Integra Med Analytics LLC v. Providence Health & Services ( 2021 )

NOT FOR PUBLICATION FILED
UNITED STATES COURT OF APPEALS MAR 31 2021
MOLLY C. DWYER, CLERK
U.S. COURT OF APPEALS
FOR THE NINTH CIRCUIT
INTEGRA MED ANALYTICS LLC, No. 19-56367
Relator,
D.C. No.
Plaintiff-Appellee, 2:17-cv-01694-PSG-SS
and
MEMORANDUM*
UNITED STATES OF AMERICA,
Plaintiff,
v.
PROVIDENCE HEALTH & SERVICES; et
al.,
Defendants-Appellants.
Appeal from the United States District Court
for the Central District of California
Philip S. Gutierrez, Chief District Judge, Presiding
Argued and Submitted February 12, 2021
Pasadena, California
Before: BOGGS,** M. SMITH, and MURGUIA, Circuit Judges.
*
This disposition is not appropriate for publication and is not precedent
except as provided by Ninth Circuit Rule 36-3.
**
The Honorable Danny J. Boggs, United States Circuit Judge for the
U.S. Court of Appeals for the Sixth Circuit, sitting by designation.
Relator Integra Med Analytics, LLC, claims that Providence Health &
Services, its affiliated hospitals, and J.A. Thomas & Associates, Inc. (JATA),
violated the False Claims Act (FCA), 31 U.S.C. §§ 3729–3733. Integra alleges that
Providence submitted claims to Medicare that it coded for more lucrative secondary
diagnoses that were not supported by patients’ conditions. Integra based its
complaint primarily on a statistical analysis of Medicare-claims data that
demonstrated Providence submitted proportionally more claims with higher-paying
diagnosis codes than comparable institutions.
Providence and JATA filed motions to dismiss before the district court. The
court granted their motions in part and denied them in part, allowing Integra’s
primary FCA claim to proceed. At Defendants’ request, the district court certified
its order for interlocutory appeal under 28 U.S.C. § 1292(b). Its certification was
based on two controlling questions of law for which there was substantial ground for
difference of opinion: (1) Did Integra adequately plead the falsity of Providence’s
Medicare claims?; and (2) Is all online information material that is “from the news
media” for the purpose of the FCA’s public-disclosure bar?
We have jurisdiction under 28 U.S.C. § 1292(b). We hold that Integra failed
to state a plausible claim for relief because its allegations do not eliminate an obvious
alternative explanation—that Providence, with JATA’s assistance, was more
effective at properly coding for better Medicare reimbursement than others in the
2
healthcare industry. Accordingly, we reverse the district court’s order denying
Defendants’ motions to dismiss, and we remand.1
I. BACKGROUND
We take the following facts from the complaint as true for the purpose of
reviewing a motion to dismiss. Curtis v. Irwin Indus., Inc., 913 F.3d 1146, 1151
(9th Cir. 2019).
Providence is “one of the nation’s largest health systems, operating 34
hospitals and 600 clinics across five states.” A significant portion of Providence’s
revenue comes from Medicare reimbursements.
Medicare reimburses hospitals on a per-discharge basis, meaning a payment
for each time a patient stays at the hospital. The payment amount depends largely
on the patient’s “diagnosis related group” (DRG). Three types of codes contribute
to the DRG: (1) A principal-diagnosis code, (2) surgical-procedure codes, and (3)
secondary-diagnosis codes. There can be multiple secondary-diagnosis codes,
which together represent “all conditions that coexist at the time of admission, that
develop subsequently, or that affect the treatment received [or] the length of stay.”
Secondary-diagnosis codes can modify the base DRG’s severity level to one of three
levels: (1) Without complication or major complication, (2) complication or
1
Because Integra did not adequately allege that Providence submitted false claims,
we do not address whether the public-disclosure bar applies.
3
comorbidity (CC), and (3) major complication or comorbidity (MCC). The
inclusion of a CC or MCC code can significantly increase the amount of
reimbursement that the hospital receives from Medicare.
Providence, like many other hospitals, has a “clinical documentation
improvement” (CDI) program that translates clinical language used by medical-
treatment providers to Medicare codes. Providence retained JATA to assist its CDI
program. JATA offers consulting services to healthcare providers through products
and services intended to improve how they document patients’ treatments.
