C.K. ex rel. P.K. v. Bassett
C.K. ex rel. P.K. v. Bassett
2023 WL 4086333 (E.D.N.Y. 2023)
June 20, 2023

Wicks, James M.,  United States Magistrate Judge

Sampling
FRE 502(d)
Protective Order
Attorney-Client Privilege
Predictive Coding
Cost-shifting
Search Terms
Proportionality
Failure to Produce
Attorney Work-Product
Waiver
Clawback
Email Threading
Download PDF
To Cite List
Summary
The Court directed Defendants to use Plaintiffs' proposed keyword search terms to locate and produce 1.2 million documents from 24 custodians. The Court also encouraged the parties to consider methods of expediting the review and production of documents, such as the use of Technology Assisted Review (TAR) and the use of contract attorneys. The Court declined to order cost-shifting for the review and production of the documents.
C.K. THROUGH his next friend P.K.; C.W. through her next friend P.W.; C.X., through her next friend P.X.; C.Y. through his next friend P.Y., for themselves and those similarly situated, Plaintiffs,
v.
Mary T. BASSETT, in her official capacity as the Commissioner of the New York State Department of Health; and Ann Marie T. Sullivan, in her official capacity as the Commissioner of the New York State Office of Mental Health, Defendants
22-CV-1791 (BMC)(JMW)
United States District Court, E.D. New York
Signed June 20, 2023

Counsel

Claire Renee Glasspiegel, Esq., Children's Rights, 197 Van Vorst St., Ste 305, Jersey City, NJ 07302, Attorney for Plaintiffs.
Valerie Achille, Esq., Children's Rights, 1702 Kenneth, Baldwin, NY 11510, Attorney for Plaintiffs
Daniele Gerard, Esq., Harry Frischer, Esq., Children's Rights, 88 Pine Street, Ste 800, New York, NY 10005, Attorneys for Plaintiffs C.K. and C.W.
Brandy Tomlinson, Esq., Disability Rights New York, 44 Exchange Boulevard, Ste 110, Rochester, NY 14614, Attorneys for Plaintiffs C.K. and C.W.
Kimberly Lewis, Esq., Martha Jane Perkins, Esq., National Health Law Program, 3701 Wilshire Blvd, Ste 750, Los Angeles, CA 90010, Attorneys for Plaintiffs C.K. and C.W.
Shiva Pedram, Esq., Jacob Wonn, Esq., Proskauer Rose LLP, Eleven Times Square, New York, NY 10036, Attorneys for Plaintiffs C.K., and C.W.
Steven H. Holinstat, Esq., Proskauer Rose LLP, Eleven Times Square, New York, NY 10036, Attorney for Plaintiffs.
Benjamin D Liebowitz, Esq., Matthew Joseph Lawson, Esq., Yuval Rubinstein, Esq., Office of New York State Attorney General, 28 Liberty Street, New York, NY 10005, Attorneys for Defendants
Rebecca Durie Katherine Culley, Esq., Office of New York State Attorney General, 120 Broadway, New York, NY 10271, Attorney for Defendants.
Wicks, James M., United States Magistrate Judge

MEMORANDUM AND ORDER

*1 Plaintiffs commenced this putative class action seeking declaratory and injunctive relief, and asserting various causes of action against Defendants related to New York State's alleged failure to provide mental health services to Medicaid-eligible children. (See DE 34.) Plaintiffs allege that Defendants have violated the Early and Periodic Screening, Diagnostic, and Treatment Services provisions of the Medicaid Act, 42 U.S.C. § 1396a(a)(10)(A); 42 U.S.C § 1396a(a)(43); 42 U.S.C. § 1396d(a)(4)(B); 42 U.S.C § 1396d(r), Reasonable Promptness provision of the Medicaid Act. 42 U.S.C. § 1396a(a)(8), Title II of the Americans with Disabilities Act, 42 U.S.C. § 12132, et seq., Section 504, 29 U.S.C. § 794, and 42 U.S.C. § 1983. (DE 34.)
Plaintiffs are four minors proceeding under pseudonyms and each through their “next friend.” (DE 34.) Defendants are Mary T. Bassett, in her official capacity as Commissioner of the New York State Department of Health, and Ann Marie T. Sullivan, in her official capacity as Commissioner of the New York State Office of Mental Health. (DE 34.) Plaintiffs challenge the scope and nature of services provided by the State of New York, specifically the intensive home and community-based mental health services provided to Medicaid-eligible children (e.g., Early and Periodic Screening, Diagnostic, and Treatment), and Plaintiffs highlight, inter alia, the alleged issues around the access to, and availability of, these services within the state. (DE 34.)
