Molly C. v. Oxford Health Insurance, Inc.
This text of Molly C. v. Oxford Health Insurance, Inc. (Molly C. v. Oxford Health Insurance, Inc.) is published on Counsel Stack Legal Research, covering District Court, S.D. New York primary law. Counsel Stack provides free access to over 12 million legal documents including statutes, case law, regulations, and constitutions.
Opinion
USDC SDNY DOCUMENT ELECTRONICALLY FILEI UNITED STATES DISTRICT COURT DOC #: SOUTHERN DISTRICT OF NEW YORK DATE FILED: _11/21/202¢ MOLLY C., etal., 21-CV-10144 (PGG) (BCM) Plaintiffs, -against- OPINION AND ORDER OXFORD HEALTH INSURANCE, INC., Defendant.
BARBARA MOSES, United States Magistrate Judge. Plaintiffs Molly C. and Naomi L. are beneficiaries of group health insurance plans administered by defendant Oxford Health Insurance, Inc. (Oxford). Beginning in 2019, plaintiffs were diagnosed with eating disorders, received outpatient nutritional counseling as part of their treatment, and submitted claims for that counseling to Oxford, which denied them on the ground that their plans did not cover those services. Because Oxford did cover nutritional counseling for diabetes, plaintiffs contend that its failure to do the same for eating disorders violated the Mental Health Parity and Addiction Equity Act (Parity Act), 29 U.S.C. § 1185a. By motion dated March 22, 2024, plaintiffs seek to certify a class consisting of all persons who were covered under an Oxford group health insurance plan in New York, were diagnosed with one of five specified eating disorders (EDs), received outpatient nutritional counseling from November 30, 2015 to the present (the Class Period), and either (a) submitted claims for that counseling, which Oxford denied on the ground that it was "not a covered benefit" (the Denied Claims Subclass), or (b) did not submit claims (the No Claims Subclass). In order to establish the required numerosity of each proposed subclass, plaintiffs rely on two expert declarations from Frank Fox, Ph.D., who estimates — based on claims data supplied by Oxford — that during the Class Period at least 458 Oxford members with EDs submitted at least 1,461 claims for outpatient nutritional counseling that were denied because the treatment was "not a covered benefit." Dr. Fox
also opines that, during the same period, 1,300 to 1,795 Oxford members per year would have received nutritional counseling for their EDs, had it been a covered benefit, and that 513 to 709 Oxford members per year actually did receive such treatment. Now before me, on referral from the Hon. Paul G. Gardephe, United States District Judge, is Oxford's motion, filed on May 10, 2024 (Dkt. 84), to exclude Dr. Fox's opinions pursuant to
Fed. R. Evid. 702 and the standards set forth in Daubert v. Merrell Dow Pharmaceuticals, Inc., 509 U.S. 579 (1993) and its progeny.1 As to the Denied Claims Subclass, defendant primarily argues that Dr. Fox merely performed "simple mathematics" that did not require any specialized expertise. As to the No Claims Subclass, defendant argues that Dr. Fox's opinions are not relevant to numerosity, and are in any event inherently unreliable. After careful review of the parties' papers, I conclude that Dr. Fox's opinions are relevant to the pending certification motion and that defendant's reliability arguments go to their weight, not their admissibility. Moreover, because the challenged expert evidence is offered in support of class certification, as to which the District Judge is both gatekeeper and factfinder, "there is no risk
of jury confusion," Simmons v. Garland, 2024 WL 1468239, at *3 (E.D.N.Y. Mar. 20, 2024), and "all doubts should be resolved in favor of admissibility." Am. Empire Surplus Lines Ins. Co. v. J.R. Contracting & Env't Consulting, Inc., 2024 WL 3638329, at *4 (S.D.N.Y. Aug. 2, 2024) (internal quotation marks and citation omitted). Consequently, as explained in more detail below, defendant's motion will be denied.
1 Plaintiff's class certification motion remains pending before Judge Gardephe. However, defendant's Daubert motion is within my referral for general pretrial management (see Dkt. 60), made pursuant to 28 U.S.C § 636(b)(1)(A) and Fed. R. Civ. P. 72(a). See Israel v. Springs Indus., Inc., 2007 WL 9724896, at *2 (E.D.N.Y. July 30, 2007) (because a Daubert motion is "nondispositive of the litigation, disposition of it by a magistrate judge in the form of an order and not a Report and Recommendation is appropriate") (collecting cases); accord BS Pac. Life Ins. Co. v. Bank of New York Mellon, 571 F. Supp. 3d 106, 111 (S.D.N.Y. 2021). I. BACKGROUND A. Plaintiffs' Allegations Oxford issues and administers health insurance plans in New York, including group plans governed by the Employee Retirement and Income Security Act of 1974 (ERISA), 29 U.S.C. §§ 1001-1461, of which the Parity Act is a part. See Am. Compl. (Dkt. 35) ¶¶ 5, 12. Plaintiffs
Molly C. and Naomi L. are beneficiaries of ERISA-governed group health insurance plans administered by Oxford in New York. Id. ¶¶ 15, 20, 23, 44. Starting in 2019, they received treatment for EDs, including outpatient nutritional counseling. Id. ¶¶ 28-30, 49-50. However, Oxford rejected their counseling claims on the ground that those services were "not a covered benefit" under their plans. Id. ¶¶ 30-31, 50-51, 55. As a result, both plaintiffs (or their parents) were required to pay out of pocket for their treatment. Id. ¶¶ 41, 56. The Oxford plans did, however, cover outpatient nutritional counseling for members diagnosed with "physical conditions such as diabetes." Id. ¶ 4; see also id. ¶¶ 27, 42, 47, 57. Thus, plaintiffs allege, Oxford violated the Parity Act's requirement that "benefits for mental health treatment [be] no more restrictive than benefits for medical and surgical treatment." Id. ¶ 25 (citing 29 U.S.C. § 1185a(a)(3)(A)).
Defendant admits that, "[p]rior to January 2021, Oxford plans issued in New York did not expressly provide general coverage for nutritional counseling services," Def. Class Cert. Opp. (Dkt. 95) at 2, and that Oxford denied some of the claims submitted by Molly C. and Naomi L. on the ground that outpatient nutritional counseling for EDs was not a covered service. Id. at 5, 6-7. In April 2021, Oxford "reconfigured its claims processing system to consistently provide coverage, and pay claims, for nutritional counseling for eating disorders with dates of service on or after January 1, 2021." Id. at 3. However, Oxford "could not" and did not "change the language of its written plan documents," which did not list outpatient nutritional counseling as a covered benefit for mental health conditions, until 2023. Id. at 3-4. B. The Class Definition In their operative complaint, filed on behalf of the named plaintiffs and "all others similarly situated," Am. Compl. ¶ 68, plaintiffs define a single proposed class, consisting of all persons covered under a group health insurance plan administered by Oxford in New York who were diagnosed with an ED "and whose request for authorization or claim for reimbursement for
nutritional counseling or dietician services was made between November 30, 2015 and the present . . . and was denied." Id. ¶ 69.
Free access — add to your briefcase to read the full text and ask questions with AI
USDC SDNY DOCUMENT ELECTRONICALLY FILEI UNITED STATES DISTRICT COURT DOC #: SOUTHERN DISTRICT OF NEW YORK DATE FILED: _11/21/202¢ MOLLY C., etal., 21-CV-10144 (PGG) (BCM) Plaintiffs, -against- OPINION AND ORDER OXFORD HEALTH INSURANCE, INC., Defendant.
