Case: 19-50818 Document: 00515431386 Page: 1 Date Filed: 05/28/2020
IN THE UNITED STATES COURT OF APPEALS
FOR THE FIFTH CIRCUIT United States Court of Appeals
Fifth Circuit
FILED
May 28, 2020
No. 19-50818
Lyle W. Cayce
Clerk
UNITED STATES OF AMERICA, ex rel., INTEGRA MED ANALYTICS,
L.L.C.,
Plaintiff–Appellant,
v.
BAYLOR SCOTT & WHITE HEALTH; BAYLOR UNIVERSITY MEDICAL
CENTER–DALLAS; HILLCREST BAPTIST MEDICAL CENTER; SCOTT &
WHITE HOSPITAL–ROUND ROCK; SCOTT & WHITE MEMORIAL
HOSPITAL TEMPLE,
Defendants–Appellees.
Appeal from the United States District Court
for the Western District of Texas
USDC No. 5:17-CV-886
Before OWEN, Chief Judge, and HIGGINBOTHAM and WILLETT, Circuit
Judges.
PER CURIAM:*
Integra Med Analytics, L.L.C., filed a qui tam suit 1 on behalf of the
* Pursuant to 5TH CIR. R. 47.5, the court has determined that this opinion should not
be published and is not precedent except under the limited circumstances set forth in 5TH
CIR. R. 47.5.4.
At the federal level, qui tam suits are those that are filed “for the person and for the
1
United States Government” and “brought in the name of the Government.” 31 U.S.C.
Case: 19-50818 Document: 00515431386 Page: 2 Date Filed: 05/28/2020
No. 19-50818
United States against Baylor Scott & White Health system and its affiliates
under the False Claims Act for allegedly using inflated codes to bill Medicare.
The district court dismissed Integra Med’s claims. We affirm.
I
The Baylor Scott & White Health system and its affiliates (Baylor)
operate a network consisting of around twenty inpatient short-term acute care
hospitals in Texas. A significant number of patients served by Baylor are
covered by Medicare. Thus, Baylor regularly submits reimbursement claims
to Medicare. In this case, Integra Med Analytics, L.L.C. (Integra Med) alleges
that Baylor submitted $61.8 million in fraudulent claims to Medicare, in
violation of the False Claims Act (FCA). 2
Medicare reimburses hospitals like Baylor on a per-discharge basis,
which means Baylor gets paid each time a patient stays at the hospital. The
exact amount that Medicare reimburses primarily depends on a hospital’s
diagnoses of Medicare-covered patients. Medicare classifies similar diagnoses
by putting them into a diagnosis related group (DRG). Each DRG is
determined by several kinds of codes, including the principal diagnosis code
and secondary diagnosis codes. The principal diagnosis code is for the
“condition established after study to be chiefly responsible for occasioning the
admission of the patient to the hospital for care.” 3 Secondary diagnosis codes
are for “all conditions that coexist at the time of admission, that develop
subsequently, or that affect the treatment received and/or length of stay.” 4
§ 3730(b)(1). Thus, in qui tam suits, the government is the real party in interest. United
States v. Tex. Tech Univ., 171 F.3d 279, 289 (5th Cir. 1999).
2 31 U.S.C. § 3729.
3 See Centers for Disease Control, ICD-9-CM Official Guidelines for Coding and
Reporting, Oct. 1, 2011 at 88, available at https://goo.gl/DC55Wx.
4 See Centers for Disease Control, ICD-9-CM Official Guidelines for Coding and
Reporting, Oct. 1, 2011 at 91, available at https://goo.gl/DC55Wx.
2
Case: 19-50818 Document: 00515431386 Page: 3 Date Filed: 05/28/2020
No. 19-50818
Reimbursement can also be affected, to a lesser extent, by other hospital-
specific factors, such as market conditions in the hospital’s city.
Integra Med’s allegations specifically concern Baylor’s use of secondary
diagnosis codes. The Centers for Medicare and Medicaid Services (CMS)
publishes a list of secondary codes each year that can modify a claim to include
a complication or comorbidity (CC) or a major complication or comorbidity
(MCC). The inclusion of CCs and MCCs can add thousands of dollars to a
Medicare reimbursement claim. Integra Med alleges that Baylor, led by its
clinical documentation improvement (CDI) program, fraudulently used higher-
value CCs and MCCs than were justified by actual medical diagnoses to
increase its revenues. Integra Med contends that Baylor’s scheme had three
main components.
