Milward v. Rust-Oleum Corp.

United States Court of Appeals
For the First Circuit

No. 13-2132

BRIAN MILWARD and LINDA J. MILWARD,

Plaintiffs, Appellants,

v.

RUST-OLEUM CORPORATION,

Defendant, Appellee.

APPEAL FROM THE UNITED STATES DISTRICT COURT
FOR THE DISTRICT OF MASSACHUSETTS

[Hon. Douglas P. Woodlock, U.S. District Judge]

Before

Howard, Chief Judge,
Thompson, Circuit Judge,
and Laplante,* District Judge.

Steve Baughman Jensen, with whom Allen Stewart, P.C., Ian
McCallister, and Kreindler & Kreindler, LLP, were on brief, for
appellants.
Francis M. Lynch, with whom Cetrulo LLP, was on brief, for
appellee.

April 25, 2016

* Of the District of New Hampshire, sitting by designation.
 HOWARD, Chief Judge. In this toxic tort case, we

previously considered the admissibility of testimony from the

plaintiffs' general causation expert. At issue in the present

appeal is whether the district court abused its discretion in

excluding the testimony of the plaintiffs' specific causation

expert. We conclude that the district court's ruling was a

supportable exercise of its discretion, and we therefore affirm

the grant of summary judgment to the defendant following that

evidentiary ruling.

I.

Background

Brian Milward worked as a pipefitter and refrigerator

technician for over thirty years. During the course of his

employment, Milward was exposed to varying levels of benzene from

paints and other products manufactured by (among others) Rust-

Oleum Corporation. In 2004, he was diagnosed with Acute

Promyelocytic Leukemia ("APL"). Three years later, Milward and

his spouse sued a number of defendants on the theory that their

negligence caused Milward's disease. The only remaining defendant

is Rust-Oleum.

To succeed against Rust-Oleum, the Milwards had the

burden of establishing, through expert testimony, general and

specific causation. In other words, they needed to show that

exposure to benzene can cause APL (general causation), and that

- 2 -
exposure to benzene was, in fact, a substantial factor in the

development of Brian's APL (specific causation). The district

court bifurcated the proceedings; it planned first to address the

admissibility of expert testimony on general causation, and then

to consider the specific causation issue.

In a 2009 ruling, the district court excluded the

Milwards' general causation expert. Accordingly, it entered

judgment in favor of the defendants without proceeding to the

second phase of the case. The Milwards appealed that decision

and, for reasons specific to their general causation expert, we

reversed. See Milward v. Acuity Specialty Prods. Grp., Inc., 639

F.3d 11 (1st Cir. 2011). We remanded the case to the district

court to proceed to the specific causation question.

Under the supervision of a different district court

judge, the parties engaged in discovery on the subject of specific

causation. Relevant here, the Milwards retained occupational

medicine physician Dr. Sheila Butler to serve as their expert

witness.1 The admissibility of her opinion testimony is at the

1 The Milwards also engaged industrial hygienist Dr. James
Stewart. Dr. Stewart evaluated Brian Milward's exposure to benzene
at various points in his career and calculated the benzene levels
in various products that he used. Based on those considerations,
Dr. Stewart estimated that Milward was exposed to benzene at a
level of 25.6 parts per million-years (the measurement of the
amount of benzene equivalent to what a person would breathe on
average each day of the year a person spent at work). The district
court found Dr. Stewart's testimony to be admissible, and Rust-

- 3 -
heart of this appeal, and thus additional background on her opinion

is in order.

Dr. Butler

Dr. Butler, an employee of the Veterans Administration,

specializes in clinical assessments of environmental and

occupational exposure in combat-exposed veterans. In her proposed

testimony, Dr. Butler presented three theories.

First, she testified that although benzene is naturally

occurring, there is no safe level of benzene exposure. This was

her predominant theory, and she consistently reiterated her

hypothesis. She emphasized that she reached this conclusion by

examining "the biology, the pathophysiology, what the substance

does to the person and the disease process." And, she noted, she

was able to do so without relying on any of the relevant

epidemiological studies. Given this no-safe level theory, Dr.

Butler maintained that Milward's exposure (as detailed by Dr.

Stewart) was likely the cause of his APL. The district court

rejected this hypothesis because it could not be properly tested

with any known rate of error. The Milwards do not meaningfully

challenge the district court's conclusion on appeal. Accordingly,

we assume that the ruling was correct and bypass further discussion

Oleum now argues that this decision was erroneous. Given our
disposition of the case, we do not reach this argument.

- 4 -
of the issue. See Mills v. U.S. Bank, NA, 753 F.3d 47, 55 (1st

Cir. 2014).

