Nguyen v. Sun Life Assurance Company of Canada et al

Filing 48

ORDER by Judge Laurel Beeler granting in part and denying in part 30 Joint Discovery Letter. (lblc3S, COURT STAFF) (Filed on 10/27/2015)

1
2
3
4
5
6
7
8

UNITED STATES DISTRICT COURT

9

NORTHERN DISTRICT OF CALIFORNIA

10

San Francisco Division

United States District Court
Northern District of California

11
ERIC NGUYEN,

12

Case No. 3:14-cv-05295-JST (LB)

Plaintiff,

13

v.

14
SUN LIFE ASSURANCE COMPANY OF
CANADA,

15

ORDER ON JOINT DISCOVERY
LETTER
[ECF No. 30]

Defendant.

16

INTRODUCTION

17
18

This is an ERISA benefits-denial case. The assigned district judge has referred it to the

19

undersigned to resolve a discovery dispute. Plaintiff Eric Nguyen was covered by an employee-

20

benefits plan for which defendant Sun Life both served as the insurer and made the claims

21

decisions. (See ECF No. 30 at 1.)1 The plaintiff made claims under the plan for short-term and

22

long-term disability benefits “based on upper[-]extremity pain syndromes, diagnosed as thoracic

23

outlet syndrome . . . and other occipital, thoracic and cervical pain conditions.” (Id.) Sun Life

24

denied the plaintiff‟s claims “and upheld those denials on appeal.” (Id.) The plaintiff has sued Sun

25

Life to recover benefits under 29 U.S.C. § 1132(a)(1)(B).

26
27
28

1

Record citations refer to material in the Electronic Case File (“ECF”); pinpoint citations are to
the ECF-generated page numbers at the tops of the documents.
ORDER (No. 3:14-cv-05295-JST (LB))

1

The administrative record, the material on which Sun Life made and then confirmed its

2

decision, appears to consist mainly of the medical records, evaluations, and opinions of the

3

plaintiff‟s numerous treating physicians. The plaintiff says that this material shows that he was

4

disabled and entitled to benefits. The record also contains the opinions of four physicians whom

5

Sun Life retained, as outside consultants, to review the plaintiff‟s claim; all four concluded that he

6

was not disabled. (See ECF No. 30 at 5-7.)

7

The plaintiff now seeks to discover material beyond the administrative record. He has

8

propounded interrogatories and requests for production to Sun Life, and proposes to take

9

depositions. (ECF No. 30-1 to -4.) His proposed discovery covers three topics: 1) The
completeness of the administrative record; 2) The “policies and guidelines” that Sun Life followed

11
United States District Court
Northern District of California

10

(or should have followed) in assessing his claim; and 3) The relationship between Sun Life and its

12

four outside medical reviewers. (See id. at 5-7.) On the last point, the plaintiff wishes to discover

13

material that he believes will show that the outside reviewers were “biased” in Sun Life‟s favor —

14

essentially, because Sun Life paid them. (See id.)

15

The undersigned concludes that, under governing Ninth Circuit law, the plaintiff is not entitled

16

to most of this discovery. (Exceptions are noted where appropriate in the Analysis below.) The

17

plaintiff has not “clearly established” that evidence outside the existing administrative record is

18

“necessary to conduct an adequate de novo review of the benefit decision.” See Opeta v. Nw.

19

Airlines Pension Plan, 484 F.3d 1211, 1217 (9th Cir. 2007) (emphasis removed). The undersigned

20

therefore partly grants and partly denies the plaintiff‟s requested discovery.

21

* * *

22

GOVERNING LAW

23

The parties have stipulated that the district judge will review de novo Sun Life‟s decision to

24

deny Mr. Nguyen benefits. (ECF No. 30 at 1.) Where de novo review applies, “no further

25

preliminary analytical steps are required,” and the district court “simply proceeds to evaluate

26

whether the plan administrator correctly or incorrectly denied benefits.” Abatie v. Alta Health &

27

Life Ins. Co., 458 F.3d 955, 963 (9th Cir. 2006) (en banc); accord Gonda v. Permanente Med.

28

Group, Inc., 300 F.R.D. 609, 613 (N.D. Cal. 2014).
2
ORDER (No. 3:14-cv-05295-JST (LB))

1

This affects the range of permissible evidence. De novo review of an ERISA benefits denial is

2

“based on the evidence in the administrative record and „other evidence as might be admissible

3

under the restrictive rule of Mongeluzo [v. Baxter Travenol Long Term Disability Benefit Plan, 46

4

F.3d 938 (9th Cir. 1995)].‟” Opeta, 484 F.3d at 1217. In Mongeluzo, the Ninth Circuit “held that

5

extrinsic evidence could be considered only under certain limited circumstances.” Opeta, 484 F.3d

6

at 1217. “[T]he district court should exercise its discretion to consider evidence outside of the

7

administrative record „only when circumstances clearly establish that additional evidence is

8

necessary to conduct an adequate de novo review of the benefit decision.” Id. (quoting in part

9

Mongeluzo, 484 F.3d at 944) (quoting in turn Quesinberry v. Life Ins. Co. of N. Am., 987 F.2d

10

1017, 1025 (4th Cir. 1993) (en banc) (emphasis in Opeta).

United States District Court
Northern District of California

11

Borrowing from Quesinberry, the Ninth Circuit in Opeta described “a non-exhaustive list of

12

exceptional circumstances where evidence beyond the administrative record could be considered

13

necessary”:

18

claims that require consideration of complex medical questions or
issues regarding the credibility of medical experts; the availability
of very limited administrative review procedures with little or no
evidentiary record; the necessity of evidence regarding interpretation
of the terms of the plan rather than specific historical facts;
instances where the payor and the administrator are the same
entity and the court is concerned about impartiality; claims which
would have been insurance contract claims prior to ERISA; and
circumstances in which there is additional evidence that the claimant
could not have presented in the administrative process.

19

Opeta, 484 F.3d at 1217 (quoting Quesinberry, 987 F.2d at 1027) (emphases added).

