United States District Court, D. Arizona
A. Teilborg Senior United States District Judge.
sets of motions are pending before the Court. First are
Defendants Baxter Healthcare Corporation
(“Baxter”), Scientific Protein Laboratories, LLC
(“SPL”), and American Capital Limited's
(“ACAS”) (collectively “Defendants”)
motions to exclude Plaintiffs Charlisa, Joshua, and Jennifer
Allen's (collectively “Plaintiffs”) expert
witnesses Ronald Moore, (Doc. 149), Suzanne Parisian, (Doc.
150), Clifford Siporin, (id.), and all three of
Plaintiffs' causation experts, (Doc. 151.) Second are
Defendants' motions for summary judgment. (Docs. 162
& 164.) Third is Plaintiffs' motion to exclude
Defense expert Dan Fintel. (Doc. 147.) All parties have
responded, (Docs. 153, 154, 155, 156, 181, 183), and replied,
(Docs. 157, 158, 159, 160, 188, 189.)
are the wife and children of Dr. Robert Allen, who they claim
was injured and killed by allegedly contaminated heparin
produced, marketed, and sold by Defendants. (Doc. 123.) They
claim that Defendants are liable under theories of strict
liability due to a manufacturing defect, negligence, breach
of implied warranty, breach of express warranty, statutory
wrongful death, and statutory survival. (Id. at
7-12.) Heparin is an anticoagulant and is one of the most
commonly used medical therapies. (Doc. 151 at 4.)
December 1, 2007, Dr. Allen, a forty-five-year-old
radiologist, entered the Mayo Clinic in Scottsdale, Arizona
(“Mayo”) with abdominal pain and self-diagnosed
gastrointestinal disease. (Doc. 155 at 3.) He continued to
suffer from this pain throughout the night. (Id. at
4.) At around 6:00 a.m. on December 2, 2007, Dr. Allen was
given a subcutaneous dose of heparin. (Id.) Between
10:15 a.m. and 10:30 a.m., Dr. Allen received a bolus dose of
heparin. (Id. at 5.) Just before 10:30 a.m., Dr.
Allen's ECG strip recorded the beginning of a myocardial
infarction, or heart attack, and his telemetry alarm began to
sound. (Id.) An echocardiogram performed at 10:34
a.m. indicated that there was significant damage to the
heart. (Id.) Dr. Allen was brought to the cardiac
catheterization laboratory at around 11:24 a.m., where he
underwent a thrombectomy to extract clots from his heart.
(Id.) At around 11:53 a.m., in the midst of the
thrombectomy, Dr. Allen received two more bolus doses of
heparin. (Id. at 5-6.) Eventually, Dr. Allen became
hypotensive, he vomited blood, and he suffered from
cardiogenic shock. (Id. at 6.) Following this
procedure, he developed heparin-induced thrombocytopenia
(“HIT”). (Id.) On December 6, 2007,
doctors replaced Dr. Allen's heart with a total
artificial heart. (Id. at 7.) Dr. Allen began to
suffer from end organ failure, and on February 27, 2008,
doctors replaced the total artificial heart with a human
transplant and replaced Dr. Allen's kidneys.
(Id.) The transplanted heart failed and, despite
further procedures, Dr. Allen died on March 7, 2008.
a limited liability company that primarily produces heparin
active pharmaceutical ingredient (“API”), a
precursor to usable heparin medication. (Doc. 162 at 3-4.) It
creates API in processing facilities in Wisconsin and in
China using crude heparin. (Id.) The primary raw
material in crude heparin is porcine intestinal mucosa, which
SPL sources from intermediaries, who in turn gather
intestines from slaughtered pigs in North America and China.
(Id.) ACAS is a majority shareholder in SPL
Acquisition Corp., the sole member of SPL. (Doc. 164 at 6-7.)
In 2007, Baxter used SPL heparin API to produce usable
heparin medication, which it distributed to wholesalers, who
passed it on to end users, such as Mayo. (Id. at 4.)
the end of December 2007, Baxter observed an increase in
adverse events from particular lots of their heparin. (Doc.
151 at 4.) Baxter notified the authorities of these issues,
who launched an investigation. (Doc. 151 5.) Public health
authorities ultimately concluded that these adverse events
were caused by a third party introducing a contaminant,
oversulfated chondroitin sulfate (“OSCS”), into
crude heparin as a cost-saving measure. (Doc. 162 at 6.)
around the country brought suits in state and federal court
for harm allegedly caused by contaminated heparin. (Doc. 151
at 8.) The federal cases were consolidated by the Judicial
Panel on Multidistrict Litigation in the Northern District of
Ohio (“MDL court”). (Id.) This case was
before the MDL court for over eight years and is the last of
the consolidated federal cases. (Id.; Doc. 155 at
Court has diversity jurisdiction under Section 1332. See 28
U.S.C. § 1332 (2012); see also U.S. Const.,
art. III, § 2. A court sitting in diversity applies
federal procedural law and state substantive law. See
Erie R. Co. v. Tompkins, 304 U.S. 64, 78 (1938). In the
case of Multidistrict Litigation (“MDL”), both
the MDL court and the non-MDL court apply the substantive law
of the non-MDL court's forum state. See In re Korean
Air Lines Co., Ltd., 642 F.3d 685, 699 & n.12 (9th
Cir. 2011) (citing In re Korean Air Lines Disaster of
Sept. 1, 1983, 829 F.2d 1171, 1176 (D.C. Cir. 1987));
Wahl v. Gen. Elec. Co., 786 F.3d 491, 494-99 (6th
Cir. 2015) (describing the majority view that cases filed
directly in an MDL apply the substantive law of the court
where the case would have otherwise been brought).
Choice-of-law decisions are substantive and must be settled
before further analysis. See Klaxon Co. v. Stentor Elec.
