion to determine eligibility or interpret the plan unless the plan language expressly confers such authority on the administrator.’ ” 47 F.3d at 142 (citations omitted).
In the present case, however, we pretermit the issue regarding which standard of review the district court should have employed in reviewing the plan administrator’s eligibility determination. We do so because, regardless of whether the district court reviewed the administrator’s eligibility determination for abuse of discretion or de novo, the nature of Lynd’s disability compelled the district court to conclude that Lynd’s long-term benefits under the plan were properly terminated.2
Section 8.0 of the instant plan includes the limitation that “Monthly Benefits for Total Disability due to mental or nervous disorders will not be payable beyond twenty-four (24) months unless you are in a Hospital or Institution at the end of the twenty-four (24) month period.” The parties do not dispute that Lynd remains disabled. Neither, however, is there any suggestion that Lynd was hospitalized or institutionalized on March 9, 1993, at the end of the two-year period during which he received long-term disability benefits. Therefore, this dispute turns on the proper characterization of Lynd’s disability; specifically, it must be determined whether or not his disability constituted a “mental or nervous disorder” within *982the meaning of this plan. We hold that the district court correctly affirmed the plan administrator’s determination that Lynd’s disability was due to a “mental or nervous disorder”; therefore, the district court’s holding is affirmed regardless of the standard of review employed by the district court in reviewing the plan administrator’s eligibility determination.
The undisputed evidence before the district court was that Lynd was diagnosed on September 19,1990, as suffering from “major depressive disorder.” This diagnosis, documented on Lynd’s benefits claims form, has remained static since that time.3
However, Lynd contends that this general diagnosis of his disability — as a “major depressive disorder” — comports with his claim that his condition is physical in nature. In support of this position, Lynd presented to the district court the deposition (taken well after benefits were denied) of his treating physician, psychiatrist Dr. Dumont, in which Dr. Dumont asserted his belief that depression is a “physical” disorder:
“Q: All right, sir. But I guess the question I am driving at is, depression itself, that is not something caused from a physical disorder itself?
A: Yes, it is. Yes, it is. We think of depression as being a chemical imbalance. It is a malfunction in the part of the brain that controls mood regulation. And we see it as usually an inefficiency of the neurotransmitters or a relative deficit of certain neurotransmitters and we try to treat that by utilization of medications that can help elevate or increase the efficiency of the neurotransmitters.
... Q: All right, sir. So depression in essence results from a nervous disorder as I understand?
A: From a disorder of the central nervous system.”
Dr. Dumont further testified that:
“Major depression is a disease and it has a physiologic basis every bit as much as diabetes, hypertension, cardiomyopathy or any other”
and that in his opinion “every major depressive disorder implicated] inefficiency of neurotransmitters in the central nervous system.”
Dr. Dumont testified regarding the symptoms experienced by Lynd as a consequence of his “major depressive disorder”:
“[Lynd] reported symptoms of lack of physical stamina, loss of energy and interest, unable to deal with pressure, unable to make decisions, he had sleep disturbance, he would wake up in a cold sweat with apprehension. He had to leave the office early one day or went to the office early one day and found anxiety so great he had to leave. He has been very anergic meaning no energy and ahdonia [sic] meaning unable to experience pleasure. He thus contacted Dr. Nichols who then referred him to me.”
Dr. Dumont expressed the view that “resolution of Mr. Lynd’s major depressive disorder would remove his disability.”
Dr. Dumont described himself as “a physician who specializes in the practice of psychiatry.” He saw Lynd on referral from Lynd’s regular physician, but received from the referring physician no “documentation” or “medical reports.” When asked if his records reflected “whether Mr. Lynd had any physical disorder or diseases,” Dr. Dumont responded, “not of any consequences that would have been connected with this, no.” Dr. Dumont treated Lynd with psychoreac-tive medication and psychotherapy.4 When asked “You did not conduct a physical exami*983nation” of Lynd, Dr. Dumont unqualifiedly responded “No, I did not.”
