From Casetext: Smarter Legal Research

Mulligan v. Provident Life Accident Insurance Co.

United States District Court, E.D. Tennessee
Jan 7, 2011
271 F.R.D. 584 (E.D. Tenn. 2011)


permitting discovery to determine whether Plan administrators, who were also employed by payor, received bonuses or other financial incentives based on company profitability

Summary of this case from Sundermeyer v. Ohio Educ. Ass'n


[Copyrighted Material Omitted] [Copyrighted Material Omitted]

         Daniel Seth Holliday, Eric Buchanan & Associates, PLLC, Chattanooga, TN, for Plaintiff.

          Tony R. Dalton, Woolf, McClane, Bright, Allen & Carpenter, PLLC, Knoxville, TN, for Defendants.


          SUSAN K. LEE, United States Magistrate Judge.

         Before the Court is Plaintiff's motion to compel responses to certain interrogatories and requests for production of documents in this ERISA disability benefits case [Doc. 22]. Plaintiff seeks evidence to support his contention that his benefits were denied because of Defendants' financial conflict of interest. Defendants do not disagree that some discovery is appropriate; they argue instead that Plaintiff's requests go too far. For the reasons stated below, Plaintiff's motion to compel [Doc. 22] will be GRANTED IN PART and DENIED IN PART.

Employee Retirement Income Security Act of 1974 (" ERISA" ), 29 U.S.C. § 1132(a)(1)(B).

         I. BACKGROUND

          In his complaint, Plaintiff alleges the ERISA plan administrator both evaluates and pays claims [Doc. 1 at ¶ ¶ 19, 20]. This allegation does not appear to be contested and, if true, it means a structural conflict of interest exists that must be weighed in determining whether the administrator met the arbitrary and capricious standard. Metropolitan Life Ins. Co. v. Glenn, 554 U.S. 105, 117, 128 S.Ct. 2343, 171 L.Ed.2d 299 (2008). Plaintiff further alleges that the administrator's " concern over its own funds ... influence[d] its decision-making" and that Defendants violated their fiduciary duty to Plaintiff of a decisionmaking process " free of influence by self-interest." [Doc. 1 at¶ ¶ 21, 32-33]. Defendants deny these allegations [Doc. 15 ¶ ¶ 19-21], although they do not dispute for purposes of this motion that the same entity both evaluated and paid claims [Doc. 28 at 3].

Plaintiff also refers to a " policy" to " take advantage" of the " potential applicability" of ERISA to his claim [Doc. 1 at ¶ 22].

         In his brief supporting the motion to compel, Plaintiff expands on his theories of bias. First, he argues the ERISA plan administrator's " incentive system ... rewards employees, including those working on claims, based, in part, on company profitability...." [Doc. 29 at 7]. Second, he argues the administrator " has engaged in an ongoing, company-wide, top-to-bottom program to save money by denying or reducing the payments on claims." [Doc. 29 at 8]. Last, Plaintiff argues the administrator's history of biased claims administration show that the denial in this case was based on self interest [Doc. 29 at 8]. Plaintiff seeks discovery to support these theories, and his discovery requests are discussed in detail below.

         II. ANALYSIS

         A. Availability of Discovery in ERISA Cases

         1. Generally

          As a general rule, an ERISA claimant may not seek discovery of matters outside the administrative record. See Wilkins v. Baptist Healthcare Sys., Inc., 150 F.3d 609, 618 (6th Cir.1998) (Gilman, J., concurring) (noting that a district court may not ordinarily consider new evidence). This general prohibition is founded on two separate principles: First, the reviewing court's role, ordinarily, is not to determine whether a claimant is eligible for benefits, but merely to determine whether the administrator's decision was defensible based on the administrative record. See Perry v. Simplicity Eng'g, 900 F.2d 963, 966 (6th Cir.1990) (holding that under either " de novo" or " arbitrary and capricious" standard, court's review is limited to the record). Second, the prohibition of discovery effectuates ERISA's " primary" goal— the " inexpensive[ ] and expeditious[ ]" resolution of disputes. Id. at 966-67.

