Dwayne Phillips et al. v. Rector and Visitors of the University of Virginia et al. — Class Certification Ruling

Jump To

Top Caption Trial Court Document The text of this document was obtained by analyzing a scanned document and may have typos.

Support FLP

CourtListener is a project of Free
Law Project, a federally-recognized 501(c)(3) non-profit. Members help support our work and get special access to features.

Please become a member today.

Join Free.law Now

April 23, 2026 Get Citation Alerts Download PDF Add Note

Dwayne Phillips et al. v. Rector and Visitors of the University of Virginia et al.

District Court, W.D. Virginia

• Citations: None known
• Docket Number: 3:22-cv-00075

• Precedential Status: Unknown Status

• Opinion

• Authorities (39)

• Cited By (0)

• Summaries (0)

• Similar Cases (3.8K)

• PDF

Trial Court Document

April, 23, 2026
IN THE UNITED STATES DISTRICT COURT aventuyeeowi
FOR THE WESTERN DISTRICT OF VIRGINIA ee ours □□□
CHARLOTTESVILLE DIVISION
DWAYNE PHILLIPS et al., )
)
Plaintiffs, ) Civil Action No. 3:22-cv-00075
)
V. ) By: Hon. Robert S. Ballou
) United States District Judge
RECTOR AND VISITORS OF THE )
UNIVERSITY OF VIRGINIA et al., )
)
Defendants. )
MEMORANDUM OPINION
Plaintiffs move for class certification in this Title VII action which claims that the
University of Virginia Health System (““UVA Health”) discriminated against them by denying
their requests for religious exemptions from the mandated COVID-19 vaccination. They seek to
certify three classes challenging the procedures UVA Health used to evaluate religious
exemption requests: (1) the “Disfavored Religions Class,” (2) the “Abortion Objectors Class,”
and (3) the “Failure-to-Accommodate Class.”
I GRANT in part and DENY in part Plaintiffs’ Motion to Certify Classes and Appoint
Class Counsel, Dkt. 221. I find that Plaintiffs satisfy the requirements of Federal Rule of Civil
Procedure 23 with respect to the Disfavored Religions Class and therefore certify that class.
However, Plaintiffs fail to meet Rule 23’s requirements for the Abortion Objectors Class and the
Failure-to-Accommodate Class, and certification of those two classes is denied.
I. Background
In the fall of 2021, UVA Health mandated COVID-19 vaccinations for Tier 1 employees,
1.e., those whose jobs required them to be present in Health System Facilities. The vaccination
policy allowed employees to submit requests for religious exemptions to a Religious Exemption

Committee (“2021 Procedure”). Training materials developed to administer religious exemption
requests under the 2021 Procedure identified six faiths with “known prohibition[s] against
vaccination” and instructed that exemption requests from employees of these faiths be approved
if “there is no reason to suspect dishonesty.” Dkt. 235-30 at 13, 15. Employees who were not

members of one of the listed faiths, but who had a sincerely held religious belief against
receiving the COVID-19 vaccination, had their requests forwarded to the Religious Exemption
Committee for further consideration. Id. at 15.
In October 2022, UVA Health replaced its 2021 Procedure with an expanded exemption
request review procedure (“2022 Procedure”). Under the 2022 Procedure, an Exemption
Coordinator was responsible for determining whether an employee had a sincerely held religious
belief that conflicted with the vaccine mandate. If she found that the employee had a sincerely
held religious belief, UVA Health then assessed the extent to which accommodating the
exemption request would impose an undue hardship on UVA Health.
Plaintiffs each sought a religious exemption to UVA Health’s vaccination requirement,

but their requests were denied. Ultimately, Plaintiffs declined the vaccine and, as a result, lost
their jobs. They filed this putative class action alleging that UVA Health’s exemption policy
violated the Free Exercise and Establishment Clauses of the First Amendment and the Equal
Protection Clause of the Fourteenth Amendment, as well as their free exercise and establishment
rights under Article I, § 16 of the Virginia Constitution. They also asserted Title VII religious
discrimination claims. I previously dismissed Plaintiffs’ constitutional claims against the
individual defendants, leaving only the Title VII claims. Dkt. 163.1

1 I also dismissed the Title VII claims asserted by certain Plaintiffs who failed to exhaust their
administrative remedies by filing a complaint with the Equal Employment Opportunity
Commission (“EEOC”). Dkt. 162 at 12–13.
On April 8, 2024, Plaintiffs filed their Third Amended Complaint, which added a new
Plaintiff, Donyll Dameron, and her employer, Defendant UVA Imaging. Dkt. 171. Dameron
alleges that she was jointly employed by UVA Imaging and UVA Health when UVA Health
implemented its vaccine mandate. Dameron, like the other named Plaintiffs, applied for a

religious exemption and was fired after her exemption was denied without explanation.
Plaintiffs now move to certify three classes with respect to their remaining Title VII
claims. Dkt. 221. The “Disfavored Religions Class” consists of approximately 210 members and
is defined as:
UVA Health employees or job applicants who (1) requested a religious
accommodation from COVID vaccination, (2) were not members of UVA’s
‘established religions,’ and (3) suffered adverse employment action from UVA as
a result of being unvaccinated.2

Dkt. 222 at 22.
The “Abortion Objectors Class” consists of approximately 120 members and is defined
as:
UVA Health employees or job applicants who: (1) submitted requests for
accommodations from COVID vaccination based on the vaccines’ association
with abortion or fetal cell lines, and (2) suffered adverse employment action as a
result of being unvaccinated.