Before the Court are two joint motions regarding the parties’ discovery disputes (DE 39; D48), which the Court construes as Plaintiffs’ motion to compel discovery, and Defendants’ motion for a protective order. (See DE 40.) At bottom, the parties are at an impasse regarding the keyword search terms that Defendants should use to locate, review, and produce responsive documents. (DE 39.) The genesis of the that dispute stems from the differing volume of documents -- the document review population -- that would result from the application of either party's proposed search terms. Relatedly, the parties disagree on how to proceed with depositions considering the ongoing dispute regarding search terms. (DE 48.) District Judge Brian M. Cogan referred this motion to the undersigned “for determination and a recommendation to [him] for any adjustments to the current case schedule necessitated as a result of that determination.” (Electronic Order dated April 18, 2023.)
For the following reasons, Plaintiffs’ motion to compel electronic discovery (DE 39) and Defendants’ cross-motion for a protective order (DE 39) are DENIED in part and GRANTED in part. Plaintiffs’ motion to compel depositions (DE 48), and Defendants’ cross-motion for a protective order (DE 48) are DENIED in part and GRANTED in part.
I. PROCEDURAL BACKGROUND
The initial complaint was filed on March 31, 2022. (DE 1.) The amended complaint was filed on October 31, 2022. (DE 34.) Defendants filed their answer on November 21, 2022. (DE 37.) The Honorable Brian M. Cogan adopted a discovery schedule on May 20, 2022, and pursuant to that schedule discovery was set to end by June 30, 2023. (Electronic Order dated May 20, 2022.) Defendants subsequently requested that the September 13, 2023 deadline by which they must (i) serve their written responses to Plaintiffs’ initial document production requests and interrogatories, and (ii) begin producing documents, be extended to October 20, 2022. (DE 32.) Plaintiffs filed a response with their own conditions. (DE 33.) The Court heard the parties’ arguments on November 2, 2022 and adopted a revised discovery schedule on November 3, 2022. (Electronic Orders dated Nov. 2-3, 2022.)
*2 The parties are still in the paper discovery phase. Plaintiffs served their First Set of Requests for the Production of Documents (Exhibit 1) on July 28, 2022 and their Second Set of Requests for the Production of Documents (Exhibit 2) on December 20, 2022. (DE 39 at 1.) As of April 17, 2022, when this dispute first arose, Defendants had already produced fourteen gigabytes of Microsoft Excel spreadsheets containing responsive data. (Id.) As for non-data documents, Plaintiffs noted that there were documents outstanding as to both sets of Plaintiffs document demands. (Id. at 2.) Plaintiffs served a Third Set of Requests for the Production of Documents on March 21, 2023. (Id. at 2 n.1.)
Since September 7, 2022, the parties have exchanged various iterations of keyword search term proposals for the review of Defendants custodians’ electronic mailboxes in an effort to begin the document review and production process. (DE 39 at 2.) However, the parties were unable to resolve the issue amongst themselves and filed a joint motion regarding their dispute on April 17, 2023. (DE 39.) District Judge Brian M. Cogan referred this motion to the undersigned “for determination and a recommendation to [him] for any adjustments to the current case schedule necessitated as a result of that determination.” (Electronic Order dated April 18, 2023.)
On April 26, 2023, the parties appeared before the undersigned for a motion hearing to discuss the discovery issues. (DE 40.) The parties were adamant that their respective proposals were the best path forward. Following discussion with counsel, the Court directed Defendants to conduct and complete a sample review of the 8,000 documents identified by the parties on or before May 12, 2023 and report the results of that sampling to Plaintiffs by that date. (DE 40.) The parties were further directed to meet and confer in a good faith effort to reach a compromise regarding the search parameters and terms that would lead to a document review population substantially below the 1.6 million document review proposed by Plaintiffs and substantially above the 166,156 proposed by Defendants. (DE 40.)
Moreover, the parties were directed to file a joint status report on or before May 19, 2023 as to the status of their meet and confer regarding search terms, including the production status of certain outstanding responsive non-email documents discussed and addressed in the parties’ joint motion. (DE 40.) Additionally, the parties were directed to proceed with the three 30(b)(6) witness depositions that were already scheduled for early May. (DE 40.) The Court reserved decision on motion (DE 39), which it construed as a motion to compel, and a cross-motion for a protective order. (DE 40.)
