OPINION

STANLEY R. CHESLER, District Judge.

This matter comes before the Court upon the renewed motion for class certification filed on September 9, 2013 by Plaintiffs Darlery Franco, David Chazen and Camilo Nelson (collectively, "Subscriber Plaintiffs"). Cigna has opposed the motion. The Court denied Subscriber Plaintiffs' first motion for class certification by Order dated January 16, 2013. The Opinion issued by the Court in connection with that Order provides a detailed factual background and defines key terms that likewise apply to this motion, for example "ONET" and "UCR." Because the Court writes for the parties only, it will continue to rely on the same background information without repetition here and will also continue to use the same abbreviated terms as defined in the January 16, 2013 Opinion. (The Opinion relating to the first motion for class certification is reported at 289 F.R.D. 121 (D.N.J. 2013) and will hereinafter be referred to as "Franco I.")

I. INTRODUCTION

In Franco I, the Court found that Subscriber Plaintiffs had adequately demonstrated that the proposed ERISA class satisfied the Rule 23(a) requirements of commonality and typicality but denied certification in large part because Subscriber Plaintiffs did not carry their burden of demonstrating that common questions of law and fact would predominate over individual issues.1 The Court identified three major deficiencies preventing the first motion for class certification from meeting Rule 23(b)(3)'s predominance requirement. In their first motion for class certification, Subscriber Plaintiffs failed to demonstrate (1) that Cigna plan language concerning UCR-based ONET benefits was uniform such that it could be applied to the entire class's claim to recover unpaid benefits; (2) that a violation of plan terms by Cigna under the abuse of discretion standard could be established based on common evidence; and (3) that damages could be calculated based on a standard methodology. The Court also found that class certification was frustrated by a class definition which failed to incorporate the identifying aspects of membership: Cigna subscribers whose plans entitled them to ONET benefits based on a UCR standard to determine the allowed amount and whose ONET claims were determined using Ingenix data.

Subscriber Plaintiffs assert that this renewed motion for class certification is not, as Cigna has contended, an untimely motion for reconsideration in disguise. The briefs submitted by Subscriber Plaintiffs, however, unequivocally demonstrate that, in large part, Cigna's characterization of the motion is correct. The briefs contain numerous instances in which Subscriber Plaintiffs refer to the Court's conclusions on the prior class certification motion as "incorrect" and "mistaken." (Mot. at 22; Reply at 8 n.3.) Nevertheless, it is clear to the Court that portions of Subscriber Plaintiffs' motion indeed raise new arguments to address the impediments to class certification discussed in Franco I. The Court will, therefore, entertain this renewed motion for class certification pursuant to Federal Rule of Civil Procedure 23(b)(3), concentrating on Subscriber Plaintiffs' revised approach to demonstrating that the redefined classes — an ERISA Class and a RICO Class — meet the (b)(3) requirements.

This motion corrects some of the problems identified in Franco I. Subscriber Plaintiffs inject greater precision into the definitions of their two proposed classes. They also narrow the UCR language to two basic formulations widely used in Cigna plans during the relevant time period.

Even as pared down, however, the classes do not present common liability issues that will predominate over individual ones. As Cigna points out, even plans that use one of the two broad UCR definitions vary as to the information that Cigna may consider in determining ONET benefits. The record shows multiple combinations of the UCR provisions and additional clauses that influence the determination of an appropriate UCR and ONET benefit. For reasons the Court will discuss, these variations would have a significant impact on a liability analysis. Yet, Subscriber Plaintiffs fail to demonstrate how a classwide trial of the ERISA and RICO claims would cohesively address such combinations and permutations of applicable plan language. Another substantial obstacle to class certification is the lack of any demonstration that injury, an essential component of liability, is capable of classwide proof. The motion makes the erroneous assumption that, if Subscriber Plaintiffs succeed in proving at trial that Ingenix was a flawed database, harm to all members of the redefined ERISA and RICO Classes necessarily follows. In the discussion below, the Court will elaborate on the reasons that Subscriber Plaintiffs' current effort falls short of passing the rigorous analysis that must be applied to ascertain whether the proposed classes actually conform with Rule 23. Marcus, 687 F.3d at 591. Each proposed class will be addressed in turn.

II. ERISA CLASS

A. Revised Class Definition

The ERISA Class has been redefined in a manner that makes class membership readily ascertainable. According to this motion, the putative class would consist of:

The revised definition includes objective criteria: specific plan language regarding the problematic UCR standard for ONET coverage, the use of Ingenix to determine the allowed amount on an ONET claim, and an allowed amount less than the provider's billed charge. Thus, in line with Subscriber Plaintiffs' theory of the case, the definition captures what the movants allege to be subscribers' rights under their Cigna plans, Cigna's conduct in violation of those rights and injury. In other words, it tailors the class to the ERISA claims at issue in this case: the alleged underpayment of an ONET claim in a manner contrary to plan language because of Cigna's use of Ingenix to determine UCR. The Court notes that the revised ERISA Class definition addresses the issue, raised in Franco I, presented by situations in which the allowed amount is less than the provider's billed charge for reasons unrelated to Ingenix, for example because Ingenix was not used at all or because the subscriber has not yet satisfied his deductible and thus the allowed amount on the claim is zero.

