ORDER ON ADMINISTRATIVE REVIEW FOLLOWING REMAND

The matter is before the Court on Administrative review. On October 31, 2006, this Court issued its Order reversing the plan’s decision to terminate plaintiffs benefits. On September 6, 2008, the Ninth Circuit Court of Appeals issued its Memorandum Decision, remanding the matter to the District Court. After this Court’s 2006 ruling, the Ninth Circuit had issued its decision in Abatie v. Alta Health & Life Ins. Co., 458 F.3d 955 (2006). On remand, the Circuit Court issued the following directive:

Igor Gunn v. Reliance Standard Life Insurance Company, 235 Fed.Appx. 553 (9th Cir.2007).

Plaintiff appeals the defendant insurer’s termination of long-term disability benefits.

I. FACTUAL AND PROCEDURAL BACKGROUND

In 2001, Plaintiff Igor Gunn (“Gunn”) was employed by UBS/Paine Webber (“PW”) as a financial advisor. At the time, PW participated in a long-term disability (“LTD”) plan (“the Plan”) which was administered by defendant Reliance Standard Life Insurance Co. (“Reliance”). Administrative Record (“AR”) at 1.

The summary plan description (“SPD”) sets forth a description of a limitation on benefits provided for periods of disability related to mental or nervous disorders:

Chandler Deck, Ex. A at 17 (emphasis added).

The policy of insurance also sets forth a description of a limitation on benefits as well:

AR at 19 (emphasis added). “Mental or Nervous Disorders” is defined by the policy as including “depressive disorders.” AR at 19.

Both the SPD and the policy reserve the authority to interpret the provisions of the Plan to the administrator.

Gunn began experiencing health problems in early 2001. On February 29, 2001, Gunn was seen in the emergency room of Cedars-Sinai Medical Center. AR at 466. His diagnosis was “severe anxiety disorder with catatonia, now resolved.” AR at 467. The attending physician noted that Gunn had “a history of anxiety disorder,” but also noted that Gunn had “a recent concern for diagnosis of multiple sclerosis for which the patient is scheduled to see a neurologist....” AR at 466.

On July 20, 2001, Gunn filed an application for LTD disability benefits under the Plan. AR at 125. Gunn claimed he was unable to work due to “periodic fainting spells, constant dizziness, MS symptoms, fatigue, balance difficulties, cognitive difficulties, depression.” AR at 125 (emphasis added). With the claim, Gunn’s physician, Dr. Cho, submitted a physician’s statement. AR at 440-41. She indicated that Gunn’s diagnosis was MS. AR at 440. Under “findings”, she noted: “double vision, balance, depression, dizziness and fatigue.” AR at 440. She noted referrals for neurology and psychiatry. AR at 440.

Reliance approved Gunn’s claim in November 2001, and began providing disability benefits. AR at 112. In the letter notifying him of the approved claim, Reliance informed Gunn that he would be eligible for benefits after twenty-four months of disability only if he were “totally disabled from performing the material duties of Any Occupation.” AR at 112. Reliance notified Gunn that before the expiration of the twenty-fourth month (i.e., before August 2003), it would conduct an investigation to determine his continued eligibility for benefits. AR at 112. Reliance also informed Gunn that he could find a more detailed description of his policy in the “LTD certificate of group insurance.” AR at 112.

An April 1, 2002, report from Dr. Gross, Gunn’s psychiatrist, reveals Dr. Gross’ opinion that Gunn is totally disabled from any work because:

AR 50-51. Dr. Gross noted a possible connection between Gunn’s symptoms of depression and his MS diagnosis:

AR at 50.

On the same date, April 1, 2002, Dr. Gross filled out a mental impairment questionnaire. AR at 741. He noted that Gunn had MS and “mood and thinking difficulty.” He stated that Gunn’s prognosis was “extremely guarded”, while noting that Gunn’s MS “eontribute[d] to mental picture and is chronic.” AR at 743.

On April 24, 2002, Reliance sent inquiries to three of Gunn’s doctors — neurologist Dr. Ronald Andiman, AR at 69, internist Dr. Cho, AR at 67, and psychologist Dr. Susan Diskin, AR at 71. The inquiries asked for help in “clarifying [Gunn]’s current medical status.” See, e.g., AR at 67. The inquiries were similar to those made by Reliance prior to its November 2001 decision to grant Gunn’s disability claim. See, e.g., AR at 78.