On May 19, 2023, the parties filed their anticipated status report. (DE 42.) Defendants reported that they had searched for and produced all non-email documents requested except any that may be attached to emails that are the subject of the parties’ keyword search proposal dispute. (DE 42.) Thus, that dispute was resolved. As to the keyword search terms, however, Defendants completed the sampling as directed, and the parties subsequently met and conferred. (DE 42.) Yet, disputes remained, and thus, each side provided further search term proposals for the Court's consideration. Plaintiffs revised terms pointed to an approximately 25% reduction from their original proposal, with a document review population of around 1.2 million. (DE 42.) Defendants’ revised terms pointed to an approximately 65% increase from their original proposal, with a document review population of around 463,932.8. (DE 42.)[1]
*3 In connection with their proposed terms, Plaintiffs urged that a three-to-four-month extension of time would be appropriate for Defendants to produce the 1.2 million documents. (DE 42.) Defendants in turn asked for an eight-month extension of all deadlines based on their proposed terms and stated that a review population of 1.2 million or anything close to that would require substantially more time. (DE 42.) Plaintiffs also noted that they had already conducted two 30(b)(6) depositions with a third scheduled for May 24, 2023. (Id.) Plaintiffs further requested that they continue to be able to take substantive depositions of Defendants’ employees without prejudice to their ability to conclude those depositions after substantial production of the email documents. (Id.) The Court allowed, and encouraged, depositions to continue. (Id.)
On May 24, 2023, the parties appeared before the undersigned once again for a status conference to discuss their revised search term proposals including the scope and timing of production and other deadlines. (DE 44.) There was extensive discussion regarding various options that could help further refine the document review population and assist Defendants in efficiently reviewing and producing the large volume of electronically stored information at issue. (DE 44.) The parties were subsequently directed to meet and confer to attempt to further narrow the scope of responsive documents based on that discussion. (DE 44.) The parties were to file independent letters addressing the parties’ respective positions, along with support, on the following topics:
The proposed keyword search terms and resulting document review population. Suggested revisions to the current discovery schedule based on the adoption of either party's revised proposed keyword search terms, along with a list of cases, if any, to support the suggested timetable for production in light of the volume to be produced. The propriety of cost-shifting, the use of a production and claw-back method or of producing documents after scanning and separately reviewing for privilege. The status of Defendants’ efforts to obtain approval for utilizing contract attorneys or a third-party vendor to assist with document review.
(DE 44.)
Subsequently, the parties filed their respective letters. (See DE 45; DE 46.) The parties’ most recent search term proposals, and respective positions on the aforementioned topics, are discussed more fully below. Thereafter, the parties also filed a second joint motion raising a dispute as to the procedure for conducting depositions given the pending document review. (DE 48.)
II. LEGAL STANDARD
Pursuant to Fed. R. Civ. P. 26:
Parties may obtain discovery regarding any nonprivileged matter that is relevant to any party's claim or defense and proportional to the needs of the case, considering the importance of the issues at stake in the action, the amount in controversy, the parties’ relative access to relevant information, the parties’ resources, the importance of the discovery in resolving the issues, and whether the burden or expense of the proposed discovery outweighs its likely benefit.
Fed. R. Civ. P. 26(b)(1).
Information “is relevant if: ‘(a) it has any tendency to make a fact more or less probable than it would be without the evidence; and (b) the fact is of consequence in determining the action.’ ” Vaigasi v. Solow Mgmt. Corp., No. 11-CV-5088, 2016 WL 616386, at *11 (S.D.N.Y. Feb. 16, 2016) (quoting Fed. R. Evid. 401). Moreover, “[t]he party seeking the discovery must make a prima facie showing that the discovery sought is more than merely a fishing expedition.” Evans v. Calise, No. 92-CV-8430, 1994 WL 185696, at *1 (S.D.N.Y. May 12, 1994); see also Mandell v. The Maxon Co., Inc., No. 06-CV-460, 2007 WL 3022552, at *1 (S.D.N.Y. Oct. 16, 2007) (“[T]he party seeking discovery bears the burden of initially showing relevance.”). To that end, the discovery sought by the parties must be, as stated by Rule 26, proportional to the needs of the case, taking into consideration such aspects as the importance of the issues, the amount in controversy, the parties’ resources and access to the information sought, and the importance of the information sought to the asserted claims or defenses. Sibley v. Choice Hotels Int'l, No. 14-CV-634 (JS)(AYS), 2015 WL 9413101, at *2–3 (E.D.N.Y. Dec. 22, 2015).
*4 Since December of 2015, “Rule 26 has defined the scope of discovery to consist of information that is relevant to the parties’ ‘claims and defenses.’ ” Pothen v. Stony Brook Univ., No. 13-CV-6170 (JFB)(AYS), 2017 WL 1025856, at *2 (E.D.N.Y. Mar. 15, 2017). “Thus, the discretionary authority to allow discovery of ‘any matter relevant to the subject matter involved in the action’ has been eliminated,” and permissible discovery under Rule 26 must be relevant “to any party's claim or defense,” and that means “proportional to the needs of the case.” Id. at *3 (citing Fed. R. Civ. P. 26(b)(1)). Proportionality goes “hand-in-hand” with relevance. New Falls Corp. v. Soni, No. 16-CV-6805 (ADS)(AKT), 2020 WL 2836787, at *2 (E.D.N.Y. May 29, 2020). That is, the more relevant the information sought is, the less likely the Court would find the subject discovery disproportionate. Id.
