King v. Dalkon Shield Trust (In re A.H. Robins Co.)

219 B.R. 161, 1998 Bankr. LEXIS 260

• Court: District Court, E.D. Virginia
• Decided: March 4, 1998
• Docket: Bankruptcy No. 85-01307-R
• Status: Published

Opinion

MEMORANDUM

MERHIGE, District Judge.

This contested matter proceeding began with the Motion to Interpret Plan filed by Daikon Shield Claimant Joyce Hurst (Docket No. 30064). In that Motion, Ms. Hurst asked this Court to rule that the presumption of causation found in § G.2 of the Claims Resolution Facility (the “CRF”) by the Court of Appeals for the Fourth Circuit in In re A.H. Robins Co. (Reichel v. Dalkon Shield Claimants Trust), 109 F.3d 965 (4th Cir.1997), and applied by the Court of Appeals in that case to the Alternative Dispute Resolution (“ADR”) program established by the Daikon Shield Claimants Trust ,(the “Trust”) under § E.4 of the CRF, also applies to the arbitration and trial of Daikon Shield Claims against the Trust under § E.5(a) and § E.5(b) of the CRF. For the reasons that follow, the Court concludes it does not.

I. BACKGROUND

A. The Reichel Decision

Section G of the CRF set out fifteen “guidelines” that “shall govern the Claims Resolution Facility and Claims Resolution Process.” (CRF-4). Among those fifteen, § G.2 provides:

2. Scheduled Compensable Claims. In determining whether an injury could have been caused by the Daikon Shield and, therefore, could be eligible for compensation, the Trust shall presume that the injuries listed in Exhibit A are eligible for compensation. The Trust shall consider on a case-by-case basis whether any injury not in Exhibit A is eligible for compensation.

(CRF-5). Exhibit A to the CRF, which was negotiated and agreed upon by the parties involved in the Robins bankruptcy along with the Plan and the terms of the GRF itself, listed the types of user and nonuser claims that could be asserted for injuries allegedly arising from use of the Daikon Shield. Within those claimant groups, Exhibit A identified the categories of injuries typically alleged by plaintiffs in prepetition Daikon Shield suits against Robins. For example, the user claims were broken down into the three main injury claims historically associated with Dai-kon Shield use: pelvic inflammatory disease; perforation or embedment of the Daikon Shield; and pregnancy occurring with a Dai-kon Shield in place. (CRF-8 and CRF-9).

In the proceeding before this Court that led to the Reichel appeal, In re A.H. Robins Co. (Reichel v. Dalkon Shield Claimants Trust), 197 B.R. 537 (E.D.Va.1994) six ADR claimants argued that CRF § G.2 created the presumption that the Daikon Shield caused the injuries listed in CRF Exhibit A; that this presumption of causation applied to the ADR program; and that, as a result, First Amended ADR Rule XII.G(2) and Second Amended ADR Rule 12.H, which specified that ADR claimants bore the burden of proving causation, were void. This Court interpreted the language of § G.2 as not creating a presumption of causation, but instead as applying only to the Trust during its evaluation of claims to require the Trust to presume that claimants who could show Exhibit A injuries were eligible for compensation during the settlement steps of the claims process.

The Court of Appeals interpreted the language of § G.2 differently. In re A.H. Robins Co. (Reichel v. Dalkon Shield Claimants Trust), 109 F.3d 965 (4th Cir.1997). It held that § G.2 “establishes a presumption of causation upon proof of use of the Daikon Shield and an injury appearing in Exhibit A.” 109 F.3d at 968. The court further concludéd, however, that the ADR Rules placing on claimants the burden of proving causation were valid, because the § G.2 presumption is a rebuttable one that vanishes from an ADR case immediately upon its refutation.

The court began its analysis by noting that the ADR Rules provide that the Federal Rules of Evidence do not apply to ADR cases. For that reason, the Court of Appeals turned to Rule 704 of the American Law Institute Model Code of Evidence (1942) to determine the effect in ADR of the presumption of causation it had found in the words of § G.2. Under Model Rule 704(1) the establishment of the basic facts of a presumption requires assuming the existence of the presumed fact unless, and-until evidence is introduced “which would support a finding of its non-existence.” Further, according to Model Rule 704(2), once the presumption is rebutted by introduction of such refutation evidence, “the existence or non-existence of the presumed fact is to be determined exactly as if no presumption had ever been applicable to the action.” The Court of Appeals described this treatment of presumptions as:

A presumption, of course, aids the party who seeks proof the presumed fact. But the presumed fact may be rebutted and upon introduction of evidence which would support a finding of the non-existence of the presumed fact, the case is left in the same situation as if no presumption had ever been applicable.

109 F.3d at 968.

Applying the vernacular of Model Rule 704 to a Daikon Shield ease, the two requisite basic facts -would be use of a Daikon Shield and having experienced an injury listed in CRF Exhibit A. The presumed fact would be that the Exhibit A injury shown was caused by the Daikon Shield. The Court of Appeals observed that a claimant’s introduction of sufficient evidence to prove these two basic facts “establishes the presumption that the woman’s injury was caused by the use of the Daikon Shield.” 109 F.3d at 968. The court continued:

If nothing else appears in the record, the finder of fact must find that the injury was caused by use of the Daikon Shield. But the presumption may be rebutted, and if evidence is introduced which would support a finding of a non-existence of cause of the injury by use of the Daikon Shield, then the existence or non-existence of causation is to be determined exactly as if no presumption had ever been applicable in this case.

109 F.3d at 968-69. With the presumption subject to rebuttal, “the ultimate burden of persuasion rests on the claimant.” 109 F.3d at 969. Thus, the Court of Appeals found the ADR Rules placing the burden of proof on the ADR claimant to be valid, because the rebuttable nature of the presumption left the burden of persuasion with .the claimant at all times. Id. . ,, •

B. The Present Motion

A maelstrom of debate and controversy emanated from the Court of Appeals’ Reichel ruling promptly after its announcement, as Daikon Shield claimants and their lawyers proclaimed its significance in the ADR program and then lobbied to extend if beyond ADR to the arbitration and trial of Daikon Shield Claims under § E.5(a) and § E.5(b) of -the CRF. The Trust reported in its Response to Ms. Hurst’s Motion that various plaintiffs have argued that Reichel creates in arbitration and trial the irrebuttable presumption that the Daikon Shield caused the injuries listed in Exhibit A; that it precludes the Trust from arguing that a plaintiffs injury stems from some other cause or from offering an expert witness to testify that the Daikon Shield did not cause an injury listed on Exhibit A; or that it imposed on the Trust the fiduciary duty simply to pay all claims rather than defend against them. The Trust resisted those efforts by staking out its position that Reichel applied only to the ADR program and not to arbitration or trial and that, even if the § G.2 presumption were extended to arbitration or trial, it could not lead to the many conclusions plaintiffs were drawing from it.

