Daniel Collins v. Unum Life Ins. Co. of America , 682 F. App'x 381 ( 2017 )

NOT RECOMMENDED FOR FULL-TEXT PUBLICATION
File Name: 17a0154n.06
Case No. 16-3918
UNITED STATES COURT OF APPEALS
FOR THE SIXTH CIRCUIT
FILED
Mar 09, 2017
DANIEL COLLINS, )
DEBORAH S. HUNT, Clerk
)
Plaintiff-Appellant, )
) ON APPEAL FROM THE UNITED
v. ) STATES DISTRICT COURT FOR
) THE NORTHERN DISTRICT OF
UNUM LIFE INSURANCE COMPANY OF ) OHIO
AMERICA )
)
Defendant-Appellee. ) OPINION
)
BEFORE: COLE, Chief Judge; STRANCH and DONALD, Circuit Judges.
BERNICE BOUIE DONALD, Circuit Judge. Almost a year after Daniel Collins fell
in his employer’s parking lot, fracturing his ankle, his foot was amputated. Pointing to Collins’
long history of diabetes and medical evidence suggesting that it led to the severity of Collins’
injury, Unum Life Insurance Company, the administrator of Collins’ ERISA plan, denied
Collins’ claim for benefits under an exclusion precluding from coverage losses caused in any
way by a disease or illness. We conclude that Unum’s medical expert based its decision on a
view of the entire record and that the decision to deny Collins’ benefits is supported by
substantial evidence, notwithstanding Unum’s inherent conflict of interest. Accordingly, we
AFFIRM the district court’s decision to uphold the benefits denial and deny Collins additional
discovery.
Case No. 16-3918, Collins v. Unum Life Ins. Co. of Am.
I.
Collins was an employee of Affymetrix, Inc. On January 6, 2012, Collins slipped and
fell while walking in Affymetrix’s parking lot, injuring his foot. His foot was amputated around
February 8, 2013. Prior to his injury and through the date of his amputation, reports from
treating physicians indicated that Collins had a history of poorly maintained diabetes.
Dr. Wissam Khoury, Collins’ treating podiatrist, examined Collins multiple times after
his initial injury, including during several procedures conducted in late 2012 to attempt to
salvage his foot, and during the eventual amputation in 2013. During this time period Dr.
Khoury’s reports often noted Collins’ longstanding history with diabetes. His assessments also
associate Collins’ Charcot ankle—a condition resulting in neuropathy, or loss of sensation—with
his diabetes and diabetic neuropathy. Importantly, Dr. Khoury’s pre- and post-operative
diagnoses for the limb-salvaging procedures and the amputation include diabetes with diabetic
neuropathy, alongside other conditions like trimalleolar ankle fracture, Charcot
neuroarthropathy, and osteomyelitits—a bone infection.
Another podiatrist, Dr. Anthony Polito, assessed Collins with diabetic sensory
neuropathy after Collins presented to him “for diabetic foot care.” Administrative R. 13-2, Page
ID 612. Though Collins showed no signs of infection from February until May 2012, Dr. Politio
consistently reported diabetic sensory neuropathy as an objective symptom.
Dr. Mark Berkowitz, another of Collins’ treating physicians, noted in a report signed on
February 12, 2013, that Collins was transferred to the hospital for “chronic right ankle and distal
tibial osteomyelitis,” and a “Charcot ankle fracture.” Administrative R., ECF No. 13-1, Page ID
277. Collins was then counselled about the risks and benefits of an amputation, and was
transferred to the operating room for that procedure.
-2-
Case No. 16-3918, Collins v. Unum Life Ins. Co. of Am.
Afterwards, Collins sought benefits under the select group insurance trust policy (the
“Plan”), which is administered by Unum and governed by ERISA. Unum both determines
eligibility for and pays benefits under the Plan. The Plan contains an exclusion for “accidental
losses caused by, contributed to by, or resulting from . . . disease of the body.”1 Id. at Page ID
180.