It is beyond peradventure that “[m]otions to compel are left to the court's sound discretion.” Mirra v. Jordan, No. 13-CV-5519, 2016 WL 889683, at *2 (S.D.N.Y. Feb. 23, 2016); see also Liberty Mut. Ins. Co. v. Kohler Co., No. 08-CV-867, 2010 WL 1930270, at *2 (E.D.N.Y. May 11, 2010) (“[A] motion to compel is entrusted to the sound discretion of the district court.”).
Parties seeking cover from discovery may avail themselves of a motion for a protective order which, in effect, is the flip side of a motion to compel. Rule 26(c) affords protection for abusive or embarrassing discovery, providing that “[a] party or any person from whom discovery is sought may move for a protective order in the court where the action is pending ... The court may, for good cause, issue an order to protect a party or person from annoyance, embarrassment, oppression, or undue burden or expense....” Fed. R. Civ. P. 26(c)(1); see Gordon v. Target Corp., 318 F.R.D. 242, 246 (E.D.N.Y. 2016) (“[T]he touchstone for determining whether to issue a protective order under Rule 26(c) lies, in the first instance, on a party's ability to establish good cause”). The burden is on the party seeking issuance of the order to show “good cause” through “particular and specific facts” as opposed to “conclusory assertions.” Rofail v. United States, 227 F.R.D. 53, 54–55 (E.D.N.Y. 2005).
“If the movant establishes good cause for protection, the court may balance the countervailing interests to determine whether to exercise discretion and grant the order.” Id. at 55. “Because of the interest in broad discovery, the party opposing the discovery of relevant information, whether through a privilege or protective order, bears the burden of showing that based on the balance of interests the information should not be disclosed.” Fowler-Washington v. City of New York, No. 19-CV-6590 (KAM)(JO), 2020 WL 5893817, at *3 (E.D.N.Y. Oct. 5, 2020) (internal quotation and citation omitted).
III. DISCUSSION
The parties’ primary dispute revolves around the collection and production of electronically stored information (“ESI”) from the electronic mailboxes of 24 custodians that the parties have agreed on. (DE 39 at 2.) More specifically, the parties disagree about the proposed keyword search terms that Defendants should use to locate responsive documents, and the appropriate time that Defendants should be afforded to review and produce those documents. The parties’ secondary dispute relates to the procedure for depositions in light of the pending electronic discovery dispute. Each of these issues are addressed below.
After the application of the deduplication technique of threading,[2] Plaintiffs’ latest proposed search terms yield a document review population of 1,065,251 documents, while Defendants’ latest proposed terms yield a document review population of 629,596 documents. (See DE 45; DE 46.) For the reasons that follow, the Court finds that Plaintiff's proposed terms are appropriate and proportionate.
*5 Defendants initially argued that Plaintiffs search terms were overbroad, and that Plaintiffs seek an unreasonable volume of documents, which is dissimilar to other cases against the State of New York requiring document productions by the Office of the New York State Attorney General (“OAG”). (DE 39.) Defendants pointed to M.G. v. Cuomo, No. 19-CV-0639 (S.D.N.Y. 2019), a class action against various New York State agencies and officials involving hundreds of individuals discharged from prison that require mental health housing upon release. (Id.) Defendants attached a transcript of the June 29, 2022 status conference in that action where Magistrate Judge Andrew E. Krause discussed a similar issue – keyword search terms -- with the parties in that case. (DE 39-11.) In M.G., defendants’ proposal involved a document review population of 211,000, and the plaintiffs’ proposal yielded a document review population of 500,000. (Id. at 4-5.)
Judge Krause noted that though the significance of the issues and complexity of the claims warranted a “substantial ESI production,” 500,000 was a “pretty extreme number.” (Id. 11 at 5-6.) Further, Judge Krause noted that 500,000 appeared “overly burdensome” but the parties should meet and confer to meet somewhere in the middle. (Id.) Additionally, Judge Krause made mention of the fact that “particularly for a government defendant” a large review like this simply will take a longer amount of time because of the difference in terms of resources and time the government can spend on a particular case so naturally such cases involve a tradeoff between the volume of documents to be received versus the speed they can be received. (Id. at 12.) The parties eventually reached agreement on a review population of slightly north of 300,000 documents approximately two weeks after the status conference. See M.G. v. Cuomo, 19-CV-0639 (S.D.N.Y. 2019), Dkt No. 228 (filed July 13, 2022).
In response to Defendants’ objections that the search terms themselves were overbroad, Plaintiffs proposed reviewing a random sample of about 8,000 documents total that would be derived from each of the search strings. (DE 39 at 3.) The idea was that after sampling, the parties could adjust the search terms depending on the percentage of responsive documents in each search string. (DE 39 at 3.) Defendants opposed this proposal on the grounds that it would delay discovery and was unnecessary to conclude that the document population universe was too large. (DE 39 at 3.)