B. Predominance of Legal and Factual Issues

Rule 23(b)(3) requires that questions common to the class predominate such that the class will "prevail or fail in unison" on its claims. Amgen, Inc. v. Connecticut Retirement Plans and Trust Funds, 133 S.Ct. 1184, 1191 (2013). Predominance is similar to Rule 23(a)(2)'s requirement of commonality in that both are concerned with ensuring that the putative class presents common questions of law of fact. Indeed, the Rule 23(a) commonality requirement is generally regarded as subsumed by the more stringent Rule 23(b)(3) predominance requirement. Sullivan v. D.B. Investments, Inc., 667 F.3d 273, 297 (3d Cir. 2011) (en banc), cert. denied, 132 S.Ct. 1876 (2012). Predominance, however, imposes a "far more demanding standard," as it "tests whether proposed classes are sufficiently cohesive to warrant adjudication by representation." In re Ins. Brokerage Antitrust Litig., 579 F.3d 241, 266 (3d Cir. 2009) (quoting Amchem Prods., Inc. v. Windsor, 521 U.S. 591, 623-24 (1997)). A class of plaintiffs seeking to try claims by representation pursuant to Rule 23(b)(3) may satisfy the predominance requirements only when the plaintiffs demonstrate that the elements of their claim are "capable of proof at trial through evidence that is common to the class rather than individual to its members." In re Hydrogen Peroxide Antitrust Litig., 552 F.3d 305, 311-12 (3d Cir. 2009); cf. Newton v. Merrill Lynch, Pierce, Fenner & Smith, Inc., 259 F.3d 154, 172 (3d Cir. 2001) ("If proof of the essential elements of the cause of action requires individual treatment, then class certification is unsuitable.").

The Court's Rule 23(b)(3) analysis must not consider whether the merits questions will be answered in a plaintiff's favor, but it does "`entail some overlap with the merits of the plaintiff's underlying claim'" to determine whether the class seeking certification has presented common liability questions which are capable of classwide adjudication. Amgen, 133 S. Ct. at 1194-95 (quoting Dukes, 131 S. Ct. at 2551); see also Sullivan, 667 F.3d at 306 (holding that the court may have to examine the elements of a plaintiff's claim on a motion for class certification, not to determine the validity of the claim but to evaluate whether the elements can be proved through evidence common to the class). In this case, to conduct a proper analysis of predominance, the Court must take a close look at the plan language on which the ERISA claims are based, as it defines the legal obligations of Cigna towards class members. At a minimum, predominance of legal and factual issues requires that the duty owed be uniform as to the class. Then, the Court must review the evidence presented by Subscriber Plaintiffs to determine whether injury is capable of classwide proof. Named Plaintiffs Franco, Chazen and Nelson wish to adjudicate their own ERISA claims and, in a representative capacity, the claims of other class member subscribers. To do this, they must demonstrate that it is possible to adjudicate the question of whether Cigna's allegedly offensive conduct resulted in harm to all class members, without individual inquiries overwhelming the common issues.

1. Plan Language Concerning UCR and ONET Benefits

In an ERISA action for recovery of unpaid benefits and/or for breach of fiduciary duty, the critical liability questions depend on plan language. Indeed, the Supreme Court has repeatedly held that adherence to the written terms of the plan in enforcing ERISA rights and obligations is of paramount importance. Heimeshoff v. Hartford Life & Accident Ins. Co., 134 S.Ct. 604, 612 (2013) (citing various other Supreme Court decisions holding that the written terms of the plan are the "linchpin" of ERISA's statutory scheme); see also Egelhoff v. Egelhoff, 532 U.S. 141, 150 (2001) (noting "ERISA's requirements that plans be administered, and benefits be paid, in accordance with plan documents."). The Third Circuit, in an ERISA breach of fiduciary duty case, stressed that "the statute dictates that it is the documents on file with the Plan, and not outside private agreements between beneficiaries and participants, that determine the rights of the parties." McGowan v. NJR Svc. Corp., 423 F.3d 241, 246 (3d Cir. 2005), abrogated on other grounds by Kennedy v. Plan Adm'r for DuPont Sav. and Inv. Plan, 555 U.S. 285 (2009); see also 29 U.S.C. § 1104(a)(1)(D) (providing that ERISA plan administrators must discharge their duties "in accordance with the documents and instruments governing the plan"). Likewise, it has stressed that "Section 502(a)(1)(B) of ERISA allows a participant to bring a claim to recover benefits due to him under the terms of the plan." Miller v. Am. Airlines, Inc., 632 F.3d 837, 845 (3d Cir. 2011) (emphasis added) (citing 29 U.S.C. § 1132(a)(1)(B)).

The ERISA claims in this case revolve around the plan term which would trigger the use of Ingenix data: UCR. In Franco I, the Court held that Subscriber Plaintiffs cannot meet the predominance standard of Rule 23(b)(3) unless they demonstrate that the same or substantially similar UCR language applies to the entire proposed class. On this renewed motion for class certification, Subscriber Plaintiffs have partially met this requirement. They have presented evidence that during the Class Period, Cigna's plans generally used one of two definitions of UCR. To adopt Subscriber Plaintiffs' nomenclature for distinguishing between the two, the Court will refer to one as the "Reasonable and Customary" or "R&C" definition and the other as the "Maximum Reimbursable Charge" or "MRC1" definition.