In a follow-up visit on March 5, 2013, Dr. Berkowitz’s notes listed Collins’ diagnoses of
diabetes mellitus and type 2 diabetes mellitus, the latter of which was “poor[ly] control[led].” Id.
at Page ID 281–83. On a form from Unum dated March 14, 2013, Dr. Berkowitz again noted
Collins’ diagnosis of “charcot arthropathy” and “chronic osteomyelitis.” Id. at Page ID 127.
Additionally, he circled “No” in response to Unum’s inquiry of whether he thought “the loss
[was] caused in any way by illness or disease.” Id. He did not provide any further explanation.
Unum pathologist Kristin Sweeney was asked to determine the extent, if any, a disease of
the body contributed to Collins’ loss. In a report dated July 10, 2013, she noted Collins’ injuries
and diabetes diagnosis and observed that “diabetics have more complications with trimalleolar
fractures.” Id. at Page ID 405–07. While not opining on whether diabetes played any role in
Collins’ amputation, Sweeney concluded that due to the limited records available, she could not
“determine the degree to which diabetes mellitus may have contributed to Mr. Collins’
complications.” Id. at Page ID 407 (emphasis added). As a result, clinical consultant Marnie
Webb, noting her discussion with Dr. Sweeney that diabetes can contribute significantly to
complications arising from ankle fractures, requested additional records that would be needed to
determine the degree, if at all, diabetes led to Collins’ loss.
1
The Plan also includes a provision excluding participants from receiving benefits for losses that occurred more
than 365 days after the date of the incident. Because we conclude that the Collins’ claim was properly denied
because diabetes contributed to his loss, we decline to address whether he would be excluded from coverage under
this provision.
-3-
Case No. 16-3918, Collins v. Unum Life Ins. Co. of Am.
Another Unum pathologist, Dr. Barbara Golder, was also asked to determine to extent, if
any, a disease of the body contributed to Collins’ loss. In a report dated November 19, 2013,
after asserting that she reviewed all medical and clinical evidence provided to her, Dr. Golder
concluded as follows:
This insured has a history of diabetes for many years as well as morbid
obesity. It is unclear from the records to what degree he had a pre-existing,
diagnosed neuropathy in the affected leg but the presence of diabetes and the
notation that it was at least intermittently difficult to control makes it likely that
there is at least a degree of pre-existing neuropathy. Charcot joints develop only
in the presence of neuropathy, which may be the result of diabetes, or other illness
such as post-chemotherapy, syphilis or leprosy. It is usually triggered by trauma,
often insignificant. To a reasonable degree of medical certainty, without the
presence of underlying diabetes (or another source of neuropathy), this insured
would not have developed a Charcot joint. Further, his course was complicated
by the seriousness of the fracture which, despite immediate and appropriate care,
did not resolve; he also developed chronic osteomyelitis. Increased risk of
infection is also a characteristic of diabetes. It was the failure of treatment and the
presence of osteomyelitis that led to the amputation of the foot.
Conversely, this was a serious, debilitating injury and from the record it
was clearly the trigger for the development of the Charcot joint as that was not
present prior to the injury. To a reasonable degree of medical certainty, but for
the ankle fracture, this insured would not have developed a Charcot joint at this
time and might never have. To a reasonable degree of medical certainty both the
underlying illness and the injury were necessary for the development of the joint
pathology that led to amputation.
Administrative R. 13-2, Page ID 807.
In a letter dated November 21, 2013, Unum denied Collins’ claim for benefits,
concluding that his loss was contributed to by his diabetes, thus barring him from coverage under
the provision excluding from coverage any “accidental losses caused by, contributed to by, or
resulting from . . . disease of the body.” Administrative R., ECF No. 13-1, Page ID 810–11.