          When a claimant makes a " procedural challenge" to the administrator's decision, however, " such as an alleged lack of due process ... or alleged bias," limited discovery is permitted in spite of the general prohibition. Johnson v. Conn. Gen. Life Ins. Co., 324 Fed.Appx. 459, 466 (6th Cir.2009) (quoting Judge Gilman's concurrence in Wilkins, 150 F.3d at 619). In such a case, the first rationale for prohibiting discovery is inapplicable; the court is not prohibited from looking outside the administrative record and must instead consider circumstances affecting the administrator's conflict of interest. Glenn, 554 U.S. at 117, 128 S.Ct. 2343. For the same reason, however, any discovery must be strictly confined to the procedural challenge by which it is justified. Johnson, 324 Fed.Appx. at 466; Moore v. LaFayette Life Ins. Co., 458 F.3d 416, 430 (6th Cir.2006). Moreover, the second rationale for prohibiting discovery remains applicable, and discovery therefore must also be tailored to facilitate the prompt and inexpensive resolution of disputes. See Price v. Hartford Life and Acc. Ins. Co., 746 F.Supp.2d 860, 865-66, 2010 WL 3998039, at *5 (E.D.Mich. Oct. 12, 2010) (courts should account for the " interests of economy, efficiency, accuracy, and fairness" when addressing scope of discovery issues).

As Glenn makes clear, the importance of an administrator's conflict of interest will vary with the circumstances of each case. A conflict will be a weightier factor where, for example, the insurer has a history of biased claims administration. Id. at 117, 128 S.Ct. 2343 . A plaintiff bears the burden to provide evidence of a conflict of interest, see Curry v. Eaton Corp., 2010 WL 3736277, at *7 (6th Cir.2010) (unpublished), and without some discovery, a plaintiff ordinarily cannot make that showing. Thus, a court cannot fairly fault the plaintiff for failing to show the contours of an administrator's conflict without first allowing some discovery. See Calvert v. Firstar Fin., Inc., 409 F.3d 286, 293 n. 2 (6th Cir.2005) (" The Court would have a better feel for the weight to accord this conflict if Calvert had explored the issue through discovery." ).

         2. Types of discovery relating to an alleged conflict of interest

          As in all civil cases, the permissible scope of discovery is drawn by reference to the parties' pleadings. See Fed.R.Civ.P. 26(b) (tying the relevance inquiry to the parties' claims or defenses or, for good cause, to the subject matter); Wilkins, 150 F.3d at 618 (Gilman, J., concurring) (additional evidence may be necessary to explore " alleged lack of due process" or " alleged bias" ) (emphases added). Relevant discovery may be had with respect to matters " reasonably calculated to lead to the discovery of admissible evidence." Fed.R.Civ.P. 26(b). As explained above, evidence outside the ERISA record may be considered only inasmuch as it relates to a procedural challenge (here, the alleged conflict of interest or bias), so discovery is similarly limited to those issues. Moore, 458 F.3d at 430.           Whether to allow discovery relating to an alleged conflict of interest is a decision entrusted to the trial court's discretion. See Bell v. Ameritech Sickness and Accident Disability Benefit Plan, 399 Fed.Appx. 991, 998, 2010 WL 4244126, at *6 (6th Cir.2010) (unpublished) (" Discovery may be appropriate to determine the weight to accord to a conflict of interest, ... but the district court retains discretion to decide when to allow such discovery." ); Johnson, 324 Fed.Appx. at 467 (noting that discovery will not be " automatically" available in every case where there is a conflict of interest). A court may deny discovery, even when relevant to the plaintiff's allegations, when it is cumulative, unnecessary, or unduly burdensome. Fed.R.Civ.P. 26(b), (c) (discovery must be limited where its burden outweighs its likely benefit and may be limited where its burden is " undue" ). See also Bell, 2010 WL 4244126, at *6 (approving the trial court's denial of discovery when it would not have aided the court's review); Murphy, 619 F.3d at 1163; Hughes v. CUNA Mut. Group, 257 F.R.D. 176, 179 (S.D.Ind.2009). And, in considering the " burden" of proposed discovery, courts may take into account ERISA's goals of expedience and thrift. See Price, 746 F.Supp.2d at 865-66, 2010 WL 3998039, at *5.

A plaintiff may not allege that a plan administrator was biased with only an uninformed hope that discovery will turn up some evidence. As always, the plaintiff's allegations must themselves be likely to find evidentiary support after a " reasonable opportunity for further investigation or discovery." Fed.R.Civ.P. 11(b)(3).