The "R&C" definition appears in Named Plaintiff Franco's plan for the years 2001 to 2005. The language in her plan provides as follows:

A charge will be considered Reasonable and Customary if:

(Pl. Ex. 88.)2 Subscriber Plaintiffs point to almost identical language in a sample "John Doe" policy drafted by Cigna in 1981 and approved by the insurance departments of over 20 states. (Pl. Ex. 89.) They have also presented an internal memo issued by Cigna's Home Office Claim Administration to Field Claim Offices, apparently intended to clarify how the field offices should communicate to policyholders and insureds how Cigna calculates allowed amounts on surgical procedure claims. The memo, dated February 2, 1987, explains the meaning of "Reasonable and Customary," which it states appears in Cigna's "standard" group plans, as follows:

(Pl. Ex. 33.) The Home Office memo also states that its explanation of Cigna's R&C methodology "applies to approximately 97% of the accounts on Connecticut General's Book of Business." (Id.)

The MRC1 definition appears in the Cigna plans of named Plaintiffs Chazen and Nelson. Their plans provide as follows:

Maximum Reimbursable Charge is the

(Pl. Ex. 76, 78.) Subscriber Plaintiffs provide evidence that Cigna began to use this language in or around 2005, apparently as required by the settlement of litigation brought against Cigna by healthcare providers concerning Cigna's managed care practices. They also provide evidence that the MRC1 definition was submitted for approval to the insurance commissioners of jurisdictions nationwide.

Subscriber Plaintiffs try to downplay the differences between R&C and MRC1, describing the switch as a mere "re-naming" of the definition. They point to representations made by Cigna that the UCR language was revised simply to redress providers' concerns that the R&C terminology suggested that amounts billed by providers were not reasonable. They also point to representations made by Cigna that, despite the terminology change, the formula for determining UCR remained the same. While the Court has no reason to doubt these explanations, they do not render the definitions identical. Indeed, a comparison of the provisions quoted above from the plans of the named Plaintiffs demonstrates that the R&C and MRC1 standards used significantly different language. For example, the Franco plan (R&C) caps reimbursement based on the "normal charge made by most providers of such service" whereas the Chazen and Nelson plans (MRC1) cap it based on the "policyholder-selected percentile of all charges made by providers of such service." The terms used by a plan to articulate UCR form a core component of liability in this action, and differences between the R&C standard and the MRC1 standard cannot be overlooked.3

Even so, the Court accepts that Subscriber Plaintiffs have sufficiently demonstrated that, during the Class Period, the UCR definitions respectively contained in the Cigna plans of Franco, Chazen and Nelson were widely used in Cigna plans generally. The Court acknowledges Cigna's representation that the R&C and MRC1 definitions which were submitted to state insurance departments were for Cigna's fully-insured plans only and that Cigna-administered self-insured plans were free to choose their own definition of UCR. In Cigna's view, this weakens Subscriber Plaintiffs' attempt to show that a standard UCR definition was in effect. However, the possibility that multiple UCR definitions would overwhelm the action with individual questions is significantly reduced by the incorporation of the UCR terms into the ERISA class definition. It limits membership to those plan subscribers whose plans "promised to pay reasonable and customary amounts defined as the charge of `most providers' in the `same geographic area' or promised a `maximum reimbursable charge' defined as a `percentile of all providers' in the `same geographic area.'" (Supra at 4.) Whether fully-insured or self-insured, a plan must contain this language for its subscriber to be included in the class. The Court can certainly appreciate that, overall, Cigna may have offered a variety of plans and that fullyinsured plans may have, as Cigna points out, covered as few as ten percent of plan participants during the Class Period. Cigna's ERISA Plans conceivably may have linked the ONET benefit to Medicare rates, set the reimbursement according to a schedule of services or even provided no ONET benefit at all. This proposed ERISA Class, however, covers solely subscribers unified by the UCR terms included in the class definition.4

Cigna does, however, raise a valid argument in pointing out that even where plans use either the R&C or MRC1 definitions, the plans vary significantly with regard to the methodology that Cigna may, or in some cases must employ to select the appropriate UCR. The variations impact the predominance inquiry in light of the standard by which Cigna's conduct must be reviewed to determine whether it violated ERISA. Any claim of ERISA liability must be proven at trial according to the abuse of discretion standard. The parties have acknowledged that Cigna plans in effect during the Class Period granted Cigna discretionary authority to apply and interpret plan terms. It is well-established ERISA law that where such discretion is granted, the administrator's decision will not be disturbed unless it is "without reason, unsupported by substantial evidence or erroneous as a matter of law." Miller v. Am. Airlines, Inc., 632 F.3d 837, 845 (3d Cir. 2011).5

According to the plans, arriving at the allowed amount is not a purely mechanical operation of determining UCR and comparing it to the provider's charge. As Cigna indicates, many plans, including the plans of the named Plaintiffs, contain a clause which authorizes Cigna to take the nature and severity of the injury or sickness into consideration in determining whether a non-participating provider's charge for the underlying service is reasonable and customary. This criterion, by its nature, calls for investigation into the individual characteristics of the medical issue underlying the claim. Where the plans contain a nature and severity clause, the ERISA claims could not be adjudicated on a classwide basis under any circumstances. The unique and claim-specific assessment Cigna would have to conduct when applying this clause is incompatible with a one-size-fits-all evaluation of the reasonableness of Cigna's decisions according to plan terms. In contrast, some plans do not contain the clause at all. Where the clause is omitted, the analysis of Cigna's benefit determination will necessarily be different than the analysis where the clause is in effect.