14
15
16
17

20

Even where such circumstances exist, however, new evidence is not “required.” Quesinberry,

21

987 F.2d at 1027. “A district court may well conclude that the case can be properly resolved on the

22

administrative record without the need to put the parties to additional delay and expense.” Id. As

23

Judge Conti of this district observed: “[T]o further ERISA‟s policy of keeping proceedings

24

inexpensive and expeditious, the Ninth Circuit has placed significant restrictions” — i.e., the

25

“narrow circumstances” outlined in Opeta — “on district courts‟ ability to consider evidence

26

outside the administrative record.” Gonda, 300 F.R.D. at 613 (citing Mongeluzo, 46 F.3d at 943).

27

The Ninth Circuit itself has “emphasized” that “[i]n most cases only the evidence that was before

28

the plan administrator at the time of determination should be considered.” Opeta, 484 F.3d at 1217
3
ORDER (No. 3:14-cv-05295-JST (LB))

1
2

(quotation omitted).
These admissibility limits also constrain discovery:

6

The Ninth Circuit‟s decision in Opeta dealt with the issue of
whether evidence outside the record is admissible rather than
whether the evidence is discoverable. Nevertheless, courts in this
district have held that in light of Opeta’s limits on admissibility of
evidence in de novo cases and the ERISA‟s policy of keeping
proceedings inexpensive and expeditious, it is appropriate to place
similar limits on discovery.

7

Polnicky v. Liberty Life Ass. Co., No. C 13-1478 SI, 2014 U.S. Dist. LEXIS 29123, *5-*6 (N.D.

8

Cal. Mar. 15, 2014) (quoting in part Rowell v. Aviza Tech. Health & Welfare Plan, No. C 10-5656

9

PSG, 2012 U.S. Dist. LEXIS 16957, *13-14 n. 26 (N.D. Cal. Feb. 10, 2012) and citing Knopp v.

3
4
5

Life Ins. Co. of N. Am. C-09-0452 CRB (EMC), 2009 U.S. Dist. LEXIS 120267, *7 (N.D. Cal.

11
United States District Court
Northern District of California

10

Dec. 28, 2009)); accord Gonda, 300 F.R.D. at 613 (“It makes little sense to allow broad and costly

12

discovery when the court‟s review of the merits is limited . . . .”); see Brice v. Life Ins. Co. of N.

13

Am., No. C 10-4204 JSW, 2011 U.S. Dist. LEXIS 77525, *1 (N.D. Cal. Jul. 18, 2011) (ERISA‟s

14

goals of cost reduction and prompt claim disposition mean that “courts cannot allow discovery in

15

ERISA actions for the denial of benefits to be as broad . . . as in other types of cases”).

16
17

The decision to allow extrinsic discovery under these rules lies in the court‟s discretion. See
Opeta, 484 F.3d at 1213.

18

* * *

19

ANALYSIS

20

The discussion below considers the plaintiff‟s requests, arranged by topic, in increasing order

21

of analytical complexity. Part 1 addresses the completeness of the administrative record; Part 2,

22

Sun Life‟s policies and guidelines. Part 3 discusses the complexity of the medical issues that this

23

case presents; it is an argument that the plaintiff raises but does not elaborate. Part 4 reaches the

24

harder question of the “bias” and lack of “credibility” that the plaintiff believes may have tainted

25

the work of Sun Life‟s outside medical reviewers. This last issue absorbs most of the plaintiff‟s

26

efforts here (see ECF No. 30 at 5-8); it is the issue on which district-court precedent, though

27

ostensibly following the same Ninth Circuit standard (Opeta), has divided most noticeably; and it

28

is this issue that has consumed most of the undersigned‟s thinking. The final Part 5 of this order
4
ORDER (No. 3:14-cv-05295-JST (LB))

1

then turns back to explain the two most salient analytical considerations that guided the

2

undersigned‟s decision on the issue of the outside reviewers‟ credibility. Essentially, Part 5 tries to

3

explain why Opeta compels the conclusions reached on the record here, and why granting the

4

requested discovery would depart, if not from the superficial, than from the substantive dictate of

5

the Ninth Circuit‟s rule in this area.
* * *

6
7

1. Completeness of the Administrative Record

8

The plaintiff does not directly dispute Sun Life‟s statement that it has “produced the entire

9

Administrative Record.” (See ECF No. 30 at 2.) He does ask for “any portions of his complete file
in this matter [that have] not yet [been] provided.” (Id. at 7.) “For instance,” he writes, “Sun Life

11
United States District Court
Northern District of California

10

alleges that the administrative record contains communications with Mr. Nguyen‟s medical

12

providers, however there is no evidence of Sun Life doctors contacting treating physicians in the

13

record.” (Id.)

14

There is no dispute that Sun Life “must produce the entire administrative record.” Duran v.

15

Cisco Sys., 258 F.R.D. 375, 382 (C.D. Cal. 2009). More fully stated, “ERISA requires that

16

claimants be given access to „all documents, records, and other information relevant to the

17

claimant‟s claim for benefits.‟” Bourland v. Hartford Life & Acc. Ins. Co., 2014 U.S. Dist. LEXIS

18

134813, *9 (W.D. Wash. 2014) (quoting in part 29 C.F.R. § 2650.503-1(h)(2)(iii)). Information is

19

“relevant” in this context if it:

20
21

i.
ii.

Was relied upon in making the benefit determination;
Was submitted, considered, or generated in the course of making the benefit
determination, without regard to whether such document, record, or other
information was relied upon in making the benefit determination;

iii.

Demonstrates compliance with the administrative processes and safeguards
required pursuant to paragraph (b)(5) of this section in making the benefit
determination; or
In the case of a group health plan or a plan providing disability benefits,
constitutes a statement of policy or guidance with respect to the plan
concerning the denied treatment option or benefit for the claimant's diagnosis,
without regard to whether such advice or statement was relied upon in making
the benefit determination.

22
23
24
25
26
27

iv.

29 C.F.R. § 2650.503-1(m)(8); Bourland, 2014 U.S. Dist. LEXIS 134813 at *9-*10.

28
5
ORDER (No. 3:14-cv-05295-JST (LB))

1

The communications that the plaintiff mentions would seem to fall under at least two of these

2

categories: material that was “relied on in making the benefit determination”; and material

3

“generated in the course of making the benefit determination.” If any such material exists and has

4

not yet been produced to the plaintiff, including communication between Sun Life or its agents

5

and the plaintiff‟s medical providers, that material is part of the administrative record and must be

6

produced.
* * *

7
8

2. Policies, Procedures, Guidelines
The plaintiff also “seeks the policies, procedures and guidelines applied to his claim.” (ECF

10

No. 30 at 8.) He correctly argues that ERISA regulations “require production of any „internal rule,

11
United States District Court
Northern District of California

9

guideline, protocol, or other similar criterion‟ relied upon in making an adverse decision.” (Id.)