Dr. Diskin responded, providing information on a Reliance form entitled “Physician’s Report Psychiatric.” AR at 314-15. Dr. Diskin opined that Gunn showed marked impairment in activities of daily living, social functioning, concentration, and adaptation to stressful conditions. AR at 315. She noted that he suffered from dizziness, nausea, balance difficulties, and fatigue. AR at 315. The form asks the physician if the impairment was caused primarily by any organic conditions; she indicated: “yes — multiple sclerosis.” AR at 315.

On February 24, 2003, an internal document in the record reveals that Reliance was considering an investigation into whether Gunn was subject to the mental disorder limitation. AR at 106 (“[I]t appears [claimant] could [return to work] from physical standpoint.... Recommend updating Medical Records on file[.] Need to determine what is [primary diagnosis]— physical (MS) or Mental (depression).”)

On April 16, 2003, as the two-year anniversary of Gunn’s initial LTD claim approached, Rebanee informed Gunn that it was reviewing his case. AR at 55-56. The company reminded Gunn that as of August 28, 2003, he would be eligible for continued benefits only if he were “disabled from performing the material duties of any occupation given [his] medical condition and vocational background.” AR at 55. The company informed Gunn that it intended to “evaluate [his] ability to perform another occupation,” and asked that Gunn complete a questionnaire. AR at 55. Despite the February 24, 2003 document regarding the intention to assess Gunn’s claim vis-a-vis the mental disorder limitation, the letter does not inform Gunn that Reliance would be examining whether his disability was related to depression. Compare AR at 106 with AR at 55-56.

Gunn returned a completed questionnaire in May 2003. Gunn wrote that he was prevented from working by “fatigue, dizziness, balancing difficulties [and] depression.” AR at 53. He listed his doctors as Cho, Diskin and Gross. AR at 53.

In an undated report that was written some time after March 2003, Gunn’s psychologist, Dr. Diskin, expressed the opinion that Gunn was completely disabled on both medical and psychological grounds. AR at 313. She noted certain symptoms of MS from which Gunn was suffering during January to March 2003: “numbness in one extremity, ... one incident of fainting, ... [d]izziness[, and] nausea.” AR at 313. Dr. Diskin also noted a link between Gunn’s physical symptoms of MS and his psychological distress:

AR at 313.

In July 2003, Reliance referred Gunn for an independent examination (“IME”) by United Review Services. AR at 241. Reliance sought the examination to determine if Gunn’s impairment was “secondary to MS, depression or neither.” AR at 246.

Dr. Carl Orfuss conducted the exam in August 2003, and at that time conducted an extensive medical file review. See AR at 220-29. Dr. Orfuss noted subjective symptoms reported by Gunn as “difficulty coordinating his lower extremities, ... frequent spasms in the lower extremities!,] ... and sluggish mentation.” AR at 221. Dr. Orfuss also noted the presence in the medical records of a “typed list of symptoms” that appeared to have been prepared by Gunn. AR at 224. With that list, Gunn complained of the following symptoms: depression, dizziness, bladder control problems, somnolence, fatigue, facial pain, decreased coordination and attention span, decreased memory and judgment, difficulty solving problems, blurred vision, shaking hands, chest pain, sweating, and balance problems. AR at 224.

Dr. Orfuss noted that that Gunn suffered from depression for several years prior to his diagnosis of MS in 2001. AR at 221. He noted that Gunn was diagnosed with MS before he saw a neurologist, and that “[rjetrospective symptoms” of MS were “dizziness and intermittent imbalance.” AR at 228. Without explanation (other than to note Gunn’s history of depression as predating the diagnosis of MS), Dr. Orfuss concluded that Gunn was “temporarily totally disabled at the present time with 99% of his disability stemming from his psychiatric problems.” AR at 228. Dr. Orfuss explained his conclusion by applauding the efforts of other MS patients who continue to work despite their symptoms:

AR at 228. Dr. Orfuss’ assessed Gunn’s physical impairment as limited to a mild gait instability that could be mitigated with an assistive gait device such as a cane. AR at 228-29.