The Court agreed that 1.6 million was simply too high. However, 166,156 was far too little especially since the concerns voiced by Defendants were not based in the relevance of the documents and instead were primarily grounded in the sheer volume of them. Following discussion with the parties at the April 26, 2023 status conference, the Court, over Defendants’ objection, ordered the sample review. (DE 40.) The parties were further directed to meet and confer in a good faith effort to reach a compromise regarding the search parameters and terms that would lead to a document review population substantially below the 1.6 million document review proposed by Plaintiffs and substantially above the 166,156 proposed by Defendants. (DE 40.)
Notably, the sampling revealed that on average, 72% of the documents identified by the search terms were responsive with a majority of the searches (16 searches) showing a responsive rate of greater than 80%. (See DE 42-1.) Nonetheless, Plaintiffs narrowed their proposed search terms even further, eliminating some search strings and tightening up others. (See DE 42-2.) Defendants’ seemingly post-hac justification for this high responsive rate is that Plaintiffs’ document requests themselves are too broad to start with – significantly, an objection not previously asserted, and thus, insufficient to bar production now (DE 42 at 8). See Cohalan v. Genie Indus., Inc., 276 F.R.D. 161, 163 (S.D.N.Y. 2011) (noting the well-settled principle that “[a] failure to respond or object to a discovery request in a timely manner waives any objection which may have been available.” (quoting UBS International Inc. v. Itete Brasil Instalacoes Telefonicas Ltd., No. 09-CV-4286 (LAK), 2010 WL 743371, at *3 (S.D.N.Y. Feb. 24, 2010)); see also Labarbera v. Absolute Trucking, Inc., No. 08-CV-4581 (DRH) (AKT), 2009 WL 2496463, at *1 (E.D.N.Y. Aug. 12, 2009) (“It is well established that by failing to respond or object to a discovery request in a timely manner, a party waives any objection which may have been available.”)
*6 Following the May 24, 2023 status conferences, the parties continued to meet and confer and narrow search terms. The parties’ latest proposals are as follows. Plaintiffs’ proposal yields 1,065,251 documents. (DE 46 at 1.) Defendants’ proposal yields 629,596 documents. (DE 45 at 2.) First, Plaintiffs point out that a review of Defendants’ documents evinces that many words and phrases are used in varying combinations such that narrower search terms would exclude far too many responsive documents. (DE 42 at 2-3.)
Though Defendants generally argue that the search terms are too broad, Plaintiffs persuasively point to a sample of documents that appear to contain highly relevant information that were picked up by Plaintiffs’ proposed search terms but would not have been picked up if Defendants’ proposed search terms were utilized. (DE 42 at 2-3.) Essentially, using Defendants’ proposed search terms creates a high risk that documents material to Plaintiffs’ claims or defense will evade discovery. But Defendants fail to meaningfully explain away the discrepancy between the parties’ terms in locating highly responsive documents.
Second, Defendants argue that the production of “every” responsive document generated by “overly broad search terms” is not proportional. (DE 45 at 2.) However, Plaintiffs have reduced their document review population by hundreds of thousands of documents such that they are certainly not seeking “every” responsive document. (DE 42; DE 45.) Additionally, the discovery sought is proportional even by comparison to the cases Defendants rely on. For example, this case involves a significantly larger putative class than in M.G. where the Court ordered discovery of 300,000 plus documents. Compare Second Am. Compl., Dkt No. 134 at ¶¶ 695-699, M.G. v. Cuomo, No. 19-CV-0639 (S.D.N.Y. 2019), (filed Aug. 7, 2020) (discussing the “hundreds” of potential class members) with (DE 34 at ¶165) (noting that a projected “200,000 children would be eligible for CFTSS Services”).
Given the undisputed importance of the issues at stake in this action, the relevance of the documents, the high responsive rates, the wide variety of terms used in the documents for the same or similar concepts, and the likelihood that highly relevant documents would escape review if Defendants proposed search terms were applied -- the Court finds that Plaintiffs proposed search terms (DE 46-1) are appropriate.
Accordingly, Defendants are directed to utilize and apply Plaintiffs’ most recent keyword search term proposal (DE 46-1) to the documents of the 24 custodians to create the document review population.
The parties’ discovery dispute was referred to the undersigned for determination, but Judge Cogan's referral also contemplated that the undersigned provide a recommendation to him “for any adjustments to the current case schedule necessitated as a result of that determination.” (Electronic Order dated Apr. 18, 2023.) The current discovery schedule is as follows: (a) fact depositions to be completed on or before July 12, 2023; (b) initial expert reports and class certification motion to be served on or before August 14, 2023; (c) opposing expert reports and opposition to class certification to be served on or before September 29, 2023; (d) reply expert reports and reply in support of class certification to be served on or before October 30, 2023; (e) all discovery to be completed on or before November 13, 2023. (Electronic Order dated Nov. 3, 2022; DE 36.)
