State of West Virginia, ex rel., West Virginia University Hospitals, Inc, and West Virginia Health Systems, Inc, d/b/a WVU Healthcare v. The Honorable Phillip D. Gaujot, Judge of the Circuit Court of Monongalia County Christopher Thomack and Joseph Michael Jenkins ( 2022 )

FILED
June 10, 2022
EDYTHE NASH GAISER, CLERK
SUPREME COURT OF APPEALS
OF WEST VIRGINIA
No. 21-0737, State of West Virginia ex rel. West Virginia University Hospitals, Inc.; and
West Virginia United Health System, Inc., d/b/a WVU Healthcare v. The Honorable Phillip
D. Gaujot, Judge of the Circuit Court of Monongalia County; Christopher Thomack; and
Joseph Michael Jenkins
Armstead, Justice, concurring in part and dissenting in part:
The petitioners in this matter, West Virginia University Hospitals, Inc., and
West Virginia United Health System, Inc., d/b/a WVU Healthcare (collectively “WVU
Hospitals”), allege that the circuit court failed to comply with our mandate in State ex rel.
W. Va. Univ. Hosps., Inc. v. Gaujot, 242 W. Va. 54, 829 S.E.2d 54 (2019) (“Gaujot II”). 1
In particular, they object to the circuit court’s findings, after remand, of commonality and
ascertainability and to the circuit court’s refusal to revisit the subject of predominance
under our more recent decision in State ex rel. Surnaik Holdings of WV, LLC v. Bedell, 244
W. Va. 248, 852 S.E.2d 748 (2020). They object, further, to the circuit court’s alleged
failure to address the ethical concerns, raised in Gaujot II, about attorneys’ participation in
the class. Id. at 64 n.16, 829 S.E.2d at 64 n.16.
1
As the majority notes, the petition requested in this matter represents the
third instance in which this Court has addressed issues related to the underlying civil action
on which the request is based. The previous request for extraordinary relief filed in 2018,
State ex rel. W. Va. Univ. Hosps., Inc. v. Gaujot, 242 W. Va. 54, 829 S.E.2d 54 (2019),
being the second such instance in which issues related to this action have come before us,
I, like the majority, refer to the 2018 action as Gaujot II.
Because I am satisfied with the circuit court’s removal of certain attorneys
from the class in order to address the ethical issues raised in Gaujot II, I agree with the
majority opinion’s refusal to grant relief on that issue. As the majority opinion observes
“[a] writ of prohibition is an extraordinary remedy” and one that “we do not grant . . .
lightly.” Nevertheless, we have also held that “[a] circuit court’s failure to conduct a
thorough analysis of the requirements for class certification pursuant to West Virginia
Rules of Civil Procedure 23(a) and/or 23(b) amounts to clear error.” Surnaik at ___, 852
S.E.2d at 750, syl. pt. 8. Because I believe that the circuit court has yet to conduct a
sufficiently thorough analysis of commonality or ascertainability for purposes of Rule
23(a) or of predominance for purposes of Rule 23(b), I respectfully dissent from those
portions of the majority opinion that find no error on these issues, and I would grant the
writ of prohibition.
Commonality.
Rule 23 authorizes “[o]ne or more members of a class [to] sue . . . as
representative parties” but “only if . . . there are questions of law or fact common to the
class . . . .” W. Va. R. Civ. P. 23(a)(2) (emphasis added). When these parties were before
us in Gaujot II, we found that the circuit court had “exceeded its legitimate powers by
certifying the class” without “conduct[ing] a sufficiently thorough analysis of the case to
determine whether the commonality required . . . under Rule 23 of the West Virginia Rules
of Civil Procedure is present.” Id. at 64, 829 S.E.2d at 64. We granted relief on that basis.
2
On remand, Christopher Thomack and Joseph Jenkins (“Class
Representatives”) conducted additional discovery, and the circuit court subsequently found
commonality by determining that WVU Hospitals’ “average cost” to produce medical
records during the period set forth in the class definition, i.e., from January 18, 2008, to
June 5, 2014 (the “Class Period”), was $2.08. The court reasoned that, “[b]ased upon that
average cost of $2.08 per request fulfilled, it appears that each and every requestor suffered
damages based on the $10.00 search fee alone, without even considering the additional
damages related to the $.40 per page/image fee that was charged for every class member’s
request.” (Emphasis added.)