Subscriber Plaintiffs suggest that this difference might be addressed by the creation of a subclass to distinguish plans containing the clause (such as Franco's 2001-2004 plans) from those without it (such as Franco's 2005 plan). This suggestion is unavailing. Where the nature and severity of the injury or illness underlying the claim may be considered, the review of a claim decision is inherently sui generis. This individuating factor, together with additional impediments to predominance, renders subclasses an ultimately inadequate way to deal with variations on the R&C and MRC1 provisions across plans.

Alternatively, Subscriber Plaintiffs urge the Court to disregard the nature and severity clause as inconsequential because, according to them, Cigna applied this type of review in the same manner to all appealed claims, whether the governing plan included the clause or not. Subscriber Plaintiffs maintain that even when presented with documentation of increased severity of the illness or injury underlying a claim, Cigna would limit its additional reimbursement to 25% of the initial Ingenix-based UCR. They note that Cigna's own Rule 30(b)(6) witness for claims appeals admitted that Cigna employed this practice of capping additional reimbursement. Thus, they appear to raise two different arguments why the clause does not destroy predominance: one, given Cigna's universal application of the nature and severity adjustments, the clause is essentially included in all plans, whether implicitly or explicitly; and two, despite any appearance that the clause's language would call for an individualized probe into the facts of each claim, Cigna simply adjusted the reimbursement amount by increasing the Ingenix figure, and never more than by 25%. For various reasons, neither argument is availing.

First, the arguments run counter to the settled jurisprudence that ERISA rights and responsibilities are dictated by the plan language, as discussed by the Court above. The view that liability on an ERISA claim to recover unpaid benefits can be established based on evaluating a common practice, without regard to the precise plan language, was recently rejected by Chief Judge Simandle in denying a motion for class certification. See Lipstein v. UnitedHealth Group, 296 F.R.D. 279 (D.N.J. 2013). In Lipstein, plaintiffs, Medicare enrollees who did not submit a claim to Medicare or whose providers did not accept Medicare, sought unpaid benefits from their private health plan, claiming that the administrator improperly determined the amount of their secondary insurance coverage by using an invalid method to estimate what the Medicare payment would have been. Id. at 282-83. The plaintiffs argued that that the ERISA claim could be resolved in common, as to all class members, because the plan administrator employed the same method of estimating the Medicare payment on all claims, regardless of the language used in different plans. Id. at 289. Judge Simandle was not persuaded, finding that the plaintiffs' argument failed to consider that liability could attach only if the method was improper according to plan terms. Id. at 289-90. He reasoned:

Id. at 290 (emphasis in original). Likewise, in this case, Subscriber Plaintiffs cannot establish predominance as to the ERISA claims without engaging the particular plan language. To echo Judge Simandle's analysis, the question is not whether Cigna's ONET claims decisions were improper; rather, it is whether they were improper according to the terms of each plan giving rise to a claim.

Second, even assuming that Cigna had a practice of capping increasing benefits on appeal based on an adjusted Ingenix figure, it does not follow from this assertion that Cigna uniformly did so each time a subscriber sought additional reimbursement based on the severity of his illness. For example, the record shows that while Franco obtained additional reimbursement on the appeal of her 2003 ONET claim due to the time and complexity of the procedure she underwent, Chazen received no additional reimbursement on his claim because, according to Cigna, he had failed to submit supporting information or documentation with his appeal.

Third, at the risk of sounding repetitive, the Court observes that the argument that uniform application of a severity-based review renders the nature and severity clause irrelevant to a predominance inquiry loses sight of the fact that the ultimate determination on the ERISA claims requires an evaluation of whether Cigna paid ONET benefits in such a way that abused its discretion under the plan or breached its fiduciary duties to subscribers. In any given situation, Cigna's decision to award reimbursement in the amount of the Ingenix UCR plus 25% may have afforded the subscriber the benefit to which the plan entitled her. In other words, after the adjustment, the ultimate allowed amount may have equaled the particular provider's normal charge for the service or amount to a "correct" UCR. Subscriber Plaintiffs propose no way to approach such important liability issues without looking at the individual claim.

Another significant difference among plans concerns the manner in which the plan documents state, if at all, the basis or source of UCR. The plans in the record combine the R&C or MRC1 language with varying statements about how the plan will determine the figure constituting that UCR, or charge of "most" or "all" providers in a geographic area for the type of service on which an ONET claim is based. Some plans provide that the charge is "as determined by CG [Cigna]" (Franco plan), others state that it is "as compiled in a database selected by CG" (Def. Ex. 8), and yet others make the explicit declaration that Cigna uses Ingenix to determine the allowed amount. (Chazen plan, Nelson plan, Def. Exs. 5, 9, 10.) Some plans specify that UCR will be determined according to a percentile of Ingenix charge data. As Subscriber Plaintiffs also recognize, and as this Court has observed in earlier opinions, some plans were subject to a New Jersey state insurance regulation which required that such prevailing fee information be obtained from the Ingenix database. While the Court is not holding that express permission or direction to use Ingenix, be it in the plan documents or in a binding regulation, would necessarily immunize Cigna from potential liability under ERISA, such language must nevertheless be considered as a relevant factor in determining whether or not Cigna's conduct constituted an abuse of discretion or a breach of its fiduciary duty. Cf. Moench v. Robertson, 62 F.3d 553, 571 (3d Cir. 1995), cert. denied, 516 U.S. 1115 (1996) (holding, in the context of an ERISA breach of fiduciary claim for imprudent investment, that because ERISA-governed employee stock ownership plans might vary, with some requiring investment in employer securities, others merely permitting it and still others setting an intermediate level of instruction, the standard of review of the fiduciary's conduct must be tailored to the applicable obligation). By analogy to the Third Circuit's decision in Moench, it might very well be the case that where a governing regulation required that Ingenix be used, an ONET determination would be subject to less scrutiny than a situation in which the governing plan did not specify the source of data to be used. In any event, Subscriber Plaintiffs do not address how the Court could manage liability questions at trial in light of these plan variations as to the data the administrator may or must consider to determine UCR.