Id. at 23.
The “Failure-to-Accommodate Class” consists of approximately 210 members and is
defined as:

2 Although Plaintiffs do not limit any of their three proposed classes by time limit or policy, all
of their claims appear to arise under the 2021 Procedure. They do not allege that the 2022
Procedure is discriminatory but, in fact, cite to the new policy as evidence that UVA Health
recognized its 2021 Procedure was discriminatory. See, e.g., Dkt. 222 at 19 (“UVA Health
applied the discriminatory policies . . . for about [a] year, until October 2022.”). This Opinion
thus assumes that each of the proposed classes includes only those members who requested
accommodations according to the 2021 Procedure.
UVA Health employees or job applicants who (1) requested accommodations
from COVID vaccination on religious grounds and (2) suffered adverse
employment action as a result of being unvaccinated.3

Id.
Defendants UVA Health and UVA Imaging filed separate memoranda in opposition to the
motion for class certification. Dkts. 234, 235. The motion is ripe for the Court’s review.
II. Analysis
A. Scope of Relief as to UVA Imaging
Plaintiffs seek class certification only against Defendant UVA Health and do not request
certification of any class claims against Defendant UVA Imaging. Although Plaintiffs allege in
their complaint that UVA Health is the employer or joint employer of all UVA Imaging
employees, Dkt. 171 ¶ 40, they do not develop that theory further. Defendants, in turn, contend
that Plaintiffs’ joint employer allegations are vague and conclusory, Dkt. 185 at 63, but do not
present evidence or argument contesting joint employment status.
Because Plaintiffs have not moved for certification as to UVA Imaging and the parties
have not provided a sufficient record to resolve the joint employment issue at this stage, I decline
to extend classwide relief to UVA Imaging at this time. Any relief sought or granted under this
class certification order is therefore limited to UVA Health and is without prejudice to UVA
Imaging’s liability. If Plaintiffs later establish a joint employer relationship, appropriate relief
against UVA Imaging may be considered at that stage. See, e.g., Butler v. Drive Auto. Indus. of

3 Although Plaintiffs define the Failure-to-Accommodate Class and the Disfavored Religions
Class differently, both classes seem to contain the same members. Plaintiffs’ theory is that all
employees and applicants who were denied accommodations and suffered adverse employment
actions (members of the Failure-to-Accommodate Class) necessarily did not belong to UVA
Health’s six established religions and are therefore also members of the Disfavored Religions
Class.
Am., Inc., 793 F.3d 404, 409 (4th Cir. 2015). The possibility that UVA Imaging employees may
later require subclassing does not provide a basis to deny certification as to UVA Health.
B. Exhaustion of Administrative Remedies
Under Title VII, an employee “seeking redress for discrimination cannot file suit until she

has exhausted the administrative process.” Balas v. Huntington Ingalls Indus., Inc., 711 F.3d
401, 406 (4th Cir. 2013) (citing 42 U.S.C. § 2000e-5(b)). Indeed, “federal courts lack subject
matter jurisdiction over Title VII claims for which a plaintiff has failed to exhaust administrative
remedies.” Id. The administrative process serves to “notif[y] the charged party of the asserted
violation . . . [and] permit[] effectuation of the [Civil Rights] Act’s primary goal, the securing of
voluntary compliance with the law.” Id. at 407 (internal quotation marks and citations omitted).
Title VII complaints are circumscribed by a plaintiff’s EEOC charge. “Only those
discrimination claims stated in the initial charge, those reasonably related to the original
complaint, and those developed by reasonable investigation of the original complaint may be
maintained in a subsequent Title VII lawsuit.” Stewart v. Iancu, 912 F.3d 693, 705 (4th Cir.

2019) (internal quotation marks and citations omitted). “However, because EEOC charges are
not normally drafted by lawyers, courts are to construe them liberally, requiring only that the
factual allegations of the charge reasonably relate to the factual allegations of the formal
litigation.” Boyd v. Miller Pipeline Corp., No. 3:16-CV-00278, 2017 WL 2990292, at *6
(W.D.N.C. July 13, 2017) (citing Harrison v. S.C. Dep’t of Mental Health, 641 F. App’x 202,
209 (4th Cir. 2015)), aff’d sub nom. Boyd v. Pipeline, 709 F. App’x 205 (4th Cir. 2018).
Defendants first argue that Plaintiffs failed to exhaust their administrative remedies as to
any pattern-or-practice claim because none of the named Plaintiffs’ EEOC charges identified a
broader discriminatory policy. Instead, Plaintiffs’ EEOC charges merely described “their
individual circumstances and requests for accommodation.” Dkt. 235 at 44. I disagree.
In the context of class actions or “pattern or practice” claims, courts are divided as to
whether individual claims of discrimination in an EEOC charge are sufficient to support a future

class action. Compare Kloos v. Carter-Day Co., 799 F.2d 397, 400 (8th Cir. 1986) (“To be
faithful to the purposes of the filing requirement, an administrative charge must allege class-wide
age discrimination or claim to represent a class in order to serve as the basis for an ADEA class
action.”), with Paige v. California, 102 F.3d 1035, 1041–42 (9th Cir. 1996) (finding that an
individual race discrimination charge could a support class action), and Fellows v. Universal
Rests., Inc., 701 F.2d 447, 450–51 (5th Cir. 1983) (explaining that the fact that an EEOC charge
and investigation did not cover class allegations was not a sufficient basis to deny the class). The
Fourth Circuit has not addressed the issue directly except to note that an EEOC charge should
“provide sufficient notice of the collective nature of [a plaintiff’s] claims.” See King v.
McMillan, 233 F. App’x 242, 244 (4th Cir. 2007). “An administrative charge of discrimination

does not strictly limit a Title VII suit which may follow.” Chisholm v. U.S. Postal Serv., 665
F.2d 482, 491 (4th Cir. 1981). What matters is whether the employer was put on notice that the
alleged discriminatory practice could affect a class of individuals, and whether a classwide
investigation could reasonably be expected to grow out of the charges. See id.
Applying these principles here, I find that Plaintiffs’ EEOC charges provided sufficient
notice to UVA Health of potential class-based discrimination claims. Plaintiffs allege that they
were denied religious exemptions for the COVID-19 vaccine. Those denials occurred under
UVA Health’s 2021 Procedure, which included a policy or practice for evaluating religious
exemption requests that favored certain religions over others. Specifically, UVA Health
instructed its Religious Exemption Committee to approve exemption requests from adherents of
six designated congregations or denominations so long as there was “no reason to suspect
dishonesty.” Dkt. 235-30 at 15. Requests citing to religious beliefs other than the six identified
faiths were forwarded to the Religious Exemption Committee “to determine whether there is

