Kabelis v. NCL (Bahamas) LTD.

UNITED STATES DISTRICT COURT SOUTHERN DISTRICT OF FLORIDA MIAMI DIVISION CASE NO. 20-21430-CIV-ALTONAGA/GOODMAN ELISA M. KABELIS, Plaintiff, v. NCL (BAHAMAS) LTD., Defendant. __________________________/ ORDER RECONSIDERING, IN PART, INITIAL COSTS-SHIFTING DISCOVERY RULING This Order is a follow-up ruling to the Undersigned’s Order [ECF No. 60] granting in part and denying in part Plaintiff’s Motion for Reconsideration concerning a discovery ruling [ECF No. 46] which authorized Plaintiff to obtain fleetwide discovery on prior incidents for three years. That earlier discovery ruling, entered after Defendant NCL (Bahamas) Ltd. submitted a declaration explaining the logistical burdens and costs associated with a fleetwide probe in the midst of a global pandemic, required Plaintiff to pay for half of the costs that NCL incurs for an outside vendor. Plaintiff Elisa Kabelis filed a motion for reconsideration, arguing that (1) the Undersigned did not expressly analyze the factors which are sometimes used to determine the propriety of requiring costshifting and (2) that she was being forced to make an uninformed decision because NCL had not provided specificity about the search-related costs. Because Plaintiff was correct about those two points, the Undersigned required NCL to submit a more-specific estimate of the costs and advised that a follow-up ruling would then be issued. [ECF No. 60]. NCL submitted the additional information [ECF No. 61] and this Order is the follow-up ruling. For the reasons outlined in greater detail below, the Undersigned modifies the initial cost-shifting ruling. Although the producing party typically absorbs the cost of production in discovery, this is not an iron-clad rule. Costs may be shifted, in whole or in part, depending on myriad circumstances. Here, NCL’s ability to locate, search, and analyze data which might arguably be responsive to discovery has been hampered by the global pandemic. NCL is operating with a significantly reduced staff, and its legal department has been similarly streamlined. Moreover, the relevance of the fleetwide discovery may not be as strong as Plaintiff urges, as the location of automatic external defibrillators (“AED”) on other ships may be different that the location on the ship at issue, especially if the vessels have dissimilar configurations. Among other allegations of negligence, Plaintiff’s lawsuit challenges NCL’s placement and use of AEDs on the NCL Joy, the cruise ship on which her husband died after collapsing in an onboard steak restaurant. Therefore, if Plaintiff wants to pursue fleetwide discovery of the prior incidents, then she will be responsible for one-quarter (i.e. 25%) of the costs generated by an outside counsel review of the materials. NCL estimates these costs to be between $8,410 and $12,615. [ECF No. 61]. This is a change from the earlier ruling, which imposed a 50% financial responsibility on Plaintiff for fleetwide discovery. I. Factual and Procedural Background Kabelis is the widow and survivor of Joseph Kabelis, and she is the personal representative of his estate. Her First Amended Complaint [ECF No. 20] alleges the following: 8. On September 21, 2019, the decedent, JOSEPH KABELIS, was a passenger on the NCL Joy. 9. On that date, JOSEPH KABELIS lost consciousness and collapsed in the Cagney Steakhouse restaurant on the NCL Joy. 10. Another passenger who was a doctor and/or other medical professional saw this happen, and immediately began administering emergency care, and said to the cruise ship staff that [a] defibrillator was needed, but the crew continued to serve dinner in [the] restaurant. 11. An emergency code was not called and/or issued until an unreasonably long amount of time after the incident, and the ship doctor and/or ship medical staff did not arrive until an unreasonably long amount of time since Decedent first showed signs of having a seizure or otherwise being in acute distress. 12. Upon information and belief, the ship’s staff (medical and non-medical) was not adequately prepared and/or set up to provide first aid, treat, and/or otherwise adequately help JOSEPH KABELIS. 13. Furthermore, the ship’s crew did not have the area adequately cleared of passengers other than passengers who were helping. 