To resolve this dispute, Daikon Shield claimant Joyce Hurst filed her Motion to Interpret Plan with this Court on September 2, 1997 (Docket No. 30064). Ms. Hurst is represented by James F. Szaller of Brown & Szaller Co., L.P.A., in Cleveland, Ohio. In her Motion to Interpret Plan, which she brought “on her own behalf and on behalf of other Daikon Shield claimants similarly situated,” Ms. Hurst asked the Court to interpret the Plan and hold that § G.2 “creates a presumption of causation in favor of the claimant in litigation and arbitration.”

The Trúst filed a Response asking the Court to rule that § G.2 does not apply to arbitration and trial.' The Trust argued further that, even if it did, the presumption of causation it would create would be rebutted in á Daikon Shield case by evidence of alternative causation or evidence disputing the causal propensities of the device. Ms. Hurst’s Reply'brief addressed primarily what can or cannot be rebutted under the presumption, rather than' with the threshold question of its applicability.

The matter came on for hearing before the Court on October 8, 1997. After arguments by both sides and other parties, the Court took the matter under advisement.

Since then, Ms. Hurst resolved her Daikon Shield Claim. As a result, Mr. Szaller and the Trust filed on January 5, 1998, an Agreement substituting three other’ litigation plaintiffs represented by Mr. Szaller (Debbie King, Valeria Griffin and Sylvia Johnson) as the moving parties (Docket No. 30166). The Court approved this substitution of parties in an Order entered on January 12,1998 (Docket No. 30174).

C. The Troutt Decision

Lurking in the background of this proceeding is this Court’s decision in In re A.H. Robins Co. (Troutt v. Dalkon Shield Claimants Trust), 197 B.R. 519 (E.D.Va.1994). There, the Court announced for cases in arbitration under § E.5(a) of the CRF the same interpretation of § G.2 as it had reached for ADR in Reichel. There, too, the Court held that § G.2 did not create a presumption of causation, but instead only a presumption of eligibility for payment. The movant in that proceeding did not appeal the ruling, so the Court of Appeals was not given the occasion to interpret § G.2 in the context of arbitration cases. It falls to this Court to do so here, in light of the Court of Appeals’ Reichel decision.

II. DECISION

A. The Court’s Jurisdiction

The import of § G.2 in arbitration and trial presents a matter of interpretation of the CRF language, both on its face and in the manner that best serves the implementation of the Plan. These tasks are within the exclusive jurisdiction of this Court to interpret the Plan and facilitate its implementation, retained in § 8.05(c) and § 8.05(d) of the Plan and paragraph 45 of the July 26, 1986 Order of this Court confirming the Plan. In re A.H. Robins Co., 88 B.R. 742 (E.D.Va.1988), Docket No. 5714, affirmed, 880 F.2d 694 (4th Cir.) cert. denied, 493 U.S. 959, 110 S.Ct. 376, 107 L.Ed.2d 362 (1989); see also In re A.H. Robins Co. (Order Limiting Attorneys Fees), 182 B.R. 128 (Bankr.E.D.Va.1995), affirmed, In re A.H. Robins Co. (Bergstrom v. Dalkon Shield Claimants Trust), 86 F.3d 364 (4th Cir.) cert. denied, — U.S. -, 117 S.Ct. 483, 136 L.Ed.2d 377 (1996); In re A.H. Robins Co. (Amended Administrative Order No. 1), Case No. 85-01307-R, Docket No. 11500 (E.D.Va. nunc pro tunc June 26,1991), affirmed, 42 F.3d 870 (4th Cir.1994); In re A.H. Robins Co. (Dalkon Shield Claimants Trust v. Reiser), 972 F.2d 77 (4th Cir.1992).

No party has questioned this Court’s subject matter jurisdiction over these Plan-related disputes. A dispute arose, however, at the October 8 hearing over whether any party with the proper standing had placed in controversy the question of § G.2’s application to arbitration. The initial movant, Joyce Hurst, had chosen litigation under CRF § E.5(b) to conclude her claim. No plaintiff pursuing a claim in arbitration under § E.5(a) had joined in the Motion. Joseph S. Friedberg, a lawyer in Minneapolis, Minnesota, who represents many Daikon Shield claimants and was involved during the Robins bankruptcy case, appeared at the October 8 hearing to suggest that Ms. Hurst did not have standing to put the arbitration question into play. He also offered his view that while § G.2 might well apply to litigation, it could not be made to apply to arbitration.

The claimants “similarly situated” to Ms. Hurst and on whose behalf she said she was bringing her Motion are those who chose litigation under § .E.5(b) and have open cases pending against the Trust. The three claimants substituted in her stead as movants after she settled her Daikon Shield Claim likewise are in litigation, not arbitration. Those movants do not have the requisite standing to ask for a ruling on whether § G.2 covers arbitration.

This technicality could have led to the inefficient result that only the litigation aspect of this puzzle could be unraveled here, at the very time the Trust and this Court are attempting to conclude all arbitration and trial cases as soon as possible so that the Trust may make the final pro rata distribution and close. The arbitration issue would be left to another day and another proceeding. In its Response, however, the Trust asked the Court to rule that § G.2 applies to neither arbitration nor litigation. As the current holder of the bankruptcy estate and as the representative of all Daikon .Shield claimants, the Trust has standing to seek a ruling on both the issues. This enables the Court to act on the subject and spares it and all parties the anomalous result of postponing the adjudication of an issue that should be concluded for all parties now.¹

B. There Should Be No Presumption of Causation in the Arbitration or Trial of Daikon Shield Claims 1. The Status of This Question

The Reichel appeal involved only the ADR program. The Court of Appeals applied the presumption it found in § G.2 only to ADR cases. Its opinion addressed first whether the words in § G.2 should be interpreted to create a presumption of causation, rather than a presumption of eligibility for payment. Concluding that § G.2 did erect a presumption of causation, the Court of Appeals then turned to the effect to be given that presumption in ADR.