On September 10, 2003, Reliance wrote a letter to Gunn which informed him that he would no longer receive LTD benefits. AR at 36-39. Noting Dr. Orfuss’ analysis, Reliance concluded that Gunn was “without a physical impairment which would prevent a return to work.” AR at 37. Reliance stated that Gunn’s inability to work was caused by a mental disorder. AR at 37 (“This information documents you are without a physical impairment.”). Reliance noted that, as a result, the mental or nervous disorder limitation applied, and he was no longer entitled to benefits under the Plan. AR at 37.

Gunn filed an appeal of the decision. Id. at 530. In support of the appeal, Gunn submitted an October 27, 2003 report from Dr. David Brandes. AR at 583-87. Dr. Brandes diagnosed Gunn as suffering from relapsing-remitting multiple sclerosis, severe depression, fatigue secondary to MS, and cognitive dysfunction secondary to MS and/or severe depression. AR at 586. He concluded that Gunn was totally disabled as a result of his MS:

AR at 586-87.

On February 26, 2004, Reliance requested another review, this time by Dr. William Hauptman. Reliance asked Dr. Hauptman specifically to “review [and] advise if MS alone is impairing.” AR at 550.

Dr. Hauptman conducted a medical file review but did not examine Gunn. AR at 532. He noted that the medical records documented diagnoses of MS and “significant psychiatric disease including depression.” AR at 532. He also noted that “[r]ecords document complaints of severe fatigue and cognitive deficits ascribed to both MS and depression.” AR at 545-46. Ultimately, however, he concluded that Dr. Orfuss’ assessment (that 99% of Gunn’s disability was due to depression) was “consistent with the entirety of the medical records.” AR at 546. He noted that the “[m]edical records clearly document consistently severe psychiatric illness characterized as severe depression.” AR at 546. He contrasted Gunn’s psychiatric illness with his physical illness: “[T]he patient’s multiple sclerosis is characterized by his neurologist, Dr. Brandes as mild.” AR at 546. Dr. Hauptman also observed that “medical records document a correlation between the patient’s level of severity of depression and response to psychiatric medication, with his subsequent level of fatigue.” AR at 546.

Reliance denied Gunn’s appeal in a March 11, 2004 letter to Gunn. AR at 524. Reliance cited the analyses of Dr. Haupt-man and Dr. Orfuss, as well as the notes from Gunn’s physicians, in upholding its initial termination of benefits. AR at 526-27.

After Reliance denied the appeal in March 2004, Gunn filed the present action.

II.ERISA PLANS

The Employee Retirement Income Security Act (“ERISA”), 29 U.S.C. §§ 1001-1461 , regulates certain employee benefit plans, including employee welfare benefit plans such as the long-term disability plan at issue in this action. 29 U.S.C. § 1002 (1), (3); § 1003(a). ERISA requires that employee benefit plans “be established and maintained pursuant to a written instrument.” 29 U.S.C. § 1102 (a)(1). The instrument must provide for one or more named fiduciaries who jointly or severally shall have authority to control and manage the operation and administration of the plan. Id.

Beneficiaries who seek benefits due under an ERISA plan must first seek relief with the plan administrator. Chappel v. Laboratory Corp. of America, 232 F.3d 719, 724 (9th Cir.2000). After the beneficiary has exhausted this remedy, he may file a civil action. 29 U.S.C. § 1132 (a)(1)(B).

III. ISSUES PRESENTED ON ADMINISTRATIVE

REVIEW

In connection with the present administrative review, the Court must decide three issues before proceeding to its review of the administrative decision. First, the Court must determine whether the SPD is a valid summary plan description under ERISA. Next, the Court must apply one of two contradictory clauses relating to “mental or nervous disorders.” In connection with that analysis, the Court must determine whether there are ambiguities in the applicable limitation. Finally, the Court must determine whether any conflict of interest affected the decision-making process in this case.

IV. VALIDITY OF SUMMARY PLAN DESCRIPTION

Gunn has submitted a PW document entitled “Disability Plan,” which he contends is a valid SPD within the meaning of 29 U.S.C. § 1022 . See Chandler Deck, Ex. A. He argues that the SPD governs his eligibility for LTD benefits. See Bergt v. Retirement Plan for Pilots Employed by MarkAir, Inc., 293 F.3d 1139, 1143 (9th Cir.2002) (holding that a valid SPD is part of the ERISA plan). Reliance argues that the SPD is not a valid plan summary plan description under either 29 U.S.C. § 1022 (b) or the relevant regulations.