BARBARA MOSES, United States Magistrate Judge. Plaintiffs Molly C. and Naomi L. are beneficiaries of group health insurance plans administered by defendant Oxford Health Insurance, Inc. (Oxford). Beginning in 2019, plaintiffs were diagnosed with eating disorders, received outpatient nutritional counseling as part of their treatment, and submitted claims for that counseling to Oxford, which denied them on the ground that their plans did not cover those services. Because Oxford did cover nutritional counseling for diabetes, plaintiffs contend that its failure to do the same for eating disorders violated the Mental Health Parity and Addiction Equity Act (Parity Act), 29 U.S.C. § 1185a. By motion dated March 22, 2024, plaintiffs seek to certify a class consisting of all persons who were covered under an Oxford group health insurance plan in New York, were diagnosed with one of five specified eating disorders (EDs), received outpatient nutritional counseling from November 30, 2015 to the present (the Class Period), and either (a) submitted claims for that counseling, which Oxford denied on the ground that it was "not a covered benefit" (the Denied Claims Subclass), or (b) did not submit claims (the No Claims Subclass). In order to establish the required numerosity of each proposed subclass, plaintiffs rely on two expert declarations from Frank Fox, Ph.D., who estimates — based on claims data supplied by Oxford — that during the Class Period at least 458 Oxford members with EDs submitted at least 1,461 claims for outpatient nutritional counseling that were denied because the treatment was "not a covered benefit." Dr. Fox
also opines that, during the same period, 1,300 to 1,795 Oxford members per year would have received nutritional counseling for their EDs, had it been a covered benefit, and that 513 to 709 Oxford members per year actually did receive such treatment. Now before me, on referral from the Hon. Paul G. Gardephe, United States District Judge, is Oxford's motion, filed on May 10, 2024 (Dkt. 84), to exclude Dr. Fox's opinions pursuant to
Fed. R. Evid. 702 and the standards set forth in Daubert v. Merrell Dow Pharmaceuticals, Inc., 509 U.S. 579 (1993) and its progeny.1 As to the Denied Claims Subclass, defendant primarily argues that Dr. Fox merely performed "simple mathematics" that did not require any specialized expertise. As to the No Claims Subclass, defendant argues that Dr. Fox's opinions are not relevant to numerosity, and are in any event inherently unreliable. After careful review of the parties' papers, I conclude that Dr. Fox's opinions are relevant to the pending certification motion and that defendant's reliability arguments go to their weight, not their admissibility. Moreover, because the challenged expert evidence is offered in support of class certification, as to which the District Judge is both gatekeeper and factfinder, "there is no risk
of jury confusion," Simmons v. Garland, 2024 WL 1468239, at *3 (E.D.N.Y. Mar. 20, 2024), and "all doubts should be resolved in favor of admissibility." Am. Empire Surplus Lines Ins. Co. v. J.R. Contracting & Env't Consulting, Inc., 2024 WL 3638329, at *4 (S.D.N.Y. Aug. 2, 2024) (internal quotation marks and citation omitted). Consequently, as explained in more detail below, defendant's motion will be denied.
1 Plaintiff's class certification motion remains pending before Judge Gardephe. However, defendant's Daubert motion is within my referral for general pretrial management (see Dkt. 60), made pursuant to 28 U.S.C § 636(b)(1)(A) and Fed. R. Civ. P. 72(a). See Israel v. Springs Indus., Inc., 2007 WL 9724896, at *2 (E.D.N.Y. July 30, 2007) (because a Daubert motion is "nondispositive of the litigation, disposition of it by a magistrate judge in the form of an order and not a Report and Recommendation is appropriate") (collecting cases); accord BS Pac. Life Ins. Co. v. Bank of New York Mellon, 571 F. Supp. 3d 106, 111 (S.D.N.Y. 2021). I. BACKGROUND A. Plaintiffs' Allegations Oxford issues and administers health insurance plans in New York, including group plans governed by the Employee Retirement and Income Security Act of 1974 (ERISA), 29 U.S.C. §§ 1001-1461, of which the Parity Act is a part. See Am. Compl. (Dkt. 35) ¶¶ 5, 12. Plaintiffs
Molly C. and Naomi L. are beneficiaries of ERISA-governed group health insurance plans administered by Oxford in New York. Id. ¶¶ 15, 20, 23, 44. Starting in 2019, they received treatment for EDs, including outpatient nutritional counseling. Id. ¶¶ 28-30, 49-50. However, Oxford rejected their counseling claims on the ground that those services were "not a covered benefit" under their plans. Id. ¶¶ 30-31, 50-51, 55. As a result, both plaintiffs (or their parents) were required to pay out of pocket for their treatment. Id. ¶¶ 41, 56. The Oxford plans did, however, cover outpatient nutritional counseling for members diagnosed with "physical conditions such as diabetes." Id. ¶ 4; see also id. ¶¶ 27, 42, 47, 57. Thus, plaintiffs allege, Oxford violated the Parity Act's requirement that "benefits for mental health treatment [be] no more restrictive than benefits for medical and surgical treatment." Id. ¶ 25 (citing 29 U.S.C. § 1185a(a)(3)(A)).
Defendant admits that, "[p]rior to January 2021, Oxford plans issued in New York did not expressly provide general coverage for nutritional counseling services," Def. Class Cert. Opp. (Dkt. 95) at 2, and that Oxford denied some of the claims submitted by Molly C. and Naomi L. on the ground that outpatient nutritional counseling for EDs was not a covered service. Id. at 5, 6-7. In April 2021, Oxford "reconfigured its claims processing system to consistently provide coverage, and pay claims, for nutritional counseling for eating disorders with dates of service on or after January 1, 2021." Id. at 3. However, Oxford "could not" and did not "change the language of its written plan documents," which did not list outpatient nutritional counseling as a covered benefit for mental health conditions, until 2023. Id. at 3-4. B. The Class Definition In their operative complaint, filed on behalf of the named plaintiffs and "all others similarly situated," Am. Compl. ¶ 68, plaintiffs define a single proposed class, consisting of all persons covered under a group health insurance plan administered by Oxford in New York who were diagnosed with an ED "and whose request for authorization or claim for reimbursement for
nutritional counseling or dietician services was made between November 30, 2015 and the present . . . and was denied." Id. ¶ 69. In their class certification motion, filed almost two years later, plaintiffs request that the Court certify two subclasses: (a) the Denied Claims Subclass, consisting of Oxford members whose claims for outpatient nutritional counseling for EDs during the Class Period were denied on the ground that such services were not covered; and (b) the No Claims Subclass, consisting of members who received outpatient nutritional counseling for EDs during the Class Period but did not submit claims for that treatment. Pl. Class Cert. Mem. (Dkts. 87, 89-1)2 at 2-3. The Class Period remains November 30, 2015 to the present. Id. C. Defendant's Interrogatory Responses In its supplemental interrogatory answers, dated October 13, 2023, Oxford acknowledged
that "1,360 claims for outpatient nutritional counseling services for eating disorders . . . with dates of service November 30, 2015 through September 23, 2023, were submitted and denied on the basis that the service was not a covered benefit," and that those claims were submitted by "427 unique Oxford members." Oxford Supp. Interrog. Resps. (Dkt. 91-1 at ECF pp. 65-78; Dkt. 93-2 at ECF pp. 65-78), at 4-5 (responses to Nos. 10 and 11). Over the same period, Oxford stated, a
2 The parties requested and received permission to file some of their motion papers under seal. Where double docket citations are provided, the first citation identifies the redacted, publicly-filed document; the second citation identifies the unredacted, sealed document. total of 8,464 claims for outpatient nutritional counseling for EDs were submitted by 1,222 unique Oxford members – many of which were allowed. Id. at 5 (response to No. 11). Oxford based its responses on an Excel file (produced in discovery as OXF0032564) containing "records of claims submitted to Oxford for nutritional counseling services for dates of service November 30, 2015 to September 23, 2023 for members with [EDs]." Id. at 3 (response to No. 2).
