Robinson v. Unum

The casenote of the month is from the Disability E-News Alert! a monthly newsletter describing new disability insurance developments.  For subscription information, e-mail Mark DeBofsky or visit www.disabilityenewsalert.com.


Robinson v. Unum Life Insur.Co. of America,

2003 U.S.Dist.LEXIS 4023 (D.N.H. 3/12/03)

(Issues: Treating Physician, Self-Reported Illness, Fibromyalgia/Chronic Fatigue Syndrome).

In this ruling, which rejected the treating physician rule, the plaintiff’s disability benefits were cut off despite flare-ups of a serious disorder, sarcoidosis, and the presence of limiting symptoms entirely consistent with that disorder.  Although Unum paid benefits for the first 24 months of disability, when the definition of disability changed at the 24 month mark to an "any occupation" standard, benefits were terminated.  This case also marks the first instance we are aware of where Unum’s self-reported limitation provision in its policy resulted in a benefit termination, with the court ruling that in the absence of objective proof, the plaintiff’s complaints of severe fatigue (which his treating physician explained were common symptoms of sarcoidosis) were self-reported; and benefits were therefore limited to a duration of 24 months.

The court explained in its ruling that under First Circuit case law, it was not bound to follow the treating doctor’s opinion.  Benefits were terminated after 24 months based on the opinions of Unum’s medical director as well as the findings of a functional capacity evaluation (although the court did not attribute much significance to those results).  Even though the treating doctor explained that a gallium scan showed that sarcoidosis remained active in the plaintiff’s skin and internally; and that sarcoidosis is well known to cause fatigue and inability to concentrate, the court favored the opinion of Unum’s in-house physician and suggested that Robinson should have undergone neuropsychological testing to validate complaints of fatigue and concentration difficulties.  The court also refused to modify a deferential standard of review based on the potential conflict of interest caused by Unum’s dual role as plan administrator and payer of benefits.  Thus, the court concluded that it could not find that Unum’s decision was irrational; therefore, it upheld the benefit termination.

Discussion: This opinion seems inconsistent with Cook v. Liberty Life Assurance Co. of Boston, 320 F.3d 11 (1st Cir. 2/5/03)(February 2003) where the court overturned a benefit termination in a claim involving fibromyalgia and chronic fatigue syndrome.  The court there reiterated that objective proof was not required in all cases.  The court also ruled that the insurer should have conducted an examination of the claimant as the basis for terminating benefits; otherwise there was insufficient evidence for the conclusion made by the insurer.

More important, though, is that this ruling illustrates the debate over the treating physician rule that will soon be resolved by the Supreme Court.  What the Ninth Circuit has done in Regula and in Nord is to utilize the treating physician rule as a means of mitigating a highly deferential standard of review by using the insurer’s failure to defer to the treating physician’s opinion as evidence of a conflict of interest that would diminish the degree of deference accorded to the insurer’s determination.  Although Robinson rejected the application of the treating physician rule as being inconsistent with ERISA’s allowance of discretionary decision-making by the plan administrator, the Supreme Court has already crossed the line as far as allowing external sources to supply a controlling opinion in a benefits claim.  The Court ruled in Rush Prudential HMO, Inc. v. Moran, 536 U.S. 355, 122 S.Ct. 2151 (2002); i.e., that a state law which requires independent review of a health insurer's medical necessity determination regarding a treatment, allows the result of that review to overcome the medical plan's own determination regardless of whether that plan was vested with discretion to determine eligibility to receive benefits.

The development of the treating physician rule in Social Security cases resulted from a recognition that [email protected] is a concept that may be statutorily or contractually defined, but requires an individualized assessment to determine whether an individual meets that definition.  Neither the Listing of Impairments (20 C.F.R. '404, Regulations No. 4, Subpart P, Appendix 1) or the Medical-Vocational Rules (20 C.F.R. '404, Regulations No. 4, Subpart P, Appendix 2) can completely resolve the question of whether an individual is incapable of engaging in any substantial gainful [email protected] (42 U.S.C. '423(d)(1)(A)-Social Security definition of [email protected]).