12

(quoting in part 29 C.F.R. § 2560.503-1(g)(v)(8)). The regulation reproduced earlier also covers

13

this topic. Most generally, policies and guidelines may be “relied upon,” or at least “considered,”

14

in “making the benefit determination.” See 29 C.F.R. § 2560.503-1(m)(8)(i)-(ii). The same

15

regulation compels production of material that constitutes a “statement of policy or guidance with

16

respect to the plan concerning the denied . . . benefit . . . .” 29 C.F.R. § 2560.503-1(m)(8)(iv).

17

Finally, the regulation makes “relevant” (and so mandates the production of) “information” that

18

“[d]emonstrates compliance with the administrative processes and safeguards required pursuant to

19

paragraph (b)(5) of this section . . . .” 29 C.F.R. § 2560.503-1(m)(8)(iii). This “paragraph (b)(5)”

20

mandates that a plan‟s “claims procedure” have “administrative processes and safeguards designed

21

to ensure and to verify that benefit claim determinations are made in accordance with governing

22

plan documents and that, where appropriate, the plan provisions have been applied consistently

23

with respect to similarly situated claimants.” 29 C.F.R. § 2560.503-1(b)(5).

24

The plaintiff‟s “policies and guidelines” request elicits a three-pronged holding. First, insofar

25

as ERISA regulations require Sun Life to give the plaintiff copies of relevant policies and

26

guidelines, and to the extent that Sun Life has not already produced this material, it is ordered to

27

do so. See Knopp, 2009 U.S. Dist. LEXIS 120267 at *10-*11; Bourland, 2014 U.S. Dist. LEXIS

28

134813 at *8-*10; Polnicky, 2014 U.S. Dist. LEXIS 29123 at *11-*12 (citing Knopp, supra, and
6
ORDER (No. 3:14-cv-05295-JST (LB))

1

Rowell, 2012 U.S. Dist. LEXIS 16957 at *11).
Second, beyond the policy documents themselves, the plaintiff is not entitled to extra-record

2
3

discovery into whether and why Sun Life followed, or failed to follow, its own policies and

4

guidelines in assessing the plaintiff‟s claims. As multiple courts have noted, this is irrelevant on de

5

novo review. Another judge in this district recently wrote: “Even if Defendants failed to follow

6

claim procedures or guidelines, that failure might reflect upon the integrity and accuracy of the

7

administrator‟s review of [the plaintiff‟s] disability claim, but that review is entitled to no

8

deference on de novo review and is therefore irrelevant.” Knopp, 2009 U.S. Dist. LEXIS 120267

9

at *10; accord Polnicky, 2014 U.S. Dist. LEXIS 29123 at *11-*12 (calling irrelevant to de novo
review whether administrator “took a more aggressive approach” under its policies and guidelines

11
United States District Court
Northern District of California

10

to avoid paying “high benefit amount”); cf. Gonda, 300 F.R.D. at 315 (“fail[ing] to see relevance”

12

of policies and history of policy changes). (The only limitation on this may lie in 29 C.F.R.

13

§ 2560.503-1(b)(5) and (m)(8)(iii). Those subsections compel the production of information that

14

“[d]emonstrates compliance with the administrative processes and safeguards” employed “to

15

ensure and to verify that benefit claim determinations are made in accordance with governing plan

16

documents and that, where appropriate, the plan provisions have been applied consistently with

17

respect to similarly situated claimants.” If Sun Life has documentary material responsive to that

18

limited requirement, it must produce it.)

19

Third, the plaintiff rolls into his request for policies and guidelines material concerning “Sun

20

Life‟s relationships with third[-]party reviewers.” (ECF No. 30 at 8.) For the reasons set out in Part

21

4, below, the plaintiff is not entitled to investigate the “bias” or “credibility” of Sun Life‟s outside

22

consultants. (At least he may not go beyond the existing administrative record to make that

23

investigation.) The material that is barred from discovery under the discussion below cannot be

24

obtained through the side door by asking for “policies and guidelines.” If the “policies and

25

guidelines” describe general administrative rules governing how and when Sun Life uses third-

26

party medical reviewers, that material may be discoverable.2 Most of what the plaintiff has asked

27
2

28

This seems to be the information the plaintiff has in mind when he points to the “written
discovery” of “procedures for oversight of third[-]party reviewers” that the court allowed in Eisner
7
ORDER (No. 3:14-cv-05295-JST (LB))

1

for, however, in connection with “Sun Life‟s relationships with third[-]party reviewers” — such as

2

how many times Sun Life has hired a specific consultant (ECF No. 30-2 at 5-6), how much Sun

3

Life has paid them (ECF No. 30-1 at 6-8; ECF No. 30-2 at 5-6), and the contracts embodying

4

these things (ECF No. 30-1 at 6-7) — would not constitute “policies” or “guidelines” discoverable

5

under this part of the analysis.

6

A particular word is needed concerning the plaintiff‟s fourth Request for Production. The

7

plaintiff discusses this request in connection with “policies and guidelines.” (ECF No. 30 at 8.) On

8

its face, though, it asks for material that would seem to be protected under the attorney-client

9

privilege or the attorney work-product doctrine. (See ECF No. 30-1 at 5.)3 The plaintiff argues that
such material “fall[s] outside the attorney[-]client privilege under the fiduciary exception.” (ECF

11
United States District Court
Northern District of California

10

NO. 30 at 8.) The issue is moot: to this request, Sun Life answers that there are “no responsive

12

documents.” (Id. at 3.)