sufficient evidence of a sincerely held religious belief to grant an exemption.” Id. at 16. The
policy’s express classification of specific religious groups created a class of similarly situated
individuals who were subjected to the same discriminatory policy based on their religious
affiliation. While Plaintiffs’ EEOC charges may not explicitly reference the 2021 Procedure,4 the
charges relate to the persistent denial of a religious exemption request under that procedure.
Moreover, the seven named Plaintiffs filed charges around the same time, each alleging religious
discrimination arising from the same exemption-review process. An investigation into Plaintiffs’
allegations could reasonably be expected to uncover the broad and sweeping implications of
denying other potentially valid exemption requests. These charges put UVA Health on notice of
the widespread impact the 2021 Procedure had on its employees and the potential for classwide

exposure from those who sought vaccine exemptions but were not members of the enumerated
religions.
Defendants also argue that Plaintiffs failed to exhaust any claim that UVA Health
maintained a policy of rejecting abortion-based religious objections. The record does not support
that argument. Several named Plaintiffs’ EEOC charges expressly stated that their religious
objections were rooted in opposition to abortion or the use of fetal cell lines in vaccine
development. See Dkt. 222 at 15–17. These allegations squarely presented UVA Health with the

4 Notably, however, Plaintiff Ehrlich’s EEOC charge alleged that he was denied an exemption
request under UVA’s Policy OCH-002 because his application was “not based on an established
religion such as Christian Scientists.” Dkt. 235-40 at 6–7.
nature of Plaintiffs’ religious beliefs and the basis for their objections. An EEOC investigation
into why UVA Health denied these abortion-based requests would naturally encompass whether
it applied a consistent rationale or practice in evaluating such objections.
Accordingly, I find that Plaintiffs have sufficiently exhausted their administrative

remedies as to each of their class claims.
C. Class Certification Under Rule 23
A plaintiff seeking to certify a class “must affirmatively demonstrate his compliance with
[Federal Rule of Civil Procedure 23].” Wal-Mart Stores, Inc. v. Dukes, 564 U.S. 338, 350 (2011).
The Supreme Court has explained that the requirements of Rule 23 are not mere pleading
standards. Id. Rather, a plaintiff must be prepared to prove that there are in fact sufficiently
numerous parties, common questions of law or fact, typicality of claims or defenses, and
adequacy of representation as required by Rule 23(a). Id. Additionally, a plaintiff must prove that
the proposed class satisfies one of the provisions of Rule 23(b). See Comcast Corp. v. Behrend, 569 U.S. 27, 33 (2013).

In evaluating a motion for class certification, a “district court has an independent
obligation to perform a rigorous analysis to ensure that all of the prerequisites have been
satisfied.” EQT Prod. Co. v. Adair, 764 F.3d 347, 358 (4th Cir. 2014) (internal quotation marks
and citation omitted). This analysis may “entail some overlap with the merits of the plaintiff's
underlying claim.” Amgen Inc. v. Conn. Ret. Plans & Tr. Funds, 568 U.S. 455, 465–66 (quoting
Dukes, 564 U.S. at 351). But Rule 23 does not give courts “license to engage in free-ranging
merits inquiries at the certification stage.” Id. at 466. Rather, “the merits of a case may be
considered ‘only to the extent . . . that they are relevant to determining whether the Rule 23
prerequisites for class certification are satisfied.’” Student A v. Liberty Univ., Inc., 697 F. Supp.
3d 496, 522 (W.D. Va. 2023) (quoting Amgen Inc., 568 U.S. at 465–66).
(1) Disfavored Religions Class
The Disfavored Religions Class is represented by all named Plaintiffs. It consists of UVA

Health employees or job applicants who (1) requested a religious accommodation from COVID
vaccination, (2) were not members of UVA’s “established religions,” and (3) suffered adverse
employment action from UVA as a result of being unvaccinated.
(a) Ascertainability
“Rule 23 contains an implicit threshold requirement that the members of a proposed class
be ‘readily identifiable.’” EQT Prod. Co., 764 F.3d at 358 (quoting Hammond v. Powell, 462
F.2d 1053, 1055 (4th Cir. 1972)). To satisfy the ascertainability requirement, “[p]laintiffs need
not be able to identify each and every class member at the time of class certification, but class
members must be identifiable by reference to objective criteria without extensive and
individualized fact-finding or mini-trials.” Williams v. Big Picture Loans, LLC, 339 F.R.D. 46,

55 (E.D. Va. 2021), aff’d sub nom. Williams v. Martorello, 59 F.4th 68 (4th Cir. 2023) (internal
quotation marks and citation omitted).
The members of this class are readily ascertainable from UVA Health’s records. UVA
Health can identify those employees who applied for religious accommodations from the
vaccination requirement and had their exemption requests denied, as well as those employees
who eventually received vaccination or refused the vaccine and were terminated.
(b) Numerosity
Numerosity is satisfied when joinder of all class members is impracticable. Fed. R. Civ.
P. 23(a)(1). “No bright line test exists for determining numerosity and the determination rests on
the court’s practical judgment in light of the particular facts of the case.” Harris v. Rainey, 299
F.R.D. 486, 489 (W.D. Va. 2014) (quoting Savani v. Wash. Safety Mgmt. Sols., LLC, No. 1:06-
CV-02805-MBS, 2012 WL 3757239, at *2 (D.S.C. Aug. 28, 2012)). “Additionally, it is not
required that the exact size of a class be established.” Id. “Plaintiff[s] need only make a

