BNSF Railway Company (Individually and as Successor-In-Interest to the Burlington Northern, Inc., Burlington Northern & Santa Fe Railway Company and Atchison Topeka and Santa Fe Railway Company) v. Leonard A. Baca

COURT OF APPEALS
SECOND DISTRICT OF TEXAS
FORT WORTH

NO. 02-17-00168-CV

BNSF RAILWAY COMPANY APPELLANT
(INDIVIDUALLY AND AS
SUCCESSOR-IN-INTEREST TO
THE BURLINGTON NORTHERN,
INC., BURLINGTON NORTHERN &
SANTA FE RAILWAY COMPANY
AND ATCHISON TOPEKA AND
SANTA FE RAILWAY COMPANY)

V.

LEONARD A. BACA APPELLEE

----------

FROM THE 48TH DISTRICT COURT OF TARRANT COUNTY
TRIAL COURT NO. 048-267301-13

----------

MEMORANDUM OPINION1

----------

1
See Tex. R. App. P. 47.4.
 In this permissive interlocutory appeal involving a claim under the Federal

Employers Liability Act (FELA), Appellee Leonard A. Baca alleges that while

working for Appellant BNSF’s predecessor in interest, he was exposed to

asbestos, causing him to develop asbestosis. See 45 U.S.C.A. §§ 51–60 (West

2007). Baca retained as an expert Dr. Alvin Schonfeld, a pulmonologist, who

provided a report in which he concluded Baca’s asbestosis was causally related

to his exposure to asbestos during his employment. BNSF moved to exclude Dr.

Schonfeld’s causation opinion as inadmissible because it was unreliable under

well-established caselaw.2 The trial court denied the motion but also granted

permission in its order for BNSF to immediately appeal, finding that the order

involved a controlling question of law as to which there is a substantial ground for

difference of opinion and an immediate appeal from the order would materially

advance the ultimate termination of this litigation. See Tex. Civ. Prac. & Rem.

Code § 51.014(d) (West 2017); Tex. R. Civ. P. 168.

BNSF filed a petition for permissive appeal, which we granted. See BNSF

Ry. Co. v. Baca, No. 02-17-00168-CV, 2017 WL 2570826, at *1 (Tex. App.—Fort

Worth June 14, 2017, no pet.) (mem. op. & order). The controlling question of

2
Included among the cases BNSF cited were Daubert v. Merrell Dow
Pharms., Inc., 509 U.S. 579 (1993); Borg-Warner Corp. v. Flores, 232 S.W.3d
765 (Tex. 2007); Merrell Dow Pharms., Inc. v. Havner, 953 S.W.2d 706 (Tex.
1997); and E.I. DuPont de Nemours & Co. v. Robinson, 923 S.W.2d 549 (Tex.
1995).

2
law the trial court identified and ruled upon in its order, and the sole issue in this

appeal, is

whether the Federal Employers Liability Act’s (45 U.S.C. §§ 51-60)
lower causation standard—i.e., whether a railroad’s negligence
played any part, even the slightest, in bringing about the injury—
makes inapplicable the expert admissibility standards expressed in
cases like E.I. DuPont de Nemours & Co. v. Robinson, 923 S.W.2d
549 (Tex. 1995) and Merrell Dow Pharmaceuticals, Inc. v. Havner,
953 S.W.2d 706 (Tex. 1997).

We answer no. Because the trial court denied BNSF’s motion to exclude based

on the opposite conclusion, we reverse the trial court’s order and remand this

case for further proceedings.

I. BACKGROUND FACTS3

FELA makes any railroad engaged in interstate commerce liable in

damages for an injury to or death of an employee sustained while employed by

the railroad if the injury or death resulted in whole or in part from the negligence

of the railroad’s employees or by reason of any defect or insufficiency in its

equipment due to its negligence. See 45 U.S.C.A. § 51; Union Pac. R.R. v.

Williams, 85 S.W.3d 162, 165 (Tex. 2002). To prevail on a FELA claim, a plaintiff

must establish the traditional common-law elements of negligence: duty, breach,

3
Baca has filed a motion to dismiss in which he argues in part that this
court lacks jurisdiction over this permissive interlocutory appeal because the
issue BNSF has presented for review in its appellant’s brief materially differs from
the issue it presented in its petition for permissive appeal and which the trial court
granted it permission to appeal. Accordingly, we set forth in detail the factual and
procedural background of this appeal, that we may appropriately address Baca’s
jurisdictional argument.