An administrator's conflict of interest is relevant only if the administrator's decision is to be reviewed under the arbitrary and capricious standard of review. See Price, 746 F.Supp.2d at 866, 2010 WL 3998039, at *6 (" If the standard of review is de novo, then the significance of the administrator's conflict of interest evaporates ...." ). Here, Plaintiff alleges that the plan documents in this case " do not grant discretion" or are " not sufficient to grant discretion" and that a de novo standard applies [Doc. 1 at ¶ ¶ 17-18]. Defendants deny such allegations and also claim that there appears to be no dispute that the Court's review of the substantive issues is limited to the administrative record based on the Plaintiff's arguments in support of his motion to compel. However, the applicable standard of review is not yet determined.

         3. Special case: Historical evidence of biased claims administration

          As described in further detail below, Plaintiff seeks to discover evidence of Defendants' allegedly biased handling of previous claims. Historical evidence will, in some circumstances, be discoverable. As Glenn instructs, a structural conflict of interest

should prove more important (perhaps of great importance) where circumstances suggest a higher likelihood that it affected the benefits decision, including, but not limited to, cases where an insurance company administrator has a history of biased claims administration.

554 U.S. at 117, 128 S.Ct. 2343. The relevance of this type of evidence, however, presents a difficult question: how extensive may discovery of an insurer's " history" of claims administration be? The line has not yet been drawn by any controlling authority. See Wilcox v. Metropolitan Life Insurance Company, 2009 WL 57053, at *2 (D.Ariz.2009) (noting the " challenging issue" of how, in the absence of controlling authority, to balance the availability of discovery against the specter of a claimant's " inquiring into hundreds of past cases in an effort to unearth evidence of ... systematic bias" ).

         a. Likelihood that the evidence will show bias

          The principle limiting factor may be derived from Glenn itself; historical evidence of bias can put a thumb on the scale of the court's review only when it is " suggest[ive]" of bias in the particular claim. 554 U.S. at 116-17, 128 S.Ct. 2343 (counting a history of bias among the " case-specific" factors that might " suggest" that bias influenced a particular decision); Marrs v. Motorola, Inc., 577 F.3d 783, 789 (7th Cir.2009) (explaining that the " likelihood that the conflict of interest influenced the decision is ... the decisive consideration" under Glenn ). Thus, a trial court may screen discovery requests by considering the likelihood that the requested information will produce evidence that the financial conflict motivated the decision at issue. See, e.g., Hughes, 257 F.R.D. at 180 (concluding that information about other claims was " not reasonably likely to lead to admissible information" about Hughes' claim).

          In determining whether historical evidence is suggestive of bias in a particular case, courts have considered a variety of factors. One such factor is the scale of the history of bias. A large body of historical evidence might lend some statistical credibility to the claim that a pervasive bias infected a particular decision. See Hackett v. Standard Insurance Company, 2010 WL 1494772 (D.S.D.2010) (stating that without " statistical information," a claim that the use of particular consulting physicians caused bias was " mere speculation" ). But see Holmstrom v. Metropolitan Life Ins. Co., 615 F.3d 758, 768 (7th Cir.2010) (alluding to " sampling problems" that might plague any attempt to draw conclusions from an insurer's " batting average" in litigation).

          Numerous instances of historical bias, however, are neither necessary nor sufficient to " suggest" bias in a particular case. First, large-scale historical evidence is not necessary because evidence of bias from a single previous claim might also be probative if, for example, the same employees and/or reviewing physicians, the same policies, or the same medical conditions were involved. See Zalkin v. Coventry Health Care of Neb., Inc., 2010 WL 1052263, at *6-9 (D.Neb.2010) (allowing discovery of investigations and results of " similar" claims). On the other hand, anecdotal evidence of bias from unrelated claims should not ordinarily be relevant to a given claim, absent some relevant (and limiting) similarity. See Branch v. Life Ins. Co. of N. Am., 2009 WL 3781217, at *5 (M.D.Ga.2009). Second, large-scale historical evidence is not sufficient to suggest bias in a particular case, because an attenuated temporal relationship between the historical evidence and the instant claim may cancel out the " suggestion" of bias. See Jones v. Unum Provident Corp., 596 F.3d 433, 438 (8th Cir.2010) (noting that courts had " heavily criticized Unum for unfair claims practices in the decade ending in 2003," but declining to modify the standard of review with respect to a claim filed in 2005). Or, as indicated by Glenn itself, the insurer's implementation of policies or practices to reduce bias may remove the suggestion of bias. 554 U.S. at 117-18, 128 S.Ct. 2343.

