IN THE UNITED STATES COURT OF APPEALS
FOR THE FIFTH CIRCUIT
_______________
No. 98-40015
_______________
KATHLEEN E. MYERS,
Plaintiff-Appellee,
VERSUS
KODAK HEALTH IMAGING SYSTEMS LONG-TERM DISABILITY PLAN, et al.,
Defendants,
KODAK HEALTH IMAGING SYSTEMS LONG-TERM DISABILITY PLAN,
Defendant-Appellant.
_________________________
Appeal from the United States District Court
for the Eastern District of Texas
(4:95-CV-261)
_________________________
March 22, 1999
Before JONES, SMITH, and EMILIO M. GARZA, Circuit Judges.
JERRY E. SMITH, Circuit Judge:*
Kathleen Myers applied for long-term disability benefits from
Kodak Health Imaging Systems Long-Term Disability Plan. The plan
administrator denied the request, and Myers sought judicial review
under § 1132(a)(1)(B) of the Employee Retirement Income Security
*
Pursuant to 5TH CIR. R. 47.5, the court has determined that this opinion
should not be published and is not precedent except under the limited
circumstances set forth in 5TH CIR. R. 47.5.4.
Act of 1974 (“ERISA”).1 The district court reversed the plan
administrator's denial of benefits because it determined that she
had abused her discretion in denying benefits. Finding no abuse of
discretion on the administrator's part, we reverse the judgment of
the district court.
I.
Myers began working for Vortech Data, a company eventually
acquired by Kodak Health Imaging Systems (“KHIS”), in April 1991.
She ultimately became KHIS's regulatory and compliance manager.
Her annual salary, which was $58,000 when she began with the
company, grew to $90,525 by early 1994. She also took advantage of
an employee benefits package that included short-term disability
benefits and membership in the Kodak Health Imaging Systems Long-
Term Disability Plan (the “LTD Plan” or “the Plan”), which is the
defendant in this case.
The version of the Plan in which Myers participated promised
to pay seventy percent of her salary should she become totally
disabled. She paid for this benefit by payroll deduction. KHIS
provided her with a handbook entitled Kodak Health Imaging Systems
Choices for Tomorrow, which included a summary (the “summary plan
description” or “SPD”) of the LTD Plan's benefits. Myers read and
relied on the benefits handbook when she received it.
1
Section 1132(a)(1)(B) authorizes a beneficiary of, or participant in, a
plan to sue “to recover benefits due to him under the terms of his plan, to enforce
his rights under the terms of the plan, or to clarify his rights to future benefits
under the terms of the plan.” 29 U.S.C. § 1132(a)(1)(B).
2
The LTD Plan itself spelled out in greater detail the terms of
the disability coverage. It provided that participants could
receive benefits in the event they became totally disabled while
covered by the Plan, and it defined “disability” to require, inter
alia, a “condition [that] results in a Participant's total and
continual inability to engage in Gainful Work.”
The LTD Plan and the SPD give different definitions of
“Gainful Work.” Under the LTD Plan, “'Gainful Work' means paid
employment for which a person is, or becomes, reasonably qualified
by education, training, or experience, and which is more than
transitory in nature, as determined by the Claims Administrator
. . . [but which] does not include Rehabilitative Employment.”
Under the SPD, “'Gainful Work' is substantial paid employment for
which [claimant is] (or [claimant] become[s]) reasonably qualified
by education, training or experience, all as determined by
MetLife.”2
In early 1994, KHIS conducted a salary study and decided that
Myers's salary of $90,525 exceeded the benchmark range for her
position. On May 25, 1994, her supervisor, Carl Alletto, informed
her that her salary would be reduced to $70,610. Thereafter, she
never reported for work.
Myers immediately began short-term disability leave, during
which she sent several detailed letters to the administrator of the
LTD Plan, concerning Myers's compensation and other matters related
to her employment. In October 1994, Myers applied for long-term
2
MetLife was designated as the claims administrator.
3
benefits under the Plan. When her short-term disability benefits
terminated on November 25, 1994, her employment ended. This date
is significant, because the LTD Plan requires that a claimant have
become “totally disabled” on or before the date his employment
ends.