JATA worked with Providence to train doctors to describe medical conditions
that would support coding for higher-paying diagnoses. For example, JATA
provided doctors “Documentation Tips” suggesting specific language conducive to
coding CCs and MCCs; for example, specifying the type and degree of malnutrition.
It also trained Providence CDI specialists to send allegedly “leading queries” to
doctors that were designed to change their initial assessments in a way that would
justify coding a CC or MCC.2 Integra alleged that this pressure “would sometimes
result in the creation of contradictory medical records,” such as an initial
documentation of “delirium” with the later addition of “encephalopathy These
2
The complaint does not provide examples of leading queries used at Providence
but claims it is “consistent with JATA’s practice at other hospital systems.”
4
queries were designed to stop short of telling doctors exactly how to document their
care because that would constitute a “noncompliant” leading inquiry.
Integra, a data-analytics company, commenced this suit after its analysis of
data received from the Centers for Medicare and Medicaid Services (CMS)
demonstrated that Providence submitted claims coded with a higher rate of MCCs
than other comparable institutions. Integra is not affiliated with Providence or JATA
and did not rely on insider information such as confidential patient medical records.
Instead, Integra’s supported its complaint primarily through a “proprietary statistical
analysis” of the CMS data.
Integra’s methodology involved comparing the rate at which Providence
coded higher-paying secondary diagnoses in connection with particular principal
diagnoses with the rates coded by other institutions. It labeled claims as false if they
were coded with an MCC at “more than twice the national rate or were used at a rate
three percentage points higher than in the other hospitals.” Thus if the national rate
of an MCC code accompanying a specific principal diagnosis was .1% but
Providence coded it .2% of the time, or if the national rate was 55% and Providence’s
rate was 59%, Integra would label the claims as false.
Integra’s second amended complaint focuses on claims submitted between
2011 and 2017 involving three categories of secondary MCC codes:
Encephalopathy, respiratory failure, and severe malnutrition. Integra refers to these
5
three MCCs as “misstated MCCs.” Providence used at least one of these three
misstated MCCs on approximately 17 percent of its Medicare claims as opposed to
a 10 percent usage rate at non-Providence institutions—1.7 times as often. Integra
asserts that its analysis controlled for inter-hospital variation caused by different
characteristics in patient populations, “such as age, gender, and race, as well as
county demographic factors such as the unemployment rate, median income, and
urban-rural differences.” Its analysis also found that “there was less than a one-
thousandth percent chance” that Providence’s greater rate of coding of these MCCs
was “due to chance.”3 (Second amended complaint, ¶ 51.)
The complaint’s supporting graphs also demonstrate that there was a steadily
increasing trend in coding rates for the three allegedly misstated MCCs from 2011
to 2017 at comparable healthcare institutions. By 2017, non-Providence entities
generally coded claims with encephalopathy, respiratory failure, and severe
malnutrition at similar rates to Providence’s in 2011 or 2012. Providence also
showed a significant increase in coding MCCs after it hired JATA to assist with
claim documentation.
3
We note, however, that the complaint states later that Integra “only considered
claim groupings where there was less than a one-in-a-thousand chance that the
difference in major complication rate at Providence versus other hospitals was due
to random causes.” (Second amended complaint, ¶ 125.)
6
II. ANALYSIS
We review de novo a motion to dismiss a claim under the FCA. United States
ex rel. Campie v. Gilead Scis., Inc., 862 F.3d 890, 898 (9th Cir. 2017). The “essential
elements” of an FCA claim are: “(1) [A] false statement or fraudulent course of
conduct, (2) made with the scienter, (3) that was material, causing (4) the
government to pay out money or forfeit moneys due.” Campie, 862 F.3d at 899
(quoting United States ex rel. Hendow v. Univ. of Phx., 461 F.3d 1166, 1174 (9th
Cir. 2006)). On appeal, Defendants contest only the first element, that Integra has
adequately pleaded false or fraudulent conduct.