13

* * *

14

In sum, insofar as ERISA regulations require Sun Life to produce its policies and guidelines,

15

and to the extent that those have not yet been produced, Sun Life must produce them. The plaintiff

16

cannot obtain discovery (beyond the administrative record itself) into whether Sun Life did or did

17

not follow its policies and guidelines in handling his claims. Nor is the plaintiff entitled to

18

extrinsic discovery into Sun Life‟s relationship with the outside medical reviewers that it hired to

19

review the plaintiff‟s claims, as discussed in Part 4, below.
* * *

20
21

3. Complex Medical Questions

22

The plaintiff briefly suggests that “this is a complicated medical case, given the overlapping

23

and comorbid conditions suffered by Mr. Nguyen, the many surgeries he has undergone over the

24

years, and the copious records describing intricate medical procedures.” (ECF No. 30 at 6.)

25
26
27
28

v. Prudential Ins. Co. of Am., No. 12-cv-01238-JST, 2013 U.S. Dist. LEXIS 65090 (N.D. Cal.
May 4, 2013). (See ECF No. 30 at 8.)
3
The fourth RFP asks for: “Any and all DOCUMENTS evidencing any and all attorney-client
COMMUNICATION and attorney memoranda, guidelines and opinions RELEVANT to the
CLAIMS at any time before the adverse decision.” (ECF No. 30-1 at 5) (capitals original).
8
ORDER (No. 3:14-cv-05295-JST (LB))

1

Presumably, the plaintiff means to invoke Opeta’s category of situations in which “complex

2

medical questions” permit extrinsic discovery. See Opeta, 484 F.3d at 1217. The court does not

3

think that the facts, or the plaintiff‟s fleeting glance off this topic, justifies discovery under Opeta.

4

The court does not doubt that this case involves complex medical issues. But there seems to be

5

ample record proof from which the district judge can adequately review Sun Life‟s decision. The

6

administrative record (according to Sun Life; and the plaintiff does not dispute this) contains over

7

6000 pages of documents. (See ECF No. 30 at 2.) By the plaintiff‟s description, the record

8

comprises an “avalanche” of material from “dozens” of physicians supporting his claim (ECF No.

9

30 at 5-6) — and, of course, it contains the contrary assessments of Sun Life‟s third-party medical

10

United States District Court
Northern District of California

11

reviewers.
The plaintiff does not elaborate his “complicated medical case” point. He does not explain

12

how the material already in the record fails to elucidate any particular medical issue, or how the

13

existing record otherwise lacks information that a court would need to decide for itself whether he

14

was disabled. To the contrary, the plaintiff‟s own argument suggests that the existing

15

administrative record permits a decision in his favor. He cites “copious records describing intricate

16

medical procedures.” (ECF No. 30 at 6.) He calls Sun Life‟s adverse decision “incredible” given

17

the “thoroughly documented medical support by [his] treating physicians.” (Id. at 5.) And he

18

describes “the support of dozens of doctors who have personally treated him — an avalanche of

19

gold-standard support by . . . specialists who have clinically demonstrated the objective and

20

subjective severity of his conditions.” (Id. at 6) (emphasis added). In other words, it is not that the

21

plaintiff feels the existing record inadequate to support a de novo decision; he just thinks that it

22

supports a decision in his favor.

23

The plaintiff has not “clearly established” that, because of “complex medical questions,”

24

evidence beyond the administrative record is “necessary” for an adequate de novo review of his

25

claim.
* * *

26
27
28

9
ORDER (No. 3:14-cv-05295-JST (LB))

1

4. Sun Life’s Relationship With Outside Medical Reviewers

2

The heart of the plaintiff‟s discovery requests asks to delve into information concerning the

3

relationship between Sun Life and the outside medical consultants that the insurer hired to review

4

(and who ultimately suggested denying) his claim. (See ECF No. 30-1 at 5-8; ECF No. 30-2 at 5-

5

8.) With these requests the plaintiff asks Sun Life to produce material reflecting (for example) how

6

many times it has hired the medical reviewers in the past five years, how much it has paid them,

7

how many times it has denied claims based on their opinions, evaluations it has made of their

8

work, and the contracts reflecting their employment. The plaintiff would also depose Sun Life on

9

these points. (ECF No. 30-3 at 5; ECF No. 30-4 at 6-7.)
* * *

10

United States District Court
Northern District of California

11

4.1 Structural Conflict of Interest & Impartiality

12

Again, “Sun Life was the insurer for the Plan and made the claims decisions.” (ECF No. 30 at

13

1.) It both paid the claims, in other words, and decided which claims would be paid. Sun Life thus

14

had what the case law calls a “structural conflict of interest.” See, e.g., Tarasovsky v. Stratify, Inc.,

15

No. C 11-03359 WHA, 2012 U.S. Dist. LEXIS 69921, *14 (N.D. Cal. May 18, 2012) (quoting

16

Montour v. Hartford Life & Acc. Ins. Co., 588 F.3d 623, 630-31 (9th Cir. 2009)); Nolan v. Heald

17

College, 745 F. Supp. 2d 916, 921 (N.D. Cal. 2010). Such a conflict exists whenever the “same

18

entity that funds an ERISA plan also evaluates claims.” Tarasovsky, 2012 U.S. Dist. LEXIS 69921

19

at *14; accord, e.g., Nolan, 745 F. Supp. 2d at 921. Because the plan administrator in such a case

20

“is also the insurer, . . . by denying benefits, the administrator retains money for itself.” Montour,

21

588 F.3d at 630. In Opeta, the Ninth Circuit recognized this as an “exceptional circumstance[]” in

22

which additional evidence “could be considered necessary.” Opeta, 484 F.3d at 1217 (“where the

23

payor and the administrator are the same entity and the court is concerned about impartiality”).

24

The plaintiff invokes this ground to bolster his primary claim that Sun Life‟s outside medical

25

reviewers were biased. He writes that Opeta permits extrinsic discovery in de novo cases that

26

present “„issues regarding the credibility of medical experts,‟ particularly involving a conflict of

27

interest such as in this case, where Sun Life acts as both the claims decision maker and plan

28

funder.” (See ECF No. 30 at 5) (quoting in part Opeta, 484 F.3d at 1217) (emphasis added).
10
ORDER (No. 3:14-cv-05295-JST (LB))

1

The court does not think that, merely by pointing to Sun Life‟s dual role as plan administrator

2

and payer, the plaintiff has “clearly established” that additional discovery is “necessary” to an

3

“adequate de novo review of the benefits decision.” Two points might be made here.