The Plan appointed MetLife to administer Myers's application3
and retained National Medical Review (“NMR”) to assist in reviewing
the application. NMR provided physicians, who were board-certified
in occupational medicine, to review the medical information Myers
had submitted in connection with her application for benefits,
analyze the information within the framework of the LTD Plan and
its SPD, and issue an opinion as to whether she had experienced
“total disability” before her termination.
MetLife sent the two physicians who reviewed Myers's files,
Drs. Robert Porter and Robert Petrie, a “Medical Review Referral
Form” stating that KHIS's “definition of disability is: must be
totally disabled from any and all occupations from the onset.”
Together, Petrie and Porter issued six opinions stating that Myers
was not totally disabled as of November 25, 1994. The district
court found that Petrie and Porter had relied on the “any and all
occupations” definition provided by MetLifeSSnot the definition in
the SPDSSin reaching their conclusions.
Based on its determination that Myers was not totally
disabled, MetLife initially denied Myers's application on
3
The terms of the LTD Plan permitted the appointment of a claims
administrator to make the initial determination of whether a claimant is eligible
for long-term disability benefits.
4
December 20, 1994, but, because it had experienced some difficulty
in obtaining information from Dr. David Buhner, one of Myers's
physicians, MetLife advised Myers that she could submit additional
information in support of her application. Upon receiving
information from Buhner, MetLife again denied the application on
February 15, 1995, and instructed Myers on how to appeal the
denial.
Myers did so on March 9, 1995. During her appeal, she
submitted to MetLife and the plan administrator voluminous
documents, including medical literature, several sets of medical
records, and statements of physicians, family members, and friends.
MetLife forwarded this information to the NMR doctors, who
continued to opine that Myers was not totally disabled and could
perform sedentary jobs. On September 19, 1995, the plan
administrator formally denied Myers's appeal, explaining that,
“[a]ccording to all of the medical and vocational information we
have received regarding Ms. Myers' claim, it has not been
established that she is totally disabled.”
Myers sought judicial review under § 1132(a)(1)(B). The
district court reversed the denial, finding that the plan
administrator “relied on the standard in the LTD plan document
[when] [t]he standard of disability which should be applied to
Plaintiff's benefits claim is the standard of disability contained
in the SPD for the LTD Plan.” The court concluded that “[t]he
decision to deny Plaintiff's benefit claim and her appeal was not
supported by substantial evidence if the definition of disability
5
in the SPD is used.” The court declined to remand to the
administrator for application of the correct standard; it simply
ordered the Plan to pay LTD benefits and awarded attorneys' fees.
II.
The Plan contends that there is substantial evidence in the
record to support the administrator's decision to deny LTD
benefits, that the administrator did not abuse her discretion in
denying Myers's application, and that the court thus erred in
reversing the administrator's decision. We agree.
A.
A somewhat complicated standard of review governs ERISA
actions in which plaintiffs appeal benefits determinations. The
district court first reviews the plan administrator's decision, and
we must then review that court's analysis. Because our review
requires us to evaluate the district court's performance, we first
delineate the standard of review the district court is to apply to
the benefits determination. We then consider the standard of
review we apply to the district court's decision.
1.
A plan administrator deciding whether to pay benefits “must
make two general types of determinations: 'First, he must determine
the facts underlying the claim for benefits . . . . Second, he
must then determine whether those facts constitute a claim to be
6
honored under the terms of the plan.'” Schadler v. Anthem Life
Ins. Co., 147 F.3d 388, 394 (5th Cir. 1998) (quoting Pierre v.
Connecticut Gen. Life Ins. Co., 932 F.2d 1552, 1557 (5th Cir.
1991)). For the administrator's factual determinations, the
district court is to apply an abuse of discretion standard.
Meditrust Fin. Servs. Corp. v. Sterling Chems., Inc., Med. Benefits
Plan, No. 98-40204, 1999 U.S. App. LEXIS 3643, at *5 (5th Cir.
Mar. 4, 1999) (citing Pierre, 932 F.2d at 1562). Abuse of
discretion review, as applied to an administrator's factual
determinations, “is limited to determining whether there is
substantial evidence in the record to support [the plan
administrator's decision].”4 Bellaire Gen. Hosp. v. Blue Cross
Blue Shield, 97 F.3d 822, 828 (5th Cir. 1996).