Mfg. Co., Inc., 313 U.S. 487, 496 (1941).
parties implicitly agree, with one exception, that Arizona
law applies. See, e.g., (Doc. 123 at 11-12;
Doc. 124 at 21, 24; Doc. 125 at 21, 24.) Therefore, Arizona
substantive law governs, except where otherwise
noted. See In re Korean Air Lines Disaster of
Sept. 1, 1983, 932 F.2d 1475, 1495 (“[C]ourts need
not address choice of law questions sua
sponte”); Ind. Lumbermens Mut. Ins. Co. v. W.
Or. Wood Prods., Inc., 268 F.3d 639, 644 n.3 (9th Cir.
2001) (applying Oregon law where the parties implicitly
agreed that Oregon law applied due to a choice-of-law
provision in the governing contract).
where “a state evidence rule is ‘intimately bound
up' with the rights and obligations being asserted,
” federal evidentiary rules apply. Wray v.
Gregory, 61 F.3d 1414, 1417 (9th Cir. 1995) (per
curiam). Neither party asserts, and this Court finds no
indication, that Arizona substantive law supplants the
Federal Rules of Evidence in this case. Therefore, the
Federal Rules of Evidence shall govern the evidentiary
matters presented in the pending motions.
Federal Rule of Civil Procedure 54(b), before final judgment,
a court may revise its earlier decisions that
“adjudicate fewer than all the claims” of the
parties. Fed.R.Civ.P. 54(b). This broad retrospective power
to modify, however, has been narrowed by the law of the case
doctrine. Under this doctrine, a court will generally not, as
a matter of practice, reexamine its earlier decisions.
Thomas v. Bible, 983 F.2d 152, 154 (9th Cir. 1993).
“For the doctrine to apply, the issue in question must
have been ‘decided either expressly or by necessary
implication in [the] previous disposition.'”
Id. (alteration in original) (quoting Milgard
Tempering, Inc. v. Selas Corp. of Am., 902 F.2d 703, 715
(9th Cir. 1990)). Courts retain discretion to apply the
doctrine, but abuse that discretion by reopening previously
decided issues falling outside of five enumerated
circumstances. Id. at 155. Thus, “[d]epending
on the nature of the case or the issue, ”
reconsideration may only occur where “(1) the first
decision was clearly erroneous; (2) an intervening change in
the law has occurred; (3) the evidence on remand is
substantially different; (4) other changed circumstances
exist; [or] (5) a manifest injustice would otherwise
United States Court of Appeals for the Ninth Circuit
(“Ninth Circuit”) has suggested, however, that
the law of the case does not bar a district court from
reconsidering its own pretrial rulings before final judgment.
Peralta v. Dillard, 744 F.3d 1076, 1088 (9th Cir.
2014) (en banc) (holding that the law of the case did not bar
a court from granting judgment as a matter of law to a party
despite previously rejecting that party's motion for
summary judgment). In so deciding, the court reasoned that
facts continue to develop as a case progresses, which may
change the appropriateness of prior rulings. Id.
Ninth Circuit has not held that the law of the case applies
to MDL cases; however, courts within this circuit have,
see, e.g., Parkinson v. Novartis Pharm.
Corp., 5 F.Supp.3d 1265, 1271-72 (D. Or. 2014), the
doctrine applies to other forms of transfer, Christianson
v. Colt Indus. Operating Corp., 486 U.S. 800, 815-16
(1988); Muth v. Fondren, 676 F.3d 815, 818-19 (9th
Cir. 2012), and the policies of judicial economy and finality
that justify MDL litigation, see Stanley A. Weigel,
The Judicial Panel on Multidistrict Litigation,
Transferor Courts and Transferee Courts, 78 F.R.D. 575,
577 (1978); 28 U.S.C. § 1407(a) (noting that transfer is
for the “convenience of parties and witnesses”
and must “promote . . . just and efficient
conduct”), all warrant applying the doctrine to MDL
cases. The parties generally accept, and the Court will
generally follow, the MDL court's holdings. The few
instances where the parties challenge such decisions are
discussed below and only in those circumstances will this
Court entertain the possibility of reconsideration.
Expert Witness Admissibility
Court acts as a gatekeeper in considering the admissibility
of expert witness testimony that involves “scientific,
technical, or other specialized knowledge.” Fed.R.Evid.
702(a); Fed.R.Evid. 104(a); Daubert v. Merrell Dow
Pharm., Inc., 509 U.S. 579, 597 (1993); Kumho Tire
Co., Ltd. v. Carmichael, 526 U.S. 137, 141 (1999). In
exercising this gatekeeping function, a court must determine
that the experts are qualified, that their opinions are
reliable, and that their testimony fits the case.
Daubert, 509 U.S. at 591-93; Fed.R.Evid. 702. The
proponent of the expert bears “the burden of proving
admissibility.” Lust v. Merrell Dow Pharm., 89
F.3d 594, 598 (9th Cir. 1996); see Daubert v. Merrell Dow
Pharm., 43 F.3d 1311, 1316 (9th Cir. 1995) (hereinafter
“Daubert II”). This analysis is meant to
be flexible. Daubert, 509 U.S. at 595. In engaging
in this flexible analysis, the Court must not supplant the
jury's fact-finding role, see Daubert, 509 U.S.
at 595; United States v. Sandoval-Mendoza, 472 F.3d
645, 654 (9th Cir. 2006); however, the Court cannot abdicate
its gatekeeping role nor delegate that responsibility to the
jury, Estate of Barabin v. AstenJohnson, Inc., 740
F.3d 457, 464 (9th Cir. 2014) (en banc). Ultimately,
“shaky but admissible evidence” is better handled
by “[v]igorous cross-examination, presentation of
contrary evidence, and careful instruction on the burden of
proof” than by judicial exclusion. Daubert,
509 U.S. at 597; see Alaska Rent-A-Car, Inc. v. Avis
Budget Grp., Inc., 738 F.3d 960, 969-70 (9th Cir. 2013).
a putative expert must be qualified as an expert by
“knowledge, skill, experience, training, or
education.” Fed.R.Evid. 702. The qualification standard
is meant to be broad and to seek a “minimal
foundation” justifying the expert's role as an
expert. Hangarter v. Provident Life & Accident Ins.