Collins filed a complaint in Ohio state court arguing he was entitled to benefits under 29
U.S.C. § 1132(a)(1)(B). Unum removed the case to the Northern District of Ohio. The district
court held that the arbitrary and capricious standard of review applied, and that the structural
-4-
Case No. 16-3918, Collins v. Unum Life Ins. Co. of Am.
conflict caused by Unum’s dual capacity as both the plan administrator and payor was
insufficient to justify the broader discovery Collins requested. On the merits, the district court
granted Unum’s motion to uphold its administrative decision, concluding that the alleged conflict
of interest alone was insufficient to overturn the denial of benefits. It further held that, though
Collins pointed to some contrary evidence, it was not enough to render the determination that
Collins’ amputation was caused, in part, by diabetes arbitrary and capricious. Collins filed a
timely notice of appeal.
II.
Collins brought this claim under Section 502(a) of ERISA, which allows a plan
participant or beneficiary “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). Because the denial of benefits here is based on an
exclusion, Unum has the burden of proving the exclusion applies. McCartha v. Nat’l City Corp.,
419 F.3d 437, 443 (6th Cir. 2005). We review the district court’s decision affirming the denial
of benefits de novo. Schwalm v. Guardian Life Ins. Co. of Am., 626 F.3d 299, 308 (6th Cir.
2010). Where, as here, the district court reviewed the benefits denial under the arbitrary and
capricious standard of review because the Plan granted Unum discretionary authority to
determine eligibility for benefits, we apply that same standard. Glenn v. MetLife, 461 F.3d 660,
665–66 (6th Cir. 2006).
“The arbitrary and capricious standard is ‘the least demanding form of judicial review of
administrative action.’” Schwalm, 626 F.3d at 308 (quoting Shields v. Reader’s Digest Ass’n,
Inc., 331 F.3d 536, 541 (6th Cir. 2003)). However, we may not simply “rubber stamp” Unum’s
decision. Glenn, 461 F.3d at 666 (quoting Jones v. Metro. Life Ins. Co., 385 F.3d 654, 661 (6th
Cir. 2004)). Rather, we must review “the quality and quantity of the medical evidence and the
-5-
Case No. 16-3918, Collins v. Unum Life Ins. Co. of Am.
opinions on both sides of the issues.” Id. (quoting McDonald v. Western-Southern Life Ins. Co.,
347 F.3d 161, 172 (6th Cir. 2003)). Yet even if there is evidence supporting a finding of
disability, if there is a reasonable explanation for the administrator’s decision to deny benefits,
the decision is neither arbitrary nor capricious. Schwalm, 626 F.3d at 308. Thus, we will uphold
the decision so long as it is the result of “‘a deliberate principled reasoning process’ and is
supported by ‘substantial evidence.’” Id. (quoting Baker v. United Mine Workers of Am. Health
& Ret. Funds, 929 F.2d 1140, 1144 (6th Cir. 1991)).
a.
Substantial evidence supports Unum’s finding that Collins’ amputation was caused, at
least in part, by his diabetes. Even prior to his injury, Collins had a history of uncontrolled
diabetes. In notes for the surgeries preceding his amputation and the actual amputation, Dr.
Khoury consistently deemed Collins’ diabetes, along with his fracture, Charcot ankle, and
osteomyelistis, relevant pre- and post-operative diagnoses, which sharply undermines Collins’
theory that his ankle fracture alone led to the amputation.
This case is analogous to the Southern District of Ohio’s decision in Morgan v. United
Omaha Life Insurance Co., No. C-1-06-837, 2008 WL 687169 (S.D. Ohio Mar. 11, 2008). A
few days after stepping on a screw, the plaintiff went to the emergency room, where he was
diagnosed with “diabetic foot infection,” “insulin-dependent diabetes mellitus, poorly
controlled,” and “diabetic neuropathy,” id. at *2, similar to Collins’ diagnoses of diabetic
neuropathy, osteomyelitis, and Charcot ankle leading up to his amputation. A month later, his
foot was amputated, and he later applied for accidental limb loss benefits under his ERISA plan.