As for the second type of determination a plan administrator
must make (i.e., interpretation and application of the plan), the
district court is to review the administrator's decision “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,” in which case the
abuse of discretion standard applies. Schadler, 147 F.3d at 394
4
An alternative characterization of our abuse of discretion review, as
applied to an administrator's factual determinations, is “arbitrary and capricious”
review. See, e.g., Penn v. Howe-Baker Eng'rs, Inc., 898 F.2d 1096, 1100 (5th Cir.
1990). Although in Pierre, 932 F.2d at 1562, we refused to equate “abuse of
discretion” and “arbitrary and capricious” standards, our analysis in that case was
anomalous, for, as we recently explained, “we decline to follow Pierre to the extent
that it rejects the use of the “arbitrary and capricious” analysis as part of abuse-
of-discretion review.” Meditrust, 1999 U.S. App. LEXIS 3643, at *9. Instead, we
are bound by Penn's holding that the abuse of discretion standard of review for
factual determinations amounts to arbitrary and capricious review. Id. This is
consistent with the great weight of authority from other circuits. Id. at *9 n.6.
7
(quoting Firestone Tire & Rubber Co. v. Bruch, 489 U.S. 101, 115
(1989)). The LTD Plan plainly vested the plan administrator with
discretionary authority to interpret plan terms and determine
eligibility,5 so the abuse of discretion standard of review applied
to the administrator's plan interpretation and eligibility
determination.
When the issue under review is plan interpretation and
application, as opposed to pure fact-finding, the abuse of
discretion standard involves a two-step inquiry:6
The court must initially determine whether the
administrator's interpretation of the plan is the legally
correct interpretation. If the administrator's
interpretation of the plan is legally correct, then the
inquiry ends because no abuse of discretion could have
occurred. However, if the court determines that the
administrator's determination is not legally correct,
then it must further determine whether the
administrator's decision was an abuse of discretion.[7]
“The fact that an administrator's interpretation is not the correct
one does not in itself establish that the administrator abused his
5
Article 9.1(c) of the LTD Plan provided:
The Plan Administrator shall have full discretionary authority in all
matters related to the discharge of responsibilities and the exercise
of authority under the Plan including, without limitation, his
construction of the terms of the Plan and his determination of
eligibility for Coverage and Benefits. It is the intent of the Plan
that the decisions of the Plan Administrator and his action with
respect to the Plan shall be conclusive and binding upon all persons
having or claiming to have any right or interest under the Plan . . . .
6
The two-step inquiry does not apply to factual determinations, which are
subject to an abuse of discretion review that essentially amounts to “look[ing] at
the administrative record and determin[ing] whether the administrator's decision was
supported by substantial evidence.” Rigby v. Bayer Corp., 933 F. Supp. 628, 632-33
(E.D. Tex. 1996). See Meditrust, 1999 U.S. App. LEXIS 3643, at *11; see also
Rutledge v. American Gen. Life & Acc. Ins. Co., 914 F. Supp. 1407, 1410-11 (N.D.
Miss. 1996).
7
Schadler, 147 F.3d at 394 n.5 (quoting Spacek v. Maritime Ass'n, ILA Pension
Plan, 134 F.3d 283, 292-93 (5th Cir. 1998). See also Wildbur v. ARCO Chem. Co., 974
F.2d 631, 637 (5th Cir. 1992).
8
discretion,” Wildbur, 974 F.2d at 638, and we have identified three
considerations that “are important” in analyzing whether there was
an abuse of discretion: “(1) the internal consistency of the plan
under the administrator's interpretation, (2) any relevant
regulations formulated by the appropriate administrative agencies,
and (3) the factual background of the determination and any
inferences of lack of good faith.” Id. (citation omitted).
The plan administrator made a factual determination (i.e.,
that Myers is not disabled) and an interpretive decision (i.e.,
that the facts warranted a denial of benefits). Both are subject
to abuse of discretion review by the district courtSSthe former to
review for substantial evidence, and the latter to review under the
two-step test articulated in Wildbur.
2.
Our review of the district court's analysis is governed by the
following principles:
On appeal from a district court's judgment in a
§ 1132(a)(1)(B) case, our traditional standards of review
apply, and we review de novo the district court's holding
on the question of whether the plan administrator abused
its discretion or properly denied a claim for benefits.
However, we will set aside the district court's factual
findings underlying its review of the plan
administrator's determination only if clearly erroneous.