Like ERISA cases, Administrative Law Judge decisions in Social Security claim are subject to judicial review, 42 U.S.C. '405(g).  That review is deferential; and the ALJ's decision is to be upheld so long as it is supported by substantial evidence which Richardson v. Perales, 402 U.S. 389, 401 (1971) defines as such relevant evidence as a reasonable mind would accept as adequate to support a [email protected] Thus, the treating physician rule was adopted as a means for courts to introduce judicial discretion into their review of the disability determinations made by the SSA's Administrative Law Judges ([email protected])[email protected]  Schneider, a Rule for the Courts: Treating Physician Evidence in Social Security Disability Determinations,@ 3 U.Chi.Law School Roundtable 391 (1996).  However, neither in the case law that preceded the promulgation of a specific regulation adopting the treating physician rule, or in the regulation itself (20 C.F.R. ''404.1527, 416.927), is there abject deference to the treating doctor's opinion.  Further, neither Nord or Regula suggested there should be unfettered acceptance of the treating doctor's opinion.

Because of concern that the treating physician Amay want to do a favor for a friend and client,@ (Stephens v. Heckler, 766 F.2d 284, 289 (7th Cir. 1985)), the treating doctor's opinion may be discounted.  As one writer noted, Among the relevant considerations are the length of the doctor/patient relationship and the presence or absence of corroborating or contradictory [email protected]  Maccaro, The Treating Physician Rule and the Adjudication of Claims for Social Security Disability Benefits,@ 41 SSRS 833 (West 1993).  However, as the Stephens decision also recognized in reliance on an earlier ruling, Allen v. Weinberger, 552 F.2d 781 (7th Cir. 1977),

When experience backed by observation is set against the Speculative [email protected] (552 F.2d at 786) of a consulting physician, substantial evidence lies on the side of the treating physician.  See also Whitney v. Schweiker, 695 F.2d 785, 789 (7th Cir. 1982)(report of the treating physician should be favored over that of a consultant who merely reviews the file and does not examine the claimant, unless the treating physician is not credible). 766 F.2d at 288-89.

A later Seventh Circuit decision also pointed out, Whereas scores of doctors may examine a claimant's present condition, going back in time to examine the claimant's past condition is difficult for even the most talented consulting [email protected]  In such a case, the ALJ must take into account the treating physician's ability to observe the [email protected] Micus v. Bowen, 979 F.2d 602, 608 (7th Cir. 1992)(citations omitted).  Hence, all courts that have accepted the treating physician rule recognize the value of a personal examination; and that the rationale for giving added weight to the treating physician's opinion Ais that he is employed to cure and has a greater opportunity to know and observe the patient as an [email protected]  Murray v. Heckler, 722 F.2d 499, 502 (9th Cir. 1983).

Now that the treating physician rule has been incorporated into regulations (20 C.F.R. ''404.1527; 416.927), the Social Security Administration has clarified the basis of the principle:

Generally, we give more weight to opinions from your treating sources, since these sources are likely to be the medical professionals most able to provide a detailed, longitudinal picture of your medical impairment(s) and may bring a unique perspective to the medical evidence that cannot be obtained from the objective medical findings alone or from reports of individual examinations, such as consultative examinations or brief hospitalizations.  20 C.F.R. '404.1527(d)(2).

Nonetheless, the treating physician's opinion will not control unless it meets carefully defined criteria.  For example, the longer the treatment relationship and the more times the treating doctor has examined the patient, the more informed is the longitudinal [email protected] of the claimant's impairment.  20 C.F.R. '404.1527(d)(2)(i).  The nature and extent of the treatment relationship is also important; however, the regulations give the example of an ophthalmologist whose patient complains of neck pain during an eye examination that doctor's opinions regarding the neck pain have less weight than the opinions of a physician who has treated the neck pain.  20 C.F.R. '404.1527(d)(2)(ii).  Of the greatest significance, though, is that for the treating physician's opinion to be given controlling weight, it must have supportability such as consistency with medical signs and laboratory findings,@ and it must be consistent with the record as a whole.  20 C.F.R. ''404.1527(d)(3) and (4).  Moreover, more weight is given Ato the opinion of a specialist about medical issues related to his or her area of specialty than to the opinion of a source who is not a [email protected]  20 C.F.R. '404.1527(d)(5).  Regardless of the foregoing, however, an opinion that an individual is [email protected] or is Aunable to [email protected] is not binding because that is an issue reserved to the [email protected]  20 C.F.R. '404.1527(e).