          It might be argued that this inquiry— the strength or scale of the " suggestion" of bias— is a question of the probative value of the evidence, not its discoverability. Such a threshold inquiry, however, is consistent with a court's obligation to confine ERISA discovery to its permissible purposes. A court must determine not only whether a discovery request is relevant, but also whether the requested information will make a difference in its review— i.e., the " likely benefit" of the evidence. Fed.R.Civ.P. 26(b); Hughes, 257 F.R.D. at 179. See also Bell, 2010 WL 4244126, at *6; Murphy, 619 F.3d at 1163. When the potential purposes of discovery are circumscribed, the likely benefits of any particular discovery request are correspondingly limited.

         b. Likelihood that evidence will be admissible

         Discovery requests seeking historical evidence of bias may also be limited if they are not " reasonably calculated" to lead to the discovery of evidence admissible under the Federal Rules of Evidence. See Fed.R.Civ.P. 26(b). The rules of evidence are not ordinarily at issue in an ERISA case, because court review is typically confined to the administrative record, and the rules of evidence do not apply to the administrative proceedings in which the record is created. See Speciale v. Blue Cross, 538 F.3d 615, 622 n. 4 (7th Cir.2008). Nonetheless, the rules of evidence govern " proceedings in the courts of the United States," Fed R. Evid. 101, not excepting ERISA actions, see Fed.R.Evid. 1101. Thus, evidence received for the first time during judicial review can be admitted only if it satisfies those rules. See Bartholomew v. Unum, 588 F.Supp.2d 1262, 1267 (W.D.Wash.2008) (excluding evidence of Unum's history of biased claims administration as hearsay); Baldoni v. UNUM Provident, 2008 WL 140716, at *4 (D.Or.2008) (considering evidence of conflict of interest outside the record " to the extent such evidence is relevant and is otherwise admissible under the rules of evidence" ); Troutman v. Unum, 2007 WL 2122649 (N.D.Cal.2007) (same).

          As explained above, evidence of a history of biased claims administration is significant only insofar as it is probative of bias in the particular claim. Such evidence is, by definition, intended to prove conformity with prior conduct, and is therefore inadmissible under Fed.R.Evid. 404 unless it is probative of an organization's " routine practice" under Fed.R.Evid. 406. See Aluisi v. Elliott Mfg. Co., Inc. Plan, 2009 WL 565544, *7 (E.D.Cal.2009). In order to show a " routine practice," evidence must show consistent conduct under " similar circumstances." United States v. Callahan, 551 F.2d 733 (6th Cir.1977). If a discovery request aimed at " historical bias" is not reasonably likely to produce evidence of an organization's conduct in similar circumstances, then it cannot be reasonably calculated to lead to the discovery of admissible evidence.

         B. Plaintiff's Discovery Requests

         Plaintiff seeks to compel responses to 23 interrogatories and requests for production, which can be usefully separated into seven categories [Doc. 23-3].

In support of his motion, Plaintiff submits discovery orders from other cases, including several non-ERISA cases [ e.g., Doc. 23-4; Doc. 31-1], in which courts have allowed discovery of requests that might fall into one or more of these categories. The Court has reviewed the orders but will not attempt to compare or distinguish them in detail here. Some of the orders contain little or no analysis of the underlying allegations or specific requests [ e.g., Doc. 23-5; Doc. 23-6; Doc. 31-1], and are consequently not useful as precedent. Although precedent from non-ERISA cases may be useful, the Court notes discovery orders from non-ERISA cases do not include consideration of ERISA's goals of speedy and inexpensive litigation. Furthermore, discovery in any case must stand on its own merits, and the Court therefore considers the requests at issue here in the specific context of this case.

         1. Definitions of terms used in the plan documents for Defendants' performance-based incentive plans

          There appears to be no dispute that Defendants reward their employees with incentives and bonuses based on company profitability and Defendants have produced the incentive plans themselves [Doc. 28 at 5]. Plaintiff seeks the definitions of several terms used in the plan documents for Defendants' performance-based incentive plans, such as: " functional area achievement factor" [Interrogatory (" Int." ) 9]; " corporate achievement," " operating earnings" [Int. 10]; " operating expense efficiency" [Request for Production (" RFP" ) 16]; " service measure" [RFP 17]; " functional area achievement" [RFP 18]; " operating earnings," " before tax operating earnings," and " net operating earnings" [RFP 19]. These terms are used, but not defined, in those plans [ see Doc. 28 at 9].