D. The Numerosity Opinions On December 21, 2023, during pre-certification discovery (see Dkt. 50), plaintiffs served Dr. Fox's initial declaration. See Fox Decl. (Dkt. 91-1 at ECF pp. 6-34; Dkt. 93-2 at ECF pp. 6- 34). On January 8, 2024, defendant served the rebuttal report of Michael J. Petron. See Petron Rep. (Dkt. 96-5; Dkt. 100-5). On February 20, 2024, with leave of the Court (see Dkt. 78), plaintiffs served Dr. Fox's supplemental declaration. See Fox Supp. Decl. (Dkt. 91-1 at ECF pp. 36-63; Dkt. 93-2 at ECF pp. 36-63). On February 26, 2024, Dr. Fox sat for deposition. See Fox Dep. Tr. I (Dkt. 85-2); Fox Dep. Tr. II (Dkt. 91-1 at ECF pp. 80-98; Dkt. 93-2 at ECF pp. 80-98). 1. Fox Declaration Dr. Fox is a consulting economist who provides healthcare planning, statistical analysis, and financial consulting to hospitals and physician practices. Fox Decl. at 1. In his initial
declaration, Dr. Fox (i) estimates the number of Oxford members in New York who received nutritional counseling for EDs between November 30, 2015 and the present (the Study Period); (ii) analyzes the claims data produced by Oxford (OXF0032564) to independently determine the number of claims for such counseling that were submitted over the Study Period, as well as those that were denied on the ground that the services were not covered; and (iii) compares the two sets of figures, offering a hypothesis as to the difference. Id. at 3-13. To estimate the "numerosity" of Oxford members in New York with EDs, and their "expected utilization of nutritional counseling," Fox Decl. at 3, plaintiffs' expert begins with the total number of Oxford insureds in New York from December 2015 to 2023, and estimates, for each year, the number of male and female insureds in various age groups, using U.S. census data for New York State. Id. at 4. For each cohort, Dr. Fox estimates one-year prevalence rates for each relevant ED, relying on national figures reported in 2020 by Deloitte Access Economics, the Harvard T.H. Chan School of Public Health, and the Academy for Eating Disorders, id. at 5-7,3
and "one-year treatment rates by condition" using figures sourced from a 2019 study by Zachary J. Ward (Harvard T.H. Chan School) and others. Id. at 7.4 For example, Dr. Fox estimates that the one-year prevalence rate for anorexia nervosa in females is 0.180%, and that 15% of females with anorexia nervosa receive treatment. See Fox Decl. at 6 tbl.3, 8 tbl.5. Because treatment for EDs does not necessarily include nutritional counseling, Dr. Fox then adjusts his treatment rates downward, using two different adjustments. First, he applies a 60% adjustment, based on a 2022 survey (of Australian eating disorder clinicians) which showed that "an average of 60% of eating disorder patients are referred to a dietitian." Fox Decl. at 7.5 Applying these rates to Oxford's New York members, Dr. Fox estimates that "between 1,300 to 1,795 Oxford
3 See Deloitte Access Econ., Social and Economic Cost of Eating Disorders in the United States of America: Report for the Strategic Training Initiative for the Prevention of Eating Disorders and the Academy for Eating Disorders (June 2020), https://www.hsph.harvard.edu/striped/report- economic-costs-of-eating-disorders. 4 See Zachary J. Ward et al., Estimation of Eating Disorders Prevalence by Age and Associations With Morality in a Simulated Nationally Representative US Cohort (JAMA Network Open 2019) (Ward Study), https://jamanetwork.com/journals/jamanetworkopen/fullarticle/2752577. Ward and his co-authors relied, in turn, on a 2007 analysis of data from the National Comorbidity Replication, which was "a nationally representative face-to-face household survey (n_= 9282) conducted in 2001-2003." See James I. Hudson et. al., The Prevalence and Correlates of Eating Disorders in the National Comorbidity Survey Replication, 61 Biological Psych. 348, 348 (2007) (Hudson Study). 5 See Caitlin M. McMaster et al., Impact of Patient Characteristics on Clinicians' Decisions to Involve Dietitians in Eating Disorder Treatment, 35 J. Hum. Nutrition and Dietetics 512 (2022) (McMaster Report). insureds a year would have received nutritional counseling treatment for an eating disorder had this been a covered benefit." Id. at 8, see also id. at 13 tbl.9 ("Baseline Estimate"). Dr. Fox recognizes, however, that the 60% factor he uses for his "baseline" estimate reflected "persons with health care insurance," and therefore "may overestimate actual utilization" of nutritional counseling among persons who did not have (or did not know they had) coverage for
such treatment. Fox Decl. at 8-9. Consequently, Dr. Fox also considers "a lower bound set of estimates based on a 24% nutritional counseling adjustment factor," id. at 10, derived from a 1989 survey of American women with EDs.6 Applying the 24% adjustment to Oxford's New York members, he estimates that "between 513 to 709 Oxford insureds a year received nutritional counseling treatment for an eating disorder." Id. at 10; see also id. at 13 tbl.9 ("Low Estimate"). Turning to Oxford's claims data, Dr. Fox agrees with Oxford that there were 8,464 claims filed for outpatient nutritional counseling services by 1,222 unique members with EDs. Fox Decl. at 10 (referencing Oxford's supplemental interrogatory responses). Of these, Dr. Fox determines that at least 1,461 claims, by 458 unique members, were denied because the services were "not a
covered benefit." Id. at 11. These figures are close to – but slightly higher than – those derived by Oxford itself from the same data. Cf. Oxford Supp. Interrog. Resps. at 5 (stating that 1,360 claims by 427 unique members were denied because the services were "not a covered benefit"). Finally, for each relevant year, Dr. Fox compares (a) the number of Oxford members who submitted claims for nutritional counseling services (22 to 314 per year) to (b) his "baseline estimate" of the number of members who "would have received nutritional counseling" for an ED "had this been a covered benefit"(1,300 to 1,795 per year), and (c) his "low estimate" of the number
6 See Joel Yager, M.D., et al., Help Seeking and Satisfaction With Care in 641 Women With Eating Disorders, 177 J. Nervous & Mental Disease 632 (1989) (Yager Survey). of members who "received nutritional counseling” for an ED (513 to 709 per year). See Fox Decl. at 13 tbl.9. Table 9 is reproduced here in full (with the three right-hand columns labelled A, B, and C for clarity): Table 9: Comparison of Count of Unique Members Estimated by Year 7 13) Op ioyueM Koreey BEI Um eh a srsnbniteice TUEP TD) Nov 30-Dec 31 2015 1,461 2016 1,579 2017 1,795 2018 L771 | 699 2019 1711 2020 1,539 2021 1,439 2022 1,346 Jan 01-Sep 22 2023 1,300 Note: 2015 and 2023 are not full calendar years, so comparisons with Fox Model estimates of annual unique members may be limited. One method of adjustment not shown above is to adjust by number of months. For example, 2015 is effectively one month. Therefore, the 577- count presented in the Fox Model (Low Estimate) column could be adjusted to reflect one twelfth of the original estimate (1.e. 577 * (1/12) = 48). If Dr. Fox is correct, the size of the No Claims Subclass (that is, the number of Oxford members who were diagnosed with EDs and actually received outpatient nutritional counseling but did not submit claims for those services) may be estimated by subtracting column A from column C. After setting out his results, Dr. Fox hypothesizes that a "key element" of the difference between the "Oxford figures" (column A) and his “estimating models" (columns B and C) ts "knowledge by Oxford insureds that over most of the Study Period, nutritional counseling for eating disorders was not a covered service. Thus, many insureds would be less likely to submit a claim for a service they expect would be denied." Fox Decl. at 13.