This Court has not previously addressed the interpretive issues raised by the allegation that the “physical” aspects of “mental” illnesses necessarily impact the construction of such qualifying phrases as “mental or nervous disorders” used in ERISA plans. However, we find the Eighth Circuit’s approach to be instructive:
“It would be improper and unfair to allow experts to define [ERISA plan] terms that were specifically written for and targeted toward laypersons. This requirement provides a source from which we may fashion a federal common law rule; the terms should be accorded their ordinary, and not specialized, meanings.
The cause of a disease is a judgment for experts, while laymen know and understand symptoms. Laymen undoubtedly are aware that some mental illnesses are organically caused while others are not; however, they do not classify illnesses based on their origins. Instead, laypersons are inclined to focus on the symptoms of an illness; illnesses whose primary symptoms are depression, mood swings and unusual behavior are commonly characterized as mental illnesses regardless of their cause.
... By focusing upon the disease’s etiology, the district court considered factors that are important to experts but not to laypersons. The court thus failed to examine the term “mental illness” as a layperson would have, which is the examination we conclude ERISA and federal common law require.” Brewer v. Lincoln National Life Ins. Co., 921 F.2d 150, 154 (8th Cir.1990), cert. denied, 501 U.S. 1238, 111 S.Ct. 2872, 115 L.Ed.2d 1038 (1991); see also Stauch v. Unisys Corp., 24 F.3d 1054, 1056 (8th Cir.1994) (observing that when an ERISA plan participant complained of depression, sleeplessness, impaired concentration and other symptoms, the court concluded that the participant “suffered from what laypersons would consider” a “mental” or “nervous” disorder).5
In its DIAGNOSTIC AND STATISTICAL MANUAL OF MENTAL DISORDERS, the American Psychiatric Association (APA) acknowledges that there is no bright-line distinction between “mental” disorders and “physical” disorders. Nevertheless, the APA also recognizes that, while “there is much ‘physical’ in ‘mental’ disorders,” the phrase “mental disorder” persists “because we have not found an appropriate substitute.” American Psychiatric Association, DIAGNOSTIC AND STATISTICAL MANUAL OF MENTAL DISORDERS xxi (Fourth Edition, 1994). Accordingly, the APA has not wavered from its classification of Lynd’s disability — “major depressive disorder” — as a “mental disorder.” Id. at 339. Thus, it is not just the lay population that holds to the view that certain disorders are properly and necessarily characterized as “mental disorders,” even though what is thus referred to may have a “physical” aspect and/or origin, as well.
The approach taken by the Ninth Circuit in Patterson v. Hughes Aircraft Co., 11 F.3d 948 (9th Cir.1993), is also instructive. In Patterson, the court confronted an ERISA plan pursuant to which benefits resulting from “mental, nervous or emotional disorders of any type” would be limited to two years. Id. at 949. The court observed that the plan did not define “mental disorder,” and held that ambiguities in the plan were to be resolved in favor of the plan participant. Id. at 950. In reaching its conclusion that the term “mental disorder” was ambiguous in this context, the court asserted its view that, when a disability was caused by “depression,” then that disability would be properly charaeter-*984ized as resulting from a “mental disorder”: “If Patterson’s disability was caused solely by his depression, ... then his condition is subject to the two-year limitation by any possible meaning of the Plan’s term ‘mental disorder.’ ” Id. at 951.6
The court ultimately remanded the case to the district court, concluding that, “[I]f Patterson’s headaches contributed to his total disability, or they are either a cause or symptom of his depression, then Patterson’s disability does not fall within the ‘mental disorder’ limitation interpreted in his favor.” Id. at 951 (emphasis added). The court reached this conclusion because the cause of Patterson’s disability had not been determined. Id7 However, the language quoted above would apparently sanction the conclusion that, even if it were established that the cause of Patterson’s disability was “depression” — and that Patterson’s headaches were only a symptom of his depression — then Patterson’s disability would nevertheless fail to constitute a “mental disorder.” We disagree with this view, particularly if the court intended that this analysis apply not only to Patterson’s headaches, but to all “physical” symptoms of “mental” disabilities. If we begin with the premise that the cause of a disability is “mental” — and the Eighth and Ninth Circuits, as well as the American Psychiatric Association, characterize “depression” as a “mental” disorder — then to find that a disability falls outside of the term “mental disorder” (as used in an ERISA plan) because the disability has “physical” symptoms would render the term “mental disorder” obsolete in this context. As the ERISA plan in the instant case pointedly refers to “mental or nervous disorders,” it would be inappropriate to effectively collapse the term “mental disorder” to include only those illnesses, if any exist, which have no “physical” manifestations. If the exclusion of disability, lasting more than twenty-four months, due to “mental or nervous disorders” is to mean anything — and we think it must— then there is no principled basis on which to exclude Lynd’s “major depressive disorders” from the reach of that exclusion.