Whether § G.2 also operated in arbitration and trial was not briefed or argued to this Court, or apparently to the Court of Appeals, in the Reichel proceedings. Because the Court of Appeals was not asked to decide the applicability of § G.2 in arbitration or trial, no specific or express direction was given on that issue in its opinion. Indeed, Ms. Hurst’s original Motion and her Reply to the Trust’s Response rest less on the language of either § G.2 or of the Court of Appeals’ decision than on her view that the mere creation in the Robins bankruptcy of a trust fund for Daikon Shield cases mandates that the § G.2 presumption be universal, regardless of what means is chosen by a claimant under the CRF either to settle or to adjudicate her claim. Ms. Hurst and her replacements thus ask the Court to apply the presumption of causation to both arbitration and trial not so much because the provisions in the CRF-or in the Reichel opinion would require it, but because they see'the presumption as implicit in the Trust’s very reason for being.

It is clear from the Court of Appeals’ opinion that the interpretation of § G.2 adopted by this Court in Reichel and Troutt is no longer the law of this bankruptcy case. Section G.2 creates a presumption of causation, not of eligibility for payment. That presumption applies to the ADR program.

To answer the question posed by Ms. Hurst’s Motion, as it would with any debate requiring interpretation of the Plan, the Court must first address whether the § G.2 presumption of causation should, under the language of the CRF and the nature of the claims processing matrix established in the Plan and CRF, apply not only to ADR but also to arbitration and trial. Then the Court must assess whether any aspect of what would be its normal analysis of that question has been preempted or altered by the Court of Appeals’ Reichel opinion.

2. The Claims Processing Options in the CRF

The question must first be placed in its proper context. The CRF, included as Exhibit C to the Plan when it was approved by Daikon Shield claimants and confirmed by this Court in 1988, provided claimants with three choices for the payment of their Dai-kon Shield Personal Injury Claims. Under Option 1 (CRF § C), users could receive $725 and non-users $300 upon signing a form verifying Daikon Shield use and an injury listed in Exhibit A to the CRF. Those who chose Option 2 (CRF § D) could receive $125 to $5,500 from among fixed amounts for Exhibit A injuries if they submitted their own affidavit and an affidavit from a doctor or medical records showing Daikon Shield use and an injury listed on Exhibit A. These dollar amounts were not specified in the CRF; they were assigned by the Trustees of the Trust pursuant to the duty and authority bestowed upon them by these sections for that purpose.

In Option 3, described in CRF § E, the claimant submitted a detailed claim form and all her medical records. Sections E.2 and E.3 of the CRF required the Trust to evaluate that information to derive an offer of compensation based on the historic value of claims against Robins and to extend that offer to the claimant. Claimants who rejected those Option 3 offers were allowed by CRF § E.4 to proceed to In-Depth Review to receive another offer. If a claimant chose In-Depth Review and then rejected the resulting final Option 3 offer, she could elect between arbitration under CRF § E.5(a) or trial under CRF § E.5(b) to adjudicate her claim.

3. The ADR Program

As an alternative to In-Depth Review, § G.4 also permitted the Trust to “invite” claimants to finish their claims through “any other voluntary alternative dispute resolution process as appropriate.” (CRF § E.4, at CRF-3). Exercising that discretion, in April 1992, the Trust began an ADR program with a $10,000 limit on the total damages recoverable. A claimant could enter ADR after rejecting either her initial offer or her final In-Depth Review offer. Each claimant who chose ADR signed an Election and- Agreement providing that her claim would be presented to a neutral referee in a two-and-a-half hour hearing under a special set of ADR Rules. The claimant and the Trust agreed in this Agreement that the referee’s decision on the claim would be final and binding.

The Trust describes ADR as a process designed to move quickly and with as few legal complications as possible. For example, the Trust did not assert a statute of limitations defense in ADR. It was represented at the hearings by Trust employees trained to be ADR advocates, but who were not lawyers. Though claimants could appear without lawyers as well and many did, over half of the ADR claimants, according to the Trust, were represented by counsel. The ADR Rules allowed no discovery and limited pre-hearing filings. An ADR case culminated with a two and a half-hour evidentiary hearing before a neutral referee appointed by the Private Adjudication Center at Duke University, which served as the Neutral Third Party throughout the ADR program.

When ADR began in April 1992, there was a $10,000 limit on the damages an ADR referee could award, a term the claimant and the Trust agreed upon in the ADR contract. The Trust increased the ADR maximum to $20,000 in September 1993. The Trust states in its Response that the program ended with its last hearing on September 25, 1997. By then, the Trust had concluded 5,684 ADR hearings on Timely Claims, Late Claims, and Breland Insurance Trust Claims. According to the Trust, a total of $42,980,347 has been awarded by referees to ADR claimants, with nearly seventy-seven percent of ADR claimants awarded some amount by the referees who heard them cases.

4. The CRF Was Designed Both to Permit and to Favor Settlement Over Litigation

This Court on many occasions has described the nature of the claims process envisioned by the Court and the drafters of the Plan in 1988. See In re A.H. Robins Co., Inc. (Nelson v. Dalkon Shield Claimants Trust), 216 B.R. 175 (E.D.Va.1997); In re A.H. Robins Co. (In re Administrative Order No. 2), 215 B.R. 112 (E.D.Va.1997); In re A.H. Robins Co., Case No. 85-01307-R, Docket No. 11500 (E.D.Va. nunc pro tunc June 26,1991), affirmed, 42 F.3d 870 (4th Cir.1994). By enjoining Daikon Shield-related litigation against all parties and instead channeling all claims to the Trust, the Plan sought to achieve “global peace” on all Dalkon Shield fronts. See In re A.H. Robins Co., 88 B.R. 742 (E.D.Va.1988) Docket No. 5714, affirmed, 880 F.2d 694 (4th Cir.), cert. denied, 493 U.S. 959, 110 S.Ct. 376, 107 L.Ed.2d 362 (1989). The Plan was designed to process and pay the thousands of Daikon Shield Claims directed to the Trust “in an orderly, fair manner, applying the same rules to all.” In re A.H. Robins Co., 88 B.R. at 753.

Resolution of disputes through the traditional model of lawsuits in an adversary system is both time-consuming and quite costly. Concluding Daikon Shield Claims by means short of full-scale litigation would avoid both the delay caused by such suits and their attendant substantial cost, thereby maximizing the necessarily finite funds available to the Trust to pay claims. Further, the sooner claims could be resolved, the sooner the Trust could wind up its affairs and close, resulting in savings of administrative and operational costs that likewise were to be passed through to deserving claimants.