2. Petron Report Oxford's expert, who is a Certified Public Accountant and a Certified Fraud Examiner, was asked to (i) review the data produced by Oxford "to opine on whether it is possible to identify the population of individuals meeting the criteria for membership in Plaintiffs' proposed class without an individualized review of claims for each potential class member," and (ii) "review, analyze, and
critique" Dr. Fox's work. Petron Rep. ¶ 8 Although Mr. Petron reviewed the same file provided to Dr. Fox (OXF0032564), he filtered the data differently, in that he categorized a claim as "allowed" rather than "denied" if even one "line item" within the claim "has an allowed amount." Petron Rep. ¶ 28. This boosted the percentage of claims for nutritional counseling that Petron reported as "allowed," both before and after the 2021 reconfiguration. See id. ¶¶ 29-32. Additionally, Petron argued that because the Oxford file "only lists one denial reason per line item," it is "possible that but for the exclusion of nutritional counseling for eating disorders pre-2021, other denial reasons would have prevented these claims from being paid." Id. ¶ 49. For this reason, Petron stated, "I do not believe that the Claims Data allows for one to identify a population of individuals whose claims for nutritional
counseling services would be covered but for the lack of a nutritional counseling benefit for eating disorder diagnoses." Id. ¶¶ 35, 61. Lastly (as relevant here), Mr. Petron characterized the assumptions that Dr. Fox used to estimate the number of Oxford members who received outpatient nutritional counseling for EDs as "flawed," and argued that they led to an "unreliable estimate[.]" Id. ¶ 11(iv).7
7 Oddly, plaintiffs never submitted the entire Petron Report to the Court. In opposition to plaintiffs' class certification motion, they submitted a partial copy, consisting of the first 30 pages of the 43- page report (exclusive of exhibits). The partial copy includes the "Summary of Opinions" section, as well as Mr. Petron's evaluation of the Oxford claims data, but does not include his fully-detailed critique of the Fox Declaration. 3. Supplemental Fox Declaration In his supplemental declaration, Dr. Fox defends his analysis against Mr. Petron's challenges. He disagrees with Petron's definition of an "allowed" claim, calling it "overly expansive." Fox Supp. Decl. at 5-7. Even using that expansive definition, Dr. Fox counts 413 unique members whose nutritional counseling claims for EDs were denied before Oxford's 2021
reconfiguration, and estimates that there were 430 (still well above the numerosity threshold) across the entire Study Period. Id. at 6 tbl.1, 8 tbl.3. Dr. Fox also "disagree[s] with Mr. Petron’s conclusion that it is not possible to estimate the population of individuals within the proposed Class from the provided Claims Data[.]" Id. at 3. He argues that Petron's "conjectured hypothetical" that some nutritional counseling claims might have been denied for other reasons is "meaningless," because, by definition, a member with an ED whose claim was denied for lack of coverage meets the criteria for inclusion in the class. Id. at 8-9. Lastly (as relevant here), Dr. Fox agrees that his utilization estimates for nutritional counseling are based on "statistics with degrees of uncertainty," Fox Supp. Decl. at 4-5, but defends his conclusions as "reliable and reasonable" in light of available data. Id. at 5. For example,
Dr. Fox defends his "overall metric" that 20.72% of individuals suffering from an ED receive treatment (of any kind) for that disorder, noting that four relatively recent studies all estimate higher treatment rates. See id. at 16 tbl.8. He acknowledges that his baseline (high) estimate – that 60% of those in treatment for EDs would receive nutritional counseling as part of that treatment (if it were a covered benefit) – relies on data from Australia, but argues that 60% is "likely an underestimate relative to standard practice in the United States." Id. at 19.8 As for his low estimate
8 Dr. Fox points out that the American Psychiatric Association (APA), unlike The Royal Australian and New Zealand College of Psychiatrists, recommends nutritional counseling (in combination with other therapies) for the treatment of EDs. Fox Supp. Decl. at 19. – that 24% of Oxford members in treatment for EDs actually received nutritional counseling – Dr. Fox agrees that it was based on "older research," from "a period in which treatment recommendations may have differed," but characterizes 24% as "reasonable if nutritional counseling was selectively excluded from coverage for eating disorder treatment, as was the case with Oxford insureds prior to 2021." Id. Further, Dr. Fox notes, Mr. Petron does not point to any
contrary data or "alternative estimates of nutritional counseling rates." Id. at 19. II. LEGAL STANDARDS A. Numerosity Class certification requires, among other things, a showing that the proposed class is "so numerous that joinder of all members is impracticable[.]" Fed. R Civ. P. 23(a)(1). Courts in the Second Circuit presume that the numerosity requirement is met if the putative class has forty or more members. See, e.g., Shahriar v. Smith & Wollensky Rest. Grp., Inc., 659 F.3d 234, 252 (2d Cir. 2011). Courts in our Circuit "generally do not require 'evidence of exact class size or identity of class members to satisfy the numerosity requirement.'" Oladapo v. Smart One Energy, LLC, 2017 WL 5956907, at *8 (S.D.N.Y. Nov. 9, 2017) (quoting Robidoux v. Celani, 987 F.2d 931, 935
(2d Cir. 1993)), adopted, 2017 WL 5956770 (S.D.N.Y. Nov. 30, 2017). "Nevertheless, a plaintiff seeking class certification 'must show some evidence of or reasonably estimate the number of class members.'" Edge v. C. Tech Collections, Inc., 203 F.R.D. 85, 89 (E.D.N.Y. 2001) (quoting Pecere v. Empire Blue Cross and Blue Shield, 194 F.R.D. 66, 69 (E.D.N.Y. 2000)). "Where the plaintiff's assertion of numerosity is pure speculation or bare allegations, the motion for class certification fails." Edge, 203 F.R.D. at 89. B. Admissibility of Expert Evidence Rule 702 provides: A witness who is qualified as an expert by knowledge, skill, experience, training, or education may testify in the form of an opinion or otherwise if the proponent demonstrates to the court that it is more likely than not that: (a) the expert's scientific, technical, or other specialized knowledge will help the trier of fact to understand the evidence or to determine a fact in issue; (b) the testimony is based on sufficient facts or data; (c) the testimony is the product of reliable principles and methods; and (d) the expert's opinion reflects a reliable application of the principles and methods to the facts of the case. Fed. R. Evid. 702. Rule 702 "embodies a liberal standard of admissibility for expert opinions," Nimely v. City of New York, 414 F.3d 381, 395 (2d Cir. 2005), such that "the rejection of expert testimony