reasonable estimate of the number of class members.” Id. (quoting Wiseman v. First Citizens
Bank & Tr. Co., 212 F.R.D. 482, 486 (W.D.N.C. 2003)).
Here, Plaintiffs estimate that there are at least 210 members of the Disfavored Religions
Class.5 I find that the numerosity requirement is met. In this circuit, class actions have been
found to meet the numerosity requirement with as few as 18 members. See Cypress v. Newport
News Gen. & Nonsectarian Hosp. Ass’n, 375 F.2d 648, 653 (4th Cir. 1967).
(c) Commonality
To meet the commonality requirement, Plaintiffs must show that “there are questions of
law or fact common to the class.” Fed. R. Civ. P. 23(a)(2). The class members’ claims must
“depend upon a common contention,” the determination of which “will resolve an issue that is

central to the validity of each one of the claims in one stroke.” Dukes, 564 U.S. at 350; see also
EQT Prod. Co., 764 F.3d at 360. One way to establish commonality is to offer “[s]ignificant
proof that an employer operated under a general policy of discrimination.” Dukes, 564 U.S. at
353 (quoting Gen. Tel. Co. of Sw. v. Falcon, 457 U.S. 147, 159 n.15 (1982)).
Plaintiffs here challenge a uniform policy that presumptively approved religious
exemptions for adherents of six favored religions while subjecting other employees to heightened

5 Excluding applicants and UVA Imaging employees, see supra Section II.A; infra Section
II.C.1.d., may reduce the total number of potential class members. Even so, Plaintiffs represent
that 200 class members were fired from UVA Health, see Dkt. 222 at 15, and this population is
still sufficiently numerous to satisfy Rule 23(a)(1).
scrutiny. UVA Health responds that its policy was not discriminatory because “[w]hile
Committee members were advised to approve [] requests [citing one of the six established
religions] unless there was reason to suspect dishonesty . . . they were still individually
assessed.” Dkt. 235 at 55 (emphasis in original). This argument is not persuasive. To the

contrary, it confirms that individuals who identified with one of the established religions were
given a presumption of sincerity that was not extended to others. Even if UVA Health
independently evaluated each exemption request, Plaintiffs may satisfy the commonality
requirement by showing a university-wide policy of discrimination. See Scott v. Fam. Dollar
Stores, Inc., 733 F.3d 105, 113–14 (4th Cir. 2013) (“[E]ven where company-wide subjective
decision-making or discretion is alleged in the employment discrimination context . . . if another
company-wide policy is also alleged, courts must also consider it.”) (citing Dukes, 564 U.S. at
353); G.T. v. Bd. of Educ. of Kanawha, 117 F.4th 193, 205 (4th Cir. 2024) (holding that plaintiffs
in an IDEA case must identify a “uniformly applied, official policy . . . or an unofficial yet well-
defined practice, that drives the alleged violation”) (citations omitted). Because UVA Health

employed the same standardized procedure to evaluate its employees’ requests, if that procedure
violated the named Plaintiffs’ rights, it violated other class members’ rights as well. See Thomas
v. FTS USA, LLC, 312 F.R.D. 407, 418 (E.D. Va. 2016); Falcon, 457 U.S. at 159 n.15 (“If
petitioner used a biased testing procedure to evaluate both applicants for employment and
incumbent employees, a class action on behalf of every applicant or employee who might have
been prejudiced by the test clearly would satisfy the commonality and typicality requirements
of Rule 23(a).”).
Statistical evidence further supports Plaintiffs’ allegations of a pattern or practice of
discrimination. Plaintiffs cite evidence that UVA Health granted all requests for exemption
submitted by individuals who identified themselves as members of one of the six listed religions
in their initial applications. In contrast, of the 397 requests for vaccine exemptions that do not
reference one of the established religions, UVA Health granted only four. Dkt. 240 at 6. This
statistical disparity “ties class members’ claims together, notwithstanding individualized

differences that might have controlled in the absence of such common proof.” Scott v. Fam.
Dollar Stores, Inc., No. 308CV00540MOCDSC, 2016 WL 9665158, at *7 (W.D.N.C. June 24,
2016) (citing Tyson Foods, Inc. v. Bouaphakeo, 577 U.S. 442, 448–52 (2016)); see also Brown v.
Nucor Corp., 785 F.3d 895, 914 (4th Cir. 2015) (explaining that statistical and anecdotal
evidence “provide precisely the glue of commonality that Wal-Mart demands” (internal
quotation marks and citation omitted)).
I find that Plaintiffs have demonstrated the existence of a discriminatory policy that raises
a central legal question common to all class members: whether UVA Health’s exemption process
violated Title VII by treating some religious beliefs as more valid than others. Because this
question can be answered “in one stroke” using common facts, it is appropriately resolved on a

class-wide basis.
(d) Typicality
Typicality is satisfied when the named plaintiffs’ claims are “typical of the claims or
defenses of the class,” ordinarily requiring that the class representatives’ claims arise from the
same event, practice, or course of conduct as the putative class members’ claims. Fed. R. Civ. P.
23(a)(3). This requirement ensures that “only those plaintiffs . . . who can advance the same
factual and legal arguments may be grouped together as a class.” Broussard v. Meineke Disc.
Muffler Shops, Inc., 155 F.3d 331, 340 (4th Cir. 1998).
Here, the named Plaintiffs and the class allege the same pattern of conduct: UVA
Health’s denial of certain religious exemptions under a single discriminatory policy. Just as in
the commonality element, the named Plaintiffs’ claims arise from the same theory as those of the
class—that UVA Health treated their exemption requests differently because their religious

beliefs were not among the six favored religions.
Defendants argue that typicality does not exist with respect to team members who
resigned, applicants, and non-UVA Health employees because those subclasses are subject to a
unique defense. “Ordinarily, where a purported class representative is subject to a unique defense
that cannot be asserted against other members of the class . . . typicality may be lacking.
However . . . [a] unique defense will render the proposed class representative’s claims atypical
only if it is likely to be a major focus of the litigation.” In re Under Armour Sec. Litig., 631 F.
Supp. 3d 285, 302 (D. Md. 2022) (internal quotation marks and citations omitted).
In this case, Defendants’ arguments do not defeat typicality for the class as a whole.
While class members who resigned must prove constructive discharge, differences in the type of