         Defendant argues that these terms were used by managerial employees and accountants, and Plaintiff has not argued that the definitions were known to the employees who administer claims. Indeed, evidence submitted by Plaintiff appears to indicate that the definitions were not known by those employees [ see Doc. 23-7 at 15 (deposition testimony explaining that the " actual calculation" of " before tax operating earnings" was not known to Defendants' employees) ]. Nonetheless, Plaintiff argues he is entitled to discover the existence of any financial incentive Defendants' employees might have had to deny his claim.

         While these requests do not appear particularly burdensome, they are not proper subjects for discovery. Whether employees are rewarded based on company profitability is not at issue; they are. And, all things being equal, an insurer will be more profitable if it pays fewer benefits claims. That is the essence of a structural conflict of interest, and discovery requests aimed at " exposing" it are simply cumulative. Plaintiff is entitled to additional discovery only to determine whether the structural conflict did in fact influence " first-level decisionmaker[s]" to deny benefits in this case. See Robbins v. Milliman USA LTD Plan, 2003 WL 22246952, at *8 (S.D.Ind.2003). The plan documents themselves, already produced, describe the performance incentives for the employees who decide claims, and these terms are not defined in those documents. There is no reason to believe that the definitions of these terms, not utilized by the persons who actually evaluated Plaintiff's claim, could have any probative value to show bias was a factor here. These definitions, in the Bell court's words, are not likely to aid the Court's review. Plaintiff's motion to compel with respect to these requests is therefore DENIED.           2. Definitions of terms used by managerial personnel to monitor claims inventory

          Plaintiff also seeks to discover the definitions of " liability acceptance rate" [Int. 12]; " benefit ratio" [Int. 13]; " net termination ratio" [Int. 14]; " historic levels," " historic measures," " historic performance," and " historic pay rate" [Int. 16]. These terms, according to Defendants, " were not derived from the claim file regarding Plaintiff's claim or the claim file of any other claimant," but are instead used by managerial personnel to " monitor and manage the company's claim inventory." [Doc. 28 at 11]. The analysis here tracks that of the first category of disputed discovery requests. Plaintiff has provided no reason to believe that these definitions have any explanatory power to show how the handling of his claim might have been infected by bias, and his motion to compel is therefore DENIED with respect to these requests.

         3. Financial reports and reserve information

          Plaintiff seeks to compel the production of " financial reports, reports of profitability, reports of reserves, or any other reports that show the financial impact of claims" [RFP 10, 31]. As Defendants argue, these requests are not appropriate for ERISA discovery. With respect to financial reports, this information has no potential to show that Plaintiff's claim might have been mishandled, except to prove the general structural conflict of interest which Defendants concede. With respect to reserve information (i.e., an insurer's estimation of its potential liabilities), an insurer's awareness of how much a claim may cost it does not show a conflict of interest. See Klein v. Central States, 346 Fed.Appx. 1, 5 (6th Cir.2009). A fortiori, the insurer's estimation of the potential cost of an aggregation of claims is insufficient to show bias. No prudent fiduciary would fail to balance its liabilities against its assets. Cf. Am. Protection Ins. Co. v. Helm Concentrates, Inc., 140 F.R.D. 448, 449 (E.D.Cal.1991) (noting that " a prudent insurer would establish reserves sufficient to pay claims based upon many factors, only one of which might be the estimate of the chances of the claimant's success" ). Accordingly, Plaintiff's motion to compel is DENIED with respect to these requests.

         4. Documents, from Plaintiff's claim and any other claim, containing certain " search" terms

          Plaintiff seeks documents from his claim and any other claim containing any of numerous search terms— advance pay and close, return to work, recovery, resolution, denial, non-compensable claim, claim closure, stretch goals, scorecard measurements, termination, termination ratio, net termination ratio, historic pay rate, historic performance, historic target, historic goals, expected resolution date, round tables, file review, operational metrics and reporting (" OMAR" ), panel review, panel discussion, and variants of these terms— from disability benefits specialists or their managers [RFP 11], assistant vice presidents, directors [RFP 12], and higher-level supervisors [RFP 13].