These truisms were as obvious in foresight in 1988 when the Plan and CRF were written, approved by this Court, and overwhelmingly approved by the claimants who voted on the Plan, as they are today in hindsight. Moreover, fresh in all minds in 1988 was the massive consumption of resources and energies wrought by the 6,000 suits then pending against Robins and other defendants that propelled Robins into bankruptcy. That crushing litigation burden, which unchecked would likely have led to the company’s demise long before all potential claimants had an opportunity for any recovery, precipitated the creation of the Trust. Whether the Dai-kon Shield had in fact caused all the ills for which it was then being blamed was by then not the primary premise of the bankruptcy. The 6,000 plaintiffs who were alleging it did, with thousands more likely to follow, were the driving force behind the bankruptcy, the willingness of Robins and AHP to pay $2.2 billion to shed themselves of the burden, and the establishment of a common, limited fund to provide an equal opportunity for all claimants to process their claims.

The framers of the Plan and CRF accordingly and justifiably made efficiency and economy the hallmarks of the CRF, determined as they and the Court were to minimize the likelihood for the litigation that had preceded it. The first section of the CRF, § A described the purpose of the CRF as providing “all persons full payment of valid claims at the earliest possible time consistent with the efficient design and implementation of the claims resolution facility----” (CRF-1). This purpose was “to be achieved by ... providing an efficient economical mechanism for liquidating claims which favors settlement over arbitration and litigation, thereby reducing transaction costs....” Id.

The CRF blueprint thus provided incentives to the conclusion of claims by voluntary agreement, rather than by adversarial means. The alternatives available to claimants were charted on a sliding scale of claim value proportionate to the level of proof required to obtain payment. As the CRF moves from Option 1 through the end of Option 3, the amount of compensation available to a claimant elevates proportionate to increasingly stringent qualifications for payment. This ensured that the larger payments would be made in the claims process to claimants with the more serious injuries and the medical evidence to prove them. It also stimulated the conclusion of claims by voluntary settlement by allowing the Trust to compensate claimants without a third-party adjudication and with provisions designed to dissuade claimants from litigating their claims.

Thus, the field on which the claimants and the Trust operated in the CRF began in Option 1' tilted heavily in favor of the claimant by provisions that diminished proof requirements and stripped the Trust of virtually all defenses to the claim. Obtaining an Option 1 payment was relatively easy, but the Trustees set those payments at $725 for users and $300 for non-users. As of October 20, 1997, the Trust had paid 129,592 option 1 claims, for a total of $84,939,775.² In Option 2, larger payments of up to $5,500 were available, but no longer was a mere affidavit from the claimant enough to qualify for payment. Instead, the claimant was required to submit her own affidavit or an affidavit of a physician, as well as medical records substantiating that the claimant actually sustained the scheduled injury for which she sought payment under Option 2; Still, though, in Option 2, the Trust was denied the majority of any defenses to the claim, including the statute of limitations. By October 20, 1997, the Trust had paid 15,793 Option 2 claimants a total of $49,428,938.

Unlike Options 1 and 2, there was no maximum limit on damages payable under Option 3, where claimants were paid amounts consistent with the historic value of such claims in prepetition suits against Robins. To be eligible for such payments, claimants were required to submit more detailed information and medical records to substantiate their injury claims. Here the Trust was allowed to take into account the “quantum and quality of the evidence supporting the claim” in determining the compensation to be offered to settle the claim, but did not rely upon other defenses, such as the statute of limitations or the contributing fault of third parties, in calculating its Option 3 settlement offers. As of October 20,1997, the Trust had settled 37,628 claims under Option 3 and had paid those claimants $1,358,124,829 in original settlements.³

Claimants were not required to accept their Option 3 settlement offers from the Trust, but those who rejected them were placed on notice by the CRF that they would face greater hurdles in their effort to obtain larger recoveries from the Trust through adversarial means. The Trust envisioned the ADR program as a streamlined adversarial process where the claimants were required to prove Daikon Shield use, causation and injury, but would not face the tasks of proving that the Daikon Shield was a legally defective product or of overcoming a statute of limitations defense to the claim. Claimants in ADR did not need a lawyer and did not face lawyers representing the Trust. They presented their cases in less than half a day, rather than the several days or weeks of evidence an arbitration or trial would entail. The trade-off to those claimants, however, was a maximum on the damages payable, which was first set at $10,000 and then increased to $20,000 per claim. When they elected ADR, claimants signed an Agreement limiting their claims to the applicable maximum amount and agreeing that the referee’s ruling on their claim would be final and binding. As of October 17,1997, ADR referees had decided 5,204 claims, awarding damages to 77.2% of all ADR claimants. Those prevailing claimants recovered a total of $40,-425,287. Breland claimants could also choose ADR; 469 Breland claims were decided by ADR referees.

Claimants who wanted neither to settle nor a limit on their damages could choose arbitration. or trial to adjudicate their claims against the Trust. The CRF, as described in this Court’s Amended Administrative Order No. 1 opinion, In re A.H. Robins Co., Case No. 85-01307-R, Docket No. 11500 (E.D.Va. nunc pro tunc June 26, 1991) affirmed, 42 F.3d 870 (4th Cir.1994), warned these claimants and their lawyers that the field between them and the Trust was leveling and that they would no longer enjoy all the advantages they had enjoyed in the settlement steps of Option 3. In arbitration under CRF § E.5(a), all factual medical and causation issues were to be decided by an impartial arbitrator in an adversarial proceeding where “all available defenses may be asserted by the Trust, other than absence of product defect.” (CRF-3). Section E.5(b) of the CRF preserved for claimants the right to a jury trial, but “[a]ll claims and defenses shall be available to both sides ...” in those cases. (CRF-4). Section G. 12 of the CRF stated specifically that “product defect, including but not limited to design and warning defect,” was not to be contested by the Trust or the subject of proof in Options 1 and 2 of the CRF and in Option 3 except for trial. (CRF-6). The CRF provisions that enfeebled the Trust in its defense of claims in Options 1 and 2 and the settlement steps of option 3 no longer apply in arbitration or trial. The adversarial field between the plaintiff and the Trust in arbitration is completely level, except that the Trust cannot contest product defect. In trial, the- CRF made the field completely level. This design reflected the effort to make arbitration less attractive to claimants than resolution of their claims by settlement, but'more appealing than trial. By making trials not only a contest, but an even contest where all issues, including product defect, could be litigated, the CRF made trial the least alluring option to claimants and the Trust.