Hence, although these are regulations promulgated by the Social Security Administration and are not incorporated within the ERISA claim regulations, they were developed as a policy that takes into account and, in large measure, was shaped by, the kinds of concerns raised by the circuit [email protected]  56 Fed.Reg. 36932, 36951 (1991).  Because those same policy concerns are equally applicable to issues arising in disability benefits claims brought under the ERISA law, these straightforward, common-sense principles are useful in adjudicating such claims because both systems recognize that Athe effects of medical conditions on individuals vary so widely, and because no two cases are ever exactly [email protected] 56 Fed.Reg. at 36934-35.  Thus, the treating physician rule strives for a greater measure of accuracy in resolving a decision-making procedure [which] includes a certain amount of subjectivity and [email protected] Schneider, 31 U.Chi.L.Sch. Roundtable 391.

Incorporating the treating physician rule in ERISA is also consistent with a principle articulated by the Seventh Circuit in Halpin v. W.W. Grainger, Inc., 962 F.2d 685, 695 n.11 (7th Cir. 1992):

Although the standards used in adjudicating social security cases are not applicable under ERISA, the guiding principles developed in those cases may be "instructive" in ERISA cases.    See, Torix v. Ball Corp., 862 F.2d 1428, 1431 & n.6 (10th Cir. 1988); Helms v. Monsanto Co., 728 F.2d 1416, 1421 (11th Cir. 1984).

Hence, while not every standard utilized in resolving Social Security disability claims need be applied in evaluating disability claims under the ERISA law, the treating physician rule should be broadly applied since it recognizes that the definition of [email protected] under an ERISA plan cannot explicitly guide the adjudication of each and every disability claim because every individual is unique.

The treating physician rule is also important with respect to enabling individuals with disabilities that are not diagnosable with laboratory tests to nonetheless qualify for benefits.  For example, Sisco v. United States Department of Health and Human Services, 10 F.3d 739, 774 (10th Cir. 1993) pointed out there is no >dipstick laboratory test for chronic fatigue [email protected]  That ruling was recently cited in an ERISA-governed disability benefits case, Cook v. Liberty Life Assur.Co., 320 F.3d 11 (1st Cir. February 5, 2003), which ordered reinstatement of benefits to a claimant suffering from chronic fatigue syndrome and rejected the insurer's termination of benefits due to a lack of objective evidence.  Likewise, in Sarchet v. Chater, 78 F.3d 305, 306 (7th Cir. 1996), the court discussed a related medical condition, fibromyalgia, and explained it was

a common, but elusive and mysterious, disease, much like chronic fatigue syndrome, with which it shares a number of features. See Frederick Wolfe et al., "The American College of Rheumatology 1990 Criteria for the Classification of Fibromyalgia: Report of the Multicenter Criteria Committee," 33 Arthritis and Rheumatism 160 (1990); Lawrence M. Tierney, Jr., Stephen J. McPhee & Maxine A. Papadakis, Current Medical Diagnosis & Treatment 1995 708-09 (1995). Its cause or causes are unknown, there is no cure, and, of greatest importance to disability law, its symptoms are entirely subjective. There are no laboratory tests for the presence or severity of fibromyalgia. The principal symptoms are "pain all over," fatigue, disturbed sleep, stiffness, and--the only symptom that discriminates between it and other diseases of a rheumatic character--multiple tender spots, more precisely 18 fixed locations on the body (and the rule of thumb is that the patient must have at least 11 of them to be diagnosed as having fibromyalgia) that when pressed firmly cause the patient to flinch.