         Defendants represent that they have already provided all correspondence containing any reference to Plaintiff's claim, and that Plaintiff has therefore already received any correspondence about his claim containing any of these terms. This request, then, seeks correspondence relating only to other claims. The Court agrees with Defendants that this asks too much. Plaintiff has not articulated any reason that the use of these terms in connection with other claims are relevant to bias in his claim. Moreover, many of these terms (e.g., file review) are likely used in most, if not all, claim files, and the burden of allowing the discovery would greatly outweigh its likely benefits. Plaintiff's motion to compel is therefore DENIED with respect to these requests.

         5. Periodic reports, including monthly trends reports and OMAR information

          Covering the time period from the month Plaintiff's claim was filed until the present, Plaintiff seeks information concerning " monthly trends reports," including the definition of that term and a list of employees with access to the reports [Int. 11], and the production of several periodic reports, including the monthly trends reports themselves, OMAR information, and other claims-tracking reports maintained by supervisors [RFP 22, 23, 26, 27, 29]. Defendants argue that these requests are not relevant to the issue of bias.

         While the discovery requests at issue are overbroad, the Court finds that they appear, in part, reasonably calculated to lead to the discovery of admissible evidence. For the reasons explained above, much of this information is unlikely to aid the Court's review because it cannot show how employees who actually worked on Plaintiff's claim might have been affected by Defendants' financial conflict of interest. Plaintiff's request for monthly trends reports, on the other hand, has the potential to show whether those employees were made aware of the effect of their claims decisions on company profitability. Accordingly, Plaintiff's motion to compel is GRANTED IN PART as follows: For the time period beginning on the month in which Plaintiff filed his claim and ending on the month his claim was denied at the final administrative level, Defendants shall answer Interrogatory 11 and produce any report named in Request for Production 22 which was available to or provided to any employee who was directly involved in handling Plaintiff's claim. Defendants are not obligated, however, to produce reports provided to employees who were not directly involved in Plaintiff's claim or to produce these reports from any other time period.

         6. Emails between employees not involved in Plaintiff's claim

          The emails at issue in RFP 33 were produced under a confidentiality order in a prior case (the " Ball " case) in which Plaintiff's counsel was involved. Plaintiff seeks the production of those same emails in this case, despite the fact that the employees who sent and received them worked in a different claim facility [ see Doc. 29-6 at 4] and, according to Defendants, were not involved in Plaintiff's claim in any way [Doc. 28 at 14]. Defendants argue the emails are not relevant, whatever they may contain. Plaintiff argues that the emails contain admissions of bias showing that Defendants' history of biased claims handling " continues to the present" [Doc. 29 at 8]. These emails, however, apparently were sent and received sometime during 2005 and 2006 [Doc. 29-6 at 3-4], and Plaintiff's claim, alleging disability beginning in September 2008, was not filed until January 2009 [Administrative Record at 2, 5]. There is no indication that these emails, about an unrelated claim processed by other employees in another claims-handling location sent in 2005 and 2006, bear any relevance to Plaintiff's claim or even to the subject matter involved in this action. Nor does the discovery appear to be reasonably calculated to lead to the discovery of admissible evidence under the circumstances, particularly given the attenuated temporal relationship. See Jones, 596 F.3d at 438. Plaintiff's motion to compel the emails produced in the Ball case is DENIED.

         7. Employee performance reviews

          Plaintiff seeks performance reviews to discover " if employees were praised or criticized for their performance on this claim or similar claims during the relevant time period." [Doc. 23 at 4]. It appears undisputed that the performance reviews at issue do not contain any mention of Plaintiff or his claim [Doc. 28-8 at ¶ 3]. Defendants' official policy, moreover, precludes the use of claims outcomes as a gauge of employee performance [Doc. 28-8 at ¶ 4].

Although Plaintiff did not provide the Court with his discovery requests pertaining to employee performance reviews, see E.D. Tenn. Local Rule 37.2, the parties have fully briefed the issue, and the Court will therefore consider it on the merits.