First, Integra Med contends that Baylor trained its physicians and CDI
employees to “upcode” MCCs. According to Integra Med, Baylor trained its
physicians to focus on key words, provided lists of high-value MCCs to
physicians to reinforce that training, and emphasized that using certain terms
would increase their performance pay. Integra Med also contends that Baylor
had its CDI employees seek opportunities to use higher-value secondary codes.
Second, Integra Med alleges that Baylor pressured physicians to alter
their original diagnoses by providing documents and asking them to “specify”
or change their diagnosis if the diagnosis did not include CCs or MCCs.
According to Integra Med, these clarification documents that requested
physicians to “specify” their diagnoses would often “suggest either specific
revenue-increasing CCs or MCCs or provide options listing several possible
CCs and MCCs.” Integra Med contends these clarification documents “reveal
a clear intent towards influencing doctors to code higher-paying CCs and
MCCs.”
Third, Integra Med alleges that Baylor provided unnecessary treatment
3
Case: 19-50818 Document: 00515431386 Page: 4 Date Filed: 05/28/2020
No. 19-50818
in order to code high-value MCCs. Specifically, Integra Med contends that
“Baylor purposefully placed and kept post-operative patients on ventilator
support” when it was medically unnecessary. Integra Med bases this
allegation on the fact “that Baylor patients undergoing major heart surgery
were placed on mechanical ventilation [at rates] over twice the national
average.”
Integra Med analyzed inpatient claims data for the 2011-2017 period
from CMS to discover that Baylor had been claiming certain MCCs
significantly above the national average for other hospitals. Specifically,
Integra Med found that Baylor coded for the MCCs of encephalopathy,
respiratory failure, and severe malnutrition at much higher rates than other
hospitals. Integra Med contends that its statistical analyses show that
Baylor’s higher rate of coding cannot be explained by patient characteristics,
county demographic data, the patient’s attending physician, or regional
differences. According to Integra Med, its “analyses prove that the excessive
rates of [certain] MCCs can be directly attributed to [Baylor’s] fraudulent
activity as opposed to external factors, indicating that the fraud was known by
the system and was intentional.”
Besides statistical data, Integra Med also relied on several statements
from a former Baylor medical coder in concluding that Baylor had defrauded
Medicare. According to Integra Med, this medical coder recalled a then-Baylor
executive “telling CDIs things that were totally not true” as a part of a
“deliberate effort to promote the coding of MCCs.” This medical coder also
allegedly received specific instructions on how to code. Integra Med claims
that this medical coder quit her job with Baylor because she was unable to
work where she “was continually getting directives to compromise her
integrity.” Integra Med also relied on certain statements about increasing
hospital revenues from a former Baylor executive’s social media.
4
Case: 19-50818 Document: 00515431386 Page: 5 Date Filed: 05/28/2020
No. 19-50818
Based on these statistics and statements, Integra Med sued Baylor
under the FCA in federal district court in April 2018. After Integra Med
amended its complaint twice, Baylor moved under Federal Rule of Civil
Procedure Rule 12(b)(6) to dismiss Integra Med’s complaint. The district court
granted Baylor’s motion to dismiss, holding that Integra Med’s complaint
failed to state a particularized claim for which relief could be granted as
required by Federal Rules of Civil Procedure 8(a) and 9(b). This appeal
followed.
II
To survive a motion to dismiss an FCA claim, Integra Med must plead
the following four elements: (1) “a false statement or fraudulent course of
conduct;” (2) that was “made or carried out with the requisite scienter;” (3)
“that was material;” and (4) “that caused the government to pay out money or
to forfeit moneys due (i.e., that involved a claim).” 5 Integra Med’s case on
appeal hinges on whether Integra Med sufficiently pleaded facts showing that
Baylor’s claims were fraudulent. Thus, we will examine each of Integra Med’s
bases for its claims, including its statistical data generally, the documents it
has gathered from Baylor, statements by a former Baylor medical coder, and
the claim that Baylor provided unnecessary medical care to boost its Medicare
reimbursements.