Co., 373 F.3d 998, 1015-16 (9th Cir. 2004) (quoting
Thomas v. Newton Int'l Enters., 42 F.3d 1266,
1269 (9th Cir. 1994)); see Ericson v. City of
Phoenix, No. CV-14-01942-PHX-JAT, 2016 WL 6522805, at
*3-4 (D. Ariz. Nov. 3, 2016) (finding a nurse practitioner
qualified to testify regarding asphyxiation, despite her lack
of independent research, due to her professional knowledge
of, and experience with, strangulation); Heck v. City of
Lake Havasu, No. CV 04-1810-PCT-NVW, 2006 WL 2460917, at
*7-8 (D. Ariz. Aug. 24, 2006) (finding that a professor's
extensive study of the effects of carbon monoxide poisoning
qualified him to testify that such poisoning caused the
decedent's death, despite the professor having no
training in forensic pathology).
the reliability of an expert's opinion is determined
almost exclusively by looking at their decision-making
process. Daubert, 509 U.S. at 593-94. In
Daubert, the Supreme Court provided four
non-exclusive factors to probe the expert's methodology.
Id. Under this scheme, a court asks whether the
expert's methodology “can be (and has been) tested,
” whether it has been subjected to peer review and
publication, what its “known or potential rate of
error” is, and whether it has been generally accepted
by others in the field. Id. A court need not
“methodologically walkthrough” each factor, as
“not every factor is relevant to reliability in every
case.” Murray v. S. Route Mar. SA, 870 F.3d
915, 924 (9th Cir. 2017); see Kumho Tire, 526 U.S.
at 142; Daubert II, 43 F.3d at 1316-17. Ultimately,
“the significance of each factor is
case-dependent.” Id. Furthermore, it is
“very significant” whether the expert is
testifying “about matters growing naturally and
directly out of research they have conducted independent of
litigation, or whether they have developed their opinions
expressly for the purpose of testifying.” Daubert
II, 43 F.3d at 1317. If the opinion is produced for
litigation, the proffering party must provide “other
objective, verifiable evidence that the testimony is based on
‘scientifically valid principles, '” such as
peer review. Id. In assessing reliability, a court
may peek beyond an expert's methodology to his or her
conclusion where “there is simply too great an
analytical gap between the data and the opinion
proffered.” Gen. Elec. Co. v. Joiner, 522 U.S.
136, 146 (1997).
fit “goes primarily to relevance.”
Daubert, 509 U.S. at 591. Testimony is relevant if
it “has any tendency to make a fact more or less
probable than it would be without the evidence, ” and
“the fact is of consequence in determining the
action.” Fed.R.Evid. 401. Fit goes beyond the bare
relevance standard, however, in that it further requires that
the testimony or evidence “speak clearly and directly
to an issue in dispute in the case” and “not
mislead the jury.” Daubert II, 43 F.3d at 1321
Court will first analyze Defendants' challenge to
Plaintiffs' product identification expert. Then it will
consider the parties' challenges to their opponents'
causation experts. Finally, it will consider Defendants'
challenge to Plaintiffs' professional standard of care
Plaintiffs' Product Identification Expert
challenge the opinion of Plaintiffs' product
identification expert- Ronald Moore, R.Ph-on three grounds.
First, Defendants claim that Mr. Moore, by claiming that the
6:00 a.m. and 10:15 a.m. doses were contaminated, contravenes
the MDL court's grant of summary judgment. (Doc. 149 at
1-2.) Second, Defendants aver, Mr. Moore's product
identification analysis is not reliable under
Daubert. (Id. at 2.) Third, Defendants
argue that Mr. Moore's product identification opinion is
based on “simple math, ” which is not helpful to
the jury. (Id.)
the Court ultimately concludes that Mr. Moore's product
identification opinion is unreliable, it does not reach the
third issue of whether Mr. Moore's product identification
opinion would be helpful to the jury.
The 6:00 a.m. and 10:15 a.m. Doses
claim that the MDL court has already determined, on summary
judgment, that Dr. Allen was not given contaminated heparin
at 6:00 a.m. and 10:15 a.m. (Id. at 1-2.) On summary
judgment, the MDL court had before it two affidavits by, and
the deposition of, Patricia Earl, Plaintiffs' MDL product
court found unpersuasive Ms. Earl's testimony that the
6:00 a.m. and 10:15 a.m. doses were contaminated. In re:
Heparin Prods. Liab. Litig., No. 1:10HC60098, 2016 WL
2868684, at *5 (N.D. Ohio May 17, 2016). The MDL court
explained that the Pyxis unit, a computerized pharmaceutical
dispensing machine, from which the heparin was drawn records
such withdrawals. Id. In particular, the
machine's records have a field for a MedId., which is
used for billing purposes, and a MedAltID, which can be
filled in by hospital staff. Id. The MedAltID for
the two doses at issue included the numbers
“46912621.” Id. The MDL court determined
that this sequence of numbers represented a National Drug
Code (“NDC”) number that identifies a drug's
manufacturer, the drug type and dosage, and the form of
packaging. Id. Usually, the MDL court explained, NDC
numbers are eleven units long and are broken up by hyphens in
the following form: “#####-####-##.” Id.
The FDA sets the manufacturer identification number
constituting the first five digits of the code. Id.