Second, Dr. Butler rather cursorily concluded that even

beyond the no-safe level hypothesis, certain epidemiological

studies have established that an individual's "relative risk" of

developing APL increases when exposed to specified amounts of

benzene. She then compared Milward's exposure levels to those

that had been found to be dangerous in that research. Since

Milward's exposure was higher than the amounts found to be

hazardous, Dr. Butler reasoned that benzene exposure was likely

the cause of his APL. Notably, she did not explain why she chose

the studies on which she relied, nor did she address any study

with contrary findings. In fact, during Dr. Butler's deposition,

defendant's counsel asked her a number of questions about her

ability and willingness to engage with the relevant

epidemiological research. For instance, counsel asked, "Are you

aware of any studies which find that there is no relationship

between benzene exposure and APL," to which she answered "Yes . .

. the literature [] has support for both." Counsel then asked,

"Do you intend in this case to weigh the different epidemiological

studies and offer an opinion as to which ones we should rely on

and which ones we should discount," to which she replied, "No."

Finally, Dr. Butler engaged in a "differential

diagnosis" to conclude that benzene exposure likely caused

- 5 -
Milward's APL. Through this method (essentially a process of

elimination) Dr. Butler "ruled out" some of the more common factors

associated with APL, among them obesity and smoking. She then

determined that since benzene exposure was a potential cause, she

could also "rule out" an idiopathic diagnosis (or, a diagnosis

without a known cause). Thus, since benzene exposure was the only

significant potential cause remaining, she concluded that it was

likely the culprit.

Procedural History

Back in court, Rust-Oleum moved both to exclude Dr.

Butler's testimony and for summary judgment. The district court

evaluated, and rejected, each of the theories that Dr. Butler put

forward to establish specific causation. For reasons discussed

below, the judge ultimately ruled that Dr. Butler's testimony was

inadmissible under Federal Rule of Evidence 702. Since the

Milwards could not establish specific causation without Dr.

Butler's testimony, the district court granted summary judgment in

favor of Rust-Oleum. Fed. R. Civ. P. 56. This timely appeal

followed.

II.

We review the district court's decision to admit or

exclude expert testimony for abuse of discretion. See United

States v. Shay, 57 F.3d 126, 132 (1st Cir. 1995) (noting that we

will only "reverse a decision . . . if (1) the district court based

- 6 -
the decision on an incorrect legal standard . . . or (2) we have

a definite and firm conviction that the court made a clear error

of judgment. . . ."). Predicate factual findings are reviewed for

clear error, while pure questions of law engender de novo review.

Milward, 639 F.3d at 13-14. As for the district court's ultimate

decision to grant Rust-Oleum summary judgment, because the

Milwards are proceeding under state-law theories of liability, we

apply Massachusetts law, see Philibotte v. Nisource Corp. Servs.

Co., 793 F.3d 159, 165 (1st Cir. 2015), and review the decision de

novo, see Samaan v. St. Joseph Hosp., 670 F.3d 21, 38 (1st Cir.

2012).

As in the district court, our admissibility inquiry is

guided by Federal Rule of Evidence 702, which provides that:

A witness who is qualified as an expert by
knowledge, skill, experience, training, or
education may testify in the form of an
opinion or otherwise if:

(a) the expert's scientific, technical, or
other specialized knowledge will help the
trier of fact to understand the evidence or to
determine a fact in issue;
(b) the testimony is based on sufficient facts
or data;
(c) the testimony is the product of reliable
principles and methods; and
(d) the expert has reliably applied the
principles and methods to the facts of the
case.

Fed. R. Evid. 702. In applying Rule 702, the district court serves

as the gatekeeper for expert testimony by "ensuring that [it] . .

- 7 -
. both rests on a reliable foundation and is relevant to the task

at hand." Daubert v. Merrell Dow Pharm., 509 U.S. 579, 597 (1993).

The party seeking to introduce the evidence has the burden of

establishing both its reliability and its relevance. See id. at

593 n.10; see also Fed. R. Evid. 702, advisory committee's note.

As noted above, the district court rejected each theory

that Dr. Butler put forward to establish specific causation. We

now focus on the two theories that the Milwards press on appeal:

Dr. Butler's relative risk conclusion and her differential

diagnosis.2

Relative Risk

The district court rejected Dr. Butler's relative risk

testimony because she had expressly disavowed her intent, and

minimized her ability, to analyze conflicting epidemiological

studies. The district court reasoned that without such analysis,

2 The Milwards broadly allege that the district court applied
the wrong legal standard when evaluating Dr. Butler's fitness to
serve as an expert witness. They note that the court "held that
Dr. Butler is unqualified because she cannot 'evaluate the relevant
studies' with the 'rigor' of an epidemiologist." This argument
misconstrues the district court's action. The court did not, in
a vacuum, conclude that Dr. Butler was unqualified to provide
expert testimony in this case because she was not an
epidemiologist. Instead, the court stated that since Dr. Butler
was unwilling to provide testimony respecting the epidemiological
literature in the context of the "relative risk" approach, the
Milwards could not rely on that method to prove specific causation.
While we provide more detail about that conclusion below, it
suffices here to say that the district court did not err as the
Milwards allege.

- 8 -
it was impossible to ensure that the studies she cited were

actually based on a reliable methodology. The Milwards challenge

this decision in three ways.