14. JOSEPH KABELIS was subsequently pronounced dead after the ship’s staff failed to timely bring a reasonable or adequate defibrillator to him and failed to reasonably or adequately use it to resuscitate him. Id. at pp. 2-3 (emphasis added). Plaintiff’s primary theory is that NCL did not timely respond to the incident (i.e., Plaintiff’s decedent lost consciousness and collapsed in an onboard restaurant) and did not timely use an AED. The Discovery Order permits discovery about prior use of, or requests for, AEDs on NCL ships or concerning onboard medical emergencies involving a passenger’s death. The Undersigned had a telephone discovery hearing on January 15, 2021. [ECF No. 44]. It lasted almost an hour and a half. [ECF No. 44]. Certain discovery rulings were made. The Undersigned scheduled a follow-up discovery hearing for January 19, 2021, at 1:30 p.m. [ECF No. 45]. At 12:30 p.m. on January 19, 2021, NCL submitted to my e-file inbox1 the Declaration of Brett Berman. NCL also served a copy by email to Plaintiff’s counsel at the same time. The parties (and the Court) discussed the Berman declaration at the January 19, 2021 discovery hearing, which lasted an hour and fifty minutes. a. The Berman Declaration and the Related Ruling Berman is the Senior Director of Passenger and Crew Claims for NCL. Although his declaration [ECF No. 54-1] does not specifically disclose that he is a licensed Florida attorney, the Undersigned is well aware of this, as he has filed other declarations in other lawsuits involving NCL. Among other points, the declaration explains that (1) NCL’s database does not 1 The Undersigned directs parties to submit discovery materials to the e-file inbox, as opposed to publicly filing them on CM/ECF. permit it to search medical records for keywords such as AED or defibrillator; (2) using certain medical diagnoses used as search terms, 1,737 cases were identified; (3) it would likely take between 58 and 87 hours to conduct the search; (4) in practical terms, it would take one person (working on nothing else for eight hours a day) approximately two weeks to complete the search; and (5) because of the pandemic, NCL is operating with limited staff and resources. More specifically, the Berman declaration further noted that its Seacare system (i.e., NCL’s medical records database) is not designed for responding to broad requests for specified historical information. The declaration also explained that the inquiry triggered by Plaintiff’s discovery request would require the exportation of every single medical record within the Seacare databases into a separate spreadsheet, and then sorting the spreadsheet by diagnosis. According to the Berman declaration, searching by diagnosis appears to be the most accurate way to find information which might reasonably uncover information about when an AED was used. In addition, the information would also need to be subjected to detailed analysis in order to provide it in a readily usable format. Based on the Berman declaration, I reconsidered my earlier discovery ruling and gave Plaintiff the power to select one of three options. Only the third option involved a cost-shifting mechanism -- the one giving Plaintiff the ability to obtain fleetwide discovery for three years. Had Plaintiff selected this option, the Order requires Plaintiff to pay for half of the cost NCL incurs from an outside vendor. Plaintiff contends that this ruling is clearly erroneous. Specifically, Plaintiff notes that the Order did not expressly analyze the seven Zubulake2 factors, which means she was being “asked to make an uninformed decision,” because NCL was not required to provide an estimated cost from its outside vendor before choosing one of the three discovery options made available to her. After evaluating Kabelis’ position, I determined that I needed to receive a specific estimate from NCL’s vendor or a reasonable dollar figure estimate for the time it would take an NCL employee to conduct the search. I therefore directed NCL to provide a more- specific prediction of how much a vendor would charge for the project and how much it would cost an in-house NCL staffer to do the required work. The Undersigned also advised that the seven-factor analysis would be undertaken after NCL submitted the additional information. The Undersigned further gave notice that the analysis would result in an appropriate ruling on whether a cost-shifting award is appropriate, and, if so, to what extent. b. NCL’s Additional Information NCL provided information on three options: (1) an outside vendor would conduct the fleetwide discovery assessment; (2) a paralegal from NCL’s outside law firm would assess the fleetwide discovery, with supervision by an associate; and (3) an NCL employee would do the assessment. NCL’s alternatives are listed below: 2 Zubulake v. UBS Warburg LLC, 217 F.R.D. 309, 322-23 (S.D.N.Y. 2003) (“Zubulake I”). Scenario #1: Assessment by Outside Vendor Performing party: DISCO Vendor contact: Dalton Cade, Account Executive Vendor quote: Approximately $35,000 Metrics used in quantifying: Approximately 1,737 separate case files to be assessed and it takes approximately two (2) to