This sliding CRF scale of benefit proportionate to difficulty was not unilateral. It did not affect the motivations of the claimants alone. It was not created to coerce claimants into surrendering their claims for paltry amounts out of fear of waging an adversarial battle _ against a well-financed defendant. Rather, the CRF provisions comprising this sliding scale served two purposes. First, by defining the level of proof necessary to obtain payment and the defenses operative in the different Options, they set the. threshold claimants were required to meet to be eligible for payment. , Second, they allowed the Trust to offer and pay compensation on that proof alone, without it or a third party making findings on limitations, defect, causation propensities, or the myriad of other issues normally litigated under traditional dispute resolution modes.

The terms of the CRF served this dual role. They allowed claimants to qualify for payment at differing levels and amounts, depending upon their proof, the’ strength and severity of their claims, and the amount of effort necessary to obtain the payment, while forewarning that the difficulty of obtaining larger payments increased when moving to the next step of the-CRF. They also allowed the Trust to make assumptions and disregard defenses to permit settlement offers, before reaching any adversary process to liquidate the claim, and to afford compensation to all claimants under uniform rules. Thus, the incentives claimants felt to settle were • achieved not only by an awareness that the path to recovery would contain more numerous and greater obstacles once an offer was rejected, but also by both allowing and requiring the Trust to offer compensation at levels high enough to make settlement more attractive than continuing on in the process.

The efforts to steer claimants and the Trust toward settlement rather than arbitration or trial proved fruitful. By October 20, 1997, 212,173 Trust and Breland claimants had settled their claims by voluntary agreement, while another 5,673 claims were concluded in ADR. Only thirty-eight cases against the Trust and four Breland claims had been adjudicated in arbitration, and fifty-three eases had been completed in litigation against the Trust under § E.5(b) of the CRF. There were twenty-three arbitration cases and sixty-one trials yet to finish as of October 20, 1997. The uncontroverted evidence before this Court at its April 27, 1995 hearing on the disallowance of attorneys fees out of pro rata payments under CRF § G.14, was that the Trust’s Option 3 offers were on the average higher than the amounts paid by Robins to settle similar claims before the bankruptcy. Moreover, the availability of excess funds to make pro rata payments means that claimants who settled for more than the de minimis $725 amount will eventually be paid twice what they originally agreed to accept in settlement of their claims. Through its allocation of benefits and burdens at differing levels in the different steps of the claims process, the CRF achieved the goal of favoring settlement over arbitration and trial.

C. The CRF Does Not Expressly Apply Section G.2 to Arbitration or Trial

Section G.2 of the CRF does not mention arbitration or trial and thus, does not specify whether the section applies to either. Nothing elsewhere in the Plan or CRF states that causation is to be presumed in arbitration or trial.

Section G opens with the preamble that “[t]he following guidelines shall govern the Claims Resolution Facility and Claims Resolution Process.” That opening salutation to the fifteen § G guidelines cannot, however, mean that every one of them applies to 'all the steps of the CRF process ranging from Option 1 through a jury trial in Option 3. Their coverage must be deduced from the context in which each provision operates.

For example, § G.15 set out the procedures for converting Late Claims to Timely Claims, a subject that obviously affected the Trust and claimants only during the claims process and not in either arbitration or trial. The order of processing of claims in § G.l could only apply to the Trust’s evaluation of claims in the steps before arbitration and trial. The same can be said of the ability given the Trust in § G.9 to rely on certain reference materials developed in the Court’s claim analysis process. Section G.10 mandated certain language for the releases required when the Trust settled claims, again something that could only have been intended to apply to the settlement steps of the CRF, though there is no such express restriction in § G.10 itself.

Other § G guidelines apply to every aspect of all claims, including any that enter arbitration or trial. The method of withholding payments prescribed in § G.3 must apply to all steps in the CRF, as does the confidentiality requirement in § G.4, the allocation of costs in § G.8, the treatment of product defect in § G-ll and of causation in § G.13, and the priorities among claims in § G.14.

Section G.ll, by its own terms, applies to “settlement or trial,” without referencing arbitration. With no similar express direction in § G.2, its outlines must be marked in the manner best serving the goal of encouraging settlement by discouraging litigation. The CRF argot can be understood only in light of that primary objective.

In Daikon Shield litigation before the Robins bankruptcy, no issues were more hotly contested than product defect and causation. Experts for both sides differed on both issues. Much of the trial of a case was devoted to testimony or other evidence on the design of the IUD, the risks it allegedly presented, what injury it could cause, and how it did so. No state’s tort law imposes liability on a manufacturer for no reason; even in a strict liability jurisdiction, the plaintiff must prove that the product was unreasonably dangerous in design, manufacture, or warnings and that its unacceptable propensity or condition proximately caused the plaintiffs injury. No Daikon Shield plaintiff could recover damages in a trial without surmounting her burden of proving those two issues to a judge or jury. Then, as now, a plaintiff was unlikely to present a jury issue on either point without admissible expert testimony in support of her cause. That legal and practical reality accounted for much of the length, expense and risk of nonrecovery inherent in prepetition Daikon Shield trials.

The writers of the CRF and this Court sought in 1988 to avoid those trials and their voracious appetites for time and money. To encourage settlement, the product defect issue was not to be debated. To promote arbitration over trial, product defect could be litigated only in the latter.

Section G.13 of the CRF directed that causation was still to be a contested issue throughout the CRF, “except as waived by the Trust.” (CRF-6). Section G.2 operates in this context. It is a prime example of the duality of roles played by the CRF incentive-building provisions. It requires claimants to prove Daikon Shield use and an injury listed on Exhibit A to obtain any payment. By requiring the Trust to presume the injuries listed in CRF Exhibit A were caused by the Daikon Shield, § G.2 permitted the Trust to make offers to and then to pay persons with those injuries and Daikon Shield use without their having to prove causation to the Trust in Options 1 and 2 and the evaluation and offer steps of Option 3. As the Court of Appeals noted, without this presumption of causation, the Exhibit A injuries would not be eligible for compensation, for there “could be no compensation for an injury not caused by the Daikon Shield.” 109 F.3d at 968. Section G.2 enacted a presumption of causation both to favor and to allow the settlement of claims by agreement by enabling the Trust to compensate in the settlement steps of the CRF claimants with proof of the listed injuries, even in the absence of a finding of causation by an arbitrator, judge, or jury.