First, they assert that in rejecting the testimony, the

district court relied on an incorrect premise: that conflicting

epidemiological studies existed. They note that there were studies

establishing an increased risk of APL after a certain level of

exposure, such as 8 ppm-years. See, e.g., Deborah C. Glass et

al., The Health Watch Case -- Control Study of Leukemia and

Benzene, 1076 Ann. N.Y. Acad. Sci. 80, 85 (2006); Richard B. Hayes

et al., Benzene and Lymphohematopoietic Malignancies in Humans, 40

Am. J. Indus. Med. 117, 120. The Milwards also acknowledge that

other studies found no increased risk of leukemia with exposure at

any level less than 40 ppm-years. See, e.g., Robert A. Rinsky et

al., Benzene and Leukemia: An Epidemiologic Risk Assessment, 316

New England J. Med. 1044 (1987). They argue, however, that since

the Rinsky study did not affirmatively find the absence of a

relationship, the studies were not actually in conflict.

While it is certainly true that, at least in some cases,

the "absence of evidence" is not the same as "evidence of absence,"

it is not similarly true that the studies must present

diametrically opposing conclusions to be in tension with one

another. Here, a number of studies have been identified that show

a correlation between APL and benzene exposure at a specific level,

- 9 -
while other studies do not show that correlation. In order to

establish specific causation by the relative risk method, Dr.

Butler was required to choose a study, or studies, to serve as a

baseline to which she could then compare Brian Milward's case.

There can be no serious question that choosing a study that showed

a correlation above a specific level (e.g., the 8 ppm-years in the

Glass study), rather than one that did not exhibit any such

correlation (e.g., the 40 ppm-years in the Rinsky study), yields

a vastly different comparison. The district court did not clearly

err in finding that the studies were sufficiently distinct from

one another such that utilizing one, rather than another, would

necessarily lead to different testimony.

The Milwards next argue, albeit summarily, that Dr.

Butler did not actually disavow her willingness to consider the

divergent studies. Instead, they allege that the district court

took her statements out of context.

We make quick work of this argument given the clarity of

the record. Dr. Butler anchored her testimony to her no-safe

threshold hypothesis, a theory that did not turn on the validity

of any of the epidemiological studies. Indeed, given that she

acknowledged that she based that theory on "the biology, the

pathophysiology, [and] what the substance does to the person and

the disease process," it was consistent for her to then state that

she had neither the need nor the intent to compare the competing

- 10 -
epidemiological literature.3 There was no error in the district

court's decision to give her statements their plain meaning.

Finally, the Milwards argue that even if the district

court did not err in these respects, Dr. Butler's testimony was

nevertheless still based on reliable evidence, and it was therefore

admissible. In support of this contention, the Milwards defend

the studies that Dr. Butler invoked in her testimony. They also

cite Schultz v. Akzo Novel Paints, LLC, 721 F.3d 426 (7th Cir.

2013), which they maintain is closely analogous to this case.

Generally, where an expert's medical opinion is grounded

exclusively on scientific literature, a district court acts within

its discretion to require the expert to explain why she relied on

the studies that she did and, similarly, why she disregarded other,

incompatible research. See, e.g., Kuhn v. Wyeth, Inc., 686 F.3d

618, 623-24 & 633 (8th Cir. 2012) (permitting testimony where the

expert witness relied on methodologically reliable studies and

provided an explanation for why those studies were chosen); Norris

v. Baxter Healthcare Corp., 397 F.3d 878, 886 (10th Cir. 2005)

(noting in the context of a general causation finding that the

expert witness's inability to address contrary views made the

3 Likewise, in discussing the statistical significance of
the reports, Dr. Butler seemingly minimized her ability to analyze
the studies when she said "and I'm not an epidemiologist if you're
going to go there. I'm just saying that to me that's fairly --
that's fairly significant."

- 11 -
opinion unreliable). It is self-evident that, when an expert

engages in a relative risk analysis in the manner that Dr. Butler

did here, the district court is on firm ground in requiring such

an explanation, since the validity of the approach depends on the

reliability of the studies chosen. See 3 Mod. Sci. Evidence §

23:27 (2014-2015 Ed.) (discussing the use of the relative risk

approach in establishing specific causation). That is, if the

expert is comparing the plaintiff's condition to a study, and the

study is based on an unreliable methodology, then the comparison

itself is futile.

Schultz, the case on which the Milwards rely, is

consistent with this view. In that case, the Seventh Circuit

reversed a district court's decision to exclude specific causation

expert testimony about an individual's exposure level to benzene.

721 F.3d at 428. The Seventh Circuit found that the testimony was

reliable because the expert "focused specifically on the amount of

benzene to which [the plaintiff] had been exposed and related this

amount to the scientific literature." Id. at 432. Importantly,

the expert in Schultz did not simply point to favorable studies

showing an increased risk of leukemia at low levels of exposure.

Instead, the expert in that case explained why he believed that a

conflicting study was unreliable and why, based on his knowledge

of the literature, he chose to rely on the studies that he did.

Id. at 432-33.