three (3) minutes to determine whether responsive information is contained within each case. The estimated total search time would take between 58 and 87 hours to complete. This cost quote includes the additional time needed to train an outside vendor on how to properly and efficiently use Defendant’s internal medical records database contained within the Seacare system. Scenario #2: Assessment by Defendant’s counsel through paralegal review with associate oversight Performing party: Cole, Scott & Kissane P.A. Estimated cost: $8,410 (low/58 hours) and $12,615 (high/87 hours). Metrics used in quantifying: In light of the costs set forth above by the vendor, the Defendant has provided the Court with another analysis discussing the estimated costs associated with the requested fleet-wide discovery project assuming the undersigned’s firm conducted the review. Approximately 1,737 separate case files to be assessed and it takes approximately two (2) to three (3) minutes to determine whether responsive information is contained within each case. The estimated total search time would take between 58 and 87 hours to complete. Assuming a paralegal would be handling the extensive review with an associate attorney overseeing the work, we estimate that it would cost between $8,410 (low/58 hours) and $12,615 (high/87 hours). Since this document will be filed with the Court and thus a matter of public record, the undersigned did not disclose the respective rates for associate attorneys and paralegals. Scenario #3: Assessment by NCL in-house staffer Performing party: In-house staff from NCL Estimated cost: Unknown at this time Metrics used in quantifying: Approximately 1,737 separate case files to be assessed and it takes approximately two (2) to three (3) minutes to determine whether responsive information is contained within each case. The estimated total search time would take between 58 and 87 hours to complete. Since the Defendant is operating with limited resources due to the effects of Covid-19, it is capable of diverting only a single employee to the handling of this task. Based on the abovementioned temporal scope, this task would take approximately two (2) weeks to complete and require diverting the complete attention of an NCL employee towards this task for 8 hours a day, further depriving the Defendant of its already limited resources. [ECF No. 64, pp. 3-4 (emphasis added)]. II. Applicable Legal Principles and Analysis There is a presumption that “the responding party must bear the expense of complying with discovery requests.” Oppenheimer Fund, Inc. v. Sanders, 437 U.S. 340, 358 (1978). However, under Rule 26(c), a district court may issue an order protecting the responding party from undue burden or expense by “conditioning discovery on the requesting party’s payment of the costs of discovery.” Id. at 358 ; see also Zubulake v. UBS Warburg LLC, 216 F.R.D. 280, 283 (S.D.N.Y. 2003) (“Zubulake III”). Further, “the responding party has the burden of proof on a motion for cost-shifting.” Id.; see also Wiginton v. CB Richard Ellis, Inc., 229 F.R.D. 568, 573 (N.D. Ill. 2004). See generally Quinby v. WestLB AG, 245 F.R.D. 94, 101 (S.D.N.Y. 2006). Cost-shifting should be considered “only when electronic discovery imposes an ‘undue burden or expense on the responding party.’” Zubulake I, 217 F.R.D. at 318 (emphasis omitted) (“[W]hether production of documents is unduly burdensome or expensive turns primarily on whether it is kept in an accessible or inaccessible format (a distinction that corresponds closely to the expense of production).” Data that is “accessible” is stored in a readily usable format that “does need to be restored or otherwise manipulated to be usable.” Id. at 320 . Conversely, data that is “inaccessible” is not readily useable and must be restored to an accessible state before the data is usable. Id. Backup tapes are considered an inaccessible format, and, thus, shifting the costs of producing data from backup tapes may be considered. Id. As outlined in Zubulake I, 217 F.R.D. at 322 , if the responding party is producing data from inaccessible sources, a seven-factor analysis is considered to determine whether shifting the cost of production is appropriate: 1. The extent to which the request is specifically tailored to discover relevant information; 2. The availability of such information from other sources; 3. The total costs of production, compared to the amount in controversy; 4. The total costs of production, compared to the resources available to each party; 5. The relative ability of each party to control costs and its incentive to do so; 6. The importance of the issues at stake in the litigation; and 7. The relative benefits to the parties of obtaining