In order to constitute a valid SPD for LTD benefits, § 1022(b) requires that a summary contain certain information, including the name and address of the plan administrator, the source of financing for the plan and the identity of any organization through which benefits are provided, and the remedies available for the redress of denied or partially denied claims. 29 U.S.C. § 1022 (b).

Defendant’s position, that the SPD is not valid, is not supported by the evidence in this case. Following remand, the parties were allowed to conduct further discovery. At her deposition Olga Torres-Villanueva, a human-resources employee of PW, was asked bring and identify all plan documents distributed by PW to its employees during the relevant period. She identified a summary plan description package, which was given to all employees in an accordian-style envelope. The package consisted of a document which identified itself as a “Summary Plan Description or SPD”, and approximately ten booklets, one for each type of employee benefit plan. Villanueva Deposition, p. 17-20; Chandler Decl. Exhs. C and E.

Additionally, at the conclusion of a hearing concerning a discovery dispute, following remand, the Magistrate Judge ordered defendant to provide to plaintiff the Summary Plan Description at issue in this case. In response, counsel produced Exhibit C. Chandler Dec. Exhs B and C. Those documents contain the information required by 29 U.S.C. § 1022 . The SPD and related booklets constitute a valid plan summary.

V. STANDARD OF REVIEW

The Supreme Court has held that a denial of benefits “is to be reviewed under a de novo standard unless the benefit plan gives the administrator or fiduciary discretionary authority to determine eligibility for benefits or to construe the terms of the plan.” Firestone Tire & Rubber Co. v. Bruch, 489 U.S. 101, 115 , 109 S.Ct. 948 , 103 L.Ed.2d 80 (1989). “When a plan unambiguously gives the plan administrator discretion to determine eligibility or construe the plan’s terms, a deferential abuse of discretion standard is applicable.” Burke v. Pitney Bowes, Inc. Long-Term Disability Plan, 544 F.3d 1016, 1023-24 (9th Cir.2008) (citing Abatie v. Alta Health & Life Ins. Co., 458 F.3d 955, 963 (9th Cir.2006) (en banc)).

In Metropolitan Life Ins. Co. v. Glenn, — U.S. -, 128 S.Ct. 2343, 2346 , 171 L.Ed.2d 299 (2008), the Supreme Court set forth a framework, similar to the one provided in Abatie , to assess whether the dual role of administering and funding an ERISA plan creates a conflict of interest, and if so, how that conflict should be considered in evaluating whether a plan administrator has abused its discretion. The Court reiterated the principles established in Firestone Tire, and noted, “[i]n ‘determining the appropriate standard of review,’ a court should be ‘guided by principles of trust law,’ ” and “[i]f ‘a benefit plan gives discretion to an administrator or fiduciary who is operating under a conflict of interest, that conflict must be weighed as a factor in determining whether there is an abuse of discretion.’ ” Metropolitan Life Ins., 128 S.Ct. at 2347-48 (emphases in original) (citing Firestone Tire, 489 U.S. at 115 , 109 S.Ct. 948 ).

A conflict of interest is clearly present “where it is the employer that both funds the plan and evaluates the claims.” Id. at 2348. The Court noted that the abuse of discretion standard of review still applied despite the structural conflict of interest. Id. at 2349-50. However, after concluding a conflict of interest exists, a reviewing court must:

Burke, 544 F.3d at 1025 (quoting Metropolitan Life Ins., 128 S.Ct. at 2350-51 ). In other words, a structural conflict of interest and the circumstances surrounding the conflict are weighed and taken into account to determine whether the plan administrator has abused its discretion.

In general, a district court may review only the administrative record when considering whether the plan administrator abused its discretion. Abatie v. Alta Health & Life Ins. Co., 458 F.3d at 970 . The administrative record generally consists of the record that was before the administrator when the decision was made. Tremain v. Bell Industries, Inc., 196 F.3d 970, 976-77 (9th Cir.1999). “The district court may, in its discretion, consider evidence outside the administrative record to decide the nature, extent, and effect on the decision-making process of any conflict of interest; the decision on the merits, though, must rest on the administrative record once the conflict (if any) has been established, by extrinsic evidence or otherwise.” Id. (citing Doe v. Travelers Ins. Co., 167 F.3d 53, 57 (1st Cir.1999)).