4

First, Sun Life‟s “structural conflict of interest” does not alone warrant extrinsic discovery. See

5

Bourland, 2014 U.S. Dist. LEXIS at *6 (“Under de novo review, [the insurer‟s] alleged conflict of

6

interest is irrelevant.”); Abatie, 458 F.3d at 963 (de novo review asks whether benefits were

7

correctly denied, “without reference to whether the administrator operated under a conflict of

8

interest”). Defendants in ERISA cases have often served as both plan administrator and plan

9

funder. The situation appears to be common. That a defendant acted as both administrator and
payer thus cannot qualify, on its own, as an “exceptional circumstance” under Opeta’s “restrictive”

11
United States District Court
Northern District of California

10

rule. The Gonda court, in this vein, refused to accept the notions that “discovery on an insurer‟s

12

conflict of interest is always permissible or that a conflict of interest has any direct bearing on the

13

merits analysis in a de novo review.” Gonda, 300 F.R.D. at 614 (discussing Waggener v. Unum

14

Life Ins. Co. of Am., 238 F. Supp. 2d 1179 (S.D. Cal. 2002)). Other cases have similarly held that a

15

plan administrator‟s motives are “irrelevant” on de novo review. See Knopp, 2009 U.S. Dist.

16

LEXIS 120267 at *10 (denying discovery into insurer‟s conflict of interest); Bourland, 2014 U.S.

17

Dist. LEXIS 134813 at *6. The undersigned largely agrees. The question on de novo review is

18

generally not why the insurer reached the decision it did — whether from misanthropy,

19

incompetence, or a miserly grip on “its” funds. The de novo question is more direct; it is “simply”

20

whether, given the administrative record, the plaintiff was entitled to benefits. See, e.g., Abatie,

21

458 F.3d at 963.

22

Second, it is far from clear that an insurer‟s “structural conflict of interest” speaks to whether

23

outside medical reviewers “biased” their own decisions in favor of the plan administrator. The link

24

that the plaintiff draws between the two (see ECF No. 30 at 5) is too casual, in any case, does not

25

hold up to closer analysis, and does not justify additional discovery under Opeta. The court is

26

aware that one decision took an approach similar to the plaintiff‟s on this point. In Knopp, too, the

27

court linked the issues of outside-reviewer bias and the plan administrator‟s structural conflict of

28

interest:
11
ORDER (No. 3:14-cv-05295-JST (LB))

1
2
3

[T]he Court concludes that Ms. Knopp has made an adequate
showing that she is entitled to some discovery which could bear
directly and substantially on the credibility of the medical reviewers
engaged by the insurer, issues potentially enhanced by the potential
conflict of interest here — the insurer/payor and plan administrator
appear to be affiliated with each other.

4

Knopp, 2009 U.S. Dist. LEXIS 120267 at *8-*9 (emphasis added). Knopp then ordered the

5

defendants to produce any “performance evaluations” of and, more generally, “information about

6

the [outside] medical consultants,” “including their relationship with Defendants”; the court called

7

the latter material “closely related to the issue of conflict of interest.” Id. at *8.

8
9

The undersigned respectfully disagrees with this aspect of Knopp. An insurer‟s conflict of
interest, on the one hand, and the “bias” and “credibility” of outside reviewers, on the other, are
distinct issues. Doctrinally, Opeta does not link medical-reviewer “credibility” to a plan

11
United States District Court
Northern District of California

10

administrator‟s “conflict of interest.” See Opeta, F.3d at 1217-18. Opeta does not suggest that the

12

latter enhances the former, so that cases involving a “structural conflict of interest” more readily

13

admit of extrinsic discovery into the “credibility” of outside medical reviewers. See id. The

14

plaintiff‟s contrary suggestion (ECF No. 30 at 5) is inaccurate.

15

Logically, too, there is no decisive link from one party‟s self-interest to another‟s credibility.

16

That is to say, from Sun Life‟s structural incentive to minimize benefit payments it does not follow

17

that its outside consultants distorted their assessments to favor the insurer. The two phenomena are

18

distinct. The Gonda court thus rightly observed that an insurer‟s “purported conflict of interest has

19

little or no bearing on whether [a] Plaintiff can introduce extra-record evidence concerning the

20

credibility of” third-party “physicians . . . retained . . . to review [the] Plaintiff‟s claims.” Gonda,

21

300 F.R.D. at 616 n. 3.

22

Obviously, there may be “exceptional” cases in which these things are tied; for example, where

23

an insurer, driven by a virulent self-regard, so distorts its obligation to rightly handle benefits

24

claims that it hires only those consultants who knowingly warp their own conclusions to protect

25

their paymaster‟s treasury. But this cannot be deemed the ordinary case — particularly in view of

26

the “restrictive” Opeta standard that bounds this whole inquiry. A case cannot fall into Opeta’s

27

“exceptional” class of “credibility” cases simply because there exists a garden-variety “structural

28

conflict of interest.”
12
ORDER (No. 3:14-cv-05295-JST (LB))

1

Sun Life‟s “structural conflict” alone does not suggest — much less “clearly establish” — that

2

its outside consultants were “biased” or otherwise lacked “credibility” so as to warrant

3

“exceptional” discovery beyond the administrative record.
* * *

4
5

4.2 Third-Party Medical Reviewers’ Credibility and Bias

6

We now reach the point on which the plaintiff puts the most weight: the notion that, because

7

the insurer paid them and perhaps had hired them in other cases, the third-party physicians whom

8

Sun Life hired to review the plaintiff‟s medical records were “biased” in Sun Life‟s favor. The

9

plaintiff writes that the conclusion that Sun Life‟s four outside consultants reached (in sum, that he
was not disabled) is contrary to the weight of the administrative record — meaning the records,

11
United States District Court
Northern District of California

10

reports, and opinions of the plaintiff‟s “dozens” of treating physicians. The consultants‟

12

“incredible” opinions provide “clear evidence of [their] bias and lack of credibility,” in the

13

plaintiff‟s view, and so justify additional discovery under Opeta. (ECF No. 30 at 5-7.)