The justification for this advantage, however, ended when a claimant rejected her opportunity to settle her claim and instead sought more damages through arbitration or trial. The CRF warned those claimants that the path to payment would no longer be as facile as it had been until then. Otherwise, there would have been no incentive to accept the Trust’s offers and every incentive instead to press forward to an arbitration hearing or trial. If claimants and their lawyers had thought that in arbitration and trial they would continue to enjoy the same presumption of causation that made them eligible for payment in the settlement steps of the CRF, they would have seen no need to find and assemble causation experts or medical proof for their trials and less reason to fear what normally would be the most difficult part of a Daikon Shield plaintiffs case-in-chief: proving that the device more likely than not caused the injury for which she sought more money than the Trust had offered her. That would have greatly enervated the behavioral incentives designed into the CRF and would not comport with the pro-settlement stance of its authors. When constructing the arbitration option to be more attractive than trial, the framers of the CRF removed product defect from the plaintiffs burden by name in two places,' § E.5(a) and § G.12. Surely if they had planned to remove causation from either, they would have been as explicit in that intention.

This leads the Court to conclude that the § G.2 presumption of causation was. not intended to and should not be allowed to apply to arbitration or trial. Options 1 and 2 and the offers generated by the Trust based on a claimant’s medical records would have lost much of their appeal if claimants had been given the same advantages in arbitration and trial as in those earlier CRF stages. A perception that the road to potentially larger recoveries in arbitration and trial was as steeply downhill and unobstructed as the road to settlement would have led more claimants to choose them, thwarting the goal of preferring settlement over both arbitration and trial. The statute of limitations defense in arbitration and trial and the prospect of having to prove product defect at trial would- have been left as the deterrents to trial. They no doubt would have dissuaded some of the 212,000 persons who settled from embarking on the arbitration and litigation paths. It is equally without doubt that far fewer of them would have decided to settle than did in the face of having to prove causation as in a normal products liability trial, if they had known or felt that causation would be presumed in their favor in arbitration and trial just as it had been until that point in the process.

A few thousand, or, even a few hundred, additional arbitrations or trials would have postponed closing the Trust for years hence and would have left far fewer excess funds to distribute to eligible claimants as pro rata payments under § G.14 of the CRF. The successful implementation of the Plan and CRF necessitates no presumption of causation in arbitration or trial. The adversarial field between the plaintiff and the Trust was designed to be level in arbitration except for defect, so it is to be level on the causation issue. The field was designed to be completely level in trial on all issues, so it is to be level on the causation issue.

1. The Incentives Designed into the CRF Remain Significant Today

It might be said that the claims resolution process and the parties have reached the stage when the CRF’s incentives to settlement have outlived their usefulness and should no longer influence the interpretation of CRF provisions. All claimants have made their choices. Fewer than eighty-five cases remain. Today there is no risk that thousands of claimants will be clamoring for arbi-trations or trials. Under this view, allowing the remaining plaintiffs a presumption of causation in front of their arbitrators or juries would not disrupt the pro-settlement CRF agenda.

That, however, is a naively quixotic view of the present circumstances. Each of the plaintiffs in those pending arbitrations and trials could still end their cases by accepting the option 3 offers made to them by the Trust on the strength of the § G.2 presumption of causation. Each case settled in that manner will hasten the end of the Trust. Each case that does not settle prolongs the life of the Trust, drains the Trust of defense and administrative costs, and thereby delays and reduces the eventual pro rata payout to eligible claimants.

Several claimants and lawyers objected to the July 31, 1998 deadline for trials and October 31, 1998 deadline for the completion of all arbitrations directed in this Court’s Administrative Order No. 2, saying they could not be ready for trial by then. Removing what has been a primary incentive for them to settle before trial would hardly improve upon that dire prediction. Maintaining the reasons for claimants to settle is as important now as it ever has been. Nor would it be fair to all those claimants who have reacted to the CRF stimuli and settled their claims rather than risk an inability to' prove causation at trial to grant now to the remaining plaintiffs a presumption of causation that could lighten their causation burden at trial. The Court will not relax in its effort to see the CRF through to its designed conclusion simply because the end of the long journey is near.

2. The Significance of a Presumption

Another slant on this situation must be considered. The Court is intrigued by the prospect that a § G.2 presumption in arbitration or trial would not, at least in theory, render those processes any more favorable to plaintiffs than they already are. A presumption is not evidence. It is a procedural tool adopted by courts or legislatures to assist in determining whether a party has met its burden of proving a particular element of a claim or defense. Presumptions operate at the decisional stage of a case, when a judge in a jury trial is deciding whether the plaintiff has set out a sufficient prima facie case to get to the jury, and in a bench trial where the judge is ruling upon the merits of the case. See, E. Clearly, McCormick on Evidence § 344, p. 973 (3d ed.1984).

The rule followed by the Court of Appeals in Reichel, Model Code Rule 704, followed the “bursting bubble” theory of rebuttable presumptions. That analysis was first promoted in Thayer, Preliminary Treatise on Evidence, Ch. 8 (1898). It was embraced in 9 Wigmore, Evidence § 2491(2) (Chadbourne Rev.Ed.1981) and by Congress when it adopted Federal Rule of Evidence 301. Under the bursting bubble theory, once the basic facts of a presumption are established by evidence, a presumption “bubble” is created that the presumed fact is assumed established as well, if nothing else happens in the case. If, however, contradictory evidence that would support a finding that the presumed fact did not exist is then introduced, or is pointed out as already in the record, the bubble is burst. Once burst, the presumption vanishes from the ease. The case is decided as if no presumption had ever existed.

The editors of the Model Code analyzed five differing views in the existing judicial opinions on the procedural effect of a presumption, ranging from the minimalist Thayerian rule, to the requirement that the party opposing the presumption must prove by a preponderance of the evidence that the presumed fact did not exist, and ultimately to a rule that makes the presumption nearly irrefutable. They rejected the more stringent requirements and instead in Rule 704 allowed a presumption to be rebutted simply by any evidence that would justify a finding contrary to the presumed fact. They did not require rebuttal by a preponderance of the evidence, one of the views in fact spurned in the Model Code.

Further, the writers of the Model Code Rule adopted by the- Court of Appeals in. Reichel made it clear that the presumption is “left in the hands of the judge to administer” and is not submitted to a jury for decision. The trial judge, not the jury, decides whether the presumption has been rebutted. Indeed, the Rule must be administered so that the “jury never hear[s] the word presumption used since it carries unpredictable connotations to different minds.” Model Code of Evidence, Ch. 8, p. 314.