adverse employment action do not necessarily defeat typicality so long as all claims arise from a
common discriminatory policy. See Carpenter v. Stephen F. Austin State Univ., 706 F.2d 608,
617 (5th Cir. 1983) (finding satisfaction of commonality and typicality where plaintiffs alleged
similar discriminatory employment practices, though the nature of their injuries differed).
Still, Defendants’ arguments regarding applicants and new hires are not without merit.
Applicants must prove that they were deterred from accepting job offers due to UVA Health’s
policy, requiring a different causation inquiry. Further, although Plaintiffs have presented
evidence that UVA Health used the same substantive criteria to evaluate exemption requests
from employees and applicants, UVA Health represents that its applicant tracking system does
not record the outcome of exemption requests. See Dkt. 235 at 42 n.19. As a result, UVA Health
cannot reliably identify which applicants were denied exemptions unless the applicant was
ultimately hired. Because this prevents the Court from determining class membership with
reasonable accuracy, applicants are excluded from the class definitions at this time. This

exclusion does not foreclose applicants’ individual claims, nor does it preclude the Court from
revisiting the issue if Plaintiffs identify a reliable method of determining outcomes for that
cohort. See Fed. R. Civ. P. 23(c)(1)(C) (“An order that grants or denies class certification may be
altered or amended before final judgment.”).
(e) Adequacy
The adequacy requirement demands that the class representatives “fairly and adequately
protect the interests of the class.” Fed. R. Civ. P. 23(a)(4). “Representation is adequate if: (1) the
named plaintiffs’ interests are not opposed to those of other class members, and (2) the plaintiffs’
attorneys are qualified, experienced and able to conduct the litigation.” Cuthie v. Fleet Rsrv.
Ass’n, 743 F. Supp. 2d 486, 499 (D. Md. 2010) (citation omitted); see also Amchem Prods., Inc.

v. Windsor, 521 U.S. 591, 625–26 (1997). Defendants do not challenge the adequacy of the
named Plaintiffs or class counsel, and there is no evidence suggesting either are inadequate. I
therefore find that the named Plaintiffs fairly and adequately protect the interests of the class.
(f) Rule 23(b)(3)
If a class meets the Rule 23(a) prerequisites, it must then fall into one of three categories
set out in Rule 23(b). Plaintiffs here seek certification under Rule 23(b)(3), which requires that
“the questions of law or fact common to class members predominate over any questions affecting
only individual members, and . . . a class action is superior to other available methods for fairly
and efficiently adjudicating the controversy.”
(i) Predominance
The “predominance inquiry tests whether proposed classes are sufficiently cohesive to
warrant adjudication by representation.” Amchem Prods., 521 U.S. at 623. For each cause of
action, the court characterizes issues as common or individual. See Tyson Foods, [577 U.S. at

453](https://www.courtlistener.com/opinion/3187592/tyson-foods-inc-v-bouaphakeo/#453). “An individual question is one where ‘members of a proposed class will need to present
evidence that varies from member to member,’ while a common question is one where ‘the same
evidence will suffice for each member to make a prima facie showing [or] the issue is susceptible
to generalized, class-wide proof.’” Id. (citing William B. Rubenstein, Newberg on Class Actions
§ 4:50 (5th ed. 2012) (internal quotation marks omitted)). The court then “asks whether the
common, aggregation-enabling, issues in the case are more prevalent or important than the non-
common, aggregation-defeating, individual issues.” Id. (quoting Rubenstein, supra, § 4:49). The
requirement that common questions of law or fact predominate “is similar to but more stringent
than the commonality requirement of Rule 23(a).” Thorn v. Jefferson-Pilot Life Ins. Co., 445
F.3d 311, 319 (4th Cir. 2006) (internal quotation marks and citation omitted).

“Rule 23(b)(3)’s [predominance] test is qualitative rather than quantitative.” Soutter v.
Equifax Info. Servs., LLC, 307 F.R.D. 183, 214 (E.D. Va. 2015) (quoting Stillmock v. Weis Mkts.,
Inc., 385 F. App’x 267, 272 (4th Cir. 2010)). “If the ‘qualitatively overarching issue’ in the
litigation is common, a class may be certified notwithstanding the need to resolve individualized
issues.” Id. (citing Ealy v. Pinkerton Gov’t Servs., Inc., 514 F. App’x 299, 305 (4th Cir. 2013)).
Thus, Rule 23(b)(3) “does not require a plaintiff seeking class certification to prove that each
elemen[t] of [her] claim [is] susceptible to classwide proof,” but “does require [] that common
questions predominate over any questions affecting only individual [class] members.” Amgen
Inc., 568 U.S. at 469 (emphasis in original) (internal quotation marks and citation omitted). As
long as individual questions will not overwhelm common ones, they do not make certification
inappropriate. See Halliburton Co. v. Erica P. John Fund, Inc., 573 U.S. 258, 276 (2014).
To prove a Title VII claim under a disparate treatment theory, a plaintiff “must
demonstrate that the employer treated her differently than other employees because of her

religious beliefs.” Chalmers v. Tulon Co. of Richmond, 101 F.3d 1012, 1017 (4th Cir. 1996)
(emphasis in original). This requires an individual employee to show: “(1) membership in a
protected class; (2) satisfactory job performance; (3) adverse employment action; and (4)
different treatment from similarly situated employees outside the protected class.” Coleman v.
Maryland Ct. of Appeals, 626 F.3d 187, 190 (4th Cir. 2010), aff’d sub nom. Coleman v. Ct. of
Appeals of Maryland, 566 U.S. 30 (2012) (citation omitted). In a class action, however, Plaintiffs
can establish liability for disparate treatment by “establish[ing] that a pattern or practice of
discrimination exists.” Muhammad v. Giant Food Inc., 108 F. App’x 757, 764 (4th Cir. 2004)
(emphasis omitted) (citing Cooper v. Fed. Rsrv. Bank of Richmond, 467 U.S. 867, 875 (1984)).
Here, Plaintiffs’ claims arise out of UVA Health’s religious exemption procedure. The