          If the requested performance reviews did contain evidence of an untoward influence by supervisory employees during the relevant time period, they would likely aid the Court's analysis of Defendants' conflict of interest. See Cirulis v. UNUM Corp., 321 F.3d 1010, 1017 n. 6 (10th Cir.2003) (stating that whether performance reviews were linked to the denial of benefits is a factor in conflict of interest analysis). A few courts, moreover, have allowed discovery of performance evaluations for just that reason. E.g., Sullivan v. Deutsche Bank Ams. Holding Corp., 2010 WL 391821, at *2 (S.D.Cal.2010) (reasoning that although a " blanket production of personnel files" would have been overbroad, the performance evaluations of the employees who worked on the plaintiff's claims were relevant to whether those employees were rewarded for providing negative assessments of the claim).

         While performance reviews may contain relevant information, however, they may also contain sensitive and private information. See Rinehart v. Life Ins. Co. of N. Am., 2009 WL 2240286 (W.D.Wash.2009) (denying Plaintiff's request for performance reviews, even though they might have contained evidence of a " larger policy or practice" to influence employees to deny claims, because of privacy concerns). Accordingly, the courts within our circuit have been reluctant to allow their discovery. E.g., Mullins v. Prudential Ins. Co. of Am., 267 F.R.D. 504, 521 (W.D.Ky.2010); Pemberton v. Reliance Standard Life Ins. Co., 2009 WL 89696, at *3 (E.D.Ky.2009); Hays v. Provident, 623 F.Supp.2d 840, 844-45 (E.D.Ky.2008); Myers v. Prudential Insurance Company of America, 581 F.Supp.2d 904, 914 (E.D.Tenn.2008). These decisions indicate that performance reviews have been sought for two separate purposes. Where they have been sought as evidence of employee credibility, training, or qualifications, courts have held that their burden exceeds their likely benefit. E.g. , Raney v. Life Insurance Company of North America, 2009 WL 1044891, at *3 (E.D.Ky.2009); Pemberton, 2009 WL 89696, at *4. The disinclination to allow discovery of performance reviews is tempered, however, where they have been sought (as here) as evidence that employees were pressured to deny claims. See Hays, 623 F.Supp.2d at 844-45. In Hays, the court rejected the request for performance reviews, but allowed discovery of " any type of incentive, bonus, or reward program or system, formal or informal," and indicated that if the former requests produced " information suggesting the influence of bias," it might revisit the issue. Id. See also Myers, 581 F.Supp.2d at 914 (" I am disinclined at this point to allow discovery of the personnel files" ) (emphasis added). While the Court is not prepared to say that performance reviews will never be discoverable, the Court agrees with those decisions which have contemplated a phased discovery process. Where an administrator's policies or programs " suggest[ ] the influence of bias," the further discovery of performance reviews may be appropriate. Hays, 623 F.Supp.2d at 845. Here, however, it appears Defendants' policies explicitly prohibit consideration of claims outcomes in performance reviews. Furthermore, Defendants have submitted uncontradicted evidence that Plaintiff's claim was never mentioned in the performance reviews he seeks. Thus, Plaintiff can pursue his motion to compel performance reviews only by theorizing that Defendants have acted in contravention of their own policies with respect to other claims. The Court is not inclined to allow the intrusive exploration of such collateral issues on the slight chance it might reveal some evidence of bias. Plaintiff's motion to compel the production of performance evaluations is therefore DENIED.

         C. Dueling Protective Orders

          Defendants' have agreed to produce job descriptions for the employees who worked on Plaintiff's claim, subject to the entry of their proposed protective order. Plaintiff argues that his proposed order better articulates the legal rights and duties of the parties under applicable law. The Court will not enter either proposed order in its current form. Instead, the parties are ORDERED to jointly submit a proposed order, agreed to in form, complying with the guidance below.

         1. Retroactive effect

         Defendants' proposed order would retroactively protect documents produced prior to the entry of the order [Doc. 28-1 at ¶ 11]. Plaintiff's proposed order does not contain such a provision. Given Defendants' voluntary production of most materials of which discovery has been properly sought, the Court agrees that confidential materials already produced should hereafter be subject to the terms of the protective order.