is the exception rather than the rule." Fed. R. Evid. 702 advisory committee's note to 2000 amendments; see also Chen-Oster v. Goldman, Sachs & Co., 114 F. Supp. 3d 110, 115 (S.D.N.Y. 2015) ("There is a presumption that expert evidence is admissible." (citing Borawick v. Shay, 68 F.3d 597, 610 (2d Cir. 1995))). Nonetheless, the Rules of Evidence "assign to the trial judge the task of ensuring that an expert's testimony both rests on a reliable foundation and is relevant to the task at hand" before that testimony is presented to the jury. Daubert, 509 U.S. at 597; accord Kumho Tire Co. v. Carmichael, 526 U.S. 137, 141 (1999). Here, defendant challenges both the relevance and the reliability of Dr. Fox's testimony.9 Relevance is governed by Rule 401 of the Federal Rules of Evidence. See Amorgianos v. Nat'l R.R. Passenger Corp., 303 F.3d 256, 265 (2d Cir. 2002) ("In fulfilling this gatekeeping role, the trial court should look to the standards of Rule 401."). Thus, relevance depends on the extent to which (1) the proffered testimony "has any tendency to make a fact more or less probable than
9 Defendant does not challenge Dr. Fox's qualifications. it would be without evidence," and (2) "the fact is of consequence in determining the action." Fed. R. Evid. 401. As the Second Circuit has noted, "the definition of relevance under Fed. R. Evid. 401 is very broad." United States v. Certified Env't Servs., Inc., 753 F.3d 72, 90 (2d Cir. 2014). Reliability depends on the indicia identified in Rule 702, "namely, (1) that the testimony is grounded on sufficient facts or data; (2) that the testimony 'is the product of reliable principles and
methods'; and (3) that 'the witness has applied the principles and methods reliably to the facts of the case.'" Amorgianos, 303 F.3d at 265 (quoting Fed. R. Evid. 702). However, "the Daubert inquiry is fluid and will necessarily vary from case to case." Id. at 266. Thus, the law grants "broad latitude" to district courts to determine reliability in context. Restivo v. Hessemann, 846 F.3d 547, 576 (2d Cir. 2017); see also Kumho Tire, 526 U.S. at 150-52 (1999) (because "there are many different kinds of experts, and many different kinds of expertise," court are granted "considerable leeway in deciding in a particular case how to go about determining whether particular expert testimony is reliable"). An expert's analysis must be "reliable at every step," meaning that "any step that renders
the analysis unreliable under the Daubert factors renders the expert's testimony inadmissible." Amorgianos, 303 F.3d at 267 (quoting In re Paoli R.R. Yard PCB Litig., 35 F.3d 717, 745 (3d Cir. 1994)). However, testifying experts need not conduct original research; they may rely on facts or data gathered by others, as long as "reasonable experts in the particular field would reasonably rely on those kinds of facts or data in forming an opinion on the subject." Fed. R. Evid. 703; see also Gussack Realty Co. v. Xerox Corp., 224 F.3d 85, 95 (2d Cir. 2000) ("The expert need not have conducted her own tests."). Moreover, the court need not determine whether the expert's conclusions are correct. See In re Paoli, 35 F.3d at 744 ("The evidentiary requirement of reliability is lower than the merits standard of correctness."). If there are flaws in an expert's reasoning, only flaws "large enough" such that "the expert lacks 'good grounds' for his or her conclusions" require exclusion. Amorgianos, 303 F.3d at 267 (quoting In re Paoli, 35 F.3d at 746). Although the Supreme Court has not definitively resolved "the extent to which a district court must undertake a Daubert analysis at the class certification stage," In re U.S. Foodservice Inc. Pricing Litig., 729 F.3d 108, 129 (2d Cir. 2013), it has suggested in dicta that such an analysis
is appropriate, see Wal-Mart Stores, Inc. v. Dukes, 564 U.S. 338, 354 (2011) ("The District Court concluded that Daubert did not apply to expert testimony at the certification stage of class-action proceedings. We doubt that is so[.]"), and courts in this district frequently undertake the inquiry prior to deciding the certification motion itself. See, e.g., In re Int. Rate Swaps Antitrust Litig., 2023 WL 8675625, at *2 (S.D.N.Y. Dec. 15, 2023); Passman v. Peloton Interactive, Inc., 671 F. Supp. 3d 417, 431 (S.D.N.Y. 2023). In so doing, however, the courts recognize that at the certification stage (as in a bench trial) the judge acts as both gatekeeper and factfinder. There is thus no "risk of jury confusion." Simmons, 2024 WL 1468239, at *3 (denying Daubert motion prior to bench trial); see also Joseph S. v. Hogan, 2011 WL 2848330, at *3 (E.D.N.Y. July 15,
2011) ("without the risk of poisoning the jury with misleading testimony of limited probative value, see Daubert, 509 U.S. at 595, the Court can take evidence freely and separate helpful conclusions from ones that are not grounded in reliable methodology"); In re LIBOR-Based Fin. Instruments Antitrust Litig., 299 F. Supp. 3d 430, 471 (S.D.N.Y. 2018) (at the class certification stage, the question "is not whether a jury at trial should be permitted to rely on the expert's report . . . but rather whether the Court may utilize it in deciding whether the requisites of Rule 23 have been met" (quoting Ge Dandong v. Pinnacle Performance Ltd., 2013 WL 5658790, at *13 (S.D.N.Y. Oct. 17, 2013))). Thus, all doubts "should be resolved in favor of admissibility." Am. Empire Surplus Lines, 2024 WL 3638329, at *4. Additionally, at the class certification stage, the Daubert inquiry extends only to the issues "necessary to the Rule 23 analysis." In re LIBOR-Based Fin. Instruments, 299 F. Supp. 3d at 471; see also Pinnacle Performance, 2013 WL at *13 ("[T]he inquiry is 'limited to whether or not the [expert reports] are admissible to establish the requirements of Rule 23.'" (quoting In re NYSE Specialists Sec. Litig., 260 F.R.D. 55, 66 (S.D.N.Y. 2009))). Thus, the Court need only consider
the portions of Dr. Fox's declarations relied on by plaintiffs to demonstrate numerosity. See Chen- Oster, 114 F. Supp. 3d at 115 (at the class certification stage, "the scope of the Daubert analysis is cabined by its purpose" and “the inquiry is limited to whether or not the expert reports are admissible to establish the requirements of Rule 23"). III. THE PARTIES' POSITIONS A. Numerosity In their class certification papers, plaintiffs rely on Dr. Fox's work to show that both the Denied Claims Subclass and No Claims Subclass are sufficiently numerous to permit certification. See, e.g., Pl. Class Cert. Mem. at 13-14. As to the Denied Claims Subclass, Oxford does not dispute numerosity. Indeed, its brief in opposition to class certification does not mention Dr. Fox's opinion