Federal Rule of Evidence 301 should operate in the same fashion, though there has been some confusion among the jurisdictions on the inconsistency between the comment to Rule 301 and the version of the Rule actually adopted. See G. Joseph & S. Saltzburg, The Federal Rules in the States, 8.3 (1989). Suits against the Trust in federal courts would be subject to Fed.R.Evid. 301, as would all arbitrations, according to Arbitration Rule 32(a). Suits in a state court normally would be governed by however the law of that state treats presumptions. Many states adhere to the Rule 301 approach. It is also likely that with the CRF providing the basis for the presumption of causation in a Daikon Shield case, this Court could exercise its power under § 8.05 of the Plan to set a Thayerian bursting bubble approach to the causation presumption that should be followed in all Daikon Shield cases, whether in arbitration or in trials in state or federal courts, so that all claimants would be subject to a uniform rule.

In theory, then, the infusion of a § G.2 presumption of causation in litigation would alter the relative burdens between the plaintiffs and the Trust so little that it would not make arbitration or trial appreciably easier .for plaintiffs, and thus not destroy the CRF incentives to settle before trial. If a Daikon Shield plaintiff could establish the basic facts — that she used the Daikon Shield and had an injury listed on Exhibit A — then the burden of production would shift to the Trust to point to evidence in the record that would support a finding that the Daikon Shield did not cause the injury, such as a suggestion in the claimant’s testimony or medical records of an alternative chuse of her injury (e.g., a sexually transmitted disease) or medical studies questioning whether the Daikon Shield elevated the risk of the injury. That would burst the bubble. The presumption would vanish from the ease, the ultimate burden of persuading the arbitrator or jury that the Daikon Shield caused the plaintiffs Exhibit A injury would remain at all times with the plaintiff.

The Court is not convinced, however, that the practical impact of a presumption of causation in arbitrations and trials would mirror this theoretical prediction. If the presumption, properly applied, were so readily rebutted, then the plaintiffs would not be urging so vociferously that it covers arbitration and trial. Plaintiffs and their lawyers want arbitrators and juries to hear of .the presumption and perceive it as the panacea for all the proof problems in their cases and as handicapping the Trust in the evidence it can offer to resist liability at trial. Plaintiffs will contend that the presumption proscribes the expert testimony that the Daikon Shield did not medically tend to 'cause any of the injuries listed in Exhibit' A, which is the very sort of evidence the Trust would offer to burst the bubble and rebut the presumption. Ms. Hurst would allow the Trust to rebut only with evidence of an alternative cause of the injury and not with evidence that questioned whether the device elevated the risk of an Exhibit A injury in all users. The bulk of the plaintiffs attempted use of Reichel to date has seemed to be an effort to turn each Daikon Shield arbitration and trial into a damages-only case.

That is neither what the CRF authors nor what this Court intended. The incentives of the CRF operate in this world of perceptions. If plaintiffs perceive arbitration or trial as easy, they will be more likely to persist in it when they should settle. The Court also prefers to avoid the collateral litigation that necessarily would stem from the debate over the meaning of the presumption in an arbitration or trial,- as the parties fight both here -and before arbitrators and state and federal judges over what it takes to rebut the presumption, whether it has been rebutted, whether it filters the evidence the Trust can offer, and whether a jury can be allowed to hear if there ever was such a presumption. The CRF, the conclusion of the Trust’s mission, and the universe of all pro rata eligible claimants will be better served if none of that occurs.

D. The Reichel Court Did Not Foreclose this Analysis

Because it was not asked to do so, the Reichel court did not discuss whether the § G.2 presumption of causation applied outside ADR to arbitration or trial. Nothing in the opinion preempts the foregoing analysis of the CRF and how § G.2 fits within that design. Indeed, the Court of Appeals characterized ADR as part of the CRF settlement process, suggesting that the court saw § G.2 as operating in the settlement steps of the CRF and not in arbitration or trial, as they clearly are not settlement steps.

In its opinion, the Court of Appeals summarized the CRF claims processing alternatives. It described ADR as a settlement mechanism and distinguished it from arbitration and trial. After reviewing Options 1 and 2, the court noted that Option 3 required claimants to Complete a detailed claim form and provide medical records or evidence of use of the Daikon Shield:

The Trust then fully evaluates the claim and makes a settlement offer. If the claimant rejects the settlement offer, the claimant may choose to proceed through In-Depth Review/Voluntary Settlement Conference or other voluntary alternative dispute resolution (ADR) process under § G.4 of the CRF.

109 F.3d at 966. The court then described how the Trust had established the ADR process beginning in 1992 and that under the First and Second Amended ADR Rules, claimants have the express burden of proving that they suffered an injury caused by the Daikon Shield. The court continued:

If claimants continue to reject a settlement offer following this initial form of alternative dispute resolution, they may then choose either binding arbitration or traditional litigation. If a claimant chooses arbitration, §§ E.5(a) and G.12 of the CRF provide that the Trust may assert all available defenses except for absence of product defect. If a claimant chooses traditional litigation, §§ E.5(b) and G.12 provide that all claims and defenses are available to both sides.

109 F.3d at 966-67. The court thus saw ADR as a settlement process housed under § E.4 of the CRF with the In-Depth Review/Voluntary Settlement Conference step. The court went so far as to see the referee’s award in ADR as a settlement offer the claimants could “continue to-reject” and then move to arbitration or traditional litigation.⁴

In reality, the ADR hearing did not generate an offer the claimant was free to reject and then proceed to arbitration or trial. Under the ADR contract signed by claimants and the ADR Rules they agreed to follow in each ADR case, the referee’s decision was final and binding and could not be rejected by a claimant. The ADR process was an alternative to arbitration or trial. The Court of Appeals’ placement of ADR in the voluntary settlement portion of the CRF process before applying to it the § G.2 presumption of causation leaves room for this Court’s reading of § G.2 as limited to the settlement steps of the CRF. It suggests that the Court of Appeals would agree with this Court’s analysis if and when the issue is fully before it. Even Ms. Hurst seems in tune with that view, for she argues in her Reply (p. 4) that the “Fourth Circuit in essence held that but for the rebuttable presumption, compensation by settlement with the Trust would be improper to anyone.” Arbitration and trial are not part of the settlement steps of the CRF.