Baxter's manufacturer number is 00641, while 00469 is
that of APP Pharmaceuticals, LLC (“APP”), one of
Baxter's competitors. Id. The MDL court found
unpersuasive Ms. Earl's claim that the MedAltId.
numbers for the two doses at issue were an internal hospital
reference number, instead finding that these were NDC codes
written in shorthand without the hyphens and zeroes-an
occurrence Ms. Earl conceded was possible. Id. &
n.5. In particular, the MDL Court cited a Defense exhibit
showing that APP produced heparin with the NDC number
“00469-1262-15.” Id. at 9. Therefore,
the MDL court held that there was no genuine issue of
material fact regarding whether the two doses at issue were
contaminated Baxter heparin-finding that the NDC number
established that these doses were produced by APP-and granted
summary judgment on that issue. Id. at *6.
now claim that their new product identification expert, Mr.
Moore, has access to a database that their previous expert
did not, and can offer proof showing that the
MedAltId. “46912621” did not represent a
current NDC number, thus introducing the possibility that
Baxter produced the two doses at issue. (Doc. 156 at 6-9.)
Mr. Moore, using information contained in the database,
determined that the NDC number 00469-1262-15, which was
contained in the exhibit cited by the MDL court, became
obsolete on April 30, 2001, which was more than six years
before Dr. Allen was given the allegedly contaminated
heparin. (Id. at 9-10.) According to Plaintiffs,
heparin has a shelf life of less than two years.
(Id. at 8 n.5.) Furthermore, Mr. Moore cited
materials provided by Mayo showing that it did not order any
drugs with the NDC number 00469-1262-15 in 2007.
(Id. at 8.) Therefore, Mr. Moore concluded that Dr.
Allen could not have been given a drug with this NDC number.
(Id. at 8-9.) Rather, Mr. Moore explained that the
MedAltID field is filled manually, and is not routinely
updated when hospitals switch manufacturers. (Id.)
This undermines the MDL court's determination that the
two relevant doses must have been produced by APP.
make two arguments as to why this Court is not bound to
follow the MDL court's rulings on the two relevant doses.
(Doc. 155 at 11.) First, Plaintiffs note that the MDL court
only formally denied summary judgment. (Id.) Second,
Plaintiffs argue that the issue is not finally decided,
because Mr. Moore's review of the database produced new
evidence that Dr. Allen was not definitively given APP
The Scope of Summary Judgment
initial argument is that the MDL court's ruling is not
binding because the MDL court only formally denied
Defendants' motion for summary judgment. (Id.)
In other words, Plaintiffs dispute Defendants'
interpretation of the conclusion of the summary judgment
order, which Defendants interpret to mean that summary
judgment was denied only on claims stated in the conclusion
and granted on all other claims discussed in the order.
decision, the MDL court expressly stated that it found
“no genuine issue of material fact as to whether Baxter
manufactured the heparin administered at 6:00 a.m. and 10:15
a.m.; it clearly did not.” In re: Heparin Prods.
Liab. Litig., 2016 WL 2868684, at *5. The MDL court
expressly excluded those doses from the paragraph detailing
which issues are appropriate for trial. Id. at *6-7.
However, the MDL court did not restate this holding in the
conclusion. Plaintiffs fail to cite authority establishing
that a court only grants summary judgment on those issues
expressly stated in the conclusion, rather than those
explicitly made in the body of the opinion.
however, Plaintiffs' argument were correct, this Court
would adopt the MDL court's reasoning, based upon the
evidence before it at that time, and hold that there was no
genuine issue of material fact as to whether the two relevant
doses were contaminated Baxter heparin.
The Law of the Case
second argument is effectively that there is substantially
different evidence before this Court justifying a departure
from the law of the case. (Doc. 155 at 11-12.)
Defendants' primary response is that this evidence is not
new; while Ms. Earl did not have access to the same database
as Mr. Moore, she did have access to public sources of
information, such as “RED BOOK” and she twice
failed to appropriately explain the information in the Pyxis
records. (Doc. 159 at 6-7.)
initial matter, the law of the case likely does not bind this
Court to follow pretrial rulings, Peralta v.
Dillard, 744 F.3d 1076, 1088 (9th Cir. 2014) (en banc);
therefore, it can discretionarily reconsider the motion for
summary judgment with no further analysis. However, even
applying the law of the case doctrine, Plaintiffs have shown
substantially different evidence justifying departure from
the prior ruling.
Moore's explanation that the sequence of numbers in the
MedAltID field represents an expired NDC number is clearly
substantially different evidence than that presented by Ms.
Earl. Ms. Earl did not testify that the code represented an
expired NDC number, but rather that the units represented an
internal Mayo reference number, which the MDL court
justifiably found unpersuasive. Given its dispositive nature,
Mr. Moore's testimony constitutes substantially different
make four arguments as to why the MDL court's ruling
stands. First, there is no new evidence justifying
reconsideration. Second, reconsideration frustrates the
purpose of the MDL procedure. Third, this Court should defer
to the MDL court's expertise. Fourth, Local Rule 7.2 bars
Plaintiffs' belated request for reconsideration.
Defendants assert, without citation, that this evidence must
be new in the sense that it was not previously discoverable
and contend that Ms. Earl had the opportunity to access
databases that detailed the expiration of the NDC number.
(Doc. 149 at 10-11; Doc. 159 at 6-7.) In making this
argument, Defendants rely primarily on equitable
considerations, arguing that they have “made strategic
decisions about fact depositions, decided not to name a
product identification expert, prepared a medical causation
expert report, filed Daubert motions, deposed
plaintiffs' experts, and have been working to prepare a
summary judgment motion” in reliance on the MDL
court's product identification holding. (Doc. 159 at 5.)
equitable force of Defendants' argument is diminished by
the permissive discovery scheduled set by this Court.
