MEMORANDUM AND ORDER RE: DISCOVERY ISSUES AS PREPATORY FOR BELLWETHER TRIAL

MICHAEL M. BAYLSON, District Judge.

The Court previously determined that the most appropriate and fair way to approach completion of this litigation is to schedule a bellwether trial. Both parties agreed on a finite and limited scope of discovery. Each side selected a limited number of patient cases for discovery purposes. The preparation for the bellwether trial contemplated that there would be discovery of this limited group of cases, and then the parties would either agree, or the Court would determine, as to the actual cases to be subject to the bellwether trial. The above events have been subject to prior orders in this case.

Relator proposed random selection of the patient cases. Relator filed a Motion on October 4, 2017 to require Defendant to select its patient cases for the bellwether trial by Relator's random method (ECF 182). The Court held this motion under advisement at the hearing it held on November 6, 2017. (Tr. at 34:3-4, ECF 200.) At the hearing, the Court contemplated that the parties might explore the criteria for case selection at the upcoming 30(b)(6) deposition:

(Id. at 34:19-35:17.)

At the November 6, 2017 hearing and since, Defendant has asserted that despite Relator's claim of random selection, Relator's selection of cases is anything but: Relator's cases include only inpatient certifications, and are drawn from a group of hospitals that have been subject to audits or other investigations where violations are more likely. The Court has no opinion on this issue, and is not revisiting the issue of hospital selection at this time. Defendant selected its own 220 cases and asserted attorney-client privilege and work product protection as to how Defendant selected its cases when its representative appeared for a 30(b)(6) deposition.

Defendant has filed a motion for reconsideration, which will be DENIED, but the Court will clarify the scope of attorney-client privilege and attorney work product protection that is provided to the parties' selection of cases. The Court will rule on this discovery dispute and will then amend the prior pretrial orders in this case to require that one half of the cases selected as subject to discovery shall be randomly selected, and one half will be selected by each party.

I. Discovery of Facts Considered in Selection Process

In the Court's Order of March 13, 2018, (ECF 227), counsel were to expand and finalize their attorney-client privilege logs to cover privileged communications concerning the selection of cases process. Counsels' responses to the Court's Order were submitted as of April 20, 2018, having been filed as ECF 238 for Relator and ECF 239 for Defendant.

Communications between a client and an attorney for the purpose of legal advice are protected attorney client communications and are not discoverable. This privilege clearly applies as to communications concerning discovery.

The Court agrees that it is customary, once litigation has started, for communications between counsel and the client not to be the subject of a privilege log, because they can occur almost daily to discuss events concerning the process of the litigation.

Defendant's brief makes the following statement: "For the reasons outlined in EHR's motion, EHR's method for selecting patient cases, and the work completed to effectuate that methodology, are core opinion work product."

Defendant cites a leading Third Circuit case, In re Cendant Corporation Securities Litigation, 343 F.3d 658, 663 (3d Cir. 2003), which held that "core" or "opinion" work product that encompasses the "mental impressions, conclusions, opinion, or legal theories of an attorney or other representative of a party concerning the litigation" is "generally afforded near absolute protection from discovery." Id. (holding work product of non-lawyer litigation consultant to fall within work product rule). Defendant also cites cases in which selection methodology has been held to be covered by the work product rule. In Sporck v. Peil, 759 F.2d 312 (3d Cir. 1985), the defendant sought mandamus relief after the district court required the defendant to disclose all documents, which had already been produced in discovery, that defense counsel had used to prepare a witness for deposition. The defendant "concede[d] that the individual documents that comprise the grouping [we]re not attorney work product, but argue[d] that the selection process itself represent[d] defense counsel's mental impressions and legal opinions as to how the evidence in the documents relate[d] to the issues and defenses in the litigation." Sporck v. Peil, 759 F.2d 312, 315 (3d Cir. 1985). The Third Circuit distinguished between "identification of the documents as a group," which would "reveal defense counsel's selection process, and thus his mental impressions," and eliciting deposition testimony about individual documents selected by the attorney and used in deposition preparation. Id. Accordingly, the court held that the selection and compilation of documents was protected work product. Id. at 316. The Court will apply a similar distinction in this case. See also James Julian, Inc. v. Raytheon Co., 93 F.R.D. 138, 144 (D. Del. 1982) (attorney's selection of otherwise discoverable documents to prepare witness for deposition was attorney work product because such selection "could not help but reveal important aspects of his understanding of the case"); U.S. ex rel. Westrick v. Second Chance Body Armor, Inc., 288 F.R.D. 222, 227 (D.D.C. 2012) (in False Claims Act case, "information regarding" defense expert's "ultimate selection" of a subset of allegedly faulty ballistics vests was work product); Vardon Golf Co. v. BBMG Golf Ltd., 156 F.R.D. 641, 647 (N.D. Ill. 1994) (stating that "[s]election of certain pieces of evidence from a mass of evidence is an important part of an attorney's preparations" and finding that choice of allegedly infringing golf clubs to cut open for internal "assessment of its potential liability" was protected work product).