- 12 -
 Here, the relevant studies were not only in tension

with one another, but expressly cast each other into doubt. See,

e.g., EPA Office of Research and Development, Carcinogenic Effects

of Benzene: An Update, at 14 (April 1998). Given that, the

district court reasonably ruled that there needed to be some

indication of why Dr. Butler utilized the studies that she did.

Indeed, her complete unwillingness to engage with the conflicting

studies (irrespective of whether she was able to or not) made it

impossible for the district court to ensure that her opinion was

actually based on scientifically reliable evidence and,

correspondingly, that it comported with Rule 702. Not only does

this render this case readily distinguishable from Schultz, but it

also justifies the district court's decision.4

Differential Diagnosis

The district court also rejected Dr. Butler's

"differential diagnosis." Although the judge did not question Dr.

4 We also note that the Milwards' position yields a further
problem. Absent Dr. Butler's testimony weighing the studies, the
only support for their reliability is the fact that they were peer-
reviewed, published works. As we have noted though, "an article
does not reach the dignity of a 'reliable authority' merely because
some editor, even a most reputable one, sees fit to circulate it
. . . [and] [m]ere publication cannot make them automatically
reliable authority." Meschino v. N. Am. Drager, Inc., 841 F.2d
429, 434 (1st Cir. 1988). Given the need for some evidence
establishing the reliability of the studies invoked, the court
likewise did not err in refusing to take judicial notice of their
reliability.

- 13 -
Butler's decision to "rule out" obesity and smoking as causes of

Brian Milward's APL, the court was concerned about the utility of

the approach given the high percentage of APL cases that are

idiopathic (according to the record, roughly 70-80% of all APL

diagnoses). The judge also stated that Dr. Butler's reasoning was

circular; she "ruled out" an idiopathic APL by "ruling in" benzene

as a cause, but she had failed to provide a scientifically reliable

method of "ruling in" benzene in the first instance. The Milwards

contend that in making this decision, the district court ignored

our case law that has blessed an expert's use of a differential

diagnosis to establish causation.

Even if the Milwards' scanty argument in their opening

brief were sufficiently developed as to avoid a waiver finding,

see United States v. Oladosu, 744 F.3d 36, 39 (1st Cir. 2014)

("[b]ecause the argument is underdeveloped, it is waived"), we

nonetheless see no abuse of discretion in the district court's

decision. The Milwards are certainly correct that a "differential

diagnosis" can be a "reliable method of medical diagnosis."

Milward, 639 F.3d at 18; see also Granfield v. CSX Transp., Inc.,

597 F.3d 474, 486 (1st Cir. 2010). But, they still must show that

the steps taken as part of that analysis -- the "ruling out" and

the "ruling in" of causes -- were accomplished utilizing

scientifically valid methods. See Ruggiero v. Warner-Lambert Co.,

424 F.3d 249, 254 (2d Cir. 2005).

- 14 -
 Since Dr. Butler was only able to "rule out" an

idiopathic APL because she had "ruled in" benzene as a cause, the

validity of her differential diagnosis turns on the reliability of

that latter conclusion. See Ruggiero, 424 F.3d at 254 (noting

that an expert must use reliable scientific methods to "rule in"

causes); see also Best v. Lowe's Home Ctrs., Inc., 563 F.3d 171,

179 (6th Cir. 2009); Glastetter v. Novartis Pharm. Corp., 252 F.3d

986, 989 (8th Cir. 2001). Indeed, the reliability of that decision

is particularly critical here given the extensive number of APL

cases that are idiopathic. Under such circumstances, eliminating

a number of potential causes -- without properly and explicitly

"ruling in" a cause -- is simply "of little assistance."

Restatement (Third) of Torts; Phys. & Emot. Harm § 28, cmt.

c(4)(2010).

Dr. Butler appears to have "ruled in" benzene exposure

solely by relying on her two other theories. But, as explained

above, the district court found both of these theories to be

unreliable. As we agree with the district court's conclusion

regarding the relative risk methodology, and since the Milwards

have not challenged the district court's no-safe threshold

determination, they have failed to show how Dr. Butler could have

reliably utilized either method to "rule in" benzene exposure.

Nor, we note, have they pointed to other evidence in the record

that Dr. Butler could have conceivably used to "rule in" benzene.

- 15 -
 Given that the record does not contain a scientifically

reliable basis to "rule in" benzene, Dr. Butler needed some other

method to "rule out" an idiopathic diagnosis. She did not provide

one. As such, the district court acted within its discretion to

conclude that the extraordinary number of idiopathic APL cases,

coupled with the lack of a reliable means to rule out an idiopathic

diagnosis here, muted Dr. Butler's ability to reliably apply this

methodology.5

III.

Once the district court excluded Dr. Butler's testimony,

it then correctly granted Rust-Oleum's motion for summary

judgment. As is well-established under Massachusetts law, "expert

testimony is required to establish medical causation." Reckis v.