(or, for that matter, Mr. Petron's contrary views) in connection with the numerosity of the Denied Claims Subclass. As to the No Claims Subclass, however, Oxford argues that Dr. Fox's estimates concerning the number of plan members who received nutritional therapy for an ED are "speculative" and "unreliable," leaving plaintiffs with "no real evidence" bearing on the size of the No Claims Subclass. Def. Class Cert. Opp. at 12. B. Admissibility of Expert Evidence Even though Oxford does not contest the numerosity of the Denied Claims Subclass, it argues, in its Daubert motion, that Dr. Fox's estimate of the size of that subclass should be excluded because he "merely added up" numbers provided by Oxford, which the Court could have done itself. See Def. Daubert Mem. (Dkt. 85) at 8-9. As to the No Claims Subclass, Oxford argues that Dr. Fox's testimony should be excluded because: (i) he never expressly opined as to whether that subclass is sufficiently numerous for certification, making his testimony "irrelevant," see Def. Daubert Mem. at 7-8; (ii) his declarations
lack the "support and rigor" required of an expert report under Fed. R. Civ. P. 26(a)(2)(B), see id. at 10-12; (iii) he "failed to validate his assumptions," some of which were based upon "admittedly insufficient data," see id. at 13-14; and (iv) some of the data he relied upon was too old or not necessarily "representative of Oxford's insured population." See id. at 14-16. Plaintiffs defend Dr. Fox on each point, and further argue that defendant's criticism "goes to the weight, not the admissibility, of Dr. Fox's testimony," and therefore is not grounds for excluding his opinions. Pl. Daubert Opp. (Dkts. 91, 93-1) at 21. IV. DISCUSSION A. The Denied Claims Subclass As noted above, there were significant disagreements between Dr. Fox and Mr. Petron
concerning the Denied Claims Subclass. Not only did Petron use a different method to filter Oxford's claims data (thereby arriving at a different count of claims allowed and claims denied); he took the position that it was not possible to estimate the size of the Denied Claims Subclass from the claims data alone. See supra Part I(D). Nonetheless, in its Daubert motion, Oxford asks the Court to exclude Dr. Fox's opinions concerning the Denied Claims Subclass because he "merely added up the total number of denied claims listed in an Excel file to arrive at his numerosity estimate." Def. Daubert Mem. at 9. In defendant's view, this was a "simple exercise," "directed solely to lay matters which a [fact finder] is capable of understanding and deciding without the expert's help," and was therefore an improper subject for expert testimony. Id. (alterations in original) (quoting United States v. Castillo, 924 F.2d 1227, 1232 (2d Cir. 1991)). Defendant's position is meritless. Combing through a lengthy data set, isolating and aggregating the data, and presenting it "in a more readily understandable format" are appropriate tasks for an expert. Louis Vuitton Malletier S.A. v. Sunny Merch. Corp., 97 F. Supp. 3d 485, 504-
05 (S.D.N.Y. 2015) (denying Daubert motion made on the ground that plaintiff's expert "did nothing more than 'simply add up' figures provided to her by Sunny"). Moreover, before he could perform any arithmetic, Dr. Fox was required to determine which claims to add up. That question is well within the province of an expert. See Passman, 671 F. Supp. 3d at 435 (denying Daubert motion made on the ground that Weir simply performed "grade-school arithmetic" because Weir also "compiled [the] data that the Court would need to conduct this grade-school arithmetic"). The fact that Oxford's expert also addressed the question of which claims to add up (and disagreed with Dr. Fox) further demonstrates that the analysis was not a "simple exercise" easily performed by a layperson. I therefore decline to exclude Dr. Fox's testimony concerning the Denied Claims
Subclass on the ground that the exercise was too simple to require expert assistance. In its reply brief, defendant adds a new line of attack: that Dr. Fox's work is "unnecessary to show numerosity of the Denied Claims Subclass." Def. Daubert Reply Mem. (Dkt. 104) at 1. As support for this argument, Oxford points to plaintiffs' assertion that this subclass "satisfies numerosity regardless of whether the Court credits Mr. Petron's or Dr. Fox's analysis." Id. at 5. This point presents a closer question. An expert report may be excluded as unnecessary when the opposing party has conceded the point for which the report was submitted. See, e.g., LaBounty v. Rivera, 1999 WL 1129063, at *5 (S.D.N.Y. Dec. 8, 1999) (explaining that plaintiff's proposed expert medical evidence was properly excluded where defendant conceded, before trial, all of the issues to which that evidence could be relevant). Here, however, Oxford has not formally conceded that the Denied Claims Subclass is numerous. It simply failed to contest the point in its class certification papers. Under these circumstances, a district judge may (and arguably should) review all of the numerosity evidence – including expert evidence – before making the required Rule 23(a)(1) finding. See, e.g., Fort Worth Emp. Ret. Fund v. J.P. Morgan Chase & Co., 301
F.R.D. 116, 129, 131 (S.D.N.Y. 2014) (noting that "[d]efendants do not dispute [the expert's] conclusion that the class is numerous and contains at least 1360 members," but nonetheless examining the expert's "numerosity calculation method," finding it reliable, and ultimately concluding that plaintiffs "satisfied the requirements of Rule 23(a)(1)"). I therefore decline to exclude Dr. Fox's testimony concerning the Denied Claims Subclass on the ground that it has become unnecessary. B. The No Claims Subclass 1. Relevance According to Oxford, Dr. Fox's opinions are irrelevant to the numerosity of the No Claims Subclass – and should be excluded on that basis – because, at deposition, he "specifically testified that he was not offering any opinion beyond [the Denied Claims Subclass]." Def. Daubert Mem.
at 7.10 Plaintiffs concede that Dr. Fox did not directly opine on the numerosity of the No Claims
10 When Dr. Fox signed his declarations, plaintiffs had not yet served their class certification motion. The only class definition available to him was the definition contained in the First Amended Complaint. At deposition, when that class definition was read to him, he agreed that it was "the class definition for which [he] attempted to estimate numerosity." Fox Dep. Tr. I at 36:1- 13. The questioning continued: Q. So your opinions in both of your reports, as it relates to numerosity, relate to the numerosity of this specific proposed class; is that correct? A. That was my scope; correct. Id. at 36:14-17. Subclass, but argue that his analysis is relevant to the question whether the No Claims Subclass is sufficiently numerous to satisfy Rule 23(a)(1). Pl. Daubert Opp. at 2, 9. Plaintiffs are correct. "To be relevant, evidence need not be sufficient by itself to prove a fact in issue[.]" United States v. Abu-Jihaad, 630 F.3d 102, 132 (2d Cir. 2010). It need only have a "tendency" to make "more or less probable" a fact "of consequence in determining the action."
