Overcoming the Technical Ediscovery Challenges in Multidistrict Litigation
There are many challenges in multidistrict litigation (“MDL”), and a number of these are technical in nature. Using the right technology can help you move from constant headaches and human error to a smooth process where legal experts can focus on the legal issues.
The Basics of MDL
Multidistrict litigation cases are lawsuits where a large number of plaintiffs in different Federal Districts have “one or more common questions of fact” which require consolidation for judicial efficiency and consistency. MDLs can range from defective products to medical injuries or anything in between.
The attorneys managing MDL cases are challenged by massive data volume, complex discovery strategies, and management of a substantial number of lawyers conducting document review. For example, one recent MDL case was comprised of approximately 30,000 cases. In another recent case, there were over 64 million pages in English and Japanese produced.
One judge highlighted the balancing act in responding to ediscovery requests where there were 1,800 underlying actions in an MDL over prescription drugs, with the Defendant having offices in the United States, Japan, and Europe. Discovery was being conducted in phases, with a focus on ESI from the U.S. and Japanese corporate Defendants. The Plaintiffs brought a motion to compel production of ESI from the Defendants’ European offices.
The Court found that the U.S. based Defendants did have “control” over ESI in Europe, because custodians in the U.S. could access documents from their European business units. However, that did not create a right for the Plaintiffs to propound omnibus discovery requests on the Defendants, but make specific discovery requests that were narrowly tailored, “well-grounded, materially relevant and non-cumulative.”
So, how can technology help manage MDL cases?
Real-time collaboration in a cloud product means that review teams across multiple firms can coordinate easily and avoid duplication of work. And search term reports and predictive coding will help you focus your document review to find the documents and data you need faster—or even eliminate thousands of documents from needing human review to begin with.
Coordinate Between Multiple Review Teams
Managing Large Volumes of Data
Since MDL cases often involve law firms in multiple locations and a small army of lawyers, having data hosted online allows everyone involved to access what they need for review and trial strategy.
And, with the right tool, the case manager can monitor review progress and pace of the different attorneys with case analytics. You can also make reports for opposing counsel or the court on estimates for when review will be completed.
Organizing Your Docs Means Organizing Your Case
With potentially thousands of plaintiffs and countless documents, what do you do when you find important documents? Manually maintain a spreadsheet, print them, put them in binders? This method is error prone and time consuming. There’s a better way.
The StoryBuilder suite of tools (Chronology and Outlines) in Everlaw can be used to replace those spreadsheets and printed docs. Whether on the plaintiff or defense side, you can use Chronology instead of Excel to create and maintain your timeline of events. It allows you to have rich descriptions of your important documents and access any work done during review (including codes, notes, etc.). Use Outlines as you prepare for depositions, associating documents with questions, all right there in the platform. Or use Outlines to organize each cause of action, dragging and dropping supporting documents in the outline. Law is inherently collaborative, and the right technology can simplify your processes in complicated cases.
Developing your Search Strategy
Ediscovery issues in multidistrict litigation can range from traditional disputes over whether production is adequate, to the more complex where a detailed workflow is outlined for the use of predictive coding.
Specific challenges in MDL cases can include determining the scope, custodians, and search terms for responsive electronically stored information that complies with proportionality under Federal Rule of Civil Procedure Rule 26(b)(1). Plus, parties need to develop search strategies that are “reasonably comprehensive.” This can include “identifying key employees and reviewing any of their files that are likely to be relevant to the claims in the litigation.”
Defendants who are facing tens of thousands of parties can develop search strategies, either as part of a case management order, or as part of their own diligent search for responsive ESI.
Consider this hypothetical: the MDL is over birth defects allegedly stemming from a prescription drug. The issue is whether the pharmaceutical company made false statements about the drug in their marketing and product labeling.
The parties first must come to an agreement on the scope of the case, including the key custodians to collect, date ranges, and other narrowing factors to collect the data for review. Once that data is collected, one option for moving forward is to train a prediction model for relevant ESI.
Training the Predictive Coding System
The Defendants have two options. The first is the “Go it Alone” model, because it is up to the Plaintiffs to claim a production is inadequate. The second is the “Let’s Cooperate on What is Relevant” model, to try to avoid a fight later in the case over production adequacy. The methodology that is used is highly dependent on the nature of the case and whether the parties are cooperative or combative.
If the parties have agreed to cooperate on determining what is relevant, data that has been tagged as relevant by the Defendant can be shared with opposing counsel as a project for a meet and confer. This would give opposing counsel the opportunity to agree with the relevancy determinations by the Defendants.
Attorneys can review data to train the predictive coding system. This data could be a random sample of the dataset, compiled from documents identified from initial disclosures, or some other strategy to review a sufficient number of records to train a predictive coding model. Using predictive coding can save you countless hours of review, not to mention arguments with the other side.
Here’s an example: in one case, the producing party agreed to allow the requesting party’s “experts” to have access to the entire sample set of records used to train the predictive coding system for determining what was relevant. The entire methodology for training the predictive coding system was detailed and subjected to a non-disclosure agreement. A similar workflow could be created with a project of a sample set for agreed-upon opposing counsel to review.
Finding Relevant Documents
After the predictive coding model has been trained, the Defendant-Producing Party can begin their review in earnest. Searches can created based upon discovery requests or allegations within the complaint, with the predictive coding predictions used to focus the review on what is likely relevant from the search term hits. And, assignments can be created from these hits for the review attorneys.
Second pass review can be conducted for quality assurance by lawyers who are handling substantive work on the case, such as conferring with the other side, motion practice, conducting depositions, or handling the case at trial.
Multidistrict litigation exists to avoid duplication of discovery in cases with questions of common facts. Using hosted repositories allows coordination of review between attorneys that otherwise would be impractical. Developing strategies leveraging predictive coding to focus targeted searches to likely responsive ESI can maximize efficiency for document review. What’s more, ESI can be organized with assignments for attorneys, so the project manager can track review progress with realistic timelines for completing document review.
So what’s the verdict? Using the right technology for your MDL cases means better results, for you and your client.