Johnson & Johnson, 28 N.E.3d 445, 461 (Mass. 2015). This applies

to both general and specific causation. Id. at 461 n.33. Without

any other medical expert evidence in the record probative on

specific causation, judgment as a matter of law was necessarily

required. Fed. R. Civ. P. 56.

5In their brief, the Milwards also argue that Dr. Butler's
position on specific causation is consistent with the latency
period in Brian Milward's case. The district court did not rest
its decision on that proposition (instead, it just noted a concern
about the issue), and we therefore need not reach the argument.

- 16 -
 Accordingly, we affirm the district court's decision to

exclude Dr. Butler's testimony and its concomitant grant of summary

judgment in favor of Rust-Oleum.

--Dissenting Opinion Follows--

- 17 -
 THOMPSON, Circuit Judge, dissenting.

Setting the Stage

Dr. Butler has quite the CV. A graduate of Wellesley

College, she has a medical degree from Howard University and a

masters of public health from Columbia University. Specializing

in occupational medicine, she is board-certified in preventive

medicine and general public health (by the American Board of

Preventive Medicine) and in anatomic pathology, clinical

pathology, and hematology (by the American Board of Pathology).6

This means (according to the American Board of Preventive Medicine)

that she has "core competencies" in, among other things,

"epidemiology" and "research into causes of disease and injury in

population groups."7 She has a pretty impressive job too, working

full time as a physician at a VA medical center that deals with

veterans ravaged by diseases after being exposed to toxins during

their service. Figuring out the causes of chronic illnesses in

patients exposed to toxic substances is what she does day in and

day out. All told, she has (in the district judge's words) over

a decade's worth of experience "as a practicing diagnostic

6
Pathology is a medical specialty focusing on the nature and
causes of diseases. See Stedman's Medical Dictionary 1332 (27th
ed. 2000) ("Stedman's," from here on). And hematology is the study
of blood-related diseases. See id. at 796.

7
Epidemiology is the study of the incidence, distribution,
and control of disease in a population. See id. at 604.

- 18 -
hematopathologist and as a consultant on occupationally-related

malignancies."

As the Milwards' specific-causation expert, Dr. Butler

testified by report, deposition, and affidavit that — based on her

review of the scientific evidence — there is no "safe" level of

benzene exposure.8 In other words, every benzene exposure

increases a person's risk of leukemia. But, she added, given our

different genetic makeups, what might be a safe exposure level for

some could be a lethal one for others. Anyway, using two accepted

causation methodologies — "relative risk" and "differential

diagnosis" — and zeroing in on Brian's benzene-exposure level (set

by Dr. Stewart at 25.6 ppm-years) Dr. Butler concluded that Brian's

"excessive" exposure to benzene caused his leukemia.9

8 Remember — the Milwards had to show that benzene exposure
can cause leukemia (general causation) and that Brian's exposure
was a substantial factor contributing to his leukemia (specific
causation). A different district judge excluded the Milwards'
general-causation expert as unreliable under Rule 702. Noting
(among other things) that the judge had taken "sides on questions
that are currently the focus of extensive scientific research and
debate — and on which reasonable scientists can clearly disagree"
— we concluded that the exclusion edict constituted an abuse of
discretion. See Milward, 639 F.3d at 22, 26.

On a different note, because there are two Milwards — Brian
and Linda — it makes sense to use a first name where necessary to
avoid confusion. Obviously I intend no disrespect.

9 As my friends in the majority note, the Milwards hired Dr.
Stewart (an industrial hygienist) to assess Brian's benzene
exposures.

- 19 -
 A quick word about how she applied these methodologies.

Starting with relative risk, Dr. Butler said that even

if there were some threshold level of benzene exposure needed to

cause leukemia, that threshold was exceeded here — and by a

considerable amount. With Brian's 25.6 ppm-years exposure level

firmly in mind, she pointed to a peer-reviewed epidemiology study

finding that workers exposed to benzene at or above 8 ppm-years

were 7 times more likely than controls to develop leukemia. And

she did not stop there. Rather, she went on to spotlight other

studies of the same caliber showing a statistically significant

increased risk of leukemia among workers cumulatively exposed to

benzene at levels below Brian's 25.6 ppm-years.10 In a deposition

she said that she is neither an epidemiologist nor a researcher.

She also agreed that some studies found no relationship between

benzene exposure and leukemia. Asked by defense counsel if she

"intend[ed] in this case to weigh the different epidemiological

studies" and comment on "which ones we should rely on and which

ones we should discount," she replied, "No" — and then added:

I'm relying on what I know about the biology,
the pathophysiology, what the substance does
to the person and the disease process. Now,

10See Deborah R. Glass et al., The Health Watch Case — Control
Study of Leukemia and Benzene: The Story So Far, 1076 Ann. N.Y.
Acad. Sci. 80 (2006); Dusica Lazarov et al., Acute Myeloid Leukemia
and Exposure to Organic Solvents: A Case-Control Study, 16 Eur.
J. of Epidemiology 295 (2000); Richard B. Hayes et al., Benzene
and the Dose-Related Incidence of Hematologic Neoplasms in China,
89 J. of the Nat'l Cancer Inst. 1065 (1997).