Defendants request nine months to complete document discovery and for the remaining deadlines to be “similarly extended” depending on the amount of time the Court recommends for document review to be completed. (DE 45 at 3-5.) Plaintiffs propose six months for Defendants to complete document discovery but propose varying extensions to the other dates. (DE 46 at 9-10.) Plaintiffs aver that depositions can continue concurrently with Defendants’ review and rolling productions of documents. (Id.) Plaintiffs propose a two-month extension to certain deadlines and an additional third month for completing discovery to account for any remaining depositions. (Id.) Defendants oppose the idea of having secondary depositions. (DE 48.)
*7 The Court afforded the parties an opportunity to provide suggested revisions to the current discovery schedule based on the adoption of either party's revised proposed keyword search terms, along with a list of cases, if any, to support the suggested timetable for production in light of the volume to be produced. (DE 44.) Plaintiffs point to cases where parties have produced more than the 1.2 million documents at issue here in time periods shorter than or comparable to 6 months. See generally Insignia Sys., Inc. v. News Am. Mktg. In-Store, Inc., No. 4-CV-4213 (JRT)(AJB), 2010 WL 11534597, at *4 (D. Minn. Feb. 4, 2010) (“seven million documents” produced by the end of the “20 months” provided for discovery); Ambac Assurance Corp. v. EMC Mortg. Corp., No. 08-CV-9464 (RMB)(THK), 2010 WL 11595698, at *7 (S.D.N.Y. Dec. 16, 2010), report and recommendation adopted in part sub nom. Ambac Assur. Corp. v. EMC Mortg. Corp., No. 08-CV-9464 (RMB)(THK), 2011 WL 566776 (S.D.N.Y. Feb. 8, 2011) (“millions of documents” produced in nine-months); In re Pfizer Inc. S'holder Derivative Litig., 780 F. Supp. 2d 336, 338 (S.D.N.Y. 2011) (over “12 million pages of documents” produced in six-months (emphasis added)); Millennium Labs, Inc. v. Allied World Assur. Co., No. 12-CV-2280 (BAS)(KSC), 2014 WL 6632765, at *2 (S.D. Cal. Nov. 21, 2014) (“1.1 million documents” produced in approximately 6-months).[3] Thus, there is certainly support for the proposition that 1.2 million documents can be produced within a 6-month period.
Defendants on the other hand, cherry pick and point to cases in which the State of New York was a producing party and where it took the Office of the Attorney General longer than 6 months to produce less documents than the 1.2 million documents at issue here. See generally Elisa W. v. City of New York., No. 15-CV-5273 (S.D.N.Y. filed Jul 8, 2015), Dkt. 393 (233,313 documents produced in approximately 11 months); Bagley v. N.Y.S. Dep't of Health, No. 15-CV-4845 (FB)(CLP) (E.D.N.Y. filed Aug. 18, 2015), Dkt. Nos. 66, 93 (173,752 documents produced in approximately 11 months); Fair Housing Justice Center, Inc. v. Cuomo, No. 18-CV-3196 (S.D.N.Y. filed Apr. 11, 2018) (28,786 documents produced in approximately 6 months); Ciaramella et al. v. Zucker, No. 18-CV-6945 (S.D.N.Y. filed Aug. 2, 2018), Dkt Nos. 177, 181 (23,094 documents produced in approximately 5 months); M.G. v. Cuomo, No. 19-CV-0639 (S.D.N.Y. filed Jan. 23, 2019) (337,804 documents produced in approximately eight months).
Given the numerous delays in discovery to date, the Court finds that time frames and approach set forth in Plaintiff's proposed schedule are the most flexible, efficient, and practical approach to completing discovery in light of the volume of documents to be reviewed. Plaintiff's proposed schedule allows the discovery process to move forward while rolling productions of electronic discovery are being made. Notably, Plaintiffs’ proposed schedule generally disturbs the discovery schedule in a less drastic manner than Defendants’ proposal. The Court is mindful that the largest production cited by the OAG was 300,000 plus documents, and the court granted approximately eight months to make that production. However, in today's world of ESI discovery, as noted above, six months has been ample time for other litigants to produce a far greater volume of documents than at issue here. Given the advances in electronic discovery technology and the general availability of other options to expedite document review, the AOG will need to adapt as all other litigants have been doing. See infra III.C. (exploring the various methods of expediting electronic discovery and decreasing expenses).