The Court clarifies that it had no intention of allowing Relator to invade "core" work product. These holdings, however, do not protect Defendant from all discovery about the selection process. Once the communications with the client have taken place, and once the attorney work product has been created, a decision is necessarily made about what discovery the client/party will or will not provide. Defendant does not dispute the cardinal legal principle that the privilege protects communications, but does not protect facts from discovery. The facts that Defendant relied on in making its decisions are discoverable.

This is no different than any litigation issue where an opposing party has made discovery requests. Discussions take place between counsel and the client about the facts, what the rules provide, what strategy is best, what the expense factor will be, which results in a decision being made as to what the party's response to discovery will be. These discussions are privileged, but the underlying facts are not.

The same process would be undertaken if a party has received what it believes is overly burdensome or improper discovery and decides to move for a protective order. The discussions with the client are privileged, and any research or opinions that the attorneys develop on this topic are protected from discovery as "core" attorney work product. However, once the decision is made, it is communicated to the opposing counsel and/or the court. The underlying facts are not protected from discovery. The Court, and other counsel, are entitled to know what facts make the request burdensome.

The same process applies in this case. Defendant and its counsel discussed the selection of cases, and those communications are privileged. The attorneys were likely preparing memoranda and perhaps reviewing specific cases and Defendant need not disclose that attorney work product. However, at some point Defendant selected cases. The facts about those cases are discoverable. As Defendant itself discloses, some cases were inpatient and some were outpatient. That is a fact. Defendant cannot stop the discovery process at that point. There are other facts about the cases that Defendant selected that Defendant must disclose. Perhaps it is the identity of the hospital. Whether the patient stayed one night or more nights in the hospital may be relevant in this case. That is a fact. Perhaps it is the identity of a physician. Perhaps it is the identity of the patient or facts about the patient's malady or treatment. These are all facts that Defendant may have relied upon in its selection process. Defendant cannot preclude discovery on these facts merely because they were the topic of privileged communications or attorney work product. Defendant must disclose the facts it considered in making its selection of cases. Defendant does not have to disclose its process of review or why it picked these cases — that would likely invade the attorney-client privilege or attorney work product, much as the Third Circuit held in Sporck. See 759 F.2d at 316.1 But the facts about those cases selected by Defendant are discoverable. These facts are likely to be important in the final selection of cases for trial, and to allow application of collateral estoppel, determination of damages, if any, etc. The Court does not know if all facts Defendant relied on will be disclosed from the documents Defendant produces. If Defendant can represent that all facts it deemed relevant are disclosed in the documents, this may shorten any deposition. If the documents themselves do not disclose relevant facts Defendant relied upon, Defendant must do so by a supplemental disclosure or at the deposition.

These disclosures should be made in writing and/or in the upcoming Rule 30(b)(6) deposition. In view of the Court's revised pretrial decision on randomization or selection, the dates for further discovery revolving around selection should be discussed between counsel, and the 30(b)(6) deposition now scheduled for early May might have to be postponed. When the 30(b)(6) witness deposition is taken, the witness should be knowledgeable as to what facts about particular cases warranted the selection of those cases, but need not disclose the reasons why those facts were important to Defendant.

II. Revision of Court's Ruling on Selection Process

After considering this issue at some length with the voluminous briefs and other source materials on bellwether trials, the Court concludes that the most appropriate way to get a fair representation of cases for trial and to arrive at a final judgment in this case, assuming it is not settled, is to have a combination of cases of some randomly selected, and others selected by each party. The Court believes that there is value in allowing each party to select some cases that it finds particularly supportive of its contentions in order to present the opportunity for a favorable verdict at trial. Defendant cites cases where a court approved this procedure. At the same time, the Court remains mindful that the cases selected should be representative, as argued by Relator in favor of randomization. Thus, the Court determines that half of the cases for discovery and most likely for trial, should be selected by each party, but the facts underlying the selection must be disclosed. The other half shall be randomly selected.

The Court finds some benefit in including inpatient as well as outpatient certifications in a bellwether trial.

Accordingly, the Court will require the following process for the selection of its cases unless otherwise agreed. Each side will select 110 cases (i.e., half of its 220) cases randomly from the full universe of 38,000 cases, which includes both outpatient and inpatient certification. Each party must then select the remaining 110 cases, such that no more than 25% of the 110 selected cases are cases in which Defendant made an outpatient certification. The parties must disclose to one another the facts underlying their selection. As to the random selection process, the Court suggests that counsel for Relator and Defendant confer to see if they can agree on a process for this, because there could be disputes about what is "random." Perhaps counsel will agree on a statistician and share the costs for professional advice or assistance.2 If necessary, the Court may employ such a statistician to ensure an appropriate random selection given the underlying data and the issues in this case, with the cost to be split evenly between the parties.

The Court will delay any decision on attorneys' fees. See attached Order.