Sweatman v. Commercial Union Ins. Co., 39 F.3d 594, 601 (5th Cir.
1994). Accord Meditrust, 1999 U.S. App. LEXIS 3643, at *7 (citing
Sweatman, 39 F.3d at 601). The key question, then, is whether the
Plan is attacking a finding of fact by the district court (i.e.,
the court's factual determination that Myers was disabled) or the
9
court's holding that the plan administrator's denial of benefits
was an abuse of discretion.8 If the former, then this court would
review the determination for clear error, for the question of
whether a claimant is disabled is “more factual in nature than
interpretive in nature,” and a district court answering that
question is therefore entitled to greater deference. Sweatman,
39 F.3d at 598. If, however, the Plan is attacking the district
court's determination that the plan administrator abused her
discretionSSa mixed question of law and factSSde novo review is
appropriate. See id. at 600.
The Plan cannot be challenging the court's factual
determination that Myers is “disabled” under the terms of the LTD
plan, because the court did not make such a finding. In its list
of twenty-two “findings of fact,” the court entered only one
finding that has anything to do with Myers's disability: In
finding number 22, the court determined that “[t]he decision to
deny Plaintiff's benefit claim and her appeal was not supported by
substantial evidence if the definition of disability in the SPD is
used.”
In other words, the court did not look at the evidence and
make the purely factual determination that Myers was disabled.
Instead, it examined the record and determined, as a matter of law,
8
The parties disagree over what standard of review applies to the district
court's determination that the Plan's decision to deny Myers's claim was not
supported by substantial evidence. The Plan contends that we review the
determination de novo; Myers asserts that the determination is factual and is
therefore subject to an abuse of discretion standard of review. We agree with the
Plan.
10
that the evidence was insufficient to support a finding of
disability under the SPD. The court thus answered either a
question of law or a mixed question of law and fact, and the proper
standard of review is therefore de novo.9 Under such a standard,
our task is to determine, upon a fresh look at the record and
without affording deference to the district court, whether that
court erred in determining that the administrator abused her
discretion in denying Myers's request for LTD benefits.
B.
The district court erred in holding that the plan
administrator abused her discretion in denying benefits. Myers had
the burden of proving entitlement to LTD benefits under the plan,10
and the adminisrator's determination that Myers had failed to
9
The Sweatman court outlined the general principles guiding the determination
of which standard of review to apply:
On appeal, our standard of review for district court decisions
reviewing plan administrators' eligibility determinations is guided by
the principles that typically guide our standard of review. Namely,
we review questions of law de novo and set aside factual determinations
only if clearly erroneous. Consistent with these principles, we review
a district court's determination of whether a plan administrator abused
its discretionSSa mixed question of law and factSSde novo.
39 F.3d at 600. Accord Meditrust, 1999 U.S. App. LEXIS 3643, at *7.
10
“A claimant under section 1132(a)(1)(B) has the initial burden of
demonstrating an entitlement to benefits under an ERISA plan, or that a denial of
benefits under an ERISA plan is arbitrary and capricious.” Perdue v. Burger King
Corp., 7 F.3d 1251, 1254 n.9 (5th Cir. 1993). Hence, Myers initially had the burden
of proving that she was totally disabled.
The district court's factual determination that the administrator and the NMR
physicians evaluated Myers's application using the wrong standard of gainful work
(and thus did not prove that she was not totally disabled) does not entail its
conclusion that she was entitled to benefits; she still had to prove disability.
Hence, evidentiary equipoise would indicate that there was sufficient evidence in
the record to sustain the administrator's denial and that the administrator thus did
not abuse her discretion.
11
shoulder that burden was not an abuse of discretion. The
administrator's interpretive decisions (i.e., her interpretation of
the plan and her decision to deny benefits given the facts found)
did not constitute abuses of discretion under the two-step inquiry
articulated in Wildbur, and her factual determinations were
supported by substantial evidence.
1.