         2. Use of confidential materials in other cases

          Defendants argue that the primary dispute over the protective order is whether confidential materials produced in this case may be used by Plaintiff's counsel in other cases [Doc. 28 at 16]. Both parties' proposed orders, however, would prohibit the use of confidential documents outside this case [Doc. 23-8 at ¶ ¶ 3, 9; Doc. 28-1 at ¶ 3, 9]. The difference, it seems, is that Defendants' proposed order would also require that confidential documents be returned or destroyed at the conclusion of the case [Doc. 28-1 at ¶ 9]. The Court will not require the return or destruction of documents produced in discovery. Indeed, both parties have indicated their assent to a provision which would allow Plaintiff's counsel to refer to a confidential document in a future case if its existence is denied [Doc. 23-8 at ¶ 3; Doc. 28-1 at ¶ 3]. Plaintiff's counsel may retain the documents for that limited purpose.

         3. Burden of challenging confidential designations

         Both parties' proposed orders would require the designating party to file a motion for a protective order to protect any document whose confidentiality is challenged, whether or not either party intends to use the material [Doc. 23-8 at ¶ 11; Doc. 28-1 at ¶ 12]. Despite the parties' agreement, these procedures have the potential to burden the Court with unnecessary motions and increase the expense of this litigation for all concerned. The revised protective order must place the burden of filing a motion to challenge a confidential designation on the party objecting to that designation. The parties are free, as they have already agreed, to allocate the burden of persuasion to the designating party.

         4. Procedures for filing confidential documents with the Court

         Finally, neither proposed order complies with the Court's procedures for filing documents under seal [Doc. 23-8 at ¶ 8; 28-1 at ¶ 8]. The revised protective order must comply with the provisions of E.D. Tenn. Local Rule 26.2 and the Court's Standing Order governing electronic case filing, In re: Electronic Case Filing Rules and Procedures at ¶ 12 ((as amended, SO-09-06) (Sep.2009)).


         For the reasons stated above, Plaintiff's motion to compel [Doc. 22] is GRANTED IN PART and DENIED IN PART, as follows:

(1) Plaintiff's Interrogatory 11 and Request for Production 22 are GRANTED IN PART. For the time period beginning on the month in which Plaintiff filed his claim and ending on the month his claim was denied at the final administrative level, Defendants shall answer Interrogatory 11 and produce any report named in Request for Production 22 which was available to or provided to any employee who was directly involved in handling Plaintiff's claim.

(2) The remainder of Plaintiff's discovery requests at issue are DENIED.

         In addition, the parties are ORDERED to jointly submit a revised protective order complying with this Order within 30 days of the entry of this Order. As set forth in the modified scheduling order [Doc. 26], Plaintiff shall submit his brief in support of his motion for judgment on the pleadings within 45 days of the entry of this Order; Defendants shall submit their responsive brief within 30 days after Plaintiff's brief is filed; and Plaintiff shall submit his reply, if any, within 10 days after Defendants' responsive brief is filed.

         SO ORDERED.

Johnson might be read to impose a heightened pleading standard for procedural challenges. 324 Fed.Appx. at 466-67 (indicating that in order to be entitled to ERISA discovery, a plaintiff must present a " colorable" claim of bias as opposed to a " mere allegation" ). Johnson, however, disclaimed any such intent, acknowledging Glenn's advice that it is neither " necessary [n]or advisable for courts to create ... special procedural or evidentiary rules" to evaluate conflicts of interests. Id. at 466 (quoting 554 U.S. at 116, 128 S.Ct. 2343). See also Murphy v. Deloitte & Touche Group Ins. Plan, 619 F.3d 1151, 1162-63 (10th Cir.2010).

Summaries of

Mulligan v. Provident Life Accident Insurance Co.

United States District Court, E.D. Tennessee
Jan 7, 2011
271 F.R.D. 584 (E.D. Tenn. 2011)

permitting discovery to determine whether Plan administrators, who were also employed by payor, received bonuses or other financial incentives based on company profitability

Summary of this case from Sundermeyer v. Ohio Educ. Ass'n

surveying cases in which Courts refused to grant discovery into employee personnel files due to privacy concerns

Summary of this case from Sundermeyer v. Ohio Educ. Ass'n
Case details for

Mulligan v. Provident Life Accident Insurance Co.

Case Details


Court:United States District Court, E.D. Tennessee

Date published: Jan 7, 2011


271 F.R.D. 584 (E.D. Tenn. 2011)

Citing Cases

Sundermeyer v. Ohio Educ. Ass'n

When it has been allowed, courts in this Circuit have consistently held that the permissible scope of…

Neubert v. Life Ins. Co. of N. Am.

Post-Wilkins fluctuations notwithstanding, one discovery principle has reigned unchallenged: discovery is…