Id. at *3. The plan in Morgan, like the Plan in this case, provided that the injury “must result in
loss independently of sickness and other causes.” Id. at *2. The administrator denied the
plaintiff’s claim because the loss was not independent of sickness or other causes. Id. at *3. The
-6-
Case No. 16-3918, Collins v. Unum Life Ins. Co. of Am.
district court affirmed, rejecting the plaintiff’s argument that because the amputation was the
result of a staph infection caused by stepping on a screw, his loss should be covered. Id. at *4.
This assertion was belied by the record indicating that diabetes was an exacerbating factor for his
loss. Id. at *4–*5. Though, unlike here, the treating physician in Morgan filled out a form as a
part of the plaintiff’s application, identifying diabetes as a contributing cause, id. at *3, that does
not render this case distinguishable because Dr. Khoury repeatedly listed Collins’ diabetes as a
diagnosis related to his increasingly severe ankle injury.
It is true that Dr. Berkowitz’s opinion that the loss was not caused in any way by a
disease lends support to Collins’ theory. Nevertheless, because there is reasonable support for
Unum’s decision to deny benefits, the fact that there exists a scintilla of evidence in support of a
loss finding does not make the decision arbitrary and capricious. See Schwalm, 626 F.3d at 308.
Importantly, though Dr. Berkowitz did not think diabetes was a cause of the amputation, he did
assess Collins with having osteomyelistis and Charcot ankle. As Dr. Golder noted in her report,
both of these conditions are associated with increased risks that are characteristic of diabetes.
Therefore, we conclude that substantial evidence supports Unum’s determination.
Collins further argues, however, that other circumstances render Unum’s denial of his
benefits arbitrary and capricious. Specifically, Collins maintains that (1) Unum suffers a conflict
of interest in its dual role of determining eligibility for benefits and paying benefits; (2) Unum’s
expert did not review Collins’ entire medical record and did not give adequate consideration to
Collins’ treating physician, who indicated that diabetes did not contribute to the amputation; and
(3) Unum erred in failing to conduct a physical exam and relying solely on a review of the
record. We address each argument in turn and conclude that, even viewed collectively, these
circumstances do not render Unum’s decision arbitrary and capricious.
-7-
Case No. 16-3918, Collins v. Unum Life Ins. Co. of Am.
i.
When a plan administrator both evaluates and pays for benefits under a plan, it creates a
conflict of interest. Metro. Life Ins. Co. v. Glenn, 554 U.S. 105, 112 (2008). Here, because
Unum both determines eligibility for benefits and pays the benefits, there exists a conflict of
interest. However, this conflict of interest does not change the standard of review or alter the
burden of proof; rather it is simply a factor to consider in analyzing the denial-of-benefits
decision. Id. at 115–17.
For instance, in Glenn v. MetLife, we gave weight to the inherent conflict of interest
where the insurance company encouraged the claimant to argue to the Social Security
Administration (“SSA”) that she could not work, but then ignored the SSA’s finding that the
claimant was totally disabled in concluding that she could, nonetheless, do sedentary work.
461 F.3d at 666–69. The Supreme Court affirmed our reversal of the denial of benefits. Glenn,
554 U.S. at 119. It emphasized that a conflict of interest was merely one of the “several different
considerations” the court should look to in determining the lawfulness of the administrator’s
denial of benefits. Id. at 117. Like any other factor, depending on the circumstances, it may
serve as a tiebreaker where the facts suggest a high likelihood that the conflict affected the
benefits decision, but may “prove less important (perhaps to the vanishing point) where the
administrator has taken active steps to reduce potential bias and to promote accuracy.” Id. So,
“a conflict of interest, standing alone, does not require reversal,” meaning a claimant “must do
more than offer general inferences of a conflict based on self-interest.” Cultrona v. Nationwide
Life Ins. Co., 748 F.3d 698, 704 (6th Cir. 2014); see also DeLisle v. Sun Life Assurance Co. of
Canada, 558 F.3d 440, 445 (6th Cir. 2009) (concluding that the claimant offered more than
conclusory allegations of bias to support his conflict-of-interest argument where the insurance
company relied on reviewers with whom it regularly contracted, and gave the reviewers
-8-
Case No. 16-3918, Collins v. Unum Life Ins. Co. of Am.