We first consider the plan administrator's interpretive
decisions, for it was her interpretation of the term “gainful work”
that most troubled the district court. The court determined, as a
matter of fact, that the administrator's denial of Myers's request
for benefits relied on the LTD Plan's “paid employment” definition
of the term “gainful work” and that the administrator had not
employed the SPD's “substantial paid employment” definition of the
term. There is sufficient support for this finding in the record,
and it is thus not clearly erroneous.11
11
The administrator's final letter denying Myers's request for benefits
references § 2.30 of the LTD PlanSSthe provision that defines “gainful work” as
“paid employment”SSand states that Myers is not totally disabled because she can
perform such work. The Plan argues that the district court erred in finding that
the administrator used the LTD Plan's definition of gainful work, because the
administrator's testimony indicates that she actually applied the SPD's definition
of the term and considered only “substantial paid employment.” Indeed, when asked
“What did you interpret the definition of gainful work under the LTD Plan to mean?”,
the administrator responded:
Work for which an individual is reasonably qualified and which was not
transitory. I interpreted it as not being menial. I interpreted it
not as being insubstantial, but definitely what someone was reasonably
qualified to do based on their education and training, experience.
[Emphasis added.]
Despite this testimony, we decline to reverse the district court's factual
determination that the administrator employed the LTD Plan's definition of gainful
work. Soon after the testimony cited above, the court interrupted
(continued...)
12
In employing the LTD Plan's definition of “gainful work,” the
administrator failed to give the Plan the legally correct
interpretation. In Hansen v. Continental Ins. Co., 940 F.2d 971
(5th Cir. 1991), we required plan administrators and courts to
apply the terms of the summary of a benefits plan when those terms
conflict with those of the full plan document.12 Here, the summary
(...continued)
examination and directly asked the administrator, “Do you consider in your mind that
there is a difference between paid employment and gainful employment?”, to which the
answer was “No.” This frank resonse supports the finding that the administrator
interpreted gainful work as the LTD Plan providedSSi.e., as paid employmentSSand the
court thus did not clearly err in finding that the administrator applied the LTD
Plan's “paid employment” standard.
12
The Plan asserts that the rule articulated in Hansen applies only “where
the summary description taken in its most natural reading would entitle a plan
participant to plan benefits and a natural reading of the plan document would
not.” But Hansen nowhere states that the rule announced therein is so limited.
The Hansen court made its holding plain:
This Court holds that the summary plan description is binding, and
that if there is a conflict between the summary plan description and
the terms of the policy, the summary plan description shall govern.
Any other rule would be, as the Congress recognized, grossly unfair
to employees and would undermine ERISA's requirement of an accurate
and comprehensive summary.
Hansen, 940 F.2d at 982. Furthermore, it stated that
[a]ny burden of uncertainty created by careless or inaccurate
drafting must be placed on those who do the drafting, and who are
most able to bear that burden, and not on the individual employee,
who is powerless to affect the drafting of the summary or the policy
and ill equipped to bear the financial hardship that might result
from a misleading or confusing document.
Id. (emphasis added).
The case the Plan cites as “explaining the limited reach of Hansen” also
fails to support the Plan's claim that Hansen covers only major conflicts. In
that case, Wise v. El Paso Natural Gas Co., 986 F.2d 929, 939 (5th Cir. 1993),
we did state that “[a] careful reading of Hansen . . . reveals that its principal
concern was with positive inconsistencies, either within the SPD or between the
SPD and the master documents.” We were not saying, however, that the
inconsistencies had to be great. The issue was whether there was an ambiguity
requiring a pro-beneficiary interpretation in an SPD that included information
about amendment and termination of the plan under the heading “OTHER IMPORTANT
INFORMATION” instead of under the heading “WHEN YOUR COVERAGE WILL END.” In
requiring “positive inconsistencies” to invoke the Hansen rule, we were simply
requiring participants to show an actual inconsistency between the text of the
(continued...)
13
defined gainful work as “substantial paid employment,” and the LTD
Plan defined the term as “paid employment . . . which is more than
transitory in nature.” There is a genuine difference between these
two definitions,13 and under Hansen, the SPD's definition applies.
Hence, the administrator's application of the LTD Plan's “paid
employment” definition of gainful work was legally incorrect.