This so-called “average cost” of $2.08 was the linchpin of the circuit court’s
analysis. The court noted that the medical records reimbursement statute, W. Va. Code §
16-29-2, had changed on June 6, 2014. After quoting both versions of the statute, the
circuit court announced, without any analysis, that both versions contained “the same
restrictions regarding the charges . . . on producing a patient’s medical record (that is, the
fees must be reasonable and based upon the expenses actually incurred).” The circuit court
also noted that WVU Hospitals had conducted a “time study” designed to “accurately
capture [WVU Hospitals’] actual costs incurred in producing medical records” under the
new statutory regime. Resulting invoices for the period of June 6, 2014, through July 31,
2014, showed that WVU Hospitals’ “average cost” to respond to a records request was
$2.08. Based on deposition testimony, the circuit court found that the results of the time
study were “transferrable” to the Class Period. The circuit court also found, based on
3
deposition testimony, that WVU Hospitals could not replicate the process of producing
individual class members’ medical records during the Class Period, and the court
concluded that any uncertainty regarding damages should be resolved against WVU
Hospitals.
I believe that the circuit court’s use of an “average cost” to find commonality
was improper. “A representative or statistical sample, like all evidence, is a means to
establish or defend against liability.” Tyson Foods, Inc. v. Bouaphakeo, 577 U.S. 442,
454–55 (2016). Sometimes, “a representative sample is the only practicable means to
collect and present relevant data establishing a defendant’s liability.” Id. at 455 (internal
quotation marks omitted). However, such evidence must be reliable. “Its permissibility
turns not on the form a proceeding takes—be it a class or individual action—but on the
degree to which the evidence is reliable in proving or disproving the elements of the
relevant cause of action.” Ibid.
In this case, the “average cost” relied upon by the circuit court was unreliable
because it rested on a flawed assumption that West Virginia Code § 16-29-2 did not
meaningfully change on June 6, 2014. During the Class Period, the relevant portion of
West Virginia Code § 16-29-2 provided as follows:
The provider shall be reimbursed by the person
requesting in writing a copy of the records at the time of
delivery for all reasonable expenses incurred in complying
with this article: Provided, That the cost may not exceed
seventy-five cents per page for the copying of any record or
records which have already been reduced to written form and
a search fee may not exceed ten dollars.
4
W. Va. Code § 16-29-2(a) (eff. 1999) (emphasis added). By contrast, during the time study,
the relevant portion of the statute contained these provisions:
A person requesting records from a provider shall . . .
pay a reasonable, cost-based fee, at the time of delivery.
Notwithstanding any other section of the code or rule, the fee
shall be based on the provider’s cost of: (1) Labor for copying
the requested records if in paper, or for placing the records in
electronic media; (2) supplies for creating the paper copy or
electronic media; and (3) postage if the person requested that
the records be mailed.
....
The labor for copying under this section shall not
exceed twenty-five dollars per hour . . . .
W. Va. Code § 16-29-2(a) & (b) (eff. 2014) (emphasis added).
These versions of the statute are substantially different. The 2014 statute
authorized WVU Hospitals to collect a “reasonable, cost-based fee” based on three things
only—labor, supplies, and postage. Ibid. Even the category of labor was restricted to
“[l]abor for copying” and “[l]abor . . . for placing the records in electronic media[.]” Under
the 1999 statute, however, WVU Hospitals had the right to seek reimbursement for each of
these costs plus every other expense that was both “reasonable” and “incurred in complying
with this article[.]” W. Va. Code § 16-29-2(a) (eff. 1999). Such reasonable expenses,
unlike those authorized by the 2014 statute, were not limited to the costs involved in
“copying” records or “placing the records in electronic media” but instead included costs
such as those incurred in searching for relevant records. W. Va. Code § 16-29-2(a) (eff.
2014) and W. Va. Code § 16-29-2(a) (eff. 1999). Indeed, the 1999 statute expressly
5
included reference to a search fee. Such costs of searching for and locating records were
not included under the 2014 version of the statute.
Because of these differences, no relevant inference can be drawn from the
amounts that WVU Hospitals charged for records between June 6, 2014, and July 31, 2014.