Subscriber Plaintiffs acknowledge that some but not all plans specify that Ingenix will be used, but argue that this poses no obstacle to class certification because the Court has already held that Cigna cannot invoke such language to foreclose Subscriber Plaintiffs' Ingenix-based ERISA claims. On the motion to dismiss the Complaint pursuant to Rule 12(b)(6), Cigna had argued that named Plaintiff Chazen had failed to state a claim for ERISA violations because Cigna was simply fulfilling its plan obligation to use Ingenix. Indeed, the Court rejected Cigna's argument. Subscriber Plaintiffs also rely on the Court's reasoning in McDonough v. Horizon, a similar UCR matter in which, like Chazen's plan, the plan of the named plaintiff was subject to a New Jersey regulation which required that the UCR on which an ONET benefit payment was based be determined according to the Ingenix database. See McDonough v. Horizon Blue Cross Blue Shield of New Jersey, Inc., No. 09-571, 2011 WL 4455994, at *3 (D.N.J. Sept. 23, 2011). The issue before the Court on this motion is entirely different than the issue analyzed in the motion to dismiss in this case and in McDonough. The Court is not concerned with the question of whether the ERISA claims are cognizable. That question has been decided. The Court's conclusion that the ERISA claims were not rendered futile by plan or regulatory language authorizing Ingenix is a far cry from holding that such language will not factor into a merits determination and thus need not factor into an evaluation of whether common questions of law and fact predominate. The possibility that Cigna may be liable when a plan is silent as to the method for determining UCR but not liable for any ERISA violation when a plan permits, or a governing regulation requires the use of Ingenix is especially pronounced in light of the deferential standard of review which applies to Cigna's claims decisions.

In the end, the uniformity Subscriber Plaintiffs believe exists by virtue of two broad UCR definitions is revealed to be illusory. The UCR language is accompanied by significant qualifiers, presenting various combinations of R&C or MRC1 with (or without) the nature and severity clause, a provision granting the plan administrator permission to choose a third party database for UCR, a specific reference to Ingenix as the designated database, and/or direction to calculate UCR at a specific percentile of the database schedules. This creates a fractured landscape for the ERISA claims of the putative class, and under the stringent requirements of Rule 23(b)(3), an inappropriate setting for class action litigation.

2. Establishing Injury on a Classwide Basis

In addition to the obstacles presented by varying plan language, establishing that issues common to the ERISA Class predominate over individual issues is hindered by a lack of evidence that injury to all class members may be proven in one stroke. Subscriber Plaintiffs have approached the predominance analysis as if a successful demonstration that Ingenix was flawed will be game, set and match for the ERISA claims of the entire class. Not so. Establishing that Ingenix is flawed is a necessary but insufficient component of proving liability in this case. The theory of Subscriber Plaintiffs' ERISA claims requires that they demonstrate, at a minimum, that Cigna's use of Ingenix constituted an abuse of its discretion, a multifactorial inquiry that must take into consideration the particular plan's language and the available alternative methods for determining UCR. Then, to establish that a subscriber warrants relief under ERISA § 502(a), they must demonstrate that Cigna's conduct in violation of the plan injured the subscriber. In other words, to prove liability, Subscriber Plaintiffs will have to demonstrate that using Ingenix to determine the allowed amount on an ONET claim was arbitrary and capricious and that it resulted in an incorrect award of benefits under the plan.

Recall that under either the R&C or the MRC1 formulation, the allowed amount on an ONET claim is the lesser of the provider's normal charge for the service or the UCR for the service. In the Complaint, Subscriber Plaintiffs had alleged that a multitude of flaws in the Ingenix database resulted in a pool of charge data that was artificially depressed. Instead of containing accurate figures reflecting amounts that providers in any given geographic area really charge patients, Subscriber Plaintiffs contended, the database was manipulated so that, by design, the UCR schedules it generated were skewed downward. According to the allegations of the Complaint, the injury consisted of an underpayment of benefits, or to put it slightly differently, reimbursement for ONET services that was inadequate because Ingenix yielded UCRs that were lower than the true prevailing fees for services. As Cigna has forcefully argued, "Plaintiffs have long since abandoned their theory that Ingenix suffers from a downward bias." (Opp'n at 48.) Subscriber Plaintiffs do not contest this assertion, and the Court therefore deems their abandonment of the downward bias theory conceded.6 This concession confirms that Subscriber Plaintiffs will not, indeed cannot, attempt to establish classwide liability based on any systematic downward bias in Ingenix UCR schedules. No other method has been proffered for establishing that, by virtue of using Ingenix, Cigna consistently underpaid ONET claims.7