“The fact that an administrator's interpretation is not the
correct one does not in itself establish that the administrator
abused his discretion.” Wildbur, 974 F.2d at 638. Indeed, “[i]f
a court concludes that the administrator's interpretation is
incorrect, the court must then determine whether the administrator
abused his discretion” in denying benefits. Id. Balancing
“(1) the internal consistency of the plan under the administrator's
interpretation, (2) any relevant regulations formulated by the
appropriate administrative agencies, and (3) the factual background
of the determination and any inferences of lack of good faith,”
id., we conclude that the plan administrator's decision to deny
(...continued)
SPD and that of the plan; we were declining to find an impermissible ambiguity
on the basis of a shoddily organized, but not inconsistent, SPD.
13
The Plan argues that there really is no difference between the definitions
of “gainful work” in the two documents. The SPD defines gainful work as
“substantial paid employment,” and the LTD Plan defines it as “paid employment
. . . which is more than transitory in nature.” The district court erred, the Plan
urges, in ignoring the “more than transitory in nature” part of the LTD Plan's
definition.
But “substantial paid employment” is not the same as “paid employment
. . . which is more than transitory in nature.” “Transitory” is a temporal measure;
it means “passing, temporary, not lasting.” NEW WEBSTER'S DICTIONARY 1014 (1992).
“Substantial” means “having real existence, not imaginary; firmly based,
. . . relatively great in size, value or importance.” Id. at 987. One could find
nontransitory paid employment that is nonetheless insubstantial. For example, one
might find a permanent part-time job. In such a case, there would be no gainful
work under the SPD, but there would be gainful work under the LTD Plan. Hence,
there is a genuine, positive conflict between the terms of the two documents.
14
benefits did not amount to an abuse of discretion, despite the fact
that she employed an incorrect interpretation of the plan.
While the administrator's interpretation of gainful work
(i.e., non-transitory paid employment) and the correct definition
under the plan (i.e., substantial paid employment) are somewhat
inconsistent, the inconsistency is not grave. Moreover, despite
the fact that there is a theoretical possibility that the
administrator could have found Myers able to perform non-
transitory, but not substantial, paid employment, the evidence
indicates that the jobs for which she found Myers qualified were,
in fact, substantial.14 Hence, any inconsistency between the
administrator's interpretation and the applicable terms of the plan
was irrelevant.
Myers has pointed to no “relevant regulations” that would
indicate an abuse of discretion, so the only factor left to balance
is the factual background of the termination and any inferences of
lack of good faithSSa factor that “really involves two separate
14
When asked, “What kinds of work specifically did you envision in
reaching your determination that Ms. Myers could engage in gainful work?”, the
administrator responded:
I considered knowing the type of work she was doing and knowing her
credentials from her resume. I considered such things as consulting
work for a contract employer that would utilize her skills and
understanding; an awful lot of computer-oriented systems,
regulations for FDA GNP, a lot of skills and experience that I feel,
based on my experience, are well sought after bySSby other companies
as they were by us.
I considered her knowledge of not only our products, but the
products of other businesses that she had worked for that she might
be able to perform a job in computer service or handling customer
complaints. Her computer skills had exposed her to many different
industries, like banking, insurance, that she might be able to,
again, provide some consultant services in that direction.
15
questions.” Wildbur, 974 F.2d at 638. There is plenty of factual
support in the record to sustain a determination that Myers was not
totally disabled, regardless of what definition of gainful work is
applied. See infra part II.B.2.
There is also no evidence suggesting bad faith on the part of
the administrator. Although she was an employee of KHIS, and we
are called to “weigh any potential conflict of interest in [our]
determination of whether [a] plan administrator abused its
discretion,” Sweatman, 39 F.3d at 599, there is no evidence that
the administrator was under any pressure to avoid paying benefits
to deserving claimants; the mere fact that an administrator is
employed by the entity that must pay benefits does not create an
inference of bad faith. See id.
Absent some concrete showing of bad faith, we will not infer
such from the mere fact that the administrator was employed by the
plan, especially given the strong evidence, discussed below,
indicating that Myers was not totally disabled. Our performance of
the two-step abuse of discretion standard for plan interpretation
and benefits determination decisions therefore reveals no abuse of
discretion in the administrator's interpretive decisions.
2.
The administrator's factual determination that Myers was not
totally disabled also survives abuse of discretion review. The
administrator determined that “Myers has the skills required to
perform a number of sedentary jobs” and is thus not “totally
16
disabled.”15 Under the abuse of discretion standard of review for
factual determinations, the district court was to determine whether
substantial evidence supports this finding. It concluded that
there was insufficient evidence to support the finding, because the
administrator “reviewed Myers' claim under the wrong standard and
relied on reports based on the wrong standard.” Regardless of what
standard the administrator applied, however, there was plenty of
evidence in support of the factual determination that Myers could
perform substantial sedentary work and was thus not totally
disabled.