It may be that these invoices, which were based on the time study, provide some insight
regarding WVU Hospitals’ actual time investment in responding to records requests during
the Class Period. However, even this insight is of doubtful relevance because the labor
costs that WVU Hospitals could lawfully recoup were narrower under the 2014 statute than
they were under the 1999 statute. 2
Furthermore, I am unsure that the “average cost” relied on by the circuit court
is useful in any manner with regard to certification of a class action in this matter. The
record shows, as WVU Hospitals has maintained all along, that the time it takes to respond
to a given records request can vary widely. According to the spreadsheet prepared by
Kathryn Crous, which served as the basis for the circuit court’s “average cost,” two minutes
of effort sometimes led to four or five images and other times led to as many as 1,240 or
5,253 images. Similarly, producing 73 images might take four minutes, while producing
2
I also disagree with the circuit court’s assumption that the results of the
time study are properly “transferrable” to the Class Period. Certainly, Christine Metheny
testified that the results of the time study were “transferrable to the same process . . . a year
earlier” and even “[a]ll the way back.” I assume “[a]ll the way back” means to the
beginning of the Class Period, but we know that is not the case because Ms. Metheny later
clarified that generalization from the time study is only possible as far back as July 1, 2008,
when the “EPIC” system became operational. The majority opinion duly applies this time
limit to the “average cost” determined by the circuit court, but the majority opinion fails to
appreciate the significance of this limitation. Being able to generalize after July 1, 2008,
is not enough to show commonality when the Class Period began on January 18, 2008.
6
75 images might take 14 minutes. The number of images-per-minute ranged from 0.6 (i.e.,
six images in ten minutes) to 11,991 (i.e., 11,991 images in one minute), 3 even as the
sample of charges ranged from $.36 to $46.25. If the underlying data is so widely
divergent, what reliable conclusion, if any, can be drawn from an average fee of $2.08?
Nevertheless, this “average cost” served as the basis for the circuit court’s finding of
commonality.
If we set aside the circuit court’s “average cost” analysis, the circuit court
cites no other realistic basis for concluding that liability or damages are capable of
resolution on a class-wide basis. The record establishes that WVU Hospitals’ expense of
producing records varied substantially from request to request. Because the statute
provided for “reimburse[ment] . . . for all reasonable expenses incurred[,]” W. Va. Code §
16-29-2(a) (eff. 1999) (emphasis added), we cannot know whether the fee that WVU
Hospitals charged to any given class member actually exceeded WVU Hospitals’ expenses
without individualized proof of what was necessary to respond to that class member’s
records request. Knowing what WVU Hospitals charged for the records does not establish
liability because WVU Hospitals’ fee is only half of the liability (and damages) equation.
Like the circuit court, the majority opinion attempts to avoid this difficulty
by saying that WVU Hospitals must be subject to “probable and inferential data” because
it failed to calculate its actual reasonable expenses during the Class Period. However, as
According to Melissa Martin, WVU Hospitals had the capacity to save
3
some matters from a records request. This would obviously reduce the time to respond to
a subsequent request for the same records.
7
noted above, such data is only acceptable proof if it is reliable, and the circuit court’s
“average cost” is not reliable. Furthermore, there is a substantial difference between
allocating the risk of uncertainty to WVU Hospitals regarding damages and allocating the
risk of uncertainty to them regarding liability. Gould v. Am.-Haw. S.S. Co., 535 F.2d 761,
782 (3d Cir. 1976) (stating that “there is a clear distinction between the measure of proof
necessary to establish the fact a plaintiff has sustained an injury, and the measure of proof
necessary to enable the jury to fix the amount of damages resulting from that injury”).
Class Representatives should not get the benefit of a less rigorous standard of proof until
they have shown that WVU Hospitals is, in fact, liable to them. J. Truett Payne Co. v.
Chrysler Motors Corp., 451 U.S. 557, 568 (1981) (“The cases relied upon by petitioner all
depend in greater or lesser part on the inequity of a wrongdoer defeating the recovery of
damages against him by insisting upon a rigorous standard of proof. In this case, however,
we cannot say with assurance that respondent is a ‘wrongdoer.’ . . . Absent such a finding,
we decline to apply to this case the lenient damages rules of our previous cases.”). This
distinction is particularly important to observe in a case where liability and damages are so
closely related. As we observed in Gaujot II, “[t]he statute is framed such that liability and
damages are two sides of the same coin[.]” Id. at 63, 829 S.E.2d at 63. Proof that WVU
Hospitals charged a class member more than its reasonable expenses is a factor of
establishing both WVU Hospitals’ liability and the extent of the class member’s damages.
Ibid.