Fed. R. Evid. 401. In this case, the consequential fact that plaintiffs seek to establish is that there were at least 40 Oxford members who received outpatient nutritional counseling for an ED during the proposed Class Period but did not submit claims for that treatment. They must therefore show that the number of members who received outpatient nutritional counseling for an ED was higher – by at least 40 – than the number who, according to Oxford's claims data, submitted claims for their nutritional counseling. Dr. Fox's work is highly relevant to this task. In his initial declaration, Dr. Fox estimates the number of Oxford members who received outpatient nutritional counseling for an ED. Fox Decl. at 3 (defining scope of work), 13 tbl.9 col. C (presenting results). While this analysis,
standing alone, is not dispositive of numerosity, it forms a necessary link in any chain leading to an estimate of the size of the No Claim Subclass. It is therefore relevant to numerosity. See Chalmers v. City of New York, 2022 WL 4330119, at *7 (S.D.N.Y. Sept. 19, 2022) (accepting expert testimony regarding similarity of certain jobs as relevant to plaintiffs' race discrimination claims, even though experts did not opine as to whether the jobs were similar "in all material respects"). Dr. Fox also provides a comparison table showing, on an annual basis: (a) how many Oxford members with EDs submitted claims for outpatient nutritional counseling; (b) his "baseline" estimate of how many members would have received such counseling, assuming coverage; and (c) his "low" estimate of how many Oxford members actually received such counseling. Fox Decl. at 13 tbl.9. From this table, the Court can easily perform the last step, which requires only simple arithmetic, and which reveals that – if Dr. Fox's estimates are reliable – there were several hundred Oxford members every year who received outpatient nutritional counseling for an ED but did not submit claims for that treatment.11
Similarly, Dr. Fox's observation that the number of unique claims for outpatient nutritional counseling jumped from approximately 700 each year before 2021 (that is, before Oxford began covering those services) to approximately 1,650 "each year after January 1, 2021," Fox Decl. at 10, 11 tbl.7, has a "tendency" to make it "more probable" that there were at least 40 Oxford members who received nutritional counseling for EDs during the earlier years but did not submit claims for those services. The relevance inquiry requires nothing more. Amorgianos, 303 F.3d at 265. 2. Reliability Oxford challenges the reliability of Dr. Fox's opinions on several different grounds. a. Rule 26(a)(1)(B) During his deposition, Dr. Fox drew a distinction between a declaration and an expert
report. See Fox Dep. Tr. I at 32:21-22. He explained that his initial declaration (at 14 pages, not including his CV and list of prior testimony) "was intended to be succinct," id. at 32:15, whereas, in his experience, expert reports are longer, "more like 40 to 80 pages," "much more in depth,
11 For example, Dr. Fox estimates that in 2019 there were at least 676 Oxford members who received outpatient nutritional counseling for EDs. See Fox Decl. at 13 tbl.9 col. C (low estimate). That same year, Oxford's claims data shows that 179 members submitted claims for that service, see id. at 11 tbl.7, 13 tbl.9 col. A. Some of those claims were denied, see id. at 12 tbl.8, while others were allowed. The difference between 676 and 179 is 497. Thus, if Dr. Fox's estimates are reliable, there were at least 497 Oxford members who – in 2019 alone – received nutritional counseling for an ED but did not submit claims for that treatment. By definition, those members are part of the No Claims Subclass. [with] much more support, much more analysis, much more exploration of why I reached the conclusions I did, more quantitative, more qualitative." Id. at 33:5-13. From this testimony, Oxford concludes that Dr. Fox has "admitted" that neither of his declarations is an expert report "as required by the Federal Rules of Civil Procedure," Def. Daubert Mem. at 10, and has "fully acknowledged" that his testimony "does not even come close to meeting the exacting standard for
reliable testimony." Id. at 12. Plaintiffs respond that both of Dr. Fox's declarations satisfy the criteria set forth in Rule 26(a)(2)(B), see Pl. Daubert Opp. at 8, and add that while his initial declaration was "deliberately succinct," his supplemental declaration (which is 27 pages long and discusses additional data sources) "expanded" on the issues "significantly." Id. (quoting Fox Dep. Tr. II at 81:12-16). Plaintiffs are again correct. While Dr. Fox may be accustomed to writing longer reports (perhaps for trial), Rule 26(a)(2)(B) does not prescribe a minimum page limit. Instead, it requires a retained expert to produce a written report containing: (i) a complete statement of all opinions the witness will express and the basis and reasons for them; (ii) the facts or data considered by the witness in forming them; (iii) any exhibits that will be used to summarize or support them; (iv) the witness's qualifications, including a list of all publications authored in the previous 10 years; (v) a list of all other cases in which, during the previous 4 years, the witness testified as an expert at trial or by deposition; and (vi) a statement of the compensation to be paid for the study and testimony in the case. Fed. R Civ. P. 26(a)(2)(B). Dr. Fox's declarations satisfy this standard. Although Oxford argues that Dr. Fox failed to "test certain assumptions," failed to "comprehensively analyze the studies he relied upon," and failed to "document his assessment" of those studies, Def. Daubert Mem. at 10- 11, these critiques go to the quality of his analysis (discussed below), not his compliance with Rule 26(a)(2)(b).12 I therefore reject defendant's contention that Dr. Fox's declarations do not qualify as an expert report. b. Insufficient Data Next, Oxford argues that Dr. Fox's estimates are unreliable because he had insufficient data
to "validate his assumptions." Def. Daubert Mem. at 13. Specifically, Oxford asserts, Dr. Fox failed to validate his "assumption that the age and sex mix of Oxford's ERISA plan members is comparable to that of the entire state [of] New York," because he "didn't have the data." Id. (quoting Fox Dep. Tr. I at 83:21-84:7). But Oxford – which does have the data – offers no reason to believe that the age and sex mix of its covered population in New York differs in any material respect from the age and sex mix of all New York residents. Moreover, as Dr. Fox explained at deposition, the U.S. census data upon which he relied is "very robust and statistically solid," such that it was "reasonable and reliable," in his opinion, to use that data where, as here, "site-specific" information was not available. Fox Dep. Tr. II at 83:11-20, 85:4-12. Oxford also faults Dr. Fox for relying on the 2019 Ward Study for his estimates of "one-
year treatment rates by condition," see Fox Decl. at 7, without "validat[ing] his assumptions." Def.
12 It is true that Dr. Fox offers no support for his hypothesis that the difference between the (estimated) number of Oxford members who received outpatient nutritional counseling for eating disorders and the smaller number who submitted claims for such services was due, at least in part, to the "knowledge by Oxford insureds that over most of the Study Period, nutritional counseling for eating disorders was not a covered service." Fox Decl. at 13. At deposition, Dr. Fox acknowledged that he did not have any "empirical data" or "validated opinions" to support that hypothesis, "other than common sense." Fox Dep. Tr. I at 204:3-12. But Dr. Fox was retained to estimate "the numerosity of persons suffering from [eating disorders] and their expected utilization of nutritional counseling," Fox Decl. at 3, not to explain the low claims rate among this group. Indeed, as defined by plaintiffs, the No Claims Subclass includes every Oxford member who received outpatient nutritional counseling services but did not submit a claim for those services – regardless of the reasons for that member's failure to claim. See Pl. Daubert Opp. at 10. Since Dr. Fox's hypothesis as to those reasons is irrelevant to numerosity, his failure to state the "basis and reasons" for that hypothesis cannot be grounds for excluding his numerosity opinions. Daubert Mem. at 13. At deposition, Dr. Fox explained that "he did not have any 'better data than the originators of the study." Id. (quoting Fox Dep. Tr. I at 125:3-19).13 According to Oxford, this rendered Dr. Fox's opinions "unreliable and subject to exclusion." Id. at 14. Again, however, defendant presents no inconsistent data or contradictory research. Nor does it suggest that experts in Dr. Fox's field cannot "reasonably rely" on prior academic research such as the Ward Study to
construct healthcare utilization models. See Fed. R. Evid. 703.14 The courts draw a distinction between insufficient data and imperfect data. If a proposed expert lacks the basic information required by professionals in his field to form an opinion, his testimony will be excluded. See, e.g., King v. Wang, 2021 WL 5237195, at *15 (S.D.N.Y. 2021)