3
foreseeability, and cause-in-fact. See Tufariello v. Long Island R.R., 458 F.3d

80, 87 (2d. Cir. 2006); Abraham v. Union Pac. R.R., 233 S.W.3d 13, 17 (Tex.

App.—Houston [14th Dist.] 2007, pet. denied). But a plaintiff’s burden to

establish a railroad’s liability under FELA is lighter than it would be in an ordinary

negligence case because FELA prescribes a relaxed standard of causation. See

Lynch v. Ne. Reg’l Commuter R.R., 700 F.3d 906, 911 (7th Cir. 2012); Abraham,

233 S.W.3d at 17. Under that relaxed causation standard, a plaintiff is entitled to

prevail on a FELA claim if the railroad’s negligence played any part, even the

slightest, in producing the injury or death for which damages are sought. See

CSX Transp., Inc. v. McBride, 564 U.S. 685, 688, 705 (2011); BNSF Ry. Co. v.

Nichols, 379 S.W.3d 378, 382 (Tex. App.—Fort Worth 2012, pet. denied). This

relaxed causation standard is often referred to as a “featherweight” standard.

See Maritime Overseas Corp. v. Ellis, 971 S.W.2d 402, 406 (Tex. 1998).

A. BACA’S FELA CLAIM

Baca sued BNSF alleging a claim under FELA.4 His theory of liability is

straightforward: he claims that his exposure to toxic substances and dusts,

including asbestos and asbestos-containing products and materials, while in the

course of his employment with BNSF caused him to develop asbestosis.

Because whether a causal connection exists between a person’s exposure to a

chemical and a disease from which he suffers is outside the common knowledge

4
Baca also alleged claims for negligence per se and negligent infliction of
emotional distress. Those claims are not at issue in this appeal.

4
and experience of lay persons, expert testimony is generally required to prove

such a causal connection. See Guevara v. Ferrer, 247 S.W.3d 662, 665 (Tex.

2007); Abraham, 233 S.W.3d at 18; Pilgrim’s Pride Corp. v. Smoak, 134 S.W.3d

880, 893–94 (Tex. App.—Texarkana 2004, pet. denied). Baca retained Dr.

Schonfeld to do just that.

B. DR. SCHONFELD’S REPORT

Dr. Schonfeld is a pulmonologist whose qualifications to make an opinion

as to whether Baca’s exposure to asbestos during his employment with BNSF

caused him to develop asbestosis are not at issue. In support of his claim, Baca

produced a report from Dr. Schonfeld, in which Dr. Schonfeld summarized his

opinion concerning that question.

Dr. Schonfeld’s report states that he interviewed and examined Baca. The

report sets forth the history of Baca’s exposure to asbestos, which was relayed to

Dr. Schonfeld by Baca. According to the report, Baca’s working career has

almost exclusively been as a railroad worker. Baca was laid off from the railroad

in approximately 1965, so for a few months he worked for the Arizona Highway

Department, where he used asbestos gloves when removing hot objects from a

stove. But other than those few months in 1965, the report says, Baca worked

for the railroad from 1964 to 1996.

Baca told Dr. Schonfeld that on an intermittent basis for about six months

in 1964, he helped demolish a roundhouse in Winslow, Arizona, and he would

sweep up asbestos without the use of a mask or respirator. Baca also told Dr.

5
Schonfeld that he rode on diesel engines that had asbestos in the ceiling and

that the asbestos dust would fall on him. Dr. Schonfeld further stated that Baca

remembered asbestos was in the brake hoses and that for about six months, he

fired the asbestos-clad steam generator on the diesel passenger units and

performed repairs on the steam generator on an as-needed basis. Baca further

told Dr. Schonfeld that he worked in refrigerator cars that were lined with

asbestos and that he would clean them out and sweep up asbestos. Finally,

Baca told Dr. Schonfeld that he was present while other employees were

changing brake shoes and that he was in other places where employees were

working on and repairing the railroad’s rolling stock.

Dr. Schonfeld noted that Baca had never smoked. He performed a

physical examination on Baca, which revealed that his lungs “were clear to

auscultation and percussion.” He also noted a report from Dr. Donald Breyer,

who on August 30, 2011, had done a B-reading of a chest x-ray that had been

performed on Baca on August 16, 2011.5 Dr. Schonfeld relayed Dr. Breyer’s

findings, stating Dr. Breyer concluded the x-ray showed irregular interstitial

infiltrates in both mid- and lower-lung zones having a “shape and size of s/s and

a profusion of 1/0” and that Dr. Breyer had not noted any pleural abnormalities.