According to the Complaint, the common contention holding the putative class together as a cohesive group was that the downward bias of Ingenix data had caused all those Cigna subscribers who were subjected to Ingenix-based ONET reimbursements to receive a lesser benefit amount than they were entitled to under the plan. As pled, the named Plaintiffs and the Cigna subscribers they purported to represent had been injured because Ingenix supplied fee schedules that were systematically too low to constitute a "true UCR." The pleading stage of this action has long passed, and Rule 23 demands more than mere allegations. At the class certification stage, the Court must assess the available evidence — not, of course, to consider the strength of Subscriber Plaintiffs' claims but to determine whether those claims "can be productively litigated at once." Dukes, 131 S.Ct. at 2551. Had Subscriber Plaintiffs proffered evidence that Ingenix data was consistently depressed or biased downward, they could have conceivably demonstrated that they were capable of proving injury as to the entire class "in one stroke." Id. The renewed Rule 23 motion, however, loses sight of the essential liability question that must be litigated. It is not whether Ingenix was flawed. The mere fact that a subscriber's claim had been determined using the Ingenix database, and the resulting allowed amount was less than the provider's billed charge, would not establish even a prima facie ERISA claim. Assuming, arguendo, that Subscriber Plaintiffs could establish, on the merits, that Ingenix data was flawed, this showing, in and of itself, would be patently insufficient to demonstrate underpayment of benefits, that is, injury to the subscriber, across the class membership. To establish liability as to any particular class member, Plaintiffs must prove that the subscriber's ONET claim was determined in a manner which deprived him of benefits to which he was entitled. The Court must ask, in evaluating predominance under Rule 23(b)(3): have Subscriber Plaintiffs demonstrated that this liability question can be answered in such a way that the proposed ERISA Class will prevail or fail in unison? The Court concludes that they have not.

C. Damages

Another hurdle to class certification continues to be the failure by Subscriber Plaintiffs to demonstrate that damages can be tried without relying on individualized proof as to each class member and each ONET claim allegedly determined in violation of ERISA. The ERISA Class cannot be certified under Rule 23(b)(3) unless Subscriber Plaintiffs establish, by a preponderance of the evidence, that the injury suffered by class members is measurable on a classwide basis using common proof. Comcast Corp. v. Behrend, 133 S.Ct. 1426, 1432-33 (2013). In Comcast, decided after this Court denied Subscriber Plaintiffs' first motion for class certification, the Supreme Court stressed that the predominance inquiry regarding the matter of damages, like other questions that must be addressed in a Rule 23 analysis, cannot be disconnected from the substantive legal matters of an action. Id. The Comcast Court held that to obtain certification under Rule 23(b)(3), a plaintiff must present a method of proving damages that is not only applicable classwide but also consistent with its liability case. Id. at 1433. It reasoned that to accept a damages methodology in the abstract, without reference to the merits of the claim, "would reduce Rule 23(b)(3)'s predominance requirement to a nullity." Id.

Notwithstanding the holding of Comcast, Subscriber Plaintiffs once again propose the billed charge model of damages they had unsuccessfully proffered in their first motion for class certification.8 The Court had rejected the model because it does not adhere to plan language. The model presupposes that a plan subscriber would be entitled to an ONET benefit based on the non-participating provider's billed charge. The ONET provisions in the named Plaintiffs' plans, quoted above, define the allowed amount on a claim by reference to the provider's "normal charge." This Court concluded that, contrary to Subscriber Plaintiffs' assertion that measuring damages would be a straightforward process of calculating the difference between the billed charge and the Ingenix-based ONET benefit, application of plan language to determine the provider's normal charge would result in claim-by-claim individualized inquiries. Such inquiries would be inconsistent with Rule 23(b)(3)'s requirement that questions of law or fact common to class members must predominate over questions affecting only individual members.

Subscriber Plaintiffs take the position that the Court's analysis reflects a mistaken interpretation of the ONET provision. They underscore that, according to Cigna plans, "[t]he term `charges' means the actual billed charges except when the provider has contracted directed or indirectly with CG for a different amount." (Mot. at 24, quoting from the plans of each of the named Plaintiffs). This assertion is correct. However, the plan language the Court is called upon to consider provides that the ONET allowed amount cannot exceed the healthcare provider's "normal charge." Nothing in the plan documents suggests that the word "normal" should be read out of the ONET provision.

In short, there continues to be a disconnect between plan language and the method proposed by Subscriber Plaintiffs to determine the class members' damages in a cohesive manner. The method cannot, therefore, survive the rigorous analysis that Comcast held must be applied to putative class actions seeking certification pursuant to Rule 23(b)(3) to ensure that the plaintiff has demonstrated not only that damages are susceptible of measurement across the entire class but also that the proffered model measures only those damages attributable to the plaintiff's theory of liability. Comcast, 133 S.Ct. at 1433.9 In an ERISA action to recover unpaid benefits and/or to seek relief for breach of a fiduciary duty, the theory of liability, and thus the awardable damages, must be grounded in the plan documents.