As with all fraud allegations, FCA claims must comply with Federal Rules of
Civil Procedure 8(a) and 9(b). See United States ex rel. Cafasso v. Gen. Dynamics
C4 Sys., Inc., 637 F.3d 1047, 1055 (9th Cir. 2011). Rule 8(a) requires the pleading
contain a plausible claim for relief, and Rule 9(b) imposes a heightened requirement
of particularity. Ibid. Integra’s complaint fails to meet the Rule 8 standard.
Rule 8 requires “a plausible claim for relief” to survive a motion to dismiss.
Ashcroft v. Iqbal, 556 U.S. 662, 679 (2009). At this stage, we “accept as true all of
the allegations contained in a complaint” but not the truth of “legal conclusions.” Id.
at 678. In evaluating plausibility, “courts must also consider an ‘obvious alternative
explanation’ for defendant’s behavior.” Eclectic Props. E., LLC v. Marcus &
Millichap Co., 751 F.3d 990, 996 (9th Cir. 2014) (quoting Iqbal, 556 U.S. at 682).
7
An allegation merely consistent with a defendant’s liability gets “the complaint close
to stating a claim, but without some further factual enhancement it stops short of the
line between possibility and plausibility . . . .” Bell Atl. Corp. v. Twombly, 550 U.S.
544, 557 (2007).
Integra argues that the existence of an alternative inference does not defeat a
plaintiff’s claim at the pleading stage. It relies on our statement in Starr v. Baca:
If there are two alternative explanations, one advanced by defendant
and the other advanced by plaintiff, both of which are plausible,
plaintiff’s complaint survives a motion to dismiss under Rule 12(b)(6).
Plaintiff’s complaint may be dismissed only when defendant’s
plausible alternative explanation is so convincing that plaintiff’s
explanation is im plausible.
652 F.3d 1202, 1216 (9th Cir. 2011) (emphasis and word break in original). But the
Starr court expressly relied on factual allegations that did not support an “‘obvious
alternative explanation,’ within the meaning of Iqbal.” Ibid. (quoting Iqbal, 556
U.S. at 682); see also Eclectic Props. E., 751 F.3d at 996–97 (discussing the above
quote from Starr and requiring “[s]omething more . . . , such as facts tending to
exclude the possibility that the alternative explanation is true, in order to render
plaintiffs’ allegations plausible” (quoting In re Century Aluminum Co. Sec. Litig.,
729 F.3d 1104, 1108 (9th Cir. 2013)).
Here, we accept the following factual allegations: Providence submitted
Medicare claims with secondary MCCs—such as encephalopathy, respiratory
failure, and severe malnutrition—at a higher rate than most other comparable
8
institutions; this increased rate was not the result of chance or variations in patient
populations; and Providence’s CDI specialists and JATA staff incentivized doctors
to use language conducive to coding higher-paying secondary diagnoses through
their documentation tips and queries. But we need not—and cannot—accept the
conclusion that these allegations resulted from fraud or that doctors recorded
unsupported medical conditions.
Integra does not rule out an obvious alternative explanation, that Providence,
with JATA’s assistance, was simply ahead of others in its industry. This situation is
unlike the competing inferences in Starr, “both of which [were] plausible.” 652 F.3d
at 1216. Integra offers only a possible explanation—that doctors lied about
underlying medical conditions—to explain a statistical trend that is consistent with
a plausible alternative (and legal) explanation.4 It is reasonable that Providence, one
of the largest healthcare systems in the country, which specifically hired consultants
to improve its Medicare billing, would be at the forefront of a national trend toward
coding these relevant MCCs at a higher rate. We need not accept the conclusion that
the defendant engaged in unlawful conduct when its actions are in line with lawful
“rational and competitive business strategy.” Twombly, 550 U.S. at 554. Therefore,
4
CMS has acknowledged that there is nothing “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.”
Medicare Program; Changes to the Hospital Inpatient Prospective Payment Systems
and Fiscal Year 2008 Rates, 72 Fed. Reg. 47,130, 47,180 (Aug. 22, 2007).
9
Integra does not state a plausible claim for relief, and its complaint must be
dismissed.5
III. CONCLUSION
Accordingly, we REVERSE the district court’s denial of the motion to dismiss
and REMAND with instructions to DISMISS the complaint.
5
We note that this conclusion does not categorically preclude statistical data from
being used to meet Rule 8(a)’s pleading requirement and, when paired with
particular details of a false claim, Rule 9(b).
10