Dr. Schonfeld also noted that Baca had undergone pulmonary function tests in

2012 and that the results were normal.

5
Baca also designated Dr. Breyer, a board-certified diagnostic radiologist
and a certified ILO “B” Reader, as an expert.

6
 Based upon all of this information, Dr. Schonfeld concluded as follows:

Given [Baca’s] history of significant exposures to asbestos in the
workplace and given an appropriate latency and given the
roentgenographic findings described above, I feel with a reasonable
degree of medical certainty that Mr. Baca is diagnosed as having
bilateral asbestosis. I feel with a reasonable degree of medical
certainty that this diagnosis is causally related to his workplace
exposures to asbestos as noted above.

BNSF subsequently filed a motion to exclude Dr. Schonfeld’s causation opinion.

II. PROCEDURAL BACKGROUND

A. BNSF’S MOTION TO EXCLUDE

In its motion to exclude, BNSF argued that Dr. Schonfeld’s causation

opinion was inadmissible because it was unreliable.6 See Daubert, 509 U.S. at

589 (stating Rule 702 of the Federal Rules of Evidence requires “that any and all

scientific testimony or evidence admitted is not only relevant, but reliable”);

Robinson, 923 S.W.2d at 550, 556–57 (holding Rule 702 of the Texas Rules of

Evidence requires the same). It advanced two grounds for why that opinion was

unreliable. First, BNSF argued Dr. Schonfeld’s causation opinion was not based

on a reliable evidentiary foundation. Second, it argued that Dr. Schonfeld’s

methodology was unreliable. In making these two arguments, BNSF

acknowledged FELA’s featherweight causation standard. But it asserted that

FELA, though relaxing the standard of causation relative to a common-law

negligence claim, does not similarly relax the procedural standards for

BNSF did not contend that Dr. Schonfeld’s opinion was inadmissible
6

because it was irrelevant.

7
determining the threshold issue of whether an expert’s causation opinion is

admissible. Thus, even though this is a FELA case, BNSF argued, Dr.

Schonfeld’s causation opinion must meet state procedural requirements for

reliability to be admissible, including the requirements of Daubert and Robinson.

See Kan. City S. Ry. Co. v. Oney, 380 S.W.3d 795, 800 (Tex. App.—Houston

[14th Dist.] 2012, no pet.) (“As a general matter, FELA cases adjudicated in state

courts are subject to state procedural rules, but the substantive law governing

them is federal.” (quoting St. Louis Sw. Ry. Co. v. Dickerson, 470 U.S. 409, 411

(1985))).

Having argued that the trial court was required to apply state procedural

standards of reliability to Dr. Schonfeld’s opinion, BNSF then discussed the

standards it believed apply in this case. Citing Austin v. Kerr-McGee Refining

Corp., 25 S.W.3d 280, 292 (Tex. App.—Texarkana 2000, no pet.), it noted that a

plaintiff in a toxic tort case must establish both general and specific causation.

BNSF focused on specific causation in particular, citing Havner, 953 S.W.2d at

714, for the proposition that “[s]pecific causation asks whether the substance at

issue caused a particular plaintiff’s injury.” It pointed to caselaw noting that

asbestosis appears to be a dose-related disease, meaning “the more one is

exposed, the more likely the disease is to occur, and the higher the exposure the

more severe the disease is likely to be.” Borg-Warner, 232 S.W.3d at 771

(quoting 3 David L. Faigman et al., Modern Scientific Evidence: The Law and

Science of Expert Testimony § 28:22, at 447 (2007)). And it also asserted that

8
medical science has established that mere background levels of exposure to

asbestos are not sufficient to cause disease; rather, pointing in part to Borg-

Warner, 232 S.W.3d at 771, BNSF stated, “[t]he accepted threshold exposure for

developing asbestosis is at least 4-5 fiber years (f/cc years), and more likely 25

to 100 fiber years.”