VI. AMBIGUITIES IN AND CONTRADICTIONS BETWEEN THE SPD AND THE POLICY

When the terms of a plan document conflict with the SPD, the terms that favor the beneficiary generally apply. Bergt v. The Retirement Plan for Pilots, 293 F.3d 1139, 1145 (9th Cir.2002) (holding that a contradiction between the master plan document and the SPD created an ambiguity as to whether a plaintiff was eligible to participate in a retirement plan.) The Ninth Circuit has determined that the “burden of uncertainty created by careless or inaccurate drafting of the [SPD] must be placed on those who do the drafting.’ ” Id. at 1145 (quoting Hansen v. Continental Ins. Co., 940 F.2d 971, 982 (5th Cir.1991)). The Ninth Circuit noted that this rule, in addition to placing the burden on the party that is most able to bear it, has the side effect of encouraging accurate drafting of SPDs, the document that must be provided to beneficiaries:

Bergt, 293 F.3d at 1145 .

Here, the SPD and the policy use different language when referring to the limitation for periods of disability related to mental or nervous disorders. The SPD notes that the 24-month limitation applies to periods of disability “due to mental or nervous disorders”; the policy states that the limitation applies to periods of disability “caused by or contributed to by mental or nervous disorders.”

The phrase “due to” is ambiguous in that it is susceptible to at least two meanings. First, the phrase could be read to mean that the limitation applies only when the inability to work is solely the result of depression. However, the phrase could also be read more broadly, and the limitation could apply when neither a physical ailment nor a mental disorder is the sole cause of the inability to work; rather, it would apply when the physical ailment and the mental disorder together create the inability to work.

The phrase “caused by or contributed to by” is susceptible to the second meaning of “due to” as set forth above, but the phrase is not susceptible to the first meaning. By adding the phrase “or contributed to by” to “caused by,” the drafters of the policy eliminated as a possible interpretation that the policy limitation applied only when the inability to work is solely the result of depression. Therefore, the Court concludes that the policy language conflicts with the SPD language.

Accordingly, the Court must apply the version of the limitation that is more favorable to the beneficiary. In this case, that version is contained within the SPD. In applying the limitation set forth in the SPD, the Court must also construe any ambiguities in favor of the beneficiary. Here, the twenty-four month limitation will not apply unless Gunn’s inability to work is solely attributable to his depression.

VII. ANALYSIS OF CONFLICT

The evidence of record in this case demonstrates that the trustee abused its discretion in terminating plaintiffs benefits. In addition to the conflict of interest inherent in the dual role played by the administrator, the following factors clearly establish that the decision to terminate benefits was a product of bias and not the result of a fair, objective, and even-handed evaluation of the plaintiffs case:

1. Reliance misrepresented the nature of its investigation into plaintiffs continuing eligibility for benefits. Although Reliance was trying to determine whether plaintiff was subject to a mental disorder limitation, it asked him for additional medical information so it could assess his “ability to perform another occupation.”

This information provided to Gunn and his doctors the incentive to show the extent of his impairment, but provided no incentive to them to provide an assessment regarding the cause of that impairment— i.e., whether his impairment was due to his MS, his depression, or both. This initially deprived Gunn of the ability to have his doctors focus on this distinction. However, Reliance was not deprived of this same opportunity, and in fact posed the question to its doctors in more precise terms: “Please clarify if impairment is secondary to MS, depression or neither.” AR at 246 (to Dr. Orfuss). “Please review [and] advise if MS alone is impairing.” AR at 550 (to Dr. Hauptman).

Gunn’s evidence that Reliance made a misrepresentation in the April 2003 letter is probative evidence that tends to show that Reliance acted in its own self interest in deciding to terminate Gunn’s LTD benefits.

2. Gunn has presented evidence that Reliance exhibited bias when it selected Dr. Hauptman to conduct a medical review of his file, notwithstanding its knowledge of the criticism Dr. Hauptman’s practices received in Conrad v. Reliance Standard Life Insurance Co., 292 F.Supp.2d 238 (D.Mass.2003). In that case, the court found that Reliance’s decision to deny LTD benefits was arbitrary and capricious because the decision was “heav[il]y depen-den[t] on ... two reports produced by Dr. Hauptman” that “betray[ed] a palpable bias in favor of rejecting the claim.” Id. at 237-38. The court noted that Dr. Haupt-man highlighted in bold-face type and underlining — presumably for Reliance’s benefit — those passages in his reports that favored a denial of benefits. Such highlighting, the court noted, “suggested that] Dr. Hauptman’s approach was that of an advocate speaking to persuade, rather than a medical professional seeking to evaluate.” Id. at 238-39. The court also took issue with what it described as Dr. Hauptman’s willingness to “repeatedly focus[] on only one or two sentences or facts” and his unwillingness “to recognize or point out the context in which those statements were made.” Id. at 240. The court concluded that it was unreasonable (and therefore arbitrary and capricious) “[f]or Reliance to base its decision on such tendentious reports.” Id.