A
We first examine the statistical data presented by Integra Med,
reviewing whether it sufficiently shows that Baylor’s Medicare reimbursement
claims were fraudulent. “[A] complaint filed under the False Claims Act must
5United States ex rel. King v. Solvay Pharm., Inc., 871 F.3d 318, 324 (5th Cir. 2017)
(quoting United States ex rel. Longhi v. United States, 575 F.3d 458, 467 (5th Cir. 2009)).
5
Case: 19-50818 Document: 00515431386 Page: 6 Date Filed: 05/28/2020
No. 19-50818
meet the heightened pleading standard of Rule of 9(b).” 6 Federal Rule of Civil
Procedure 9(b) provides, “[i]n alleging fraud or mistake, a party must state
with particularity the circumstances constituting fraud or mistake.” 7
Although the particularity Rule 9(b) demands “differs with the facts of each
case,” 8 it does generally require that a complaint detail “the who, what, when,
and where . . . before access to the discovery process is granted.” 9 Rule 9(b)’s
particularity requirement supplements Rule 8(a)’s demand that “a complaint
must contain sufficient factual matter, accepted as true, to ‘state a claim to
relief that is plausible on its face.’” 10 Rule 8(a) prohibits any claims that are
merely conceivable rather than plausible. 11 A claim is merely conceivable and
not plausible if the facts pleaded are consistent with both the claimed
misconduct and a legal and “obvious alternative explanation.” 12
Here, Integra Med’s statistical analysis is consistent with both Baylor
having submitted fraudulent Medicare reimbursement claims to the
government and with Baylor being ahead of most healthcare providers in
following new guidelines from CMS. In 2007, CMS reduced the standardized
amount paid out to hospitals for Medicare reimbursement claims but increased
the number of secondary diagnoses identified as CCs and MCCs, and coding
6 See, e.g., United States ex rel. Grubbs v. Kanneganti, 565 F.3d 180, 185 (5th Cir.
2009) (first citing United States ex rel. Russell v. Epic Healthcare Mgmt. Grp., 193 F.3d 304,
308-09 (5th Cir. 1999), abrogated on other grounds by United States ex rel. Eisenstein v. City
of New York, 556 U.S. 928 (2009); and then citing United States ex rel. Karvelas v. Melrose–
Wakefield Hosp., 360 F.3d 220, 228 (1st Cir. 2004), abrogated on other grounds by Allison
Engine Co. v. United States ex rel. Sanders, 553 U.S. 662 (2008)).
7 FED. R. CIV. P. 9(b); see also Kanneganti, 565 F.3d at 185-86.
8 Hart v. Bayer Corp., 199 F.3d 239, 247 n.6 (5th Cir. 2000) (citing Guidry v. Bank of
LaPlace, 954 F.2d 278, 288 (5th Cir. 1992)).
9 Id. (alteration in original) (quoting Williams v. WMX Techs., Inc., 112 F.3d 175, 178
(5th Cir. 1997)).
10 Ashcroft v. Iqbal, 556 U.S. 662, 678 (2009) (quoting Bell Atl. Corp. v. Twombly, 550
U.S. 544, 570 (2007)); see also Kanneganti, 565 F.3d at 185.
11 Iqbal, 556 U.S. at 680 (quoting Twombly, 550 U.S. at 570).
12 Id. at 682 (quoting Twombly, 550 U.S. at 567).
6
Case: 19-50818 Document: 00515431386 Page: 7 Date Filed: 05/28/2020
No. 19-50818
more CCs and MCCs can increase hospital reimbursements. 13 In response to
public comments expressing concern that the new rules would lead to lower
reimbursements, CMS stated that it expected reimbursements to increase
under the system. 14 CMS believed it was “clear” that hospitals would “change
their documentation and coding practices and increase case mix consistent
with the payment incentives that are provided by the” then new coding
system. 15 In fact, CMS encouraged hospitals to adopt CDI programs “in order
to increase reimbursement” and highlighted an article touting the
effectiveness of CDI programs at increasing Medicare reimbursement rates.16
CMS unequivocally stated in its guidelines that, “[w]e do not believe there is
anything inappropriate, unethical or otherwise wrong with hospitals taking
full advantage of coding opportunities to maximize Medicare payment that is
supported by documentation in the medical record.” 17
The conclusion that Baylor was simply ahead of the healthcare industry
in following CMS guidelines is supported by the data in Integra Med’s own
complaint. Integra Med’s complaint shows that the rate at which non-Baylor
hospitals were using the MCCs for encephalopathy, respiratory failure, and
severe malnutrition was increasing every year. These increases were causing
the MCC usage rates of both Baylor and non-Baylor hospitals to converge.