central question is whether this procedure included a policy that facially discriminated against
certain religious beliefs in violation of Title VII. This question affects every class member
identically. Because Plaintiffs have proffered evidence of a pattern or practice of discrimination
that could establish liability and entitle all class members to a presumption of individualized
harm, I find that the predominance standard is satisfied. See Scott, 2016 WL 9665158, at *8.
While individual determinations may be necessary to assess whether each class member
would have received a religious exemption under a neutral policy, such determinations do not
defeat commonality. Any individualized inquiries regarding sincerity of belief or actual
entitlement to accommodation may be addressed at the remedial stage. See Brown, 785 F.3d at
914 (recognizing second damages stage) (citing Int’l Bhd. of Teamsters v. United States, 431
U.S. 324, 361 (1977)). So long as common questions predominate as to liability, “the need for
individualized proof of damages alone will not defeat class certification.” Gunnells v. Healthplan
Servs., Inc., 348 F.3d 417, 429 (4th Cir. 2003) (emphasis in original) (citations omitted). In fact,

Rule 23 explicitly envisions class actions with such individualized damage determinations. See
Fed. R. Civ. P. 23 advisory committee’s note to 1966 amendment, subdiv. (c)(4) (noting
that Rule 23(c)(4) permits courts to certify a class with respect to particular issues and
contemplates possible class adjudication of liability issues with “the members of the
class . . . thereafter . . . required to come in individually and prove the amounts of their respective
claims”).
(ii) Superiority
The superiority analysis under Rule 23(b)(3) requires the court to consider whether
alternatives to a class action would better “dispose[] of the dispute before it. Soutter, 307 F.R.D.
at 217–18. “In determining whether the class mechanism is truly superior, the court should

consider (1) the interest in controlling individual prosecutions; (2) the existence of other related
litigation; (3) the desirability of concentrating the litigation in the forum; and (4) manageability.” Id. (internal quotation marks and citation omitted).
Plaintiffs contend that class litigation would be more manageable than individual suits
because each class member would need to present the same evidence of UVA Health’s
discriminatory policy. Defendants, in turn, argue that predominating individual issues render
class treatment inferior. For the reasons discussed above, I find that the Disfavored Religion
Class’s claims are susceptible to common proof such that allowing the action to proceed as a
class action would promote judicial economy. Individualized damages calculations alone do not
defeat certification where common issues predominate on liability. Moreover, the alternative to
class treatment—dozens of individual cases raising identical legal questions about the same
policy—would create far greater inefficiency and potential for inconsistent outcomes than a
single class proceeding with individualized damages determinations.

Defendants also contend that the damages available under Title VII provide sufficient
incentive for individual litigation, evidenced by three individual class members’ suits. Dkt. 235
at 63. However, the relevant inquiry is not whether some individuals have sufficient incentive to
sue but whether class treatment better serves the interests of justice and efficiency for the
controversy as a whole. The existence of three individual suits among a much larger affected
group suggests that most potential class members face practical barriers to individual litigation
that class treatment would overcome. See Scott, 2016 WL 9665158, at *8 (noting that class
treatment would “allow class members to seek relief . . . that they would otherwise be unable to
pursue because of financial limitations or fear of retaliation”). Plaintiffs have therefore satisfied
the superiority requirement.

(2) Abortion Objectors Class
The Abortion Objectors Class is represented by five of the named Plaintiffs. It consists of
UVA Health employees or job applicants who (1) submitted requests for accommodations from
COVID vaccination based on the vaccines’ association with abortion or fetal cell lines, and (2)
suffered adverse employment action because they did not receive the vaccination.
(a) Ascertainability
As with the Disfavored Religions Class, I find that this class is readily ascertainable from
UVA Health’s records.
UVA Health argues that a class can only contain members who have sincere religious
beliefs because individuals with insincere religious beliefs are not protected by Title VII. But
ascertainability requires only that class membership be defined by “objective criteria,” EQT
Prod. Co., 764 F.3d at 358, not that those criteria be entirely divorced from the underlying

liability inquiry. Because membership in the Abortion Objectors Class turns on objective,
administratively determinable facts—whether the individual submitted a religious-exemption
request on a stated basis and suffered adverse action—ascertainability is satisfied even if
sincerity must later be evaluated on the merits.
(b) Numerosity
Plaintiffs estimate that there are at least 120 members of the Abortion Objectors Class.
This is sufficient to satisfy the numerosity requirement.
(c) Commonality
“[I]n resolving an individual’s Title VII claim, the crux of the inquiry is the reason for a
particular employment decision.” Dukes, 564 U.S. at 352 (internal quotation marks and citation

omitted). A procedure that requires individualized discretion in decision-making may defeat
commonality “[w]ithout some glue holding the alleged reasons for all those decisions together.” Id. (emphasis in original).
Here, Plaintiffs allege that UVA Health fired employees who would not adhere to its
official religious doctrines on vaccines and abortion. Unlike the Disfavored Religions Class,
Plaintiffs in the Abortion Objectors Class do not point to any facially discriminatory policy.
Instead, they cite to UVA Health’s “companywide religious orthodoxy about abortion and
vaccines,” Dkt. 222 at 22, and the fact that UVA Health fired 99 percent of employees who acted
according to a different view, id. at 4.
This evidence does not establish the “significant proof” of a common discriminatory
policy that Dukes requires. Plaintiffs fail to provide evidence of a common policy or standard
that governed UVA Health’s decisions beyond general denial, nor do they identify any consistent
criteria used by UVA Health in rejecting requests related to abortion or fetal cells. To the

contrary, the record shows that UVA Health evaluated these exemption requests for sincerity like
any other requests that did not reference one of the six established religions in the 2021
Procedure. See, e.g., Dkt. 235-34 (noting that Tyson’s belief was insincere because she
previously received other vaccinations).
Plaintiffs’ statistical evidence does not bridge the gap between an individual employee’s
discrimination claim and “the existence of a class of persons who have suffered the same injury.”
Falcon, 457 U.S. at 157–58. Without evidence linking high denial rates to a specific
discriminatory practice, the statistics cannot supply the “glue” that Dukes demands. See Dukes,
564 U.S. at 356–57 (finding that statistical proof does not demonstrate commonality where
plaintiffs have not identified a “specific employment practice”). The critical question is not how