14

The court disagrees. That these reviewers reached conclusions different from those reached by

15

the plaintiff‟s treating physicians does not stand as evidence that the outside physicians were

16

biased in Sun Life‟s favor, much less that they “clear[ly]” lack credibility. This is true even if, as

17

the plaintiff suggests, the reviewers‟ reasoning and opinions cut against the overwhelming weight

18

of the administrative record. That would not mean that they distorted their work to serve Sun

19

Life‟s financial interests. It would not mean that they “summarily ignored” (ECF No. 30 at 6) the

20

plaintiff‟s existing medical records. That their opinions do not jibe with the opinions of “dozens”

21

of the plaintiff‟s own physicians means only that they took a minority view. A view, perhaps, with

22

little support in the wider administrative record. These last points might in turn suggest, at most,

23

that Sun Life‟s reviewers are incorrect. Not biased, or lacking credibility, but wrong. A de novo

24

review can account for all this without imagining that the reviewers were shills who warped their

25

analyses to serve Sun Life‟s pecuniary interests. All the items that the plaintiff cites as indicating

26

“bias” and “lack of credibility” (ECF No. 30 at 6) really just take issue with the soundness of the

27

medical reviewers‟ reasoning or the rightness of their conclusions.

28

The plaintiff ‟s argument at the hearing on this matter confirmed this last point. The plaintiff
13
ORDER (No. 3:14-cv-05295-JST (LB))

1

there urged that what was “exceptional” about this case, and what demonstrated the medical

2

reviewers‟ bias, was that they had summarily dismissed or just “ignored” what the plaintiff

3

considers to be “gold standard” support demonstrating that he has thoracic outlet syndrome. The

4

proposed deposition could inquire into that bias — or, more precisely, into why the reviewers had

5

“ignored” decisive material. Which is to say that the plaintiff would not inquire into the reviewers‟

6

“bias,” so much as invoke that putative bias to justify a further inquiry into the soundness of their

7

reasoning and conclusions. But, judging again from the plaintiff‟s own statements, the existing

8

record should provide enough information from which the district court may assess the reviewers‟

9

substantive work and reach its own de novo decision on whether the plaintiff is disabled.
Furthermore, consider the plaintiff‟s likely position if he were allowed to depose Sun Life‟s

11
United States District Court
Northern District of California

10

consultants — and these consultants elaborated defenses of their work. Would the plaintiff

12

consider that material as outside the administrative record and thus beyond consideration? At all

13

lengths, the plaintiff‟s “bias” argument seems a dispute with the quality of the reviewers‟ work.

14

The plaintiff sees the bias as stemming from the fact that Sun Life paid these outside

15

reviewers. (See ECF No. 30 at 5-7.) These “paid consultants,” the plaintiff writes, were “hired to

16

render opinions that support Sun Life‟s terminations. Accordingly, additional discovery is

17

necessary to investigate and expose the extent of this bias and its effect on the credibility,

18

objectivity, and thus reliability of their opinions . . . .” (ECF No. 30 at 7) (emphasis in original).

19

The court again disagrees. The court sees better reasoning in those decisions holding that mere

20

compensation does not itself justify extrinsic discovery under Opeta’s “credibility of medical

21

experts” head. See Polnicky, 2014 U.S. Dist. LEXIS 29123 at *7 (“mere fact that a physician

22

receives compensation from a plan administrator for performing medical reviews is insufficient by

23

itself to be probative of bias”) (citing Lavino v. Metro. Life Ins. Co., 779 F. Supp. 2d 1095, 1104

24

(C.D. Cal. 2011), Nolan, 745 F. Supp. 2d at 923, Kludka v. Qwest Disability Plan, No. CV-08-

25

1806, 2010 U.S. Dist. LEXIS 34572, *2 (D. Ariz. Apr. 7, 2010), and Brice, 2011 U.S. Dist. LEXIS

26

77525, at *3. The plaintiff does not challenge the reviewers‟ qualifications; nor does he point to

27

misrepresentations or similar problems in their work. See Polnicky, 2014 U.S. Dist. LEXIS, at *7-

28

*8; Brice, 2011 U.S. Dist. LEXIS 77525, at *3. He does not show that they were biased, or
14
ORDER (No. 3:14-cv-05295-JST (LB))

1

otherwise lack credibility, to warrant additional discovery under Opeta.

2

Some cases have reached a different conclusion. Facing essentially the same “bias” and

3

“credibility” arguments that the plaintiff makes here, these cases have held that Opeta sanctions

4

additional discovery into the relationship between a plan administrator and its outside medical

5

reviewers, “because it goes to the weight the court will assign those [reviewers‟] opinions in its de

6

novo review.” Rowell, 2012 U.S. Dist. LEXIS 16957, at *11; accord Gonda, 300 F.R.D. at 614-15;

7

Bourland, 2014 U.S. Dist. LEXIS 134813, at *7; Reynolds v. UNUM Life Ins. Co. of Am., No.

8

2:10cv2383, 2011 U.S. Dist. LEXIS, *3 (D. Ariz. Aug. 12, 2011); see also Eisner, 2013 U.S. Dist.

9

LEXIS 65090 at *4-*9 (analyzing outside Opeta). But Opeta’s mandate to limit additional
discovery to “exceptional circumstances,” including on the issue of medical-reviewer credibility,

11
United States District Court
Northern District of California

10

leads the undersigned to a different view. This points to a more general consideration of Opeta’s

12

rule and the way in which it has guided this decision. That consideration occupies the last part of

13

this analysis.
* * *

14
15
16

5. Key Analytical Points
Two operative points have guided the conclusions reached above, especially on the issues of

17

Sun Life‟s “structural conflict of interest” and the possible “bias” of its outside medical reviewers.