- 20 -
 if there are studies that support it then
that's even better, but without the studies
based on what I know there is a very — it's
more likely than not that benzene contributes
to the development of [the type of leukemia
Brian suffers from].11

And she later said that "one doesn't just rely on literature" in

formulating a specific-causation opinion.

Turning, then, to differential diagnosis (aptly

described by the majority as "essentially a process of

elimination"), Dr. Butler "ruled out" possible causes of Brian's

leukemia, like smoking and obesity, leaving only benzene. She

talked about "'idiopathic' leukemia" too — "idiopathic" being

another way of saying medical professionals do not know why a given

person has the disease. "[E]very case of leukemia has some

cause[]," she explained, and only "[t]hose cases with unidentified

causes" get hit with the "'idiopathic'" tag. But given her

conclusion that Brian's "benzene exposures were a substantial

factor causing his [leukemia]," she could "also 'rule[] out' that

his [leukemia] was . . . 'idiopathic.'"

The district judge, however, would have none of Dr.

Butler's talk about benzene being the specific cause of Brian's

leukemia. Given her concession that she is "'not an

epidemiologist'" and "'not a researcher,'" and given her

11Pathophysiology is the study of the functional changes that
accompany a particular disease. See Stedman's 1333.

- 21 -
"professed inability to engage with conflicting epidemiological

literature" (these are quotes from the judge's rescript), the judge

excluded her relative-risk analysis as unreliable under Rule 702.12

That meant that Dr. Butler's differential-diagnosis analysis —

through which she "'ruled out' an idiopathic origin of [Brian's]

leukemia by 'ruling in' benzene" (these too are quotes from the

judge's order) — was unreliable too (because she is, the judge

concluded, not qualified to say whether benzene exposure at Brian's

level could have caused his leukemia). And with the Milwards'

specific-causation expert out of the picture, all that was left

for the judge to do was enter summary judgment against them — which

the judge did.

Fast-forward to the present, with the majority spying no

abused discretion here because Dr. Butler's "complete

unwillingness to engage with the conflicting studies (irrespective

of whether she was able to or not) made it impossible for the

[judge] to ensure that her opinion was actually based on

scientifically reliable evidence" as required by Rule 702. Call

me unpersuaded. As I see things, the complaints about Dr. Butler's

12The "conflicting" study that everyone focuses on is Robert
A. Rinsky et al., Benzene and Leukemia: An Epidemiologic Risk
Assessment, 316 New Eng. J. of Med. 1044 (1987), which found no
increased risk of leukemia in workers exposed to less than 40 ppm-
years of benzene.

- 22 -
specific-causation opinion go to weight, not admissibility — as I

now explain.13

My Take on the Matter

(a)
The Standard of Review Explained

Abuse-of-discretion review is "respectful," certainly.

Corp. Techs. v. Harnett, 731 F.3d 6, 10 (1st Cir. 2013). But

"respectful" does not mean we must throw up our hands and simply

affirm every discretionary call. See, e.g., Negron-Almeda v.

Santiago, 528 F.3d 15, 21 (1st Cir. 2008). Review under this

standard does involve review, after all. See, e.g., Dopp v.

Pritzker, 38 F.3d 1239, 1253 (1st Cir. 1994). And we will not

hesitate to find abuse where, for example, the district judge based

his decision on clearly erroneous facts, made a serious legal

error, or suffered a significant lapse of judgment, see, e.g.,

Cent. Pension Fund of the Int'l Union of Operating Eng'rs &

Participating Emp'rs v. Ray Haluch Gravel Co., 745 F.3d 1, 5 (1st

Cir. 2014); Riva v. Ficco, 615 F.3d 35, 43 (1st Cir. 2010); Ruiz-

Troche v. Pepsi Cola of P.R. Bottling Co., 161 F.3d 77, 83 (1st

Cir. 1998) — a point made each time we have reversed the exclusion

13 Because the majority jettisons the case by upholding the
judge's decision to exclude Dr. Butler's testimony, I (obviously)
focus my energy on that issue. So, like the majority, I make no
comment on Rust-Oleum's other arguments — i.e., that the judge
should have excluded Dr. Stewart's testimony and that the Milwards
cannot show that the failure to provide certain warnings about
benzene proximately caused Brian's injuries.

- 23 -
of expert testimony, see, e.g., Milward, 639 F.3d at 13-14, 23-

25; Ruiz-Troche, 161 F.3d at 79, 83-86.

(b)
A Short Primer on Expert Opinion

Rule 702 governs the admission of expert-opinion

testimony, with the offering party required to show that such

testimony is relevant and reliable. See, e.g., Kumho Tire Co. v.

Carmichael, 526 U.S. 137, 149 (1999) (relying on Daubert, 509 U.S.

at 592); Ruiz-Troche, 161 F.3d at 80 (same). Expert-opinion

testimony is relevant if it will assist the factfinder in

understanding and deciding a fact. See, e.g., Daubert, 509 U.S.

at 592. And it is reliable if it has "a reliable basis in the

knowledge and experience of [the pertinent] discipline."14 Id.