Defendants became aware of Mr. Moore's opinion on June 9,
2017. (Doc. 126 at 2.) Defendants then had until July 7, 2017
to make their expert disclosures, August 18, 2017 to depose
Mr. Moore, September 1, 2017 to file their Daubert
motions, and October 6, 2017 to file their motions for
summary judgment. (Doc. 144 at 2-3.) Thus, Defendants had
nearly a month to assess Mr. Moore's opinion-which
plainly introduces a genuine issue of material fact regarding
the status of the 6:00 a.m. and 10:15 a.m. doses-and
determine whether it would be prudent to retain a product
identification expert. Additionally, Defendants took Mr.
Moore's deposition on August 30, 2017-over two months
after they received his report-and did not once inquire about
his opinions regarding the two earlier doses. See
(Doc. 149-2.) Furthermore, Defendants had nearly three months
to address the earlier doses in their Daubert motion
and four months to address the issue on summary judgment.
Simply, Defendants squandered multiple opportunities, over a
span of months, to address Mr. Moore's opinions regarding
the 6:00 a.m. and 10:15 a.m. doses, despite awareness that
discovery was still ongoing and that Plaintiffs had offered
Mr. Moore to testify on the earlier doses.
Defendants' contention that the primary purpose of
multidistrict litigation-to economize-would be frustrated by
reconsidering these issues is unavailing. Multidistrict
litigation encourages consolidation to “promote the
just and efficient conduct” of actions, 28 U.S.C.
§ 1407(a); this goal is not met by the MDL court
adjudicating issues of fact and law unique to a single
action. This reasoning is bolstered by the Judicial Panel on
Multidistrict Litigation's opinion consolidating the
heparin cases due to “common questions of fact”
involving allegedly contaminated Baxter heparin, which would
serve to “eliminate duplicative discovery; avoid
inconsistent pretrial rulings; and conserve the resources of
the parties, their counsel and the judiciary.” In
re Heparin Prods. Liab. Litig., 559 F.Supp.2d 1403, 1404
(J.P.M.L. 2008). Discovery and pretrial rulings on whether
Dr. Allen received contaminated Baxter heparin at particular
times do not bear on the common questions of fact that
justify multidistrict litigation.
Defendants' appeal to the MDL court's specialized
expertise, gained over a period of years adjudicating cases
involving allegedly contaminated heparin, does not squarely
bear on this issue. While the MDL court surely obtained an
impressive understanding of the issues in this case,
including product identification, its decision regarding the
earlier doses was grounded in a different set of evidence
than that before the Court today. The MDL court did not have
the benefit of Mr. Moore tracking down the defunct NDC number
and his equally important explanation as to why that number
remained in the MedAltID field. Thus, even without the MDL
court's extensive experience with scores of heparin
cases, this Court is confident that Mr. Moore has unsurfaced
a genuine issue of material fact regarding whether the
earlier doses were contaminated Baxter heparin.
the Local Rules do not bar Mr. Moore's testimony for two
reasons. First, this Court need not even address
Defendants' invocation of the Local Rules, because
Defendants raise this argument for the first time in a reply
brief, depriving Plaintiffs of an opportunity to respond.
See Sonoran Res., LLC v. Oroco Res. Corp., No.
CV-13-01266-PHX-JJT, 2015 WL 11089497, at *7 n.4 (D. Ariz.
Apr. 17, 2015); MJG Enters., Inc. v. Cloyd, No.
CV-10-0086-PHX-MHM, 2010 WL 3842222, at *6 n.1 (D. Ariz.
Sept. 27, 2010). Second, good cause provides an exception to
the timing requirements of Local Rule 7.2(g)(2). See
LRCiv 7.2(g)(2). Good cause exists for Plaintiffs'
belated filing, given the irregular procedural structure of
this proceeding under which the parties were subject to
multiple periods of discovery.
next argue that Mr. Moore's opinion is unreliable for
four reasons. First, Mr. Moore's turnaround time
calculation is unreliable. (Doc. 149 at 11.) Second, Mr.
Moore improperly cherry-picked the data that he relies upon.
(Id. at 15.) Third, Moore's opinion fails to
satisfy the Daubert factors. (Id. at
15-16.) Fourth, Mr. Moore's experience does not make his
opinion reliable. (Id. at 17.)
Moore provided two different conclusions regarding the
turnaround time of heparin. First, Mr. Moore concluded that
“it was more likely than not” that Dr. Allen
received contaminated Baxter heparin from 1, 000U/ML in 10ML
vials for order numbers 115 and 116. (Doc. 149-1 at 5.)
Second, he concluded that it was “more likely than
not” that Dr. Allen received contaminated Baxter
heparin from 5, 000U/ML in 1ML vials for order numbers 68 and
102. (Id.) To reach these conclusions, Mr. Moore had
to opine regarding the amount of time the wholesaler
(“Cardinal”) held onto Baxter heparin, which lot
numbers of heparin that Cardinal shipped to Mayo, and the
amount of time Mayo stored heparin before using it.
Moore's opinion is based on a number of sources. First,
Mr. Moore relies on his experience in the pharmaceutical
industry, with emphasis on his experience at Cardinal in
particular. As part of this experience, Mr. Moore claims that
pharmaceutical wholesalers and hospitals follow two industry
standards: just-in-time inventory practice, under which
product is ordered as needed rather than stockpiled, and
first-in, first-out (“FIFO”), where drugs are
shipped and utilized in chronological order to prevent
expiration. (Id. at 13-14.) Second, he relies on
four spreadsheets-one from Baxter describing shipments to
Cardinal, one from Cardinal describing all Cardinal purchases
of heparin, one from Cardinal describing all heparin
shipments to Mayo, and one from Mayo showing all purchases of
heparin from Cardinal. (Id. at 10-11.) The Mayo and
Cardinal spreadsheets include the manufacturer name, the NDC
number, the amount of heparin in each package, the order
quantity, and the shipment quantity. (Id.) The
Baxter spreadsheet contains the same information with the
addition of lot numbers. (Id. at 11.) Third, Mr.