The District of Massachusetts’ criticism of Dr. Hauptman is not relevant to the substance of his analysis of Gunn’s impairment in this action. Rather, this criticism is illuminating with respect to Reliance’s actions — or lack thereof — in response to that criticism. Conrad gave notice to Reliance of Dr. Hauptman’s practices with respect to providing analyses regarding the insured’s degrees of impairment. Here, Reliance continued to engage the services of Dr. Hauptman without attempting to correct the perceived bias. In fact, the evidence reflects that Reliance used the services of Dr. Haupt-man 167 times in 2003 and 110 times in 2004. Indeed, the only response to the criticism was Reliance’s agreement with Dr. Hauptman’s suggestion that he stop highlighting in bold-face type (and under lining) the passages in his report that most justified a denial of a claim. Concentrating on form only, this response did nothing to address the substance of the court’s criticism.

The Conrad court’s criticism focused on the fact that Dr. Hauptman appeared to be a man with a mission — to find a way to justify a denial of benefits. See id. at 239 (“Dr. Hauptman’s two reports leave the impression of an examiner who has embarked on his investigation determined to find evidence that [the beneficiary] is not in as much pain, either physically or mentally, as he claims to be.”). Here, Reliance set Dr. Hauptman off on a new mission, informing him that Gunn’s “claim had been paid for [the] duration of [the] policy on psych,” and asking him to “please review [Gunn’s medical records and] advise if MS alone is impairing.” AR at 550.

Gunn’s evidence that Reliance continued to rely on Dr. Hauptman’s analysis after the Conrad court’s criticism without addressing the substance of the Court’s criticism is probative evidence that tends to show that Reliance acted in its own self interest in deciding to terminate Gunn’s LTD benefits.

3. Reliance accepted whole-heartedly the opinions of its own physicians, that plaintiffs disability was caused 99% by his depression and only 1% by his multiple sclerosis, in spite of all medical evidence to the contrary offered by plaintiff.

VIII. CONCLUSION

The decision to terminate benefits was an abuse of the administrator’s discretion. Therefore, the Court REVERSES the Plan’s decision. Plaintiff is to prepare and lodge a proposed Judgment within ten days of the date of this Order.