Basically then, district judges are supposed to weed out

nonsense opinions by junk scientists. But in doing so, they must

keep a bunch of things in mind — including the following:

 The rule on expert-opinion testimony is notably "liberal,"

with the evidence considered presumptively admissible. See

4 Jack B. Weinstein & Margaret A. Berger, Weinstein's

14Because everyone focuses on whether Dr. Butler's testimony
is reliable, I will do likewise.

- 24 -
 Federal Evidence § 702.02[1], at 702-5 (Joseph M. McLaughlin

ed., 2d ed. 2013) ("Weinstein's," to save some keystrokes).

 Proponents of expert testimony must show that the proposed

witness is able — through her education, training, or

experience — to offer a meaningful opinion on the issue in

play. Id. § 702.04[1][c], at 702-57.

 An expert can rely, then, on "clinical instinct" — i.e.,

"what experience adds to scientific knowledge and training"

— which is a well-known and accepted part of today's medical

practice. Mueller v. Auker, 700 F.3d 1180, 1191 (9th Cir.

2012) (quoted approvingly in Weinstein's § 702.05[2][c], at

702-103 n.46).

 Judges abuse their discretion if they "exclude testimony

that would otherwise" help the factfinder "understand a

fact in issue, simply because the expert does not have the

specialization" that the judges think "most appropriate."

Pagés-Ramírez v. Ramírez-González, 605 F.3d 109, 114 (1st

Cir. 2010) (internal quotation marks omitted); see also

Ralston v. Smith & Nephew Richards, Inc., 275 F.3d 965, 970

(10th Cir. 2001) (explaining that so long as the expert

keeps "within the reasonable confines of [her] subject area,

. . . a lack of specialization does not affect the

- 25 -
 admissibility of [her] opinion, but only its weight"

(internal quotation marks omitted)).

 Also, an expert need not have epidemiological studies at

the ready to get her opinion in. See Milward, 639 F.3d at

24 (holding that "[e]pidemiological studies are not per se

required as a condition of admissibility regardless of

context"); see also Daubert, 509 U.S. at 593 (explaining

that "[p]ublication . . . is not a sine qua non of

admissibility").

 And an opinion, by the way, does not have to conclusively

prove causation to be admissible. "[M]edical knowledge,"

we can all agree, "is often uncertain. The human body is

complex, etiology is often uncertain, and ethical concerns

often prevent double-blind studies calculated to establish

statistical proof." United States v. Sandoval-Mendoza, 472

F.3d 645, 655 (9th Cir. 2006). But that "does not preclude

the introduction of medical expert opinion testimony when

medical knowledge permits the assertion of a reasonable

opinion." Id. (internal quotation marks omitted).

 Critically too, deciding "which of several competing

scientific theories has the best provenance" is none of the

judges' business — which is just another way of saying that

- 26 -
 judges must focus on the expert opinion's admissibility,

not its correctness. Ruiz-Troche, 161 F.3d at 85.

 Here's a biggie: That the parties' experts disagree (they

often do, unsurprisingly) goes to weight, not

admissibility. See, e.g., Feliciano-Hill v. Principi, 439

F.3d 18, 25 (1st Cir. 2006); see also Weinstein's

§ 702.05[3], at 702-112 n.58 (collecting a cornucopia of

additional cases).

 Here's another biggie: An expert's backers "do not

necessarily have the burden" of disproving a study

championed by the other side — that is what a case the

majority relies on says. See Kuhn, 686 F.3d at 626. Again,

the proponents must "show that [their expert] arrived at

[her] contrary opinion in a scientifically sound and

methodological fashion." Id. And if they do, "the question

becomes one for the jury to decide." Id.

(c)
The Instances of Abused Discretion

The ruling my colleagues affirm — that Dr. Butler "is

'not an epidemiologist' and 'not a researcher'" who "professed" an

"inability to engage with conflicting epidemiological literature"

and "thus" is "unqualified" to say whether Brian's level of benzene

exposure could cause his leukemia (quotes lifted from the district

- 27 -
judge's order) — is filled with errors. And these errors rise to

the level of an abuse of discretion.

Take the district judge's fixation on her saying that

she was "not an epidemiologist" and "not a researcher." Time and

again we have said that one "need not be a specialist in a

particular medical discipline to render expert testimony relating

to that discipline." Gaydar v. Sociedad Instituto Gineco–

Quirurgico y Planificacion Familiar, 345 F.3d 15, 24 (1st Cir.