*8 Moreover, depositions shall occur concurrently with the rolling document productions. This topic was first raised at the April 26, 2023 status conference when Plaintiffs informed the Court that they intended to move forward with depositions irrespective of the electronic discovery production delays. The Court encouraged this approach, and no objection was raised. (See DE 40 (“The parties will proceed with the three 30(b)(6) witness depositions as scheduled in early May.”).) Plaintiffs noted that they have already served six deposition notices regarding individuals identified in Defendants’ initial disclosures. (DE 48 at 1.) Plaintiffs seek to compel these depositions before the current July 12, 2023 deadline. (DE 48.) Defendants on the other hand do not oppose depositions commencing alongside document production, however, they question the wisdom in having witnesses appear for dual depositions if it can be avoided. (DE 48.)
Defendants lay out a sensible approach of having the documents of the noticed witnesses reviewed and produced first, and then commencing then depositions. The Court does not find this would unduly delay discovery. Thus, Defendants should prioritize reviewing the documents of the witnesses that have been noticed for deposition. Once the documents of a given custodian that has been noticed for deposition are reviewed, the parties are to arrange for the deposition to be scheduled forthwith.
Having reviewed the parties’ proffered authority, and considered the parties’ arguments and briefing, the Court respectfully recommends to Judge Cogan that the discovery schedule be adjusted as follows:

Defendants aver that discovery will take ages and exceed their current budget. (DE 45 at 2.) However, during the two status conferences, the Court explored a variety of methods with the parties that may expedite Defendants’ review and production of documents, and/or decrease the expense burden. (See DE 40; DE 44.) First, the Court asked Defendants whether they have access to and can use Technology Assisted Review (“TAR”) including but not limited to “predictive coding” and “continuous active learning,” or other electronic discovery resources to reduce either the time or manpower required for the review. Defendants simply advised that in their view, these tools have not proved to be successful in narrowing the process in similar types of policy cases.
Nonetheless, TAR is often helpful in reducing the burden of electronic discovery. See Rio Tinto PLC v. Vale S.A., 306 F.R.D. 125, 127 (S.D.N.Y. 2015) (“In fact, we understand that the technology industry now considers predictive coding to be widely accepted for limiting e-discovery to relevant documents and effecting discovery of ESI without an undue burden.”) (quoting Dynamo Holdings Ltd. P'Ship v. Comm'r of Internal Revenue, 143 T.C. 9, 2014 WL 4636526 (T.C. Sept. 17, 2014)). Especially in cases involving copious amounts of data. See Moore v. Publicis Groupe, 287 F.R.D. 182, 193 (S.D.N.Y. 2012) (“What the Bar should take away from this Opinion is that computer-assisted review is an available tool and should be seriously considered for use in large-data-volume cases where it may save the producing party (or both parties) significant amounts of legal fees in document review.”). Thus, Defendants should reconsider whether the use of TAR may in fact expedite the review considering the volume of documents to now be reviewed.
Second, the parties discussed whether a claw-back procedure pursuant to the Federal Rules of Evidence Rule 502(d) may substantially decrease the time required to conduct the review. See Fed. R. Evid. 502(d) (“A Federal court may order that the privilege or protection is not waived by disclosure connected with the litigation pending before the court—in which event the disclosure is also not a waiver in any other Federal or State proceeding.”). The parties passionately disagree as to the propriety of producing documents without having eyes on every single document for privilege review before it goes out the door. (DE 45; DE 46.)
*9 Plaintiffs have suggested that Defendants can, for example, run their documents through privilege search terms and produce the remainder under a claw-back agreement such that all privileged or non-responsive documents will be returned. (DE 46 at 7.) Defendants can include attorneys’ names, and law firms, and other likely privileged terms in the search to weed out highly privileged material. That is non-starter for Defendants, who do not believe that a claw-back approach to production would save any significant time based on their review of the sampling and aver that privilege search terms and/or attorneys’ names would not suffice to replace manual privilege review. (DE 45 at 5-6.) Moreover, Defendants stress that this approach would undermine attorney-client privilege and lead to adverse results because once Plaintiffs get their eyes on a privilege document, even if they return the document, the information contained within may influence the course of this litigation or other litigations. (Id.) In short, the proposed method in their opinion is too risky.
The Court notes that “[i]ndeed, many parties to document-intensive litigation enter into so-called ‘claw-back’ agreements that allow the parties to forego privilege review altogether in favor of an agreement to return inadvertently produced privileged documents.” Zubulake v. UBS Warburg LLC, 216 F.R.D. 280, 290 (S.D.N.Y. 2003). Plaintiffs point out, as the Court has previously noted, these kinds of claw-back procedures have been ordered in cases involving a handsome volume of electronic discovery. (See DE 46 at 7 (collecting cases).) A similar claw-back agreement in this case would undoubtedly move the process along.