Moore relies on a document by Baxter detailing the
contaminated shipment lot numbers. (Id.) Fourth, Mr.
Moore relies upon a memo produced by a Baxter employee
estimating the amount of uncontaminated Baxter product
remaining in the field. Fifth, Mr. Moore relies on written
Mayo policies detailing drug purchase times, inventory
review, and drug expiration disposal. (Id. at
12-13.) Sixth, Mr. Moore relies on a Mayo document describing
how much heparin they sent back in the recall. (Id.
aver that Mr. Moore's calculations for the turnaround
times for the heparin at Cardinal and Mayo are based upon
speculation. (Id. at 11.) Generally, Defendants
claim, Mr. Moore guesses as to how long it took for Cardinal
to ship heparin to Mayo, and then guesses as to how long it
took Mayo to use the heparin upon receipt. Defendants argue
that Mr. Moore's reliance upon documents to justify his
guesses is unfounded because he “draws unreasonable
conclusions from them.” (Id.) Thus, Defendants
conclude, Mr. Moore's testimony is unreliable.
1, 000U/ML in 10ML Vials
Moore claims that it is more likely than not that Dr. Allen
received two doses of contaminated Baxter heparin contained
in 1, 000U/ML in 10ML vials. To reach this conclusion, Mr.
Moore tracks contaminated Baxter heparin, via the lot numbers
and shipment information provided by Baxter, to Cardinal.
(Id. at 16.) Once at Cardinal, Mr. Moore can no
longer track lot numbers, but instead bases his opinion on
industry practices-namely, just-in-time inventory and
FIFO-the shipment times, the quantity, and the NDC numbers of
heparin moving from Cardinal to Mayo, and the Baxter employee
memo detailing the amount of uncontaminated Baxter heparin
remaining in the field. (Id.) Ultimately, Mr. Moore
concludes that because Cardinal only had contaminated heparin
for the five weeks preceding December 2, 2007, and because
Cardinal and Mayo each stored that heparin for two weeks, Dr.
Allen more likely than not received contaminated Baxter
heparin. (Id. at 15-17.)
Defendants argue that Mr. Moore's calculation of the
turnaround time of two weeks for the 1, 000U/ML in 10ML vials
is unreliable, because it is based, in part, upon a
post-recall Baxter document discussing average Baxter heparin
supplies. (Doc. 149 at 11-12.) This Baxter memo prepared by
E. DeLuise, a Baxter employee, (“DeLuise Memo”)
discusses four wholesalers, including Cardinal, who make up
85% of Baxter's heparin sales for the relevant dose
sizes. (Doc. 149-1 at 17.) The relevant portion of this memo
[F]rom a historical perspective, there is approximately 2
weeks inventory stored on average at the end user location .
. . . I estimate that the existing field inventory of 10 [M]L
vials to be a total of 2.2 weeks of usable Baxter product
inventory at the wholesaler and end user locations.
first contend that the DeLuise Memo does not support Mr.
Moore's conclusion, because it was written after
Baxter's recall of heparin, and both fails to discuss
pre-recall supplies at wholesalers and hospitals and
establishes a shorter turnaround time due diminishing
post-recall supplies. (Doc. 149 at 11.) Plaintiffs respond
that the memo expressly discusses historical averages, thus
implicating pre-recall supplies of heparin. (Doc. 156 at 14.)
memo temporally distinguishes its conclusions as to past
data-that end users, like Mayo, historically and on average
have two weeks of heparin on hand-with its estimations as to
the current post-recall amounts of heparin held by both
wholesalers, like Cardinal, and end users. Thus, the fact
that there was a recall does not affect the memo's
conclusion as to the two week time period for end users at
the time Dr. Allen received heparin. With regard to
wholesalers, however, the memo only estimates post-recall
turnaround times of 2.2 weeks.
contend that Defendants are “disingenuous” for
making the argument that there would be less inventory due to
the recall, because the recall occurred one day before the
memo went out. (Doc. 156 at 14 n.9.) The memo, however, takes
into account “usable Baxter product, ”
meaning that it does not count any potentially-contaminated
recalled products. (Doc. 149-1 at 17 (emphasis added.)) The
practical difficulties in implementing a recall do not bear
on this analysis, because the memo analytically takes into
account only non-recalled product.
next contend that the memo does not provide support for Mr.
Moore's conclusion as to wholesalers, because it averages
the data from four different wholesalers to reach the 2.2
week turnaround time for wholesalers. (Doc. 149 at 12.)
Plaintiffs do not respond to this contention. See
(Doc. 156.) This certainly does cut against the
appropriateness of Mr. Moore's reliance on the report,
given that Cardinal's inventory could have varied greatly
from that of other wholesalers, which would not be clearly
reflected in the average. Although unraised by Defendants,
the same logic seems to corrupt the reliability of the
memo's estimation of end user inventory as well.