2003); see also Pagés-Ramírez, 605 F.3d at 116-17. And not only

have we talked the talk, but we have walked the walk — reversing

as an abuse of discretion expert-exclusion rulings premised on an

expert's missing the type of specialization the judges think

necessary, even though the testimony would have helped the jury

understand a disputed issue. See, e.g., Pagés-Ramírez, 605 F.3d

at 116-17. And given her training and experience — don't forget,

(a) her board certification in preventive medicine shows she has

competency in epidemiology and research into causes of disease,

and (b) she analyzes specific-causation issues as a routine part

of her job — the judge's ruling faulting Dr. Butler for not being

able to "evaluate the relevant studies" with the "rigor" of an

epidemiologist fits that category of error. The majority tries to

downplay the district judge's comments about her not being an

epidemiologist by playing up how concerned he was with her

"unwilling[ness]" to analyze the "conflicting" literature. But

- 28 -
the fact remains that the judge did add her non-epidemiologist

status to his decisional mix, which (for the reasons just

discussed) is an abuse of discretion, plain and simple.

Now, as for the judge's belief — shared by the majority

— that Dr. Butler "professed [an] inability to engage with the

conflicting epidemiological literature," there are problems

galore.

For starters, I espy no conflict. To repeat a point I

made a few paragraphs ago: The studies Dr. Butler relied on show

that benzene-exposure levels below the 25.6 ppm-years endured by

Brian can cause leukemia. The Rinsky study — the supposedly

"conflicting" study — also shows that benzene exposure at certain

levels can cause leukemia, though the authors found no increased

risk of leukemia among workers exposed to less than 40 ppm-years

of benzene. According to the district judge, because the Rinsky

study did not find any increased risk of leukemia at lower exposure

levels, there is a "conflict" and "debate within the

epidemiological literature" that can only be put to rest by someone

with epidemiologist credentials. Not only did the judge get the

epidemiologist-credentials part wrong (as I just noted); he got

the "conflict" part wrong too. For a true conflict to exist, the

Rinsky study would have to show that benzene-exposure levels of

- 29 -
25.6 ppm-years or lower cannot cause leukemia. And the Rinsky

study does no such thing.

Anyhow, even assuming there is a conflict, the judge

still erred in two important ways. For one thing, despite what

the judge said, Dr. Butler hardly copped to being unable to engage

with the literature. By my lights, the judge could only say what

he said by misreading her deposition. Questioned (recall) by

defense counsel about whether she "intended in this case to weigh

the different epidemiological studies and offer an opinion as to

which ones we should rely on and which ones we should discount,"

Dr. Butler said, "No." That is because, she stressed, (a) one

need not rely just "on literature" and (b) her experience with

"biology," "pathophysiology," and "the disease process" provided

the specialized knowledge to support her specific-causation

testimony. Statements (a) and (b) square with our caselaw. See,

e.g., Milward, 639 F.3d at 24 (emphasizing how "[e]pidemiological

studies are not per se required as a condition of admissibility").

And just as importantly, nothing she said there intimated even a

possible whisper of a hint of a suggestion that she could not take

on the relevant literature. Put differently, she did not say that

she lacks the know-how to assess Rust-Oleum's preferred studies —

only that she did not need to in formulating her expert opinion.

- 30 -
The word "intend" — I hope we can all agree — does not imply

"can't."

For another thing, despite what the judge indicated,

neither Dr. Butler nor the Milwards had any burden to explain why

the Rinsky study is wrong. Think back to the primer: The

proponents of expert testimony, I noted, are not reflexively

obliged to "discredit" a study pushed by their opponents. Kuhn —

a case highlighted by the majority — says as much. Sure, the

plaintiffs' expert there tried to poke holes in a study relied on

by the defendants. And, deeming the criticisms insubstantial, the

judge excluded the expert from testifying. Significantly for

present purposes, though, the circuit court wrote that the expert

did not have to debunk the study; he only had to show that he

reached his conclusion via a sound methodology.15 See Kuhn, 686

F.3d at 626. Dr. Butler did that in spades, using two recognized

techniques for identifying causes (relative risk and differential

diagnosis) and relying in part on studies that (as best I can tell)

neither the district judge nor the majority has any problems with.

15So instead of supporting the majority's position that Dr.
Butler had to explain why she disagreed with "incompatible"
studies, Kuhn rejects that position. And Norris — another case
cited by the majority — is not a difference maker for the majority
either. The court there upheld the exclusion of expert testimony
because the experts did not confront the reality that their
opinions were "flatly contrary to all of the available
epidemiological evidence," see 397 F.3d at 885-86 — which is worlds
apart from our case.

- 31 -
And having met her burden, a jury should get to decide which

studies to believe (hers or Rust-Oleum's), if any, see id. — just

like a jury would get to do if faced with dueling experts (instead

of dueling studies), see Feliciano-Hill, 439 F.3d at 25.

Summing Up

Because, as discussed, the judge made serious judgment

errors in excluding Dr. Butler's expert testimony — a ruling (in

my view) inconsistent with the "liberal thrust of the Federal Rules

and their general approach of relaxing the traditional barriers to

'opinion' testimony," see Daubert, 509 U.S. at 588 — I would

reverse his ruling as an abuse of discretion. And because the

majority, though conscientious, has decided otherwise, I

respectfully — but unequivocally — dissent.

- 32 -