the information. The Zubulake I Court weighed the factors in descending order, the first factor being the most important consideration and the seventh factor the least important. Id. at 323 ; see also Quinby, 245 F.R.D. at 102 (explaining Zubulake I). Florida district courts have followed Zubulake I and have recognized that courts have “the discretion, however, to shift all or part of the costs of production to the requesting party.” Classic Soft Trim, Inc. v. Albert, No. 6:18-cv-1237, 2020 WL 6734369 , at *9 (M.D. Fla. June 15, 2020); see also FDIC v. Brudnicki, 291 F.R.D. 669, 676 (N.D. Fla. 2013) (stating courts have held that Rule 26(c) permits cost shifting as part of enforcing proportionality limits). Cf. Peskoff v. Faber, 251 F.R.D. 59, 61 (D.D.C. 2008) (finding court has discretion to shift all or part of costs to requesting party). See generally Fed. R. Civ. P. 26(b)(2)(B) (“The court may specify conditions for the discovery.”); see also Stewart v. Cont’l Cas. Co., No. 12-00532, 2104 WL 12600282 (S.D. Ala. Jan. 9, 2014). The Undersigned will assess the appropriateness of any cost-shifting measures for a fleetwide search of data by engaging in the seven-factor methodology. Before doing so, however, the Undersigned notes that “we do not just up add the factors.” Zubulake I, 217 F.R.D. at 322 . That is because the “central question” of any cost-shifting evaluation is “does the request impose an ‘undue burden or expense’ on the responding party.” Id. at 322-23 . Or, as the Zubulake I Court explained, another way of phrasing the central question is “how important is the sought-after evidence in comparison to the cost of production?” Id. Although the Zubulake I Court established the seven factors, it also noted that “the test cannot be mechanically applied at the risk of losing sight of its purpose.” Id. The first two of the seven factors are the most important. On the other hand, the sixth factor -- the importance of the litigation itself -- “will only rarely come into play.” Id. at 323 . When it does, however, it “has the potential to predominate over the others.” Id. Moreover, the “absolute wealth of the parties is not the relevant factor.” Id. at 321 . Instead, “the focus should be on the total cost of production as compared to the resources available to each party.” Id. at 321 . Framed by these overarching considerations, the Undersigned’s assessment is below: i. Extent to Which the Request is Specifically Tailored to Discover Relevant Information Plaintiff’s requests are not entirely tailored to discover relevant information. As briefly noted earlier in this Order, fleetwide discovery of prior incidents is necessarily likely to be overbroad, at least in part, because the structure and design of other nonsister ships in the entire fleet will yield information inapplicable to the instant ship. On the other hand, some of the prior incidents might concern ships with similar designs, which means the placement of AEDs on other ships could have a bearing on the instant case. These factors slightly tip the scale in favor of some cost-shifting. ii. The Availability of Such Information from Other Sources It is unlikely that Plaintiff would be able to obtain information about delays in providing AEDs to other passengers from any other source. This factor weighs against cost-shifting. iii. The Total Cost of Production Compared to the Amount in Controversy This is a wrongful death lawsuit and the damages may well turn out to be substantial. RCCL estimates that the second alternative is roughly between approximately $8,500 and $12,500. This expense is “a drop in the bucket” for a defendant like RCCL. This factor favors Plaintiff, and it cuts against cost-sharing. iv. The Total Cost of Production Compared to the Resources Available to Each Party The pandemic has caused RCCL to operate with limited resources. If it decided to use an in-house employee for the project, RCCL would need to monopolize the schedule of one employee for two weeks. This factor is, for all practical purposes, a wash -- it is neutral and does not favor either party. v. The Relative Ability of Each Party to Control Costs and Its Incentive to Do So As explained by RCCL, the information requested by Plaintiff is not maintained in readily usable format. RCCL cannot merely search its database with the “AED” search term and locate prior incidents in which a passenger needed or requested an AED. The Undersigned already ordered RCCL to, at its own expense, provide similar incident history materials for the subject ship for three years or fleetwide for three years. Instead of accepting one of those alternatives, Plaintiff is pushing for fleetwide discovery -- but does not want to pay any part of the