The administrator's personal observations of Myers supported
the conclusion that she was not totally disabled. The
administrator had numerous opportunities to observe Myers. They
worked on the same floor from late 1993 until January 1994, and
saw each other several times a day. From January 1994 until Myers
began short-term disability in May 1994, the two worked on
different floors but still saw or interacted with each other twice
or thrice a week. At no time during these interactions did the
administrator observe any manifestations of Myers's claimed
disability. Plan administrators may consider their personal
observations of claimants in reviewing applications for disability
benefits.16
15
The question whether a claimant is disabled is “more factual in nature than
interpretive in nature.” Sweatman, 39 F.3d at 598.
16
See Sweatman, 39 F.3d at 597 (plan administrator relying on, among other
things, claim investigator's observations of participant's ability to engage in
commonplace activities).
17
The administrator also considered the fact that Myers never
had reported what Myers now casts as long-standing and severe
health problems to the KHIS Human Resources Department, which the
administrator directed. As Human Resources Director, the
administrator was kept apprised of the debilitating health
conditions and chronic absentee or tardiness problems of all KHIS
employees, and the fact that she never heard anything about Myers's
alleged disabilities is evidence that Myers was not as disabled as
she now claims.17 In addition, Myers's direct supervisor informed
the administrator that he was unaware that Myers was suffering from
any extended illness or disabling condition.
The circumstances surrounding Myers's departure on leave also
contributed significantly to the substantial evidence underlying
the administrator's factual determination that Myers was not
totally disabled. Myers left KHIS the very day she was told her
salary would be cut from $90,500 to $70,610. Her annual benefits
under the LTD Plan would total 70% of $90,500, or $63,350SSquite
close to the salary she would have earned after the reduction.
This “fishy” departure, in combination with the administrator's
observations of Myers at work and Myers's failure to report any
serious ailments to Human Resources, indicates that Myers may have
overstated her disabling conditions, choosing to live off
17
Myers correctly asserts that the absence of reports of her health
problems to KHIS's Human Resources Department does not establish the absence of
such health problems. But Myers had the burden of proving total disability; the
Plan did not have the burden of proving the lack thereof. See Perdue, 7 F.3d at
1254 n.9. The administrator thus could rely on the lack of reports to Human
Resources in determining that Myers had not shouldered her burden.
18
disability payments instead of a reduced salary.
The correspondence Myers sent the administrator during the
time Myers was on short-term disability leave provided additional
evidentiary support for the administrator's factual determinations.
The administrator testified,
I was receiving documents from [Myers] that seemed to
demonstrate her ability, as I recall, to write a well-
constructed letter with details. It was lucid, well-
constructed. And I felt from that that it supported my
feeling that there was work that Ms. Myers could do. It
also seemed to demonstrate to me her ability to type.
Myers asserts that the letters she sent the administrator while on
short-term disability leave provided no evidence of a lack of total
disability, because Myers was assisted in constructing the letters
and had received a “speaking computer” from KHIS. The
administrator, however, did not know about either of these things.
When asked, “[D]id you have any information from Ms. Myers that
she, indeed, had not typed or constructed the letters that you
received?”, the administrator replied, “No, absolutely not.” The
evidence before the administrator when she denied Myers's
application, then, supports the view that Myers could craft
persuasive written documents and thus was not totally disabled.
The objective medical findings in the NMR physicians' reports
also support the administrator's decision. Even if the NMR doctors
did use the wrong disability standard in reaching their ultimate
conclusion that Myers was totally disabled, their six reports
contain numerous objective medical observationsSSstatements that in
no way relied on the doctors' understanding of gainful
workSSindicating that Myers was capable of substantial paid
19
employment and was not totally disabled, even under the SPD's
definition of gainful work.18
Myers insists that the NMR reports cannot provide sufficient
evidence to support a denial, because the physicians used the wrong
standard of disability and did not opine that she was capable of
“substantial paid employment,” only of “sedentary work.” But the
reports do show that Myers has not established fibromyalgia, carpal
tunnel syndrome, severe arthritis, or chronic pain. See supra
note 18. Moreover, the reports imply that the “sedentary work”
Myers could perform was substantial.19
Myers also contends that the NMR physicians' reports cannot
constitute evidence that she was not disabled, because they are
18
For example, the December 7, 1994, report concluded that Myers possessed
the skills and ability to perform a number of different jobs that “could range from
part-time employment in the home using a computer and modem to returning to her
present place of employment.” The February 10, 1995, report stated that (i) “Ms.