incomplete and potentially prejudicial information—that the claimant was terminated for
cause—that portrayed the claimant in a negative light); Pflaum v. Unum Provident Corp., 175 F.
App’x 7, 10 (6th Cir. 2006) (“Because [Claimant] points to nothing beyond the mere existence of
a conflict of interest to show that UNUM’s decision was motivated by self-interest, we give no
further consideration in the arbitrary and capricious analysis to the possibility that the conflict
affected UNUM’s decision-making.”).
Here, Collins “has pointed to nothing more than the general observation that [Unum] had
a financial incentive to deny the claim.” See Judge v. Metro. Life Ins. Co., 710 F.3d 651, 664
(6th Cir. 2013). This is insufficient. We find nothing else in the record that would lead us to
conclude that Unum’s potential for biased decision-making rendered its benefits denial arbitrary
and capricious. Further, as discussed above, the weight of the evidence supports Unum’s
decision. Accordingly, we conclude that Unum’s conflict of interest did not make its decision
arbitrary and capricious.
ii.
Collins also argues that Dr. Golder did not fully review his medical record, and did not
adequately consider Dr. Berkowitz’s conclusions. Plan administrators must “provide a full and
fair review of claim denials.” Glenn, 554 U.S. at 115 (citations omitted) (internal quotation
marks omitted). Moreover, they cannot arbitrarily refuse to credit a participant’s reliable
evidence. Black & Decker Disability Plan v. Nord, 538 U.S. 822, 834 (2003). However,
administrators need not afford special weight to a participant’s treating physician, nor need they
carry the burden to provide an explanation “when they credit reliable evidence that conflicts with
a treating physician’s evaluation.” Id. This is particularly true when the findings reached by the
treating physician are unsupported by any analysis or explanation. See Storms v. Aetna Life Ins.
Co., 156 F. App’x 756, 758–59 (6th Cir. 2005) (discounting the disability finding of the
-9-
Case No. 16-3918, Collins v. Unum Life Ins. Co. of Am.
claimant’s personal physician where the physician’s “conclusory finding was not supported by
objective medical data, useful analysis, or the other opinions in the record.”)
In Calvert v. Firstar Finance, Inc., we reversed the plan administrator’s denial of the
claimant’s benefits, concluding its decision was arbitrary and capricious. 409 F.3d 286, 297 (6th
Cir. 2005). There, we considered relevant the fact that the SSA determined that the claimant was
totally disabled based on objective evidence in the record, and that the administrator did not
perform a physical exam even though it expressly reserved the right to do so in the plan. Id. at
293–95. We concluded that the administrator’s denial of benefits, based on its determination that
there was no objective evidence in the record supporting that the claimant was disabled, was not
based on a reasoned review of the record. Id. at 296. Though there was “nothing inherently
objectionable about a file review,” the review by the administrator’s expert was inadequate
where he did not describe the data he reviewed and did not mention the SSA disability
determination. Id. Thus, we decided that the expert’s conclusion that the claimant was
exaggerating (despite not having examined her) and that there was no objective data supporting
her disability (despite the wealth of objective evidence available), combined with the
administrator’s inherent conflict of interest, rendered the denial decision arbitrary and capricious.
Id. at 296–97.