Accordingly, because commonality is an essential prerequisite for class
certification and because I believe that, despite our clear guidance in Gaujot II, the circuit
8
court has yet to truly resolve the fundamental legal and factual issues necessary to decide
commonality in this case, I dissent on this issue and would grant a writ of prohibition on
this basis.
Ascertainability.
I further dissent from the majority opinion’s finding that we have no basis to
grant a writ of prohibition based on an absence of ascertainability. We have held that
“[b]efore certifying a class pursuant to Rule 23 of the West Virginia Rules of Civil
Procedure, it is imperative that the class be identified with sufficient specificity so that it
is administratively feasible for the court to ascertain whether a particular individual is a
member.” Syl. Pt. 3, State ex rel. Metro. Life Ins. Co. v. Starcher, 196 W. Va. 519, 474
S.E.2d 186 (1996) (emphasis added). The majority opinion quotes at length the circuit
court’s purported findings on this issue. However, much of the circuit court’s rationale for
its findings in relation to ascertainability is based upon its related findings regarding
commonality. As discussed above, I believe such findings regarding commonality and the
reliance of the circuit court on the “time study” to establish commonality are in error. The
majority, in reaching its determination that ascertainability has been established, points to
finding 8 of the circuit court’s order. However, finding 8 states, in pertinent part, that “the
Defendants’ assertion that an individual assessment of each class member’s damages will
be necessary has not been demonstrated under the facts presented to this Court” and further
finds that “the Defendants’ expert analysis set forth in the ‘time study’ concluded that
such individual charges would be the same as those charged under the subsequent version
9
of the statute, thus providing a means of proving such individual cost . . . .” (Emphasis
added.) Accordingly, the circuit court’s finding as to ascertainability is inextricably
intertwined with its flawed analysis of commonality, which in turn is based upon the
erroneous assumption that the “time study” is a valid means of determining commonality.
Beyond this flawed analysis, the circuit court’s ascertainability findings amount merely to
conclusory statements about the ability of the claims process, patient spreadsheets and the
proposed class notice to render the class ascertainable. Such findings are inadequate to
support certification.
Furthermore, I disagree with the majority opinion’s conclusion that EQT
Prod. Co. v. Adair, 764 F.3d 347 (4th Cir. 2014), is distinguishable from the facts in this
case. The court in EQT remanded the underlying case for reconsideration of
ascertainability issues because “numerous heirship, intestacy, and title-defect issues
plague[d] many of the potential class members’ claims to the gas estate . . . [and] pose[d]
a significant administrative barrier to ascertaining the ownership classes.” Id. at 359.
According to EQT,
[t]he fact that verifying ownership will be necessary for the
class members to receive royalties does not mean it is not also
a prerequisite to identifying the class. Without even a rough
estimate of the number of potential successors-in-interest, we
have little conception of the nature of the proposed classes or
who may be bound by a potential merits ruling. Lacking even
a rough outline of the classes’ size and composition, we cannot
conclude that they are sufficiently ascertainable.
Id. at 359–60. As EQT explains, “plaintiffs need not be able to identify every class member
at the time of certification. But “[i]f class members are impossible to identify without
10
extensive and individualized fact-finding or ‘mini-trials,’ then a class action is
inappropriate.” Id. at 358 (quoting Marcus v. BMW of N. Am., LLC, 687 F.3d 583, 593 (3d
Cir. 2012)). I believe that similar difficulties are present in this case because each plaintiff
would be required to produce individualized proof that he or she “reimbursed” WVU
Hospitals for more than their “reasonable expenses incurred[.]” W. Va. Code § 16-29-2(a)
(eff. 1999). Accordingly, I believe that it would be difficult to establish ascertainability
without addressing the specific facts in each case, and I respectfully dissent from the
majority opinion’s conclusion that no relief is warranted on this issue.
Predominance.
We have held that class certification requires a circuit court to “determine
that the party seeking class certification has satisfied all four prerequisites contained in
Rule 23(a) . . . and has satisfied one of the three subdivisions of Rule 23(b).” Syl. Pt. 8, in
part, In re W. Va. Rezulin Litig., 214 W. Va. 52, 585 S.E.2d 52 (2003), holding modified
by Surnaik, 244 W. Va. 248, 852 S.E.2d 748. Subdivision (3) of Rule 23(b) states, in
relevant part, “that the questions of law or fact common to the members of the class [must]
predominate over any questions affecting only individual members . . . .” W. Va. R. Civ.