Nonetheless, based on the reports of the claimant's treating physician, a rheumatologist, benefits were awarded.  Another condition, myofascial pain of unknown etiology, was found compensable in an ERISA disability benefits claim, Gaylor v. John Hancock Mut.Life Insur., 112 F.3d 460, 466 (10th Cir. 1997), based on a recognition that Medicine is, at best, an inexact science, and we should not disregard the great weight of the evidence merely because objective laboratory diagnostic findings either are not yet within the state of the art, or are [email protected]

Likewise, with respect to mental impairments, laboratory testing simply does not exist; and the opinions of treating physicians are that much more important.  As an illustration, in  Hackett v. Xerox Corp. Long Term Disability Income Plan, 315 F.3d 771 (7th Cir. 2003), the court overturned a benefit termination based on a consultant's examination which contradicted the reports of a treating physician (Dr. Gerber), whose opinions were corroborated by several other examiners, and held:

[The termination procedures were arbitrary and capricious. After twelve years of paying out disability benefits, Xerox terminated those benefits simply on the basis of an examination by Dr. Holeman, whose conclusion that Hackett was able to work was contrary to numerous prior opinions. Dr. Holeman provided no explanation for his departure from the opinions of the previous doctors, and Xerox provided no explanation for believing Dr. Holeman's opinion over the opinions of the previous doctors. There was no weighing of the evidence for and against, and there were no articulated reasons given for Xerox's rejection of the evidence that Hackett was unable to work. Conclusions without explanation do not provide the requisite reasoning and do not allow for effective review.  Id. at 693. We are left without explanation as to why Dr. Holeman's opinion is different from Dr. Gerber's. Had Dr. Holeman referenced the previous opinions and explained his deviation from them, we could have readily reviewed this case. Any non-arbitrary explanation could show that he had weighed the evidence for and against. We could, therefore, assume that any decision by the administrator took these factors into consideration. 315 F.3d at .

Accordingly, without giving deference to the treating physician, claimants may unfairly suffer the denial or termination of benefit payments under plans and insurance policies containing discretionary language.  Although the treating physician rule as such was not applied in Hackett or in Cook, in many cases, claimants lack a reasonable opportunity to protect against the denial or loss of benefits so long as some physician certifies non-disability even in the absence of an examination.  Without the treating physician rule, the treating doctor's opinion can be simply disregarded; and under a deferential standard of review, it would make no difference since a court may rule that the other evidence is sufficient to support the benefit denial.

Nord offers a clear example of such a situation.  Metropolitan Life Insurance Company had Nord examined by a neurologist, Dr. Antoine Mitri, who concurred in the treating doctors= diagnosis of degenerative spinal disc disease, but disagreed as to whether that condition caused a disability.  However, Nord supported his claim with opinions from his primary treating doctor, which were corroborated by opinions of two other treating doctors.  Instead of articulating a reason for disagreeing with the treating doctors, the Ninth Circuit found the administrator appears merely to have preferred to rely upon the more favorable conclusions of its own [email protected]  296 F.3d at 831.  The Ninth Circuit earlier ruled in Regula v. Delta, supra. that the treating physician [email protected] in the context of ERISA required that the plan administrator could reject the treating physician's reports, but it had to show its decision was supported by specific, legitimate reasons that are based on substantial evidence in the [email protected]  266 F.3d at 1147.  Instead, in Nord, the plan administrator A simply asserted at every turn, and again before this Court, that it was under no duty to consider evidence that was unfavorable to its determination, whether coming from Nord's physicians or from its own human resources [email protected]  266 F.3d at 831.  The treating physician rule guards against such actions.

Regula further highlights the value of the treating physician rule in ERISA cases.  After experiencing an on the job injury, Regula qualified to receive long term disability benefit payments in 1987.  Every three months thereafter, in order to continue receiving benefits, Regula was required to submit an updated attending physician's report.  The plan accepted reports from the same treating physician until 1995 when the plan suddenly terminated benefit payments.  Even though the plan relied on examination results from two separate doctors, the court found that the termination decision was deficient due to its failure to explain why the treating physician's opinion was suddenly ignored when it had been accepted thirteen times over an eight year period.  Although the Ninth Circuit applied the treating physician [email protected] to ERISA claims, it made it clear that the deference to the treating physician Ais not [email protected]  266 F.3d at 1140.  Moreover, when other physicians provide evidence consistent with the evidence in the record overall,@ the treating physician's opinion would not control.  Id.