See (Doc. 149-1 at 17 (“[T]here is
approximately 2 weeks inventory stored on
average at the end user location”) (emphasis
added.)) Thus, without more, Mr. Moore's reliance on the
memo was misplaced and lessens the reliability of his expert
contend that Mr. Moore independently corroborated the
memo's conclusions through personal experience working at
Cardinal and independent review of records from Cardinal and
Mayo showing two week turnaround times. (Doc. 156 at 13.) As
Defendants note, however, Mr. Moore's experience does not
directly support his opinion, because he was not involved in
Cardinal's sale of heparin to end users during 2007, he
did not request information from Cardinal or Mayo on how long
they stored heparin, and he was not aware of how long it took
Cardinal to ship heparin to Mayo. (Doc. 149 at 17.) While his
experience working at Cardinal gave him some specialized
knowledge as to Cardinal's internal practices, his lack
of familiarity with heparin, Cardinal's storage of
heparin, and Mayo's storage practices detract from the
corroborating power of his opinion. Additionally, as
discussed below, Mr. Moore's independent analysis of
Cardinal and Mayo turnaround times is unreliable.
next contend that Mr. Moore improperly used the DeLuise Memo
to determine that Cardinal had a two week turnaround time.
(Doc. 149 at 12.) As Defendants note, and as discussed above,
the memo estimates that end users historically had a two week
turnaround time, while end users and wholesalers had 2.2
weeks of inventory at the time the memo was written. (Doc.
149-1 at 17.) Mr. Moore-without explanation-concluded that
Cardinal, a wholesaler, had a two-week inventory of heparin,
despite the report estimating 2.2 weeks. (Id.)
Defendants argue that that this unexplained rounding down has
a practical impact, because Cardinal had received a shipment
of uncontaminated heparin “just four days before the
shipment that Moore identifies as containing Robert
Allen's dose.” (Doc. 149 at 12.) Plaintiffs respond
that this rounding down does not affect Mr. Moore's
conclusion, because he “identifies as likely containing
Dr. Allen's dose” contaminated shipments
“received by Cardinal on November 12 and 16,
2007.” (Doc. 156 at 14.) Plaintiffs go on to say that
the prior two shipments, on November 5, 2007 and October 26,
2007, were also both contaminated. (Id.) Defendants
correctly respond that Mr. Moore never limited Dr.
Allen's doses to November 12 and 16, but rather concluded
that Dr. Allen likely received “Baxter lot number
087095 or 087092, ” which totaled six shipments to
Cardinal, and lot number 087095 was received as early as
October 26, 2007. (Doc. 159 at 11; Doc. 149-1 at 17.) Just
four days before that date, Cardinal received a shipment of
uncontaminated Baxter heparin. (Doc. 194-2 at 76.) Plaintiffs
note that Mr. Moore testified that Cardinal's practices
would put them “at the lesser end of average rather
than the greater.” (Doc. 156 at 14.) Mr. Moore
admitted, however, that if one used the unrounded 2.2 weeks
from the DeLuise Memo, it was “within the realm of
possibility” that the lot 087092 would not have even
been received by Mayo in time for use in Dr. Allen's
procedure, (Doc. 130- 31.) The uncertainty in Mr. Moore's
opinion arises from him “not knowing . . . the
specifics.” (Id. at 131.) Uncertainty
regarding reasonably obtainable information weighs against
this point was Mr. Moore's testimony that it was possible
that “Cardinal's inventory supply [of heparin]
could be 15, 16, 17, 18 days.” (Doc. 149-2 at 108.) Mr.
Moore testified that this higher end was not likely, because
Cardinal wanted to maintain a “thin” supply to
obtain a “maximum return on” investment.
(Id. at 109.) But Mr. Moore still must rely on
speculation, based upon his experience with Cardinal in the
past, in a different department, involving a different region
of the country, with different drugs, to reach this
conclusion. Thus, Mr. Moore's methodology does not
reliably provide the basis for the conclusion that it was
more likely than not that Dr. Allen received contaminated
rather than uncontaminated heparin from the 1, 000U/ML in
5, 000U/ML in 1ML Vials
Moore concluded that Dr. Allen more likely than not received
two doses of contaminated Baxter heparin from 5, 000U/ML in
1ML vials. Mr. Moore's analysis of these doses is similar
to that of the 1, 000U/ML in 10ML vials, and results in Mr.
Moore finding that there is a 21-day turnaround period from
Cardinal receiving the heparin from Baxter to the patient
receiving the heparin at Mayo. (Doc. 149-1 at 19.) Mr. Moore
determined that the two lots Baxter shipped to Cardinal
closest to 21 days before Dr. Allen received heparin at Mayo
were lots of contaminated Baxter heparin. (Id.) One
of these shipments, received November 6, 2007, was the last
contaminated shipment containing NDC number 0641-0400-25.
(Id.) The next shipment, which Cardinal received on
December 4, 2007, was also contaminated and included a
different NDC number. (Id.) The heparin with the new
NDC number was first received by Mayo on December 11, 2007,
which led Mr. Moore to conclude that the turnaround time from
Cardinal to Mayo was one week. (Id. at 26.)
contend that Mr. Moore's testimony that Cardinal turned
around heparin in a week conflicts with the evidence that
Cardinal did not receive any heparin for a month, suggesting
that Cardinal did not completely turnover its heparin within
a week. (Doc. 149 at 13; Doc. 149-2 at 89.) Mr. Moore's
failure to explain his departure from this evidence,
Defendants contend, makes his opinion unreliable.
attempt to defend Mr. Moore's opinion on this front by
noting that Mr. Moore opined that there was a supply problem,
and that past data showed a one week turnaround time. (Doc.
156 at 16.) But Mr. Moore's suggestion of a supply
problem is pure speculation. According to Mr. Moore:
[There was] some problem that I guess Baxter had
recognized. Because their shipments up to that point had been
fairly regular, and then all of a sudden, you have something
that's changed. Because the next entry after that, almost
28 days, is a brand-new NDC number, which that would have
meant they were having to - they were depleting the inventory
down even further . . . . [T]he only thing I can
speculate on is that in that time period, knowing
that they had a limited supply coming in, that they would
have modified their shipping. Mayo Health being one of their
probably primary customers, they would have begun
parceling that and holding that inventory in stock longer
than normal to ship out. But I'm having to speculate
on that because I don't have that information.