With the foregoing in view, BNSF turned to discuss Daubert and

Robinson’s reliability standards. It first contended Daubert and Robinson require

expert testimony to be based upon a reliable evidentiary foundation. And in the

context of an asbestosis case, BNSF argued, evidence of the amount—or

dose—of asbestos the plaintiff was exposed to is a necessary evidentiary

foundation for an expert to reliably opine that the plaintiff’s exposure to asbestos

caused him to develop asbestosis because without knowing the dose, an expert

has no reliable basis upon which to conclude that the plaintiff’s exposure to

asbestos met or exceeded the scientifically-accepted exposure threshold that is

necessary to cause that disease. To support that contention, BNSF relied on

Borg-Warner, 232 S.W.3d at 773, as well as Abraham, 233 S.W.3d at 21, in

which one of our sister courts stated that “[k]nowledge of the extent of exposure

to a potentially harmful substance is essential to any reliable expert opinion that

the particular substance caused a disease.”

Second, BNSF contended Daubert and Robinson require the methodology

underlying the expert’s testimony to be reliable. See Daubert, 509 U.S. at 592–

93; Robinson, 923 S.W.2d at 557. Conducting that inquiry, BNSF argued,

9
requires the trial court to consider whether the expert’s methodology (1) has

been subjected to peer review and publication, (2) has a high known or potential

rate of error, (3) has standards controlling its operation, and (4) enjoys general

acceptance within a relevant scientific community. See Daubert, 509 U.S. at

593–94; Robinson, 923 S.W.2d at 557. Pointing to Havner, 953 S.W.2d at 715,

BNSF acknowledged that in a toxic-tort case such as this one, a plaintiff simply

may not be able to obtain reliable, direct evidence of the amount of the plaintiff’s

exposure to the toxin. But again pointing to Havner, 953 S.W.2d at 720, BNSF

stated that in such cases, the expert can utilize epidemiological studies to

circumstantially establish that the plaintiff’s exposure or dose levels were

comparable to or greater than the levels of the subjects in those studies, but the

expert must exclude any other plausible causes of the plaintiff’s injury or

condition with reasonable certainty.

BNSF then argued that Dr. Schonfeld’s opinion did not satisfy either of

these two reliability requirements.

B. BACA’S RESPONSE

In response to BNSF’s motion, Baca acknowledged that because he

brought his FELA claim against BNSF in state court, the trial court was required

to apply federal substantive law but state procedural law in considering BNSF’s

motion. He also acknowledged that decisions concerning the relevance and,

consequently, the admissibility of expert testimony are generally based upon

procedural rules of evidence. However, Baca argued, whether an expert’s

10
testimony is relevant and, therefore, admissible is a question that necessarily

depends upon the causation standard that applies to the claim under

consideration. Citing Brown v. Western Railway of Alabama, 338 U.S. 294, 298

(1949), Baca maintained that state procedural rules cannot be applied in such a

way as to impose unnecessary burdens upon the rights of recovery authorized by

FELA and argued that FELA’s featherweight causation standard should

“significantly influence a determination of the admissibility of an expert’s

causation testimony.”

Baca argued that FELA’s featherweight causation standard impacted state

procedural standards governing the admissibility of expert testimony in two ways.

First, Baca argued, FELA’s lower causation standard rendered Borg-Warner and

Havner wholly inapplicable in FELA cases. Second, Baca insisted that given

FELA’s featherweight causation standard, the admissibility standards set forth in

Daubert and Robinson are relaxed in FELA cases such that the trial court should

more leniently apply Daubert and Robinson to expert testimony in a FELA case

than it would in a non-FELA case.

C. THE TRIAL COURT’S RULING

The trial court considered BNSF’s motion to exclude by submission. On

January 20, 2017, the trial court issued a letter ruling indicating that it had denied

the motion. Additionally, in the letter, the trial court explained the basis of its

ruling: it stated that in denying the motion, it “went with the plaintiff on the issue

of the application of FELA causation standards to . . . Dr. Schonfeld’s opinion on

11
causation.” The trial court ultimately signed an order denying BNSF’s motion to

exclude and granting it permission to seek a permissive interlocutory appeal

because it found the order involved a controlling question of law as to which

there is substantial ground for difference of opinion and an immediate appeal

would materially advance the ultimate termination of this litigation. In pertinent

part, the trial court’s order provides,

The controlling question of law as to which there is substantial
ground for difference of opinion is whether the Federal Employers
Liability Act’s (45 U.S.C. §§ 51-60) lower causation standard—i.e.
whether a railroad’s negligence played any part, even the slightest,
in bringing about the injury—makes inapplicable the expert
admissibility standards expressed in cases like E.I. DuPont de
Nemours & Co. v. Robinson, 923 S.W.2d 549 (Tex. 1995) and
Merrell Dow Pharmaceuticals, Inc. v. Havner, 953 S.W.2d 706 (Tex.
1997). The Court answered this question in the affirmative, holding
that FELA’s lower causation standard contemplates a lower
admissibility threshold for causation evidence.