often UVA Health denied abortion-based requests, but why and whether those reasons were
common across the class.
Likewise, the commentary accompanying UVA Health’s denial of accommodation
requests, see Dkt. 222 at 13, also does not establish a pattern or practice. While the comments
suggest that some UVA Health employees were resistant to exemption requests based on
abortion or fetal cells, they reveal only “isolated or accidental or sporadic discriminatory acts”
that do not support a pattern or practice. Teamsters, 431 U.S. at 336 (internal quotation marks
omitted).
Finally, the personal nature of Plaintiffs’ religious claims further undermines
commonality. Although this class collectively challenges a set of rejections based on abortion-
related objections, the specific beliefs, explanations, and supporting statements varied
significantly among class members. Some employees objected based on complicity in abortion,

others objected to the use of fetal cell lines in research, and still others framed their objections in
broader theological terms. See Dkt. 235 at 25–35. Addressing these issues requires employee-by-
employee inquiries into individual beliefs, communications, and justifications. See Ellison v.
Innova Health Care Servs., 692 F. Supp. 3d 548, 555–56 (E.D. Va. 2023) (explaining that
plaintiffs’ class claims “would require the Court to ask whether each putative class member’s
abortion-based objection to the COVID-19 vaccine was based on their own religious beliefs”
(emphasis in original)); Falcon, 457 U.S. at 157–59 (finding a lack of commonality and
typicality where plaintiff failed to establish a company-wide policy of discrimination).
Because Plaintiffs have not presented meaningful evidence of a uniform policy or
practice governing UVA Health’s treatment of abortion-based objections, and because

adjudicating these claims would require individualized inquiries into each employee’s beliefs
and the reasons for denial, I find class-wide resolution to be unworkable.
(d) Typicality
Plaintiffs have failed to establish commonality, so the Court need not analyze the
remaining Rule 23(a) requirements. Nevertheless, as discussed in the commonality analysis,
Plaintiffs seek to litigate not a uniform policy, but a set of discretionary decisions. Because the
reasons for denial varied among class members and turn on individualized facts, the named
Plaintiffs’ experiences therefore are not necessarily typical of the class.
(e) Adequacy
While adequacy is likely satisfied, it cannot overcome the absence of commonality and
typicality.
(3) Failure-to-Accommodate Class
The “Failure-to-Accommodate” class is represented by all named Plaintiffs. It consists of

all UVA Health employees or job applicants who (1) requested accommodations from COVID
vaccination on religious grounds and (2) suffered adverse employment action because of
being unvaccinated.
(a) Ascertainability
For the reasons discussed earlier, I find that this class is readily ascertainable from UVA
Health’s records.
(b) Numerosity
Plaintiffs estimate that there are at least 210 members of the Failure-To-Accommodate
Class. This is sufficient to satisfy the numerosity requirement.
(c) Commonality

Unlike the Disfavored Religions Class, which challenges a discriminatory policy, the
Failure-to-Accommodate Class seeks to litigate claims arising from UVA Health’s
individualized assessments of religious exemption requests.
UVA Health argues that Plaintiffs cannot satisfy a class claim for failure-to-
accommodate because individualized issues of religiosity, sincerity, and hardship predominate.
Plaintiffs respond that they do not have to address religiosity, sincerity, and hardship because
UVA Health had a pattern or practice of denying requests for accommodation as “insincere,” and
because UVA later conceded that most requests were in fact sincere. Neither of Plaintiffs’
arguments establishes commonality.
The “pattern-or-practice” framework applies only to disparate treatment claims. See
United States v. City of New York, 717 F.3d 72, 83 (2d Cir. 2013) (“[A] pattern or practice case is
not a separate and free-standing cause of action . . . but is really merely another method by which
disparate treatment can be shown.”).6 Members of the Failure-to-Accommodate Class must

therefore establish the elements of a failure to accommodate claim. For religious accommodation
cases, a plaintiff must prove that “(1) he or she has a bona fide religious belief that conflicts with
an employment requirement; (2) he or she informed the employer of this belief; and he or
she was disciplined for failure to comply with the conflicting employment requirement.” EEOC
v. Firestone Fibers & Textiles Co., 515 F.3d 307, 312 (4th Cir. 2008). If the plaintiff meets this
burden, the employer can avoid liability showing that it could not reasonably accommodate the
plaintiff’s needs without undue hardship. Id. (citing Chalmers, 101 F.3d at 1019).
To prove a “bona fide religious belief,” the plaintiff must show that her belief is both
sincere and religious. In evaluating sincerity, the court “seeks to determine an adherent’s good
faith in the expression of [her] religious belief.” Barnett v. Inova Health Care Servs., 125 F.4th

6 Courts have sometimes analyzed failure-to-accommodate cases under a pattern-or-practice
framework where there was clear evidence of a policy or pattern or practice and no consideration
was given to individual requests for accommodation. See, e.g., United States v. New York City
Transit Auth., No. 04-CV-4237, 2010 WL 3855191, at *1 (E.D.N.Y. Sept. 28, 2010) (“At all
times relevant to this action, the TA has had formal uniform policies.”); Equal Emp. Opportunity
Comm’n v. JBS USA, LLC, 115 F. Supp. 3d 1203, 1225 (D. Colo. 2015) (“JBS concedes . . . that
its conduct related to religious accommodations was a pattern or practice.”). In this case, the
record does not show that UVA Health maintained a formal, uniform policy of denying all
religious exemption requests, nor does UVA Health concede that its conduct was a pattern or
practice. Plaintiffs cite Doster v. Kendall, 54 F.4th 398 (6th Cir. 2022), but Doster is
distinguishable on its basic undisputed facts. In Doster, the Air Force “denied all religious-
exemption requests unless a service member . . . agreed to leave within a certain time,” creating a
general policy that the Air Force could either justify or not. Id. at 437. Here, by contrast, UVA
Health evaluated each request individually, considering religious affiliation and applying a
sincerity and hardship analysis to requests outside the six enumerated faiths. Where individual
discretion is exercised, a high denial rate reflects the aggregate of many individual decisions, not
a single common policy. See Dukes, 564 U.S. at 355–58.
465, 470 (4th Cir. 2025) (citation omitted). This analysis requires an individualized assessment.
The evidence in the record shows that employees seeking religious exemptions from UVA
Health’s vaccine mandate sent individual applications describing their unique beliefs, history
with their faith, and other factual circumstances. The Court would need to examine each class