Myers's main problem is fatigue related to her obesity”; (ii) “patients with
fibromyalgia are, in fact, best to continue with productive lifestyles and, in fact
rest and avoidance of socialization has a negative effect on these individuals”; and
(iii) Myers's medical history does not “demonstrate any evidence of severe arthritis
or other conditions that may limit her mobility.”
The June 7, 1995, report states that (i) although Myers's medical history is
“suggestive of a diagnosis of fibromyalgia . . . there is insufficient evidence in
the records to indicate that the diagnosis of fibromyalgia has been made using
strict clinical criteria”; (ii) Myers's medical records indicate “a fleeting C1
neuropathy, which is not consistent with carpel [sic] tunnel”; and
(iii) “individuals with fibromyalgia do benefit from returning to employment and a
productive lifestyle.” And one of Dr. Porter's reports, which reviewed the medical
records of Myers's primary care physician, noted:
The records do support significant obesity and hypertension on a
chronic basis, but there is no suggestion that these diagnoses should
prohibit sedentary work. It is interesting that there are no
complaints of chronic pain or diagnoses of fibromyalgia in these notes
dating over a six and a half year time frame. There is, likewise, no
documentation of carpal tunnel syndrome or other neurological
conditions.
19
Dr. Petrie's December 7, 1994, report concluded that Myers possessed the
skills and ability to perform a number of different jobs “which could range from
part-time employment in the home using a computer and modem to returning to her
present place of employment.”
20
refuted by the opinions of her own treating and examining
physicians. But plan administrators may rely on the opinions of
consulting physicians who review only a claimant's medical
records.20 Consequently, the objective medical observations in the
NMR reports contribute to the substantial evidence underlying the
decision to deny LTD benefits.
Finally, Myers's medical records from her own doctor support
the denial of benefits. Dr. Buhner, Myers's personal physician,
examined her four times before and four times after November 25,
1994, the date by which she must have been totally disabled. On
October 19, 1994, Buhner concluded that “until and unless [Myers]
achieves a good response to medication, she will remain permanently
disabled due to the severity of her pain and fatigue.”
Buhner's next few examinations revealed that Myers was
responding well. His October 25, 1994, examination showed that she
had experienced “about a 40 percent decrease in her pain while on
the medication [Prednisone].” He noted in early December 1994
(just two weeks after the critical November 25, 1994, date) that a
combination of the medications Soma and Flexeril had given Myers
relief from her symptoms, allowed her to sleep better seventy-five
percent of the time and did not create hangovers or other side
effects. He stated that as late as January 1995, Myers “was still
taking Soma and Flexeril and still sleeping better and feeling
somewhat better.” Buhner had stated in October that Myers would be
20
See Sweatman, 39 F.3d at 601-02; see also Salley v. E.I. Dupont de Nemours
& Co., 966 F.2d 1011, 1014 (5th Cir. 1992) (rejecting the notion that plan
administrators must give deference to a claimant's treating physician).
21
totally disabled “unless and until she achieves a good response to
medication,” and his subsequent examinations of her showed that she
had, in fact, achieved such a response.
In short, numerous items of evidence before the
administratorSSthe administrator's personal observations of Myers,
reports about Myers at work, the circumstances of her departure on
leave, correspondence from her, objective observations in the NMR
reports, and the medical records from her own physicianSSsupported
the factual determination that “Myers has the skills required to
perform a number of sedentary jobs” and is thus not “totally
disabled.”
While it is possible that none of these pieces of evidence, in
isolation, would support the plan administrator's conclusion, when
taken together they provide substantial evidence for the conclusion
that Myers was capable of substantial paid employment. They
therefore indicate that the administrator did not abuse her
discretion in finding that Myers was not totally disabled and thus
was ineligible for LTD benefits.
REVERSED.
22