Although the briefing in Nord has expressed concern about the honesty of the treating physician, of greater concern to the court in Regula was

the conflict of interest inherent when benefit plans repeatedly hire particular physicians as experts.  Especially in cases such as this one, where the plan administrator is also the funding source, these experts have a clear incentive to make a finding of >not disabled= in order to save their employers money and to preserve their own consulting arrangements.  266 F.3d at 1143.

Moreover, Regula explained that courts rejecting the application of the treating physician rule have done so in the context of health benefits claims, where the physician may have a financial motive to assure payment, unlike disability benefits cases where the patient receives the payment and not the physician pursuant to assignments of the right to receive insurance payments.  266 F.3d at 1143.  The court's differentiation between health benefits claims and disability claims is borne out by the medical articles cited by Petitioner.  For example, in Wynia, et al., A Physician Manipulation of Reimbursement Rules for Patients: Between a Rock and a Hard Place,@ Journal of the American Medical Association 283:1858 (2000), the authors limited their presentation only to medical benefits and still found manipulation of reimbursement rules relatively uncommon;@ and even among those who admitted to it, the majority still did not believe it is an ethical [email protected]  Another article, Freeman, et al., Lying for Patients: Physician Deception of Third-Party Payers,@ Archives of Int.Med. 159: 2263 (1999), was also restricted to health insurance reimbursement and concluded that deception was correlated with the clinical severity of the patient's [email protected] None of these studies provide any useful guidance, though, that would support an argument that a treating physician would furnish an opinion in a disability benefit claim that was not warranted by the evidence.  Furthermore, the caveats to the application of the treating physician rule operate as a check against such a practice if the physician's opinion was inconsistent with the record as a whole; or if that opinion could be disputed by reliance on other clinically supported evidence

Since the issuance of Regula and Nord, the Sixth Circuit has added further support for the utilization of the treating physician rule in ERISA-governed disability benefits claims.  In Darland v. Fortis Benefits Insur.Co., 317 F.3d 316 (6th Cir. 2003), the court concurred with Regula=s determination that an insurer that rejects a treating doctor's opinion in favor of opinions of a consultant who merely reviews the medical evidence and does not examine the claimant is evidence of a conflict of interest.  The court found that utilization of Network Medical Review, a company providing reviews of disability benefits claims for insurers, illustrates that Fortis has a >clear incentive= to contract with a company whose medical experts were inclined to find in its favor that Darland was not entitled to continued LTD [email protected] 317 F.3d at .  Relying heavily on Regula, but clarifying that deference to the treating doctor's opinion is not absolute, Darland held the district court should have deferred to the opinions of Darland=s treating physicians absent substantial evidence in the record contradicting those [email protected]  317 F.3d at .  The court further explained:

The treating physician rule has particular applicability to the factual circumstances of this case where there is such a stark dichotomy between the opinions of the treating physicians who possessed an abundance of first-hand knowledge of the patient's medical condition and those of the non-treating medical consultants who were hired by a company selected by Fortis, the plan administrator, which had a financial stake in the matter as the insurer who ultimately pays the benefits.  Although we do not dispute the concurrence's assertion that the treating physician's opinion does not >trump= all other evidence, we do maintain that the treating physician's opinion is entitled to deference particularly when, as in the matter at hand, there is an absence of substantial evidence to the contrary.  317 F.3d at .

These cases therefore illustrate that the treating physician rule merely complements the conflict of interest factor elucidated in Firestone, supra.  Although the Court was unclear with respect to the practical import of remediating the conflict, it has been well-recognized that there is an inherent conflict between the roles assumed by an insurance company that administers claims under a policy it [email protected] since it Apays out to beneficiaries from its own assets rather than the assets of a trust, its fiduciary rule lies in perpetual conflict with its profit-making role as a [email protected]  Brown v. Blue Cross & Blue Shield of Ala., 898 F.2d 1556, 1561 (11th Cir. 1990).