18

The first point is that Opeta must be taken at its “restrictive” word. “Exceptional circumstances”

19

cannot too readily be found without nullifying Opeta. The second is that the directness of a de

20

novo review obviates most extra-record discovery into the motives and putative shortcomings of

21

those who assessed the plaintiff‟s benefits claims. In circumstances like the present, at least,

22

further inquiry into such matters is not “clearly . . . necessary” to a de novo review.
* * *

23
24

5.1 Opeta Must be Applied “Restrictively”

25

There is tension in the heart of Opeta. On the one hand, to advance ERISA‟s goals of keeping

26

claims procedures prompt and inexpensive, Opeta expressly describes a “restrictive” standard,

27

under which discovery beyond an administrative record may be allowed “only” in “limited” and

28

“exceptional circumstances.” Opeta, 484 F.3d at 1217. The de novo plaintiff who seeks such
15
ORDER (No. 3:14-cv-05295-JST (LB))

1

discovery must “clearly establish[]” that extrinsic evidence is “necessary to conduct an adequate

2

de novo review of the benefit decision.” Id. (emphasis in original). Even then, a court may

3

conclude that further discovery is not “required.” Quesinberry, 987 F.2d at 1027. “In most cases,”

4

the Ninth Circuit has “emphasized,” “only the evidence that was before the plan administrator at

5

the time of determination should be considered.” Opeta, 484 F.3d at 1217 (quotation omitted).

6

On the other hand, Opeta counts among the “exceptional circumstances” that “could” warrant

7

extrinsic discovery situations that arise frequently. As we have already noticed, defendants in

8

ERISA benefits-denial suits have often served as both plan administrator and plan funder, setting

9

up the “structural conflict of interest” that might cause a court to question the insurer‟s
“impartiality” in denying benefits. See id. Furthermore, judging from the case law, ERISA plan

11
United States District Court
Northern District of California

10

administrators often hire third-party physicians to assess claimants‟ cases. No one would expect

12

such consultants to work for free, and the fact that they are paid, and are perhaps employed

13

repeatedly by the same administrator, will reliably prompt ERISA plaintiffs to challenge their

14

opinions as “biased” and lacking “credibility.” See id.

15

This tension between Opeta’s ostensibly “restrictive” rule and the common situations that it

16

describes has led district courts to contradictory decisions. Thus, some cases have agreed that an

17

insurer‟s structural conflict justifies extrinsic discovery (see Knopp), while others have held that

18

conflict irrelevant to de novo review (Gonda). Decisions have split more numerously over whether

19

to permit additional discovery into the “bias” and “credibility” of paid outside medical reviewers.

20

Some cases have swiftly concluded that such evidence would help the de novo court to weigh the

21

reviewers‟ opinions (e.g., Rowell); while others — correctly, in the undersigned‟s view — hold

22

that the mere fact of compensation does not alone raise sufficient “credibility” issues under Opeta,

23

and that the directness of the de novo inquiry makes such considerations irrelevant (e.g., Polnicky).

24

The undersigned thinks that Opeta must be taken at its “restrictive” word. If it justified

25

extrinsic discovery merely to note the existence of a structural conflict of interest, or to observe

26

that consultants were paid, then the situations in which Opeta authorizes such discovery would be

27

routine rather than “exceptional.” Opeta would erase its own rule. The undersigned thus

28

respectfully concludes that decisions permitting such discovery — once a claimant points to a
16
ORDER (No. 3:14-cv-05295-JST (LB))

1

“structural conflict of interest,” or wonders if an outside reviewer is “biased” and lacks

2

“credibility” because the plan administrator paid for her opinion — drift away from Opeta’s

3

explicitly “limited” rule. A plaintiff who wants discovery beyond the existing administrative

4

record in a de novo case must point to something more than a mere structural conflict of interest or

5

the fact of compensation to third-party medical reviewers. That is more in keeping with the letter

6

and substance of Opeta.
* * *

7
8

5.2 De novo review obviates most discovery into “conflict” and “bias”

9

Adhering closely to Opeta’s “significant restrictions,” Gonda, 300 F.R.D. at 613, is also
consistent with the (perhaps more basic) insight that a de novo review, by its nature, need not

11
United States District Court
Northern District of California

10

concern itself with the motives of plan administrators or their paid third-party medical reviewers.

12

The only question on de novo review, as we have repeatedly said, is “simply . . . whether the plan

13

administrator correctly or incorrectly denied benefits.” Abatie, 458 F.3d at 963. Why the

14

administrator reached a putatively incorrect decision — whether because its self-interest inclined it

15

to hold on to funds; or because it relied on the “biased” opinions of lackey consultants — that

16

question is supplanted and made “irrelevant” by the directness and simplicity of the de novo

17

inquiry.

18

Going further, one can say that de novo review can account for (and not merely avoid) some

19

issues that often lead ERISA plaintiffs to seek additional discovery. Consider Tarasovsky, supra,

20

which both parties cite. There, too, a plan administrator‟s outside reviewers concluded that the

21

plaintiff was not disabled. See Tarasovsky, 2012 U.S. Dist. LEXIS 69921 at *1-*2, *6-*9. The

22

Tarasovsky court denied summary judgment (to both parties) partly because the administrative

23

record left “unanswered questions” about the relationship between the insurer and its outside

24

reviewers. Id. at *15-*19. “The unknown connection” between the defendant and its consultants,

25

Tarasovsky reasoned, was “especially concerning . . . where so many other physicians concluded

26

that [the] plaintiff was disabled.” Id. at *16. The plaintiff here draws a parallel to Tarasovsky; he

27

argues that Sun Life‟s outside reviewers found no disability despite an “avalanche” of proof from

28

his treating physicians. (ECF No. 30 at 6.) This impugns the reviewers‟ bona fides, in the
17
ORDER (No. 3:14-cv-05295-JST (LB))

1

plaintiff‟s view, so that further discovery should be allowed into the “connection” between Sun

2

Life and its consultants. (See id.)

3

That is the wrong conclusion. First, Tarasovsky used a different legal standard. Though it
declined to decide which standard applied to its merits review, Tarasovsky effectively used the

5

“more complex” variant of the abuse-of-discretion standard. See Tarasovsky, 2012 U.S. Dist.

6

LEXIS 69921 at *11-*15. That review is conceptually wider-ranging than de novo review. See id.