Moreover, for severe malnutrition, non-Baylor hospitals were coding it at a
higher rate in 2017 than Baylor was in 2015. Similarly, for respiratory failure,
non-Baylor hospitals were coding it at a higher rate in 2017 than Baylor was
13 See Medicare Program; Changes to the Hospital Inpatient Prospective Payment
Systems and Fiscal Year 2008 Rates, 72 Fed. Reg. 47,130, 47,135-39 (Aug. 22, 2007) (final
rule).
14 See Medicare Program; Changes to the Hospital Inpatient Prospective Payment
Systems and Fiscal Year 2008 Rates, 72 Fed. Reg. at 47,180-82.
15 Id. at 47,182.
16 Id.
17 Id. at 47,180.
7
Case: 19-50818 Document: 00515431386 Page: 8 Date Filed: 05/28/2020
No. 19-50818
in 2011. These show that the healthcare industry as a whole was following
Baylor in its trajectory and by 2017, other hospitals’ coding was within a few
percentage points of Baylor’s.
These facts strongly indicate that a legal and “obvious alternative
explanation” for the statistical data presented by Integra Med is that Baylor
was simply ahead of the healthcare industry at implementing the Medicare
reimbursement guidelines supplied by CMS. 18 We note that this conclusion
does not exclude statistical data from being used to meet the pleading
requirements of Federal Rule of Civil Procedure 8(a) and, when paired with
particular details, Rule 9(b). 19 Our conclusion merely means that statistical
data cannot meet those pleading requirements if, among other possible issues,
it is also consistent with a legal and obvious alternative explanation. 20
Insofar as Integra Med purports to give specific examples of fraudulent
claims, it also fails to meet the pleading requirements of Rules 8(a) and 9(b).
Integra Med’s examples simply give some identifying patient information and
pair it with a diagnosis. No example gives any indication about what makes it
a false claim. The claims of falsity are simply conclusory. 21
18 See Ashcroft v. Iqbal, 556 U.S. 662, 682 (2009) (quoting Bell Atl. Corp. v. Twombly,
550 U.S. 544, 567 (2007)).
19 See, e.g., United States ex rel. Customs Fraud Investigations, LLC v. Victaulic Co.,
839 F.3d 242, 247-48, 258 (3d Cir. 2016) (concluding, in the Rule 8(a) and 9(b) context, that
statistical data about the lack of markings on a company’s pipe fittings was sufficient to state
an FCA claim for avoiding import duties when paired with an expert’s declaration analyzing
the facts of that case, specific examples of unmarked pipes with photographs, a witness
statement about receiving improperly marked pipes, and detailed records about the
shipments at issue); Boykin v. Georgia-Pac. Corp., 706 F.2d 1384, 1390-94 (5th Cir. 1983)
(concluding, in the Rule 8(a) context, that plaintiff’s presentation of statistical data
successfully stated a prima facie case of racial discrimination).
20 See Iqbal, 556 U.S. at 678.
21 See Taylor v. Books A Million, Inc., 296 F.3d 376, 378 (5th Cir. 2002) (“[C]onclusory
allegations or legal conclusions masquerading as factual conclusions will not suffice to
prevent a motion to dismiss.” (quoting S. Christian Leadership Conference v. Supreme Court
of the State of La., 252 F.3d 781, 786 (5th Cir. 2001))).