. The SPD provides: The Plan Administrator ... has the sole discretion and authority to apply, construe, and interpret all provisions of each of the Plans, to grant or deny all claims for benefits, and to determine all benefit eligibility issues. Chandler Errata Decl. at 216. Similarly, the policy provides: Reliance Standard Life Insurance Company shall serve as the claims review fiduciary with respect to the insurance policy and the Plan. The claims review fiduciary has the discretionary authority to interpret the Plan and the insurance policy and to determine eligibility for benefits. AR at 12. . The Reliance Policy defines a person as "totally disabled” if he cannot perform the material duties of his "regular occupation.” AR at 9. This definition applies for only two years. After two years of receiving disability benefits, the person is considered totally disabled only if he cannot perform the material duties of "any occupation.” Id. . Drs. Cho and Andiman also appear to have responded, but their responses consisted of treatment notes without narrative or conclusions. AR at 317-34, 349-91. . Reliance criticizes Gunn’s selection of Dr. Brandes, implying that Gunn shopped for a more favorable opinion than that offered by his previous neurologist, Dr. Andiman. Reli ance points to handwritten notes in the record from one of Gunn’s doctors (it is not clear from which), which states that “a neurologist says his problem re work psychiatric’ and refused to continue disability.” AR at 746. Someone (again, it is not clear who) interli-neated on these notes the name "Andiman.” AR at 746. Dr. Andiman, for his part, has submitted a declaration that disavows ever having expressed the opinion that Gunn’s disability was only psychiatric in nature and states that he did not ever refuse to certify Gunn’s disability. Ordinarily, the court may not consider evidence outside the administrative record; the only exception to that rule does not apply here. See infra note 5. Here, the admission of Dr. Andiman’s declaration is not necessary. The notation in one doctor’s treating notes of another doctor’s conclusions is of limited probative value. Dr. Andiman’s opinion, as reported by him, regarding the nature of Gunn’s disability is not part of the record. For that reason, the Court does not consider the opinion of Dr. Andi-man, whatever that opinion might be. . Dr. Hauptman's deposition is part of the record. Ordinarily, on administrative review, the Court does not consider extrinsic evidence; however, the Court may consider evidence outside the administrative record where the evidence is offered as proof of a conflict of interest. Tremain v. Bell Industries, Inc., 196 F.3d 970, 976-77 (9th Cir.1999). Reliance is the only insurance company for which Dr. Hauptman works, and he derives approximately one-third of his income from his work with Reliance. Hauptman Depo. (attached to the Chandler Decl. as Ex. C) at 20, 32. Reliance prohibits Dr. Hauptman from contacting a beneficiary’s treating physicians to discuss those physicians’ opinions unless he first receives permission from Reliance. Hauptman Depo. at 37. In 2003, Hauptman was criticized by the District of Massachusetts in a published opinion for an obvious bias in favor of rejecting a claim for LTD benefits. Reliance's sole response to this criticism was to note its agreement with Dr. Hauptman’s suggestion that he stop using bold-faced type in certain areas because that was perceived by the Massachusetts court as evidence of bias. Hauptman Depo. at 42. . By addressing this argument, the Court does not mean to imply that where a summary plan description fails to meet one or more of the technical requirements of 29 U.S.C. § 1022 (b), it is not binding on an ERISA plan. The SPD is the document that must be provided to participants and beneficiaries, and they justifiably rely on the SPD as a statement of their rights under an ERISA plan. 29 U.S.C. § 1022 (a); see also Bergt, 293 F.3d at 1143 (noting that the SPD is the "primary source of information regarding employment benefits” for participants and beneficiaries). . When language in an ERISA document susceptible to two different meanings, it is considered to be ambiguous. Bergt, 293 F.3d at 1143-44 . . After the initial denial letter was sent, Gunn was informed that the termination of his benefits was based on this limitation, and he was able to focus medical evidence on the relevant question of causation. . The Court notes, however, that there is some evidence in the record in this action that suggests that Dr. Hauptman is still willing to base his conclusions "almost exclusively on the most optimistic data available.” Conrad, 292 F.Supp.2d at 238 . For example, in justifying his conclusion, Dr. Hauptman contrasts Gunn’s history of "severe depression” with his "mild” MS. AR at 546. However, Dr. Hauptman chose to quote a passing notation of a diagnosis of "mild” relapsing-remitting multiple sclerosis found in a treatment note from Gunn's follow-up visit to Dr. Brandes, rather than an extensive discussion by Dr. Brandes regarding the nature and cause of Gunn’s impairment found in the "Comprehensive Neurological Evaluation.” Compare AR 583-87 (Dr. Brandes’ October 27, 2003, "Comprehensive Neurological Evaluation” that states that Gunn was "totally disabled as a result of his multiple sclerosis”) and AR at 538 (Dr. Hauptman’s report, noting his review of Dr. Brandes’ October 27 report and his treatment notes from December 23, 2003) with AR at 619 (Dr. Brandes’ December 23 treatment note regarding Gunn's diagnosis: "RRMS — mild”) and AR at 546 (Dr. Hauptman's report, justifying his conclusion that Gunn’s MS is not disabling by referring to the December 23 treatment note but making no mention of the October 27 comprehensive report). Additionally, Dr. Hauptman was not specifically asked to comment on the opinion offered by the doctor performing the IME. See AR at 550. Nevertheless, Dr. Hauptman emphatically stated that "[i]t is [his] opinion within a reasonable degree of medical certainty that the evaluation, analysis and conclusion as set forth by Dr. Orfuss are consistent with the entirety of the medical records.” AR at 546. Although this ringing endorsement alone is not damning evidence, when viewed in light of the Conrad court's criticism, the other example noted above, and the fact that only Dr. Orfuss’ and Dr. Hauptman’s reports express the opinion that Gunn’s MS is only 1% impairing, the seemingly unsolicited opinion that emphatically endorses the insurer’s IME becomes suspect.