Footnote 4 to the Court of Appeals opinion does cause some pause, however:

In that connection, we note that the district court in a previous case, In re A.H. Robins Co. Inc., 131 B.R. 292, 297 (E.D.Va.1991), reversed on other grounds, 972 F.2d 77 (4th Cir.1992) construed § G.2 as we do here. And in In re A.H. Robins Co. Inc., 197 B.R. 519 (E.D.Va.1994), it construed § G.2 as it did in this case. As our opinion sets out, we prefer its first construction.

109 F.3d at 968 n. 4. In Troutt, the Court had read § G.2 as creating only a presumption of eligibility for payment, and not of causation. In Reiser, the Court had observed that for purposes of the definition in § 1.85 of the Plan of Unreleased Claims of medical malpractice, the injuries listed in CRF Exhibit A are presumed to be Daikon Shield Claims and not independent malpractice claims. Thus, this Court surmises that the Court of Appeals was not disapproving of Troutt because it held that § G.2 did not apply in arbitration, but rather because of the portion of the opinion in which the Court had ruled, as it had in Reichel, that the language of § G.2 did not erect a presumption of causation. This left the coverage issue to be decided here today.

E. There are No Other Grounds for Applying § G.2 to Arbitration or Trial

Ms. Hurst argued in her Motion and brief that “reason and logic” warranted applying § G.2 to arbitration and trial. The Court concludes today that there are ample reasons and compelling logic in the CRF for the contrary. Ms. Hurst also contended that allowing ADR claimants the benefit of a presumption while denying it to arbitration and trial plaintiffs would violate § 1123(a)(4) of the Bankruptcy Code, 11 U.S.C. § 1123(a)(4). That section requires that Chapter 11 plans:

(4) provide the same treatment for each claim or interest of a particular class, unless the holder of a particular claim or interest agrees to a less favorable treatment of such particular claim or interest.

11 U.S.C. § 1123(a)(4). This section legislates a cardinal rule of bankruptcy and of this Court’s approach from the day the Robins reorganization began — that of equality of treatment of similarly situated creditors.

Daikon Shield Personal Injury Claims are Class 3A claims under § 3.03 of the Robins Plan. The holders of those claims voted overwhelmingly to approve the Robins Plan. It provided their claims would be treated as set out in the CRF. The CRF prescribed different options for the payment of Daikon Shield Personal Injury Claims.- It fixed the differences between the settlement steps of the CRF and arbitration or trial, as well as between arbitration and trial. Claimants in the settlement steps of the CRF are favored with various advantages. The Court of Appeals saw ADR as a settlement step and gave ADR claimants a § G.2 presumption. The holders of Daikon Shield Claims agreed to the different treatment given claimants among the different CRF options. Equality of treatment was preserved by ensuring that all claimants had equal opportunity to elect among the CRF Options and to the application of the same rules to all who had chosen the same Option.

Under the CRF, arbitration claimants face a different process than trial plaintiffs; both arbitration and trial plaintiffs act in arenas vastly different from ADR. Those differences are part of the CRF design. Not applying a presumption of causation in arbitration and trial no more violates § 1123(a)(4) than does carving product defect out of arbitration while leaving the issue in trials. If Ms. Hurst’s argument were accepted, the CRF would have had only one Option. The Trust would have needed decades to review them all.

III. CONCLUSION

The unfinished Daikon shield cases should be concluded in the manner the CRF and this Court’s previous rulings have established. Those plaintiffs are to follow the same rules adhered to by all the claimants who came before them. The CRF and this Court’s Amended Administrative Order free plaintiffs in arbitration from the burden of proving product defect, but otherwise place plaintiffs and the Trust on an equal footing in an arbitration case. Plaintiffs and the Trust stand completely equal in litigation. The plaintiffs in the thirty-seven arbitration cases and fifty-one litigation cases already adjudicated to date were on that level field; they prevailed in twelve of the arbitration cases and twenty-one of the litigation ones. Neither the CRF nor this Court’s opinions should be rewritten now to favor the plaintiffs in the remaining cases in any respect or to undermine at this late stage the goal of encouraging settlement of claims. The Courts, the entire group of all pro rata eligible claimants, and even the still litigating plaintiffs would be far better served if all the unfinished cases were settled under the Option 3 criteria. Then the Trust could distribute its remaining funds as pro rata payments and close, and all plaintiffs could put this long chapter of their lives behind them. The Court continues to pursue those goals and sees today’s ruling as a significant final step in their direction.

An appropriate Order shall issue.

1 . Though resort to the power was rendered unnecessary by the Trust's request for a complete decision, this Court’s retained jurisdiction in §§ 8.05(c) and (d) of the Plan to interpret and implement the Plan and CRF should be sufficient to permit the Court to decide sua sponte whether extension of a § G.2 presumption of causation to either arbitration or litigation comports with the design of the Plan and the efficient and successful implementation of that design, even in the absence of a specific request by a claimant or the Trust. See In re A.H. Robins Co. (Order Limiting Attorneys, Fees), 182 B.R. 128 (Bankr.E.D.Va.1995), affirmed, In re A.H. Robins Co. (Bergstrom v. Dalkon Shield Claimants Trust), 86 F.3d 364 (4th Cir.) cert. denied, - U.S. ——, 117 S.Ct. 483, 136 L.Ed.2d 377 (1996).

2 . This and the other data in this opinion were furnished to the Court in Exhibits 6-8 filed by the Trust at the October 21, 1997 hearing on the Trust’s Motion for Administrative Order No. 2, unless otherwise noted.

3 . The design of the CRF also motivated Breland claimants to resolve their claims with the Bre-land Insurance Trust by voluntary agreement, rather than by more time-consuming and expensive means. The Breland Trust paid 18,631 claimants a total of $3,726,200 under its Option Q, which allowed a quick payment of $200 to any claimant who signed a release and a statement that she or he used the Daikon Shield. The Breland Trust paid $2,279,691 to 3,077 Option A claimants; $6,570,586 to 2,012 Option B claimants; and $52,356,389 to 2,607 Option C claimants. Breland Options A, B, and C corresponded to the CRF Options 1, 2, and 3.

4 . The court's characterization of ADR as a settlement mechanism, and indeed as a preliminary step that could lead to arbitration or litigation, had been foreshadowed in the court's opinion in In re A.H. Robins Co. (Bergstrom v. Dalkon Shield Claimants Trust), 86 F.3d 364 (4th Cir.), cert. denied, - U.S. -, 117 S.Ct. 483, 136 L.Ed.2d 377 (1996). There, the court wrote that if settlement were not reached in ADR, the claimant must elect arbitration or trial. 86 F.3d at 369.