8
Case: 19-50818 Document: 00515431386 Page: 9 Date Filed: 05/28/2020
No. 19-50818
B
1
We next examine whether Integra Med’s allegations that Baylor trained
and pressured its physicians and CDI employees to “upcode” MCCs are
sufficient to establish that Baylor was engaging in a scheme to submit
fraudulent claims to Medicare. We conclude that they are not. In publishing
the new DRG coding rules, CMS explicitly expected hospitals to work with
their physicians and medical coders, including through training, to “focus on
understanding the impact of the revised CC list.” 22 According to Integra Med,
Baylor trained physicians to focus on keywords, provided tip sheets reminding
physicians of how to report high-value MCCs, had CDI employees look for
opportunities where high-value MCCs might be present, and would sometimes
send physicians documents asking them to clarify their diagnoses. Integra
Med argues that these practices show Baylor was involved in a scheme to
defraud Medicare. But CMS encouraged hospitals to employ practices like
these after it implemented the new DRG rules. 23 Far from a fraudulent
scheme, Baylor’s implementation of such practices is entirely consistent with
the new DRG rules. 24
For example, Baylor’s use of tip sheets is consistent with the fact that
coding and clinic terminology are often different. Tip sheets help hospitals
align the two. Likewise, non-leading documents asking physicians to clarify
their diagnoses are also consistent with implementing the new DRG rules since
22 See Medicare Program; Changes to the Hospital Inpatient Prospective Payment
Systems and Fiscal Year 2008 Rates, 72 Fed. Reg. at 47,182 (“[H]ospitals may focus on
understanding the impact of the revised CC list, training and educating their coders, and
working with their physicians for any documentation improvements required to allow the
reporting of more specific codes where applicable.”).
23 See id.
24 Id.
9
Case: 19-50818 Document: 00515431386 Page: 10 Date Filed: 05/28/2020
No. 19-50818
the new DRG rules moved hospitals away from focusing on general diagnoses
and codes to frequently using more specific diagnoses and codes. 25 Physicians
were likely still accustomed to the old, more general system. These
clarification documents had numerous suggestions, a simple box to check to
decline clarification, and a disclaimer not to take implications from the fact
clarification was asked for. Additionally, some of the clarification documents
provided by Integra Med in its complaint show that clarification was requested
in instances in which physicians wrote down symptoms but failed to provide a
diagnosis for the cause of those symptoms. These clarification documents also
did not ask leading questions. Considering diagnoses are critical for Medicare
reimbursements and these specific clarification documents were not leading,
they are consistent with Baylor engaging in legal activity.
Therefore, we conclude that these allegations are also consistent with a
legal and “obvious alternative explanation.” 26
2
In its complaint, Integra Med also cites the statements of a medical coder
who said that a then-Baylor executive told “CDIs things that were totally not
true” as a part of a “deliberate effort to promote the coding of MCCs.”
According to Integra Med, this medical coder said she was given specific
instructions on how to code, and that medical coders “receive[d] pressure
directly from . . . leadership to code unethically.” This medical coder also
allegedly quit her job because she “was continually getting directives to
compromise her integrity.” But these allegations fail to satisfy the heightened
pleading standards required by Federal Rule of Civil Procedure 9(b) because
they fail to state the content of these allegedly unethical and fraudulent
25See id. at 47,130-82 (Aug. 22, 2007) (final rule).
26See Ashcroft v. Iqbal, 556 U.S. 662, 682 (2009) (quoting Bell Atl. Corp. v. Twombly,
550 U.S. 544, 567 (2007)).
10
Case: 19-50818 Document: 00515431386 Page: 11 Date Filed: 05/28/2020
No. 19-50818
directives, trainings, and guidance. 27 Thus, the district court correctly
dismissed the claim based on these conclusory allegations.
C
We next look at Integra Med’s allegations that Baylor provided
unnecessary treatment to patients in order to use higher-value MCCs.
Specifically, Integra Med contends that “Baylor purposefully placed and kept
post-operative patients on ventilator support” when it was medically
unnecessary. The allegations here are based solely on the fact “that Baylor
patients undergoing major heart surgery were placed on mechanical
ventilation over twice the national average.” These allegations do not
withstand the heightened pleading requirements for fraud under Rule 9(b).
Integra Med fails to plead particular details of a scheme to defraud
Medicare. Even when plaintiffs in an FCA case use statistics, which can be
reliable indicia of fraud, they must still plead particular details of a fraudulent
scheme for each claim. 28 Here, Integra Med’s complaint contains a conclusory
27 See Hart v. Bayer Corp., 199 F.3d 239, 247 n.6 (5th Cir. 2000) (concluding that to
meet the pleading requirements of Rule 9(b) a complaint must state “the who, what, when,
and where” of a claim. (quoting Williams v. WMX Techs., Inc., 112 F.3d 175, 178 (5th Cir.
1997))). Integra Med claims that the situation here is “strikingly similar” to the situation in
United States ex rel. Integra Med Analytics, LLC v. Creative Solutions in Healthcare, Inc., No.