Though the Court is not inclined to direct the parties to enter into such a claw-back agreement -- which effectively compels the parties to produce privileged documents with the right to claw-back -- Defendants are nevertheless free to consider this option and weigh its utility in helping them meet their production obligations. See, eg., Capitol Recs., Inc. v. MP3tunes, LLC, 261 F.R.D. 44, 51 (S.D.N.Y. 2009) (“[I]f [defendant] wishes to ensure that it is able to make a timely production despite the need to review its emails for privilege issues, it remains free to confer with the Plaintiffs with respect to a clawback agreement to be ‘so ordered’ by the Court.”).
Third, the Court inquired about the use of contract attorneys to help expedite the review. The use of contract attorneys is undoubtedly a common practice in the legal industry to help efficiently tackle large document reviews. Cf. In re WorldCom, Inc. Sec. Litig., No. 02-CV-3288 (DLC), 2004 WL 2591402, at *21 n.48 (S.D.N.Y. Nov. 12, 2004) (“The extensive use of contract attorneys was justified by the need to review over ten million pages of documents and was a far more efficient way of proceeding than giving the task to more highly compensated counsel.”). On this front, Defendants noted that they have taken “proactive measures to obtain approval for contract attorneys” and have started constructing a document review protocol for anticipated contract attorneys to use to assist in document review. (DE 45 at 6.) Defendants represent that upon finalization of the search terms, they intend to “move expeditiously” to finalize the document review protocol and secure the necessary approvals to hire contract attorneys including budgetary approval. (DE 45 at 6.)
These are among the myriad of options available to the parties to ensure the review and production of the subject documents occurs within a reasonable timeframe. The Court leaves it to the parties to confer and work out mutually desirable methods of achieving their presumably common goal of completing document discovery.
The Court declines to order cost-shifting for the review and production of the subject documents. The status quo is that the producing party bears the financial burden of production even where electronic evidence is involved. See Zubulake v. UBS Warburg LLC, 217 F.R.D. 309, 317 (S.D.N.Y. 2003) (noting that there is a “presumption [ ] that the responding party must bear the expense of complying with discovery requests” and “[a]ny principled approach to electronic evidence must respect this presumption.”). Cost-shifting is a mechanism that is typically appropriate only where production of electronic discovery imposes an undue expense or burden, which, in turn primarily depends on whether the electronically stored information is inaccessible. See id. at 324. (“A court should consider cost-shifting only when electronic data is relatively inaccessible, such as in backup tapes.”); Bailey v. Brookdale Univ. Hosp. Med. Ctr., No. 16-CV-2195 (ADS) (AKT), 2017 WL 2616957, at *4 (E.D.N.Y. June 16, 2017) (noting that the Zubalke factors should be considered after “a threshold determination is made that the electronic data sought is ‘relatively inaccessible’ ”).
*10 “Accessible data is stored in a readily usable format that does not need to be restored or otherwise manipulated to be usable.” Est. of Shaw v. Marcus, No. 14-CIV-3849 (NSR) (JCM), 2017 WL 825317, at *3 (S.D.N.Y. Mar. 1, 2017) (internal quotation marks omitted). Whereas data is considered inaccessible when it “is not readily usable and must be restored to an accessible state before the data is usable.” Id. Accessibility is indisputably not an issue in this case. The relevant documents from the 24 custodians have already been gathered, the only issue that remains is what terms will be applied to the collected documents to create the document review population. Thus, although the undersigned raised the issue of cost-shifting and invited briefing, upon consideration, cost-shifting is not warranted under these circumstances.
CONCLUSION
For the reasons set forth herein, Plaintiffs’ motion to compel electronic discovery (DE 39) and Defendants’ cross-motion for a protective order (DE 39) are DENIED in part and GRANTED in part. Plaintiffs’ motion to compel depositions (DE 48), and Defendants’ cross-motion for a protective order (DE 48) are DENIED in part and GRANTED in part.
Defendants are directed to utilize and apply Plaintiffs’ most recent keyword search term proposal (DE 46-1) to the documents of the 24 custodians to create the document review population. The parties are directed to continue with depositions and Defendants shall prioritize reviewing the documents of custodians noticed for deposition.
Further, the undersigned respectfully recommends that the discovery schedule be amended as noted above. See supra III.B.

Footnotes

These figures were calculated based on the parties’ expected reduction of approximately 20% based on deduplication and threading. (DE 42 at 4) (plaintiffs projected 1.2 million after deduplication and threading); DE 42 at 8 (defendants projected 579,916 documents with an expected reduction of 20% after deduplication and threading).
Threading, as envisioned by the parties, is the process of “reviewing and producing only the most inclusive version of each email thread.” (DE 42 at 8.)
The parties collected the document volume and production information from a combination of various docket filings. Defendants even note that some of these numbers are extrapolated from a combination of docket filings, the Defendants’ supporting declaration, and records maintained by Defendants. (DE 45 at 3 n.3.) Thus, the Court incorporates by reference the citations in the parties’ letters. (DE 45; DE 46.)