This point, of course, squarely illustrates the difficulty in the Court's application in Firestone of trust law principles onto benefit plans governed by contractual agreements rather than trusts.  For that reason, several courts have found that decisions made by conflicted fiduciaries require a heightened level of scrutiny even if the plans (or insurance policies) contain the deferential language that Firestone ruled may be included in such plans in order to trigger an arbitrary and capricious standard of review.  In addition to Brown, the Fourth Circuit, in Doe v. Group Hospitalization & Med.Servs., 3 F.3d 80 (4th Cir. 1993), reached such a conclusion, as did the Fifth Circuit in Vega v. National Life Ins.Co., 188 F.3d 287 (5th Cir. 1999), the Sixth Circuit in Darland, supra., the Eighth Circuit in Armstrong v. Aetna Life Ins.Co., 128 F.3d 1263 (8th Cir. 1997), the Ninth Circuit in Atwood v. Newmont Gold Co., Inc., 45 F.3d 1317 (9th Cir. 1995), and the Tenth Circuit in Pitman v. Blue Cross & Blue Shield of Okla., 24 F.3d 118 (10th Cir. 1994) to name some of the more significant rulings on this issue.  Nonetheless, some circuits require a specific showing that the conflict infected the decision reached before the deference applied to the claim decision is diminished.  Those circuits include the Second in Whitney v. Empire Blue Cross & Blue Shield, 106 F.3d 475 (2d Cir. 1997), the Seventh Circuit in Mers v. Marriott Intl. Group Accidental Death and Dismemberment Plan, 144 F.3d 1014 (7th Cir. 1998).

To harmonize these rulings, the Supreme Court will need to be clear as to what evidence constitutes a demonstration of a conflict of interest since, like employers who commit acts of discrimination, insurers are unlikely to admit to acting pursuant to a bias.  Nord merely supplies a means of inferring a bias; i.e., when a plan simply chooses to prefer an opinion differing from the treating doctor's opinion without articulating an adequate basis for rejecting the treating doctor's opinion.  Regula was even more explicit and referenced an unsettling pattern of inconsistency and insufficiency in the plan administrator's reasons for terminating the appellant's [email protected]  266 F.3d at 1146.  That was shown by a Asudden [email protected] of benefits which came abruptly with no evidence alleged of a significant change in [Regula's] [email protected]  Id.  Additionally, the plan disregarded the treating physician's evidence without providing specific reasons for their [email protected]  266 F.3d at 1147.  Other grounds for finding a conflict of interest include a self-serving selectivity in the analysis of evidence such as crediting only a portion of a medical report or rejecting an internal recommendation that a claim be paid.  Pinto, supra. 214 F.3d at 393.

Accordingly, Nord=s consideration of the treating physician rule in the context of a conflict of interest remediates the inherent bias existing when the plan both administers and pays benefits out of its own assets.   Given the parameters of such a rule, which does not, as Petitioner and amici argue before the Supreme Court, allow the treating physician to trump all other opinions, utilization of the treating physician rule in the context of ERISA benefits in the same manner as the rule is applied in Social Security disability claims is far more in harmony with ERISA's goal of protecting... the interest of participants in employee benefit plans and their [email protected] (29 U.S.C. '1001(b)) than any alternative.  So long as decisions made by insurers and plan administrators are largely insulated from meaningful review based on the inclusion of discretionary language in insurance policies and benefit plans, claimants are denied the protections of the ERISA law, just as claimants were denied the protection of the Social Security law prior to the adoption of a treating physician rule.  Therefore, the Court should adopt a standard such as the McDonnell Douglas burden shifting standard applied to discrimination cases; i.e., once evidence of conflict is shown, the burden shifts to the plan administrator to show that the decision was not influenced by the conflict of interest.  Rather than wrestling with conflicting rules about how the conflict is to be resolved, a rule placing the burden on the conflicted insurer or plan administrator to explain its actions is simple and consistent in its application.

This note appeared in the March 2003 issue of the Disability E-News Alert! For subscription information, please go to www.disabilityenewsalert.com.

Notice: Neither by accessing this site or by reviewing its contents has an attorney-client relationship been formed or established; and nothing contained in this site shall constitute the giving or rendering of legal advice or be construed as a legal opinion, or guarantee of a particular resolution of a legal problem. This information is provided as a public service, and is not intended to be a substitute for competent legal counsel. The information provided is general in nature and may not apply to your circumstances, particularly if you are not in the State of Illinois. Under no circumstances should you make legal decisions solely based upon the information provided on this web site. You should consult an attorney before making any important decision involving a legal matter.
Copyright 2002 Daley, DeBofsky & Bryant.