expense. Factor five tips in favor of cost-sharing. vi. The Importance of the Issues at Stake in the Litigation There is nothing necessarily unique about this wrongful death claim, and the Undersigned will adopt the Zubulake I comment about this factor usually being neutral. vii. The Relative Benefits of Obtaining the Information The fleetwide discovery at issue here would not provide direct or circumstantial evidence of the critical allegations concerning what happened on the subject ship on the day of the incident. However, the fleetwide discovery could (though not necessarily) generate discovery of constructive notice of a dangerous condition -- i.e., an inadequate number of AEDs on the ship or the strategically problematic locations of the AEDs (e.g., not close enough to a popular onboard restaurant). The Undersigned considers this factor to be substantially speculative at this point, and I therefore classify it as a neutral factor. Given the assessment above, the Undersigned concludes that a modest amount of cost-sharing is appropriate should Plaintiff decide to seek fleetwide discovery of information difficult to locate during the midst of a pandemic. III. Conclusion In ordinary, pre-pandemic times, the Undersigned would likely not have considered a cost-sharing provision for the fleetwide discovery provided here. But the pandemic has hit the cruise ship industry particularly hard, leading to RCCL’s need to significantly streamline its staff and legal department employees. Before the pandemic, RCCL could have spared an in-house employee to conduct the search and assessment necessary to locate the requested discovery. That is no longer the situation. Evaluating the seven factors, the Undersigned will require Plaintiff to absorb 25% of the cost for the second alternative should she decide to pursue the fleetwide discovery. If Plaintiff obtains the fleetwide discovery and uncovers discovery from incidents on similarly constructed ships in which AEDs were not timely used or were not close enough to the need, then the expense of $2,000 to $3,000 would likely be a strategically wise expenditure. Plaintiff may, of course, still pursue Objections to this discovery ruling. But magistrate judges are afforded “broad discretion” in discovery disputes. Johnston v. Aetna Life Ins. Co., 282 F. Supp. 3d 1303, 1312 (S.D. Fla. 2017). A district judge may modify or set aside a magistrate judge’s discovery rulings which are clearly erroneous or contrary to law, a standard which is “extremely deferential.” Sun Capital Partners, Inc. v. Twin City Fire Ins. Co., Inc., No. 12-CV-81397, 2015 WL 11921411 , at *1 (S.D. Fla. July 6, 2015) (internal quotation omitted). “A finding is clearly erroneous only if ‘the reviewing court, after assessing the evidence in its entirety, is left with a definite and firm conviction that a mistake has been committed.’” Id. (quoting Krys v. Lufthansa German Airlines, 119 F.3d 1515, 1523 (11th Cir. 1997)). Or, as the Seventh Circuit has put it: “[t]o be clearly erroneous, a decision must strike us as more than just maybe or probably wrong; it must . . . strike us as wrong with the force of a five-week-old, unrefrigerated dead fish.” Parts & Elec. Motors v. Sterling Elec., Inc., 866 F.2d 228, 233 (7th Cir. 1988). “The mere fact that a reviewing court might have decided the issue differently is not sufficient to overturn a decision when there are two permissible views of the issue.” Pendlebury v. Starbucks Coffee Co., Case No. 04-80521, 2007 WL 4592267 , at *2-3 (S.D. Fla. Dec. 28, 2007) (internal citation omitted). “[T]he ‘clear error’ exception must be rarely invoked.” Cox Enters., Inc. v. News-Journal Corp., 794 F.3d 1259, 1272 (11th Cir. 2015). Given this broad discretion, the Undersigned makes the following observations (in general) about discovery rulings. To be sure, another magistrate judge could reach a different decision about the modest cost-sharing involved here, but the mere fact that a reviewing court may decide the issue “differently is not sufficient to overturn a decision when there are two permissible views of the issue.” Sun Capital Partners, 2015 WL 11921411 , at *1 (internal quotation omitted) (denying objections to magistrate judge’s discovery order). See generally Am. Family Mut. Ins. Co. v. Roth, No. 05C 3839, 2010 WL 3397362 , at *2 (N.D. Ill. Aug. 25, 2010) (internal citations omitted) (“[I]ndeed, on virtually identical facts, two decision makers can arrive at opposite conclusions, both of which constitute appropriate exercises of discretion.”). DONE AND ORDERED in Chambers, in Miami, Florida, on March 2, 2021. Jgnathan Goodman UNITED STATES MAGISTRATE JUDGE Copies furnished to: The Honorable Cecilia M. Altonaga All counsel of record