Here, unlike Calvert, Dr. Golder’s conclusion that diabetes contributed to Collins’ loss
was based on a wealth of objective evidence and nothing suggested that she was willfully blind
to evidence supporting Collins’ claim. While true that Dr. Golder did not describe each piece of
evidence she reviewed in reaching her conclusion, she did state that she had reviewed all of the
medical records provided to her. Notably, Collins does not point to any evidence that Dr. Golder
allegedly failed to review that supported his theory of the case. Furthermore, there are
- 10 -
Case No. 16-3918, Collins v. Unum Life Ins. Co. of Am.
indications that Dr. Golder reviewed Dr. Berkowitz’s assessment. Dr. Golder cites Charcot
joints and bone infections, diagnoses given by Dr. Berkowitz, but explains how these diagnoses
were likely contributed to by Collins’ diabetes. Particularly in light of the substantial evidence
supporting her conclusion, Dr. Golder reasonably discredited Dr. Berkowitz’s conclusory
assertion—indicated only by a circle around the word “no”—that an illness did not contribute to
Collins’ loss. See Storms, 156 F. App’x at 758–59. Accordingly, we conclude that Dr. Golder
adequately reviewed the record and appropriately discounted the conclusory finding of one of
Collins’ treating physicians.
iii.
Finally, Collins points us to Calvert in support of his assertion that Unum should have
conducted a physical exam before making the decision about whether to grant him benefits. In
Calvert, we held that the reliance on a file review, standing alone, does not render the
administrator’s decision arbitrary and capricious. 409 F.3d at 295. But we noted that “the
failure to conduct a physical examination—especially where the right to do so is specifically
reserved in the plan—may, in some cases, raise questions about the thoroughness and accuracy
of the benefits determination.” Id.
We see no reason for Unum to have conducted a physical examination under these
circumstances. First, the right to a physical exam was not reserved in the Plan. Furthermore,
unlike Calvert, where the administrator’s expert discounted the claimant’s subjective reports of
continuing pain without examining her, id. at 296–97, as Unum notes, we see no reason why an
examination after the amputation could have rendered more accurate the determination of what
factors leading up to the amputation contributed to it. This is especially true in light of the fact
that Collins’ treating physician, based on objectively verifiable facts, included diabetes as one of
- 11 -
Case No. 16-3918, Collins v. Unum Life Ins. Co. of Am.
the diagnoses for the severity of Collins’ injuries. The failure to conduct a physical examination
of Collins did not make Unum’s denial decision arbitrary and capricious.
III.
Lastly, Collins argues that the trial court abused its discretion in not allowing discovery
because of Unum’s conflict of interest. We review a district court’s decision regarding the scope
of discovery for abuse of discretion. Board of Trustees v. Moore, 800 F.3d 214, 223 (6th Cir.
2015).
Typically, a court’s factual review is limited to the administrative record. Schwalm,
626 F.3d at 308. However, if the claimant offers evidence “in support of a procedural challenge
to the administrator’s decision, such as an alleged lack of due process afforded by the
administrator or alleged bias on its part,” discovery may be warranted. Johnson v. Conn. Gen.
Life Ins. Co., 324 F. App’x 459, 466 (6th Cir. 2009) (citation omitted). Though we have
identified a conflict of interest as a type of bias that may warrant additional discovery, “[t]hat
does not mean, however, that discovery will automatically be available any time the defendant is
both the administrator and the payor under an ERISA plan.” Id. at 467. So, the claimant must
put forth a factual foundation to establish that he has done more than merely allege bias. See
Pearce v. Chrysler Grp., LLC Pension Plan, 615 F. App’x 342, 350 (6th Cir. 2015).
As detailed above, Collins has not set forth evidence establishing more than a mere
allegation of bias based on the inherent conflict of interest. Thus, we conclude that the district
court did not abuse its discretion in denying additional discovery based on Collins’ mere
allegations that Unum was biased against him.
IV.
For the foregoing reasons, we AFFIRM the district court’s judgments.
- 12 -