13 Where Ward did not have discrete data regarding treatment rates for a specific cohort diagnosed with a specific ED, he made assumptions based on the data he did have for other cohorts or other EDs. For example, Ward assumed that the "12-month treatment probability" for males with diagnosed with OSFED (Other Specified Feeding or Eating Disorder) was 15%, that is, the same as the 12-month treatment probability for females diagnosed with OSFED. See Ward Study, Supplementary Online Content, at 11. Dr. Fox incorporated these figures into his model, which likewise assumed that the one-year treatment rate for both males and females with OSFED was 15%. See Fox Decl. at 8 tbl.5. 14 In its reply brief, Oxford cites several cases for the proposition that an expert report should be excluded where the expert fails to "validate" the data upon which he relies. See Def. Daubert Reply Mem. at 8-9. They are all inapposite. For example, in Forte v. Liquidnet Holdings, Inc., 675 F. App'x 21, 24 (2d Cir. 2017), where the plaintiff alleged that her former employer paid women less than men, her expert "simply input the numbers he was given by Forte and used them to calculate pay discrepancies," id. at 23-24, without doing anything to verify that the accuracy of the numbers. And in United States v. Tuzman, 2017 WL 6527261, at *14-17 (S.D.N.Y. Dec. 18, 2017), Judge Gardephe excluded testimony by a proposed ink-dating expert because, during his testing of the notebook in question, he failed to use any of the "basic quality control protocols" set forth in the scientific papers he relied upon, including maintaining written records of his "validation process," that is, the tests he claimed to have performed on ink samples of known ages to confirm the accuracy of his process. Nothing in these cases – or any of the cases cited by defendant – suggests that that an expert witness is required to independently "validate" the third-party academic research that he relies upon to build a healthcare utilization model. See N.Y.C. Transit Auth. v. Express Scripts, Inc., 588 F. Supp. 3d 424, 445 (S.D.N.Y. 2022) (rejecting similar failure-to-validate argument and noting that that the Court was unable to find any "precedent requiring experts to validate the data underlying each source on which they rely in order to satisfy the reliability threshold for admissibility"), reconsideration denied, 2022 WL 3577426 (S.D.N.Y. Aug. 19, 2022), objections overruled, 2022 WL 20056027 (S.D.N.Y. Oct. 12, 2022). (excluding portions of expert art appraisal based on "a single prior sale"). Imperfect data, however, goes to the weight of the expert's opinion, not its admissibility, and is not grounds for exclusion. See In re Int. Rate Swaps, 2023 WL 8675625, at *4 (rejecting argument that financial expert's report should be excluded because expert failed to validate the Bloomberg data he used as an "imperfect proxy input" when estimating the value of an interest rate swap); Express Scripts, 588
F. Supp. 3d at 445 (denying Daubert motion because expert's failure to "validate the data she used to form her opinion . . . goes to the weight of [her] testimony rather than to its admissibility"); King, 2021 WL 5237195, at *15. (admitting remaining portions of art appraisal, even though expert's conclusions were based on "very few comparables" and on "information from Mainland Chinese auction houses," which "frequently is unreliable"). As explained in In re Int. Rate Swaps, "[t]he question for Daubert purposes is not whether the expert employed the best possible input data to form her testimony, but whether 'the testimony is based on sufficient facts or data.'" 2023 WL 8675625, at *4 (quoting Fed. R. Evid. 702). Here, defendant has failed to show either that Dr. Fox's opinions were based on
"insufficient data," in violation of Rule 702, or that he relied upon facts or data beyond those that experts in his field "would reasonably rely on," in violation of Rule 703. Consequently, any gaps in the studies he relied on to build his model go to the weight of his testimony, not its admissibility. c. Unrepresentative Studies Lastly, Oxford contends that the studies upon which Dr. Fox relied rendered his opinions unreliable because they were too old, not specific to New York, or "cherry-picked" to support Dr. Fox's conclusions. Def. Daubert Mem. at 14. For example, Oxford questions Dr. Fox's reliance on the "first nationally representative study of eating disorders in the United States," see Hudson Study at 9, because he "did not perform any analysis to confirm that this data was representative of recent nutritional counseling rates in New York for eating disorders." Def. Daubert Mem at 14. But Dr. Fox did not rely on the Hudson Study to determine "recent nutritional counseling rates in New York," He relied on it (in part) to determine overall "treatment rates" for EDs. See Fox Decl. at 7. And Oxford offers no reason to believe that overall ED treatment rates in the State of New York are materially different from those in the rest of the country. The research that Dr. Fox actually relied on for his "low estimate" (that 24% of the Oxford
members in New York who were in treatment for their EDs received nutritional counseling as part of their treatment) was the Yager Survey, published in 1989, based on a survey of 641 American women with eating disorders. See Fox Decl. at 10 n.10. Oxford characterizes that data as "outdated" and argues that it "cannot be used to reliably forecast current eating disorder treatment trends." Def. Daubert Mem. at 15. Dr. Fox agrees that the Yager Survey is "based on older research," from "a period in which treatment recommendations may have differed." Fox Supp. Decl. at 19. He points out, however, that more recent studies show that providers prescribe nutritional counseling to their ED patients at much higher rates, and that the APA now recommends nutritional counseling for the treatment of EDs. See id. at 17-19. These data points
suggest, if anything that the 24% rate that Dr. Fox used as an "alternative lower bound," Fox Decl. at 10, may have underestimated the proportion of Oxford-insured ED patients in New York who actually received nutritional counseling.15
15 Oxford notes that Dr. Fox did not rely on a "conflicting study," cited in the McMaster Report, "that indicated that only 6% of eating disorder patients who present to their general practitioner are referred to nutritional counseling." Def. Daubert Mem. at 15-16. The cited study, however, surveyed only "primary care practitioners" (to the exclusion of psychiatrists, psychologists, and other clinicians who treat patients with ED), and reported only direct referrals from the primary care practitioner to the nutritionist. See Lorraine Ivancic et al., Prevalence and Management of People With Eating Disorders Presenting to Primary Care: A National Study, 55 Austl. & N.Z. J. Psych. 1089, 1089 (2021). It is therefore not a particularly useful gauge when trying to determine what percentage of persons who are in treatment for ED receive nutritional counseling as part of the treatment. At best, Oxford has demonstrated that reasonable minds could disagree as to whether Dr. Fox drew the most reasonable conclusions from the best available data. Under Daubert, however, "an expert need not base his or her opinion on the best possible evidence, regardless of availability, but upon 'good grounds, based on what is known.'" In re Fosamax Prod. Liab. Litig., 645 F. Supp. 2d 164, 178 (S.D.N.Y. 2009) (quoting Daubert, 509 U.S. at 590); see also Amorgianos, 303 F.3d
at 266 (an expert need not "back his or her opinion with published studies that unequivocally support his or her conclusions"); In re Int. Rate Swaps, 2023 WL 8675625, at *4 (an expert need not have "employed the best possible input data to form her testimony"). Thus, while Oxford has raised questions about some of the assumptions made by Dr. Fox, it has not shown that his reliance on the Ward Study, the Hudson Study, or the Yager Survey is "so 'speculative,' 'conjectural,' or 'unrealistic and contradictory' that they render his estimate of [ED treatment utilization] fundamentally unreliable." In re Int. Rate Swaps, 2023 WL 8675625, at *4 (quoting Zerega Ave. Realty Corp. v. Hornbeck Offshore Transp., LLC, 571 F.3d 206, 214 (2d Cir. 2009)). It is significant, moreover, that Oxford's objections to Dr. Fox's utilization model "focus almost entirely
on his factual inputs and assumptions rather than his methodology." SourceOne Dental, Inc. v. Patterson Companies, Inc., 2018 WL 2172667, at *6 (E.D.N.Y. May 10, 2018). "These objections go to the probative value of [Dr. Fox's] testimony, not its admissibility[.]" Id. V. CONCLUSION For the foregoing reasons, defendant's motion to exclude Dr. Fox's opinions (Dkt. 84) is DENIED. This means only that the Court "may utilize" his testimony "in deciding whether the requisites of Rule 23 have been met." In re LIBOR-Based Fin. Instruments, 299 F. Supp. 3d at 471 (quoting Pinnacle Performance, 2013 WL 5658790, at *13). I express no opinion as to whether that testimony, together with the other evidence adduced by plaintiffs, adequately demonstrates the numerosity of the proposed subclasses. That determination is for the District Judge. See id. ("[T]o the extent that flaws in expert testimony proffered at class certification do not warrant that testimony's exclusion by the Court as gatekeeper under Daubert at the threshold, those flaws may nonetheless be considered in the Rule 23 analysis undertaken by the Court as trier of fact."). Dated: New York, New York SO ORDERED. November 21, 2024 Yeates \ BARBARA MOSES United States Magistrate Judge
Related
Cite This Page — Counsel Stack
Molly C. v. Oxford Health Insurance, Inc., Counsel Stack Legal Research, https://law.counselstack.com/opinion/molly-c-v-oxford-health-insurance-inc-nysd-2024.