SA-17-CV-1249-XR, 2019 WL 5970283 (W.D. Tex. Nov. 13, 2019). We disagree. In Creative
Solutions, the employee witness interviews actually revealed the contents of a specific
fraudulent scheme. Id. at *4. That opinion notes, “a physical therapist at Fairfield recalled
being instructed to allot 15 minutes for evaluation, even though it required 45 minutes, with
the rest of the evaluation session charged at therapy rates.” Id. (internal quotation omitted).
The interview responses given by Integra Med here, while alleging a vague scheme to
“promote the coding of MCCs,” do not provide the who, what, when, and where of such scheme
as required by Rule 9(b). The vague allegation here contrasts with the Creative Solutions
interview responses, which included the requisite particularity and specificity.
28 United States ex rel. Grubbs v. Kanneganti, 565 F.3d 180, 190 (5th Cir. 2009) (“We
hold that to plead with particularity the circumstances constituting fraud for a False Claims
Act § 3729(a)(1) claim, a relator’s complaint, if it cannot allege the details of an actually
submitted false claim, may nevertheless survive by alleging particular details of a scheme to
submit false claims paired with reliable indicia that lead to a strong inference that claims
were actually submitted.”); see also United States ex rel. Nunnally v. W. Calcasieu Cameron
Hosp., 519 F. App’x 890, 893 (5th Cir. 2013) (“We established that a relator could, in some
11
Case: 19-50818 Document: 00515431386 Page: 12 Date Filed: 05/28/2020
No. 19-50818
allegation that Baylor was providing unnecessary treatment to its patients and
supports it with a single statistic—that Baylor patients undergoing major
heart surgery were put on a mechanical ventilator at a rate over twice the
national average. Integra Med does not present sufficient particular details of
this alleged fraud claim. The district court correctly dismissed the FCA claim
based on Integra Med’s allegation that Baylor provided unnecessary treatment
to patients to increase its Medicare reimbursements.
In conclusion, Integra Med has failed to meet its pleading requirements
under Rules 8(a) and 9(b). The district court did not, as Integra Med contends,
view the complaint in the light most favorable to Baylor—it simply correctly
held Integra Med to the higher pleading standard required for an FCA claim.
III
Integra Med contends that the district court improperly held its
allegations to a more rigorous scienter requirement than was required by the
FCA. But we need not address scienter because the district court correctly
dismissed Integra Med’s claims for failing to meet the pleading requirements
required by Rules 8(a) and 9(b) for pleading the FCA’s element that there be
“a false statement or fraudulent course of conduct.” 29
Integra Med also contends that the district court improperly applied a
probability standard at the pleadings stage instead of a plausibility standard.
But regardless of whether the district court mistakenly applied a probability
circumstances, satisfy Rule 9(b) by providing factual or statistical evidence to strengthen the
inference of fraud beyond mere possibility, without necessarily providing details as to each
false claim. This standard nonetheless requires the relator to provide other reliable
indications of fraud and to plead a level of detail that demonstrates that an alleged scheme
likely resulted in bills submitted for government payment.” (emphasis and citations
omitted)).
29 United States ex rel King v. Solvay Pharm., Inc., 871 F.3d 318, 324 (5th Cir. 2017)
(quoting United States ex rel. Longhi v. United States, 575 F.3d 458, 467 (5th Cir. 2009)).
12
Case: 19-50818 Document: 00515431386 Page: 13 Date Filed: 05/28/2020
No. 19-50818
standard rather than a plausibility standard, our conclusion is the same. 30
Since “[we] may affirm the district court on any grounds supported by the
record and argued in the court below,” any misapplication that might have
occurred here would not require us to vacate or reverse the district court’s
judgment. 31
* * *
For these reasons, the district court’s judgment is AFFIRMED.
30 See Ashcroft v. Iqbal, 556 U.S. 662, 678 (2009) (“Where a complaint pleads facts that
are ‘merely consistent with’ a defendant’s liability, it ‘stops short of the line between
possibility and plausibility of “entitlement to relief.”’” (quoting Bell Atl. Corp. v. Twombly,
550 U.S. 544, 557 (2007))).
31 Maria S. ex rel. E.H.F. v. Garza, 912 F.3d 778, 783 (5th Cir. 2019) (citing Doctor’s
Hosp. of Jefferson, Inc. v. Se. Med. All., Inc., 123 F.3d 301, 307 (5th Cir. 1997)).
13