D. Demands of Rule 23

Recent class action jurisprudence has been clear: In Dukes, the Supreme Court emphasized that "[t]he class action is `an exception to the usual rule that litigation is conducted by and on behalf of the individual named parties only.'" Dukes, 131 S. Ct. at 2550 (quoting Califano v. Yamasaki, 442 U.S. 682, 700-01 (1979)). It has directed that a district court addressing a Rule 23 motion must conduct a rigorous analysis that considers factual and legal issues comprising the plaintiff's cause of action insofar as they affect the Rule 23 requirements. Comcast, 133 S. Ct. at 1432. Third Circuit class certification precedent reinforces this view. The Court of Appeals has held that, when conducting a Rule 23 analysis, "the court cannot be bashful. It `must resolve all factual or legal disputes relevant to class certification, even if they overlap with the merits—including disputes touching on elements of the cause of action.' Rule 23 gives no license to shy away from making factual findings that are necessary to determine whether the Rule's requirements have been met." Marcus, 687 F.3d at 591 (citation omitted) (quoting Hydrogen Peroxide, 552 F.3d at 316). Unless the movant satisfies all requirements of Rule 23(a) and one subsection of Rule 23(b), by a preponderance of the evidence, the action is not an appropriate use of the exceptional class action procedure. Dukes, 131 S.Ct. at 2551-52; Hayes v. Wal-Mart Stores, Inc., 725 F.3d 349, 354 (3d Cir. 2013); Marcus, 687 F.3d at 590. Subscriber Plaintiffs' failure to satisfy the predominance requirement of Rule 23(b)(3), pursuant to which they seek class certification, renders their ERISA claims incompatible with classwide litigation.

The stringent nature of the predominance requirement reflects the paramount Rule 23 concern with the manageability of a class action trial. See Sullivan, 667 F.3d at 302-03 (noting that "the concern for manageability is a central tenet in the certification of a litigation class"). In Sullivan, the Third Circuit emphasized this is a "`key' distinction between certification of a class for settlement versus certification for purposes of litigation." Id. at 303. It reasoned that while the Rule 23 requirements must be satisfied by any class seeking certification, be it for purposes of settlement or trial, the challenges inherent in proving claims with legal and factual variations disappear when the proposal is for certification of a settlement class. Id. at 303-04, 322 n.56. Guided by the Supreme Court's holding in Amchem and its own analysis in In re Warfarin Sodium Antitrust Litigation, 391 F.3d 516 (3d Cir. 2004), the Sullivan court held that, when no trial of the class action is anticipated, manageability issues that might have been presented by differences or inconsistencies among putative class members' claims drop out of the certification analysis. Id. at 303-04. Not so, however, when a court must predict how the varying claims will be established at trial. Id. at 303. The Third Circuit's discussion in Sullivan, distinguishing between a class postured for settlement and one intending to try its claims, makes it apparent that manageability concerns impact the predominance requirement of Rule 23(b)(3) in the litigation context. Id. at 303-04.

The Court wishes to be abundantly clear about its conclusion. It is not expressing any opinion as to the merits of the ERISA claims, and it is certainly not concluding that putative class members have sustained no compensable harm. However, the analysis that must be performed to determine if a claim for ONET benefits was properly decided according to plan language cannot be performed en masse. Indeed, even among the three named Plaintiffs, the governing plan language and the factual circumstances of the underlying ONET claims vary widely.10 Subscriber Plaintiffs give no indication that the deficiencies in meeting the predominance standard could be cured by dividing the ERISA Class into appropriate subclasses pursuant to Rule 23(c)(5). The facts present multiple permutations of essential aspects of the ERISA claims: available options for determining UCR as required by the plan; plan and/or regulatory requirements concerning Ingenix; adjustments made to the initial claim determination in the appeals process; whether the Ingenix UCR was less than the provider's normal charge for the underlying service; and review of Cigna's decision according the arbitrary and capricious standard. Addressing these matters may be a feasible endeavor in individual trials. For the reasons discussed, they are not suited for classwide resolution.

III. RICO CLASS

The renewed attempt to certify a RICO Class requires very little discussion. The mail and wire fraud on which the RICO claim in this case is predicated turn on the alleged underpayment of benefits under the applicable Cigna plan; that is, Cigna's alleged failure to perform as promised in the plan "would be evidenced by . . . [Cigna's] denial of benefits due under the insurance arrangement." Maio v. Aetna, 221 F.3d 472, 490 (3d Cir. 2000). Moreover, an essential element of a RICO claim is "injury to the business or property" of the plaintiff. Ins. Brokerage Antitrust Litig., 579 F.3d at 269. As Subscriber Plaintiffs themselves acknowledge in their briefs, there can be no RICO injury unless Cigna's predicate acts resulted in reduced ONET benefits. In the Court's view, this underpayment — reimbursement below the amount the Cigna plan represented it would pay — is required at a minimum to establish RICO injury. The Court has previously held that redressable injury sufficient to confer standing under RICO would also require proof that the subscriber in fact overpaid their ONET provider for the service, that is, incurred a tangible, out-of-pocket expense, for example, in the form of a balance bill from the provider. Subscriber Plaintiffs disagree with this holding, arguing that the concrete loss consists of subscribers' payment of premiums for inferior health insurance. Apart from the futility of rearguing an issue based solely on their disagreement with the Court's prior decision, Subscriber Plaintiffs' argument regarding RICO injury makes no difference to the certification question. The contention that overpaid premiums constitute the injury turns on the alleged inferiority of the Cigna plans, which turns on the alleged underpayment of ONET benefits per the terms of the plans. For the reasons set forth at length in both this Opinion and Franco I, Subscriber Plaintiffs have not demonstrated that either the occurrence of fraud or existence of an injury redressable by RICO could be proven as to the entire class based on common proof. In other words, they have not shown that common issues would predominate over individual matters on the RICO claims. Therefore, the revised RICO Class cannot be certified under Rule 23(b)(3).11