P. 23(b)(3) (emphasis added).
Whether common questions “predominate” over individual questions is an
issue that requires “thorough analysis[.]” Surniak at ___, 852 S.E.2d at 750, syl. pt. 7, in
part. According to Surnaik,
thorough analysis of the predominance requirement of West
Virginia Rule of Civil Procedure 23(b)(3) includes (1)
11
identifying the parties’ claims and defenses and their
respective elements; (2) determining whether these issues are
common questions or individual questions by analyzing how
each party will prove them at trial; and (3) determining whether
the common questions predominate. In addition, circuit courts
should assess predominance with its overarching purpose in
mind—namely, ensuring that a class action would achieve
economies of time, effort, and expense, and promote
uniformity of decision as to persons similarly situated, without
sacrificing procedural fairness or bringing about other
undesirable results. This analysis must be placed in the written
record of the case by including it in the circuit court’s order
regarding class certification.
Ibid. WVU Hospitals object that the circuit court, on remand, failed to conduct a
preponderance analysis with the level of thoroughness that Surnaik requires, despite their
request for the court to do so. Indeed, the majority recognizes that the circuit court has not
addressed preponderance in an order since the class certification order that was entered on
April 16, 2014.
The circuit court’s April 16, 2014 certification order does not approach the
level of analysis that Surnaik requires. The certification order merely announces that Class
Representatives “have satisfied the predominance requirement because the issue of the [sic]
whether the amount charged to patients for copies of their medical records is ‘reasonable’
under the applicable statute predominates all issues affecting individual members.” I do
not believe that this conclusory language even satisfies the preponderance standard that
applied when the certification order was entered. See Rezulin at 72, 585 S.E.2d at 72 (“A
conclusion on the issue of predominance requires an evaluation of the legal issues and the
proof needed to establish them. . . . The predominance requirement is not a rigid test, but
12
rather contemplates a review of many factors . . . .” (emphasis added) (internal quotation
marks and citation omitted)).
The majority opinion excuses this lack of analysis for two reasons, first,
because our opinion in Gaujot II allegedly did not require “the circuit court to address any
of the Rule 23 factors other than commonality” and, second, because “the circuit court
conducted its predominance evaluation eight years ago under the standard . . . in effect” at
that time. I do not believe either of the explanations offered by the majority excuse the
lack of findings by the circuit court to establish preponderance.
I agree that, in Gaujot II, our express basis for granting the writ of
prohibition, as moulded, was the circuit court’s failure to conduct a “sufficiently thorough
analysis” of commonality. Gaujot II at 64, 829 S.E.2d at 64. That is not to say, however,
that other sufficient grounds for relief were not present; we did not decide that question
either way. On the contrary, we limited our analysis to commonality because the circuit
court’s lack of analysis, on that issue alone, was an entirely sufficient basis for granting
relief. Ibid. Nevertheless, we expressly “urge[d] the circuit court to determine whether the
requirements of Rule 23, particularly as they relate to commonality, have been met and, if
so, to craft a class definition consistent with such findings.” Ibid. (emphasis added). While
we admittedly emphasized commonality, we clearly urged the circuit court to determine
whether the “requirements” of Rule 23 had been met, not whether a single “requirement,”
namely commonality, had been met.
Moreover, the fact that the circuit court had addressed the preponderance
criterion several years before did not prevent the circuit court from addressing it again after
13
Surniak or cure the fact that the circuit court’s previous predominance analysis was
inadequate when it was issued. As the majority acknowledges in another context, class
“certification is conditional and may be altered, expanded, subdivided, or vacated as the
case progresses toward resolution on the merits.” Syl. Pt. 2, in part, State ex rel. Metro.
Life Ins. Co. v. Starcher, 196 W. Va. 519, 474 S.E.2d 186 (1996). Nothing in Surnaik
suggests that we intended its holdings to operate only prospectively. Moreover, because
we urged the circuit court in Gaujot II to make an analysis of the Rule 23 factors, the circuit
court had a duty to review the preponderance issue under the prism of Surnaik, which was
decided in 2020, after Gaujot II but before the circuit court entered the order under appeal
in this case in July 2021. Because compliance with Rule 23(b) is an essential prerequisite
for class certification and because I believe that the circuit court’s preponderance analysis
was inadequate under either a pre-Surnaik or post-Surnaik standard, I dissent as to this
issue and would grant a writ of prohibition on this basis as well.
For these reasons, I concur in Part III.D of the majority opinion, and I
respectfully dissent as to Parts III.A., III.B., and III.C.
14