(Doc. 149-2 at 90, 93-94 (emphasis added.))
Moore, through speculation, invents a supply chain problem to
explain away contrary evidence, which diminishes the
reliability of his expert opinion. See Cabrera v. Cordis
Corp., 134 F.3d 1418, 1423 (9th Cir. 1998) (“An
opinion based on . . . unsubstantiated and undocumented
information is the antithesis of the scientifically reliable
expert opinion admissible under Daubert and Rule
702.”); United States v. Scholl, 959 F.Supp.
1189, 1197 (D. Ariz. 1997) (excluding as unreliable expert
testimony that “barely r[ose] above mere
further attack Mr. Moore's opinion regarding the 5,
000U/ML in 1ML vials by claiming that he failed to account
for Cardinal receiving a large shipment of uncontaminated
heparin “just four days before a smaller shipment of
contaminated heparin containing the lot that Moore
alleges” that Dr. Allen received. (Doc. 149 at 14
(emphasis removed.)) Plaintiffs respond by arguing that Mr.
Moore did consider this larger shipment and that their
disagreement with his conclusion is a matter appropriate for
cross-examination rather than exclusion. (Doc. 156 at 16.)
Cardinal received 320 packages of uncontaminated heparin on
October 22, 2007. (Doc. 149-1 at 18.) Mr. Moore opines that
Dr. Allen could have received heparin from a contaminated
shipment of 160 packages that Cardinal received on October
26, 2007, just four days later. (Id.) When asked why
he concluded that it was more likely than not that Dr. Allen
received the contaminated heparin from the smaller shipment
rather than the uncontaminated heparin from the larger
shipment, Mr. Moore replied that the larger shipment
“would have been at the front end of anything that they
were distributing out to other locations, which
that's - that's information that - that is not
available to me . . . . [I]t just seems reasonable that
the only thing left to be distributed were two other
[contaminated] lots.” (Doc. 149-2 at 98 (emphasis
added.)) Defense counsel probed Mr. Moore by asking
“You say it seems reasonable. But without knowing
Cardinal's inventory and what other customer Cardinal was
shipping” this form of heparin to “you don't
know with any certainty whether Cardinal's shipment to
Mayo came from” the contaminated lot “versus a
prior lot?” (Id.) Mr. Moore replied that based
upon Cardinal's “distribution pattern . . . it just
seems reasonable that was the only thing they would have had
to ship at that point.” (Id. at 99.)
Moore never explains, beyond relying on just-in-time
inventory and FIFO, why it was more likely than not that Dr.
Allen received heparin from the smaller contaminated shipment
rather than the larger uncontaminated shipment. Compounding
this uncertainty is Mr. Moore's admission that he is
unaware of the number of end users to which Cardinal sells
heparin. Thus, Mr. Moore's methodology does not reliably
provide the basis for the conclusion that it was more likely
than not that Dr. Allen received contaminated rather than
uncontaminated heparin from the 5, 000U/ML in 1ML vials.
All Vial Sizes
regard to both vial sizes, Defendants contend that Mr.
Moore's turnaround analysis fails to account for changes
in shipment size. (Doc. 149 at 14.) Plaintiffs respond that
Mr. Moore did account for changes in shipment size by noting
that shipment size fluctuates with expected sales. (Doc. 156
at 16.) Countering this reasonable explanation, however, is
the following exchange between Mr. Moore and Defense counsel,
where Mr. Moore agreed that size of a shipment affects its
relevancy to the analysis:
Q. I assume you thought it was relevant and appropriate to
list the September 18 shipment in Table 2 in this analysis of
the 5, 000-unit dose?
A. Right. That's a good - that's a fairly high
quantity, 320 times 25. That's a lot of tainted heparin
to have on their shelves.
(Doc. 149-2 at 73.)
Mr. Moore fails to take one approach to the size of shipments
in his analysis. On the one hand, Mr. Moore claims to treat
all shipments consistently, regardless of size, but on the
other hand, Mr. Moore highlighted certain shipments as more
important than others due to their size. This inconsistent
approach suggests deficiencies in his methodology.
with regard to both doses, Defendants contend that Mr.
Moore's concession that he could only say more likely
than not that Dr. Allen received heparin from either of two
contaminated lots, rather than one specific lot, “alone
warrants exclusion.” (Doc. 149 at 15; Doc. 159 at 12.)
The Court disagrees that this form of conclusion diminishes
the reliability of Mr. Moore's methodology.
Daubert does not require certainty.
further contend that Mr. Moore cherry-picked the data for his
analysis to produce a litigation-driven result. (Doc. 149 at
15.) Specifically, Mr. Moore's report includes two tables
that individually list shipments from Baxter to Cardinal of
5, 000U/ML in 1ML vials and 1, 000U/ML in 10ML vials
respectively. (Id.) These tables utilize different
start dates and end dates that Defendants contend were chosen
to exclude uncontaminated shipments. (Id.)
respond that this is a matter appropriate for
cross-examination and that Mr. Moore chose as relevant start
dates the time Cardinal originally received contaminated
heparin from Baxter. (Doc. 156 at 12.) He did not include
earlier uncontaminated shipments, Plaintiffs claim, because
he did not believe that such shipments would be available for
Dr. Allen on December 2, 2007. (Id.)
Moore did effectively cherry-pick data. While he determined
that the turnaround times for the 5, 000U/ML in 1 ML vials
and 1, 000 U/ML in 10ML vials were three weeks and four weeks
respectively, he chose to display a longer period of data for
the 5, 000U/ML in 1 ML vials, which had a shorter turnaround
time. The effect of this was to exclude prior shipments of
uncontaminated heparin from the 1, 000 U/ML in 10ML vials
table, which had a longer turnaround time. Regardless of
whether the lots were actually available to Dr. Allen on