IV. ISSUE CERTIFICATION

Subscriber Plaintiffs request, as an alternative to certification of the ERISA and RICO claims, that the Court certify specific core liability issues respecting those claims under Rule 23(c)(4). They identify three ERISA issues they believe can be tried in common: (1) whether the Ingenix database was so flawed as to be incapable of supporting a reliable reimbursement methodology; (2) whether Cigna's reliance on Ingenix data was a breach of its fiduciary duty for purposes of the ERISA § 502(a)(3) claim; and (3) whether Cigna's denial of benefits under ERISA § 502(a)(1)(B) was an interpretation that was not reasonably consistent with unambiguous plan language. On the RICO claims, Subscriber Plaintiffs assert that the issues of whether Cigna and Ingenix formed an enterprise and whether Cigna engaged in a pattern of mail and wire fraud in the conduct of that enterprise may be certified under Rule 23(c)(4).

Rule 23(c)(4) states that "[w]hen appropriate, an action may be brought or maintained as a class action with respect to particular issues." Issue certification is a matter left to the Court's discretion, but the decision to certify a particular issue, like any other certification decision under Rule 23, "must be supported by rigorous analysis." Gates v. Rohm and Haas Co., 655 F.3d 255, 272 (3d Cir. 2011). In Gates, the Third Circuit set forth various factors a court should consider when deciding whether or not to certify an issue class:

Id. at 273.

There is no doubt that the issue of whether the Ingenix database itself was flawed is a complex issue, and moreover, one that impacts the ERISA claims of the entire class. Certification of this issue, however, would not materially advance the liability issues at the core of Plaintiff's § 502(a)(1)(B) and § 502(a)(3) claims. As the Court discussed at length above, even assuming Subscriber Plaintiffs succeed in establishing that Ingenix was defective, resolving the question of Cigna's liability to class members entails individualized, plan-specific and claimspecific inquiries to determine whether, by reference to various other language affecting the ONET benefit Cigna was obligated to pay, its determination breached fiduciary duties (as defined by the plan) or constituted an arbitrary and capricious application of plan terms. Class adjudication of the validity of Ingenix would not only leave critical remaining liability issues unanswered but also fail to achieve any efficiency in the resolution of class members' claims. A court would be required to conduct mini-trials to establish the elements of the ERISA claims for unpaid benefits and breach of fiduciary duty in light of the nuances of each plan's instructions to the administrator.

The other ERISA issues for which Subscriber Plaintiffs seek certification are even less suited to common, classwide resolution. It is not only the issue of quantifying the appropriate damage award to each class member which presents the need to conduct an individualized analysis of plan language and underlying facts. Rather, the very occurrence of an ERISA violation will vary from class member to class member, and for each class member, from one ONET claim to another. Moreover, like the damages calculation, determining whether Cigna has underpaid an ONET claim and/or breached its fiduciary duty to discharge plan obligations requires consideration of the plan language, entitling plan subscribers to an ONET allowed amount that does not exceed the provider's normal charge. Where "liability is inseverable from other issues that would be left for follow-up proceedings," the Court may decline to certify a liability-only issue class. Gates, 655 F.3d at 273.

This Court agrees with the conclusion reached by Judge Debevoise in Clark v. Prudential Insurance Company, a putative class action at a similar procedural juncture to this action. See Clark v. Prudential Ins. Co. of Am., 940 F.Supp.2d 186 (D.N.J. 2013). The plaintiffs in Clark, former policyholders of the defendant health insurance company, sought relief for the defendant's alleged misrepresentation of the reason it increased premiums on the plaintiffs' policies. Id. at 188. They moved for reconsideration of the court's order denying their motion for class certification, and, in the alternative, requested that the Court certify only the liability issues and handle individual damages issues in a separate trial. Id. at 189. The court denied the request for bifurcation after engaging in a Gates analysis. Id. at 192-93. Judge Debevoise reasoned that certification of liability issues is not appropriate where "the individualized issues which arise in the calculation of damages and damages in fact are so inextricably linked that bifurcation would be judicially inefficient." Id. at 194.

Certifying only certain RICO issues is similarly inefficient. Even if the question of whether Cigna and Ingenix formed an enterprise within the meaning of the statute could be resolved as to all members of the putative RICO Class, substantial liability questions would remain unanswered without a member by member and claim by claim analysis of the alleged failure to perform as promised by the plan.

The alternative request for class treatment of particular issues, pursuant to Rule 23(c)(4), will accordingly be denied.

V. CONCLUSION

This motion fails to satisfy the requirements for certification of a Rule 23(b)(3) class as to either the proposed ERISA Class or the proposed RICO Class. For the reasons discussed, Subscriber Plaintiffs have not demonstrated that they could litigate their claims through evidence that is common to the class. Moreover, issue certification, as permitted by Rule 23(c)(4), would not advance the resolution of the claims, and thus this action does not warrant exercise of the Court's discretion to certify certain issues.

An Order denying Subscriber Plaintiffs' renewed motion for class certification in its entirety will be filed.