member’s application to determine whether they contain a good-faith expression of religious
belief to establish UVA Health’s liability to that class member. See Meinert v. Port Auth. of
Allegheny Cnty., No. 2:22-CV-01736-RJC, 2025 WL 903815, at *10 (W.D. Pa. Mar. 25, 2025)
(finding that commonality was not satisfied because “[d]etermining the religious sincerity of
each putative class member’s religious beliefs is an inherently individual inquiry that would
require the Court to consider the exemption request and genuine motivations of every Plaintiff”);
Chavez v. San Francisco Bay Area Rapid Transit Dist., No. C 22-06119 WHA, 2024 WL
332881, at *4 (N.D. Cal. Jan. 28, 2024).7 Similarly, the Court would need to examine each
application to determine whether each applicant’s belief was religious in nature. The evidence
shows that employees sought religious exemptions for a wide range of beliefs, and there is no

evidence from which the Court can conclude that these beliefs were universally religious.
Accordingly, religiosity also requires an individualized assessment. See O’Hailpin v. Hawaiian
Airlines Inc., No. CV C22-00532 JAO-RT, 2023 WL 8600498, at *9 (D. Haw. Dec. 12, 2023)

7 In support of their claims, Plaintiffs cite Kincannon v. United Airlines, Inc., 168 F.4th 713 (5th
Cir. 2026). In Kincannon, the court found that the sincerity issue did not undermine commonality
because employees were required to submit a letter from a third party who would attest to their
religious beliefs. Id. at 728–30. These circumstances are not present here, as UVA Health did not
rely on any uniform, third-party attestation process and instead evaluated each employee’s
sincerity individually.
(explaining that distinguishing between a sincerely held belief and secular preference “involves
individualized inquiries that are not susceptible to common proof”).8
Further, whether UVA Health would suffer an undue hardship requires an additional
individualized inquiry. An undue hardship is one that results in “more than a de minimis cost” to

the employer. Firestone Fibers & Textiles, 515 F.3d at 312 (citations omitted). In reviewing
requests for undue hardship, UVA Health asserts that it considered “patient-safety concerns” and
“whether there would be multiple team members in the same department with an exemption.”
Dkt. 235 at 23. That calculus is different for every plaintiff. See Chavez, 2024 WL 332881, at *5
(“[T]he undue hardship inquiry requires consideration of the particular accommodations at
issue.” (internal quotation marks and citation omitted)). The employees who requested vaccine
exemptions had a variety of occupations and played different roles at UVA Health. Necessarily,
then, the hardship endured from accommodating each employee’s requests for exemptions
varies.
The type of discretionary, individualized decision-making at issue in Plaintiffs’ claims is

not amenable to resolution through common answers. As the Supreme Court has emphasized,
“[w]hat matters to class certification . . . is not the raising of common questions—even in
droves—but rather, the capacity of a class-wide proceeding to generate common answers apt to
drive the resolution of the litigation.” Dukes, 564 U.S. at 350 (emphasis in original) (internal
quotation marks and citation omitted). Because resolution of the claims in the Failure-to-

8 Plaintiffs’ argument that UVA Health cannot contest sincerity because it later approved many
exemption requests does not eliminate individualized issues. The relevant question for Rule 23 is
whether the original denials were made for a common reason. A post-hoc decision does not
retroactively supply the “glue” that is required. Additionally, at no point does UVA Health
concede the issue of religiosity or sincerity. Instead, UVA Health explains that, in 2022, it
“decided to be more lenient in its assessment of religiosity and sincerity, given different
pandemic conditions.” Dkt. 235 at 50 n.23.
Accommodate Class would require a plaintiff-by-plaintiff inquiry into sincerity, reasonableness,
and hardship, the Failure-to-Accommodate Class fails to satisfy Rule 23(a)(2)’s commonality
requirement. See, e.g., Ellison, 692 F. Supp. 3d at 564–66 (striking class allegations because
plaintiffs’ failure-to-accommodate theory would require individualized assessments of religious

beliefs and accommodations).
(d) Typicality
For the same reasons discussed above, the claims of the named Plaintiffs—whose beliefs,
job roles, and exemption rationales vary—are not typical of the class as a whole.
(e) Adequacy
While adequacy is likely satisfied, it cannot overcome the absence of commonality and
typicality.
III. Conclusion
For the reasons stated above, Plaintiffs’ motion for class certification is GRANTED as to
the Disfavored Religions Class and DENIED as to the Abortion Objectors Class and the Failure-

to-Accommodate Class. The Disfavored Religions Class is certified and defined as “UVA Health
employees who (1) requested a religious accommodation from COVID vaccination, (2) were not
members of UVA’s ‘established religions,’ and (3) suffered adverse employment action from
UVA as a result of being unvaccinated.” Relief granted through class certification shall apply
only to Defendant UVA Health and not to Defendant UVA Imaging.
Rule 23(c)(1)(B) requires that a class certification order appoint class counsel under Rule
23(g). Plaintiffs request that their current counsel be appointed to represent the certified class.
After considering the factors set forth in Rule 23(g)(1)(A), I find that Plaintiffs’ counsel will
fairly and adequately represent the interests of the class. Accordingly, I appoint Plaintiffs’
counsel as class counsel for the certified Disfavored Religions Class.
An appropriate order shall issue.
Entered: April 23, 2026

Robert S. Ballou
Robert S. Ballou
United States District Judge