Lynd suffers from “major depressive disorder”. There has been no suggestion that Lynd’s major depression is in some relevant aspect unusual, nor that his disability is caused by anything other than this disorder. As noted, Dr. Dumont testified that “resolution of Mr. Lynd’s major depressive disorder would resolve his disability.” Instead, Lynd maintains that his condition falls outside the phrase “mental or nervous disorder” simply because “every major depressive disorder,” according to Lynd’s psychiatrist, has “physical” origins and symptoms. Based on this evidence, the district court was compelled to affirm the plan administrator’s eligibility determination, regardless of the standard of review employed by the district court in reviewing this determination.
Accordingly, the judgment of the district court is
AFFIRMED.
. The plan at issue in Duhon addressed the discretionary authority of the administrator in two passages: (1) "The decisions of the Plan Administrator shall be final and conclusive with respect to every question which may arise relating to either the interpretation or administration of this Plan;” and (2) “After you undergo the necessary physical examination(s) and upon review of all facts in the case, the Plan Administrator will make the decision to authorize or deny payments.” Id. at 1305.
. Lynd concedes that, if he is found to be disabled as a result of a "mental or nervous disorder,” then he cannot recover under the plan; he allows that "it makes absolutely no difference” which standard is employed to review the administrator’s determination of eligibility should.his disability be so characterized.
. The following characterizations of Lynd's disability were attached to Reliance’s motion for summary judgment:
(1) In a letter of October 22, 1990, Lynd's treating physician, psychiatrist Dr. Arthur Dumont III, described Lynd’s disability as "major depression”;
(2) In a letter of April 24, 1991, Dr. Dumont described Lynd's disability as "a severe treatment resistant depression";
(3) In a letter of January 14, 1992, Dr. Dumont described Lynd's disability as “a severe major depressive disorder”;
(4) In a letter of May 12, 1993, Dr. Dumont characterized Lynd's disability as a “major depressive illness.”
. The medication included Prozac, Wellbutrin, Lithium, Desyrel, Klonopin, Buspar, Anafranil, Depakote, Cylert, Zoloft, and Serzone.
. In identifying the “causes” and "symptoms” of illnesses, it seems that an argument could always be fashioned that the illness itself should be viewed as a “symptom" of some underlying "physical” cause; this is particularly true if one is willing to trace the origins of the illness ad infinitum. An illustration of this is provided by Dr. Dumont's testimony that depression represents a "chemical imbalance” and stems from "an inefficiency of the neurotransmitters ..." The Eighth Circuit reasonably concludes that, within this analytical framework, laymen will look to the “symptom" of an illness in order to characterize that illness, and the symptom “depression” is indicative of a “mental" illness.
. The court's view that disabilities caused by depression fall within the classification of "mental disorders” is further underscored by the following:
"First, the Plan does not specify whether a disability is to be classified as ‘mental’ by looking to the cause of the disability or to its symptoms. Since the ambiguity is to be resolved in Patterson’s favor, his disability is not a mental disorder subject to the two-year limitation on payments if it is either manifested by headaches though caused by depression, or caused by headaches but manifested by depression.
Second, the Plan does not make clear whether a disability qualifies as a ‘mental disorder' when it results from a combination of physical and mental factors. Patterson’s disability may result solely from depression, or solely from headaches, or from a combination of the two. Since this ambiguity must also be resolved in Patterson's favor, he is not within the limitation for mental disorders if his disability is caused in any part by headaches.” Id. at 950 (citations omitted).
. Patterson filed his claim for benefits under the ERISA plan at issue in that case “for disability benefits due to headaches.” Id. at 949.