7

(discussing Montour, 588 F.3d at 629-31). Under the “more complex” inquiry, “all the facts and

8

circumstances” surrounding a benefits decision “must be considered”; unlike de novo review, this

9

innately admits of inquiry into how the administrator‟s or its consultants‟ motives affected that

10

decision. See id. at *13-*15. Thus, a court conducting a “complex” abuse-of-discretion analysis

11
United States District Court
Northern District of California

4

“may consider evidence beyond that contained in the administrative record.” Id. at *15 (quoting

12

Abatie, 458 F.3d at 970). But that is not our case. Second, the de novo standard that applies here

13

can substantively account for the fact that paid consultants reach decisions that go against the

14

weight of the administrative record. As we observed earlier, in deciding whether a benefit decision

15

was correct, a de novo court can consider the facts that Sun Life‟s outside medical reviewers took

16

a minority position, and that (according to the plaintiff) their opinions find little or no support in

17

an “avalanche” of record material. The de novo review can, in this way, account directly for the

18

contrarian position of Sun Life‟s consultants without imagining that they or Sun Life acted from

19

improper motives, and without launching additional discovery into those essentially ancillary

20

questions.

21

Consider too the case of Barteau, supra. The plaintiff writes that Barteau “found evidence

22

regarding claims[-]reviewer bias necessary to its de novo review.” (ECF No. 30 at 7 (citing

23

Barteau, 2009 U.S. Dist. LEXIS at *52). (The word “necessary” is a bit of an overstatement;

24

Barteau did not mention Opeta’s “necessary” limitation but instead wrote that “when the evidence

25

in question pertains to the bias of the reviewing physician, such evidence is permitted under a de

26

novo standard of review.” Id.) The Barteau court did consider extrinsic evidence into whether the

27

insurer-administrator‟s outside consultant “was an advocate for” the insurer. Id. The present point

28

is to notice that Barteau need not have done so to carry out its de novo review; or, more to the
18
ORDER (No. 3:14-cv-05295-JST (LB))

1

point, that the present plaintiff need not obtain additional discovery to address the similar

2

problems he perceives in the work of Sun Life‟s third-party consultants. The Barteau court was

3

concerned that the insurer‟s medical reviewer had reached conclusions contrary to the plaintiff‟s

4

historical medical records, that his analysis was internally inconsistent, and that he proffered

5

statements that were flatly incorrect. See id. at *26-*28. The plaintiff here makes similar

6

complaints about the work of Sun Life‟s consultants. (ECF No. 30 at 5-6.) But these challenges do

7

not justify a new investigation into the motives of Sun Life‟s reviewers. Assume that the plaintiff

8

is correct in all that he says about the outside reviewers‟ reasoning and conclusions. The de novo

9

inquiry can account for all this directly. If a medical reviewer seems to have “summarily
ignore[d]” a wealth of existing information; if his analysis is internally inconsistent; if he proceeds

11
United States District Court
Northern District of California

10

under factual statements that are demonstrably incorrect — all this can weigh into whether, on de

12

novo review, the district court concludes that the reviewer, and the insurer who relied on his

13

opinions, “correctly or incorrectly denied benefits.” Invoking such concerns, in other words, does

14

not “clearly establish” that additional discovery is “necessary” to an “adequate” de novo review.

15

* * *

16

The undersigned thinks that, in light of Opeta’s “restrictive” standard, and ERISA‟s concern

17

that claims be resolved inexpensively and expeditiously, it cannot be too quick to perceive

18

“exceptional circumstances” that permit discovery beyond the administrative record. Particularly

19

where de novo review, being direct, negates or accounts in its own way for many such problems.

20

This “restrictive” approach is more than consistent with Opeta; it ensures the basic integrity of

21

Opeta‟s framework and rule, which otherwise threatens to swallow itself.

22

The plaintiff has pointed only to the existence of a structural conflict and the fact that Sun Life

23

paid its outside medical reviewers. From those facts he then offers his suspicion and suggestion

24

that Sun Life and the outside reviewers may have reached their conclusions from inappropriate

25

motives. He has not “clearly established” that discovery into these matters is “necessary” to an

26

adequate de novo review.
* * *

27
28

19
ORDER (No. 3:14-cv-05295-JST (LB))

1

CONCLUSION

2

The decision whether additional discovery is warranted in an ERISA benefits-denial case is

3

discretionary and contextual. If Opeta means what it says, then it applies to only “exceptional

4

circumstances,” where further discovery is “clearly . . . necessary” to an adequate de novo review,

5

and not to “most cases.” But the pertinent circumstances here are not exceptional. They are

6

ordinary. The ultimate question is whether the administrative record gives the district court enough

7

information to decide de novo whether the plaintiff is entitled to benefits. For the reasons

8

discussed throughout this order, the existing record here does provide enough information to make

9

that decision.

10

United States District Court
Northern District of California

11
12

The court thus partly grants and partly denies the plaintiff‟s requested discovery. Sun Life is
ordered to produce the material described in the foregoing order and summarized as follows:
Sun Life must produce any parts of the administrative record, defined as “relevant” under the

13

applicable ERISA regulations, that have not yet been given to the plaintiff. See supra, Analysis,

14

Part 1.

15

Insofar as ERISA regulations require Sun Life to produce the policies and guidelines that

16

applied to the plaintiff‟s claims, and to the extent that those have not yet been produced, Sun Life

17

is ordered to produce them. Beyond the policy documents, the plaintiff cannot obtain further

18

discovery into whether Sun Life followed its policies and guidelines in handling his claims. If Sun

19

Life has any material that “demonstrates [its] compliance with the administrative processes and

20

safeguards” required for the specific purposes of 29 C.F.R. § 2560.503-1(b)(5), it must produce

21

that material. See supra, Analysis, Part 2.

22
23

The plaintiff is not entitled to additional discovery into Sun Life‟s relationship with the thirdparty medical reviewers that it hired to assess the plaintiff‟s claims. See supra, Analysis, Part 4.

24

The plaintiff will not be permitted to conduct depositions on any of this material. Cf. Johnson

25

Leung v. Cigna Life Ins. Co. of N. Am., 2011 U.S. Dist. LEXIS 9109, *7 (N.D. Cal. Jan. 24, 2011)

26

(“Depositions are generally not warranted or approved.”) (abuse-of-discretion case).

27
28
20
ORDER (No. 3:14-cv-05295-JST (LB))

1

IT IS SO ORDERED.

2

Dated: October 27, 2015

3

______________________________________
LAUREL BEELER
United States Magistrate Judge

4
5
6
7
8
9
10

United States District Court
Northern District of California

11
12
13
14
15
16
17
18
19
20
21

22
23
24
25
26
27
28
21
ORDER (No. 3:14-cv-05295-JST (LB))