An immediate appeal of this interlocutory Order would materially
advance the ultimate termination of this litigation because the
Court’s ruling allows the jury to consider causation evidence that
would otherwise be excluded. The answer to this question,
therefore, determines whether this case may be resolved without a
trial on the merits.

D. BNSF’S PETITION FOR PERMISSIVE APPEAL AND BACA’S RESPONSE

BNSF subsequently filed a petition for permissive appeal. In its petition,

BNSF stated that the sole issue it was presenting for appeal was “[w]hether

FELA’s lower causation standard—i.e., whether a railroad’s negligence played

any part, even the slightest, in bringing about the injury—makes Texas’s expert

admissibility standards inapplicable?”

12
 At our request, Baca filed a response to BNSF’s petition. In his response,

Baca stated that he agreed that the trial court’s order met the jurisdictional

standard for a permissive appeal set forth in civil practice and remedies code

section § 51.014(d), but he stated that he disagreed “with the way in which the

issue has been framed by both the trial court and BNSF” because the way the

trial court and BNSF had framed the issue for permissive review did not “fully

encompass the nature of the legal issues which are implicated by this appeal.”

Baca suggested his own phrasing of the question presented for review, a

phrasing that he believed more accurately reflected the issue presented in this

appeal: “Whether the FELA’s lower causation standard—i.e. whether a railroad’s

negligence played any part, even the slightest, in bringing about the injury—may

be modified by applying state expert admissibility rules to impose a higher

causation standard.” Baca then concluded by stating he did not “oppose

interlocutory review of the issue raised by this appeal” but that he requested that

we “grant review on the issue as [he had] more accurately presented it” in his

response.

We then granted BNSF’s petition for permissive appeal. See Baca,

2017 WL 2570826, at *1.

III. JURISDICTION

In conjunction with filing his appellee’s brief, Baca filed a motion to dismiss,

challenging this court’s jurisdiction over this appeal. Baca argues we lack

jurisdiction for two reasons. First, he argues that this appeal no longer involves a

13
controlling question of law as to which there is a substantial ground for difference

of opinion. See Tex. Civ. Prac. & Rem. Code § 51.014(d). And second, Baca

contends this court has already concluded that it is not proper to review by

permissive, interlocutory appeal a trial court’s decision whether to exclude an

expert’s opinion as insufficiently reliable under Robinson. See Blakenergy, Ltd.

v. Oncor Elec. Delivery Co., No. 02-14-00241-CV, 2014 WL 4771736, at *1 (Tex.

App.—Fort Worth Sept. 25, 2015, no pet.) (mem. op.). Neither of these

arguments is persuasive.

A. APPLICABLE LAW

We have jurisdiction to consider a permissive, interlocutory appeal by

virtue of a statutorily authorized exception to the general rule that we have

jurisdiction over final judgments only. See Tex. Civ. Prac. & Rem. Code Ann.

§ 51.014(d); Eagle Gun Range, Inc. v. Bancalari, 495 S.W.3d 887, 889 (Tex.

App.—Fort Worth 2016, no pet.). Under that exception, we may permit an

appeal from an interlocutory order that otherwise would not be immediately

appealable if (1) the order involves a controlling question of law as to which there

is a substantial ground for difference of opinion and (2) an immediate appeal

from the order will materially advance the ultimate termination of the litigation.

Tex. Civ. Prac. & Rem. Code Ann. § 51.014(d), (f).

14
 B. A CONTROLLING QUESTION OF LAW RESOLVED BY THE TRIAL COURT
REMAINS AT ISSUE

We first address Baca’s argument that this appeal no longer involves a

controlling question of law. As Baca reads it, the trial court’s order involves only

a very narrow controlling question of law: whether one specific Texas case—

Havner—applies to a FELA claim. He gets to this reading by stating that he has

conceded Robinson applies to a FELA claim. Baca maintains that having so

conceded, there is no “substantial disagreement” between the parties as to

whether Robinson applies. Therefore, argues Baca, since Havner is the only

other case referenced in the trial court’s order, the trial court’s order only involves

the narrow controlling question of law of whether Havner applies to a FELA

claim. With that premise in place, Baca turns to BNSF’s appellant’s brief,

contending that BNSF did not argue in its brief that Havner applies to a FELA

claim; it only argued that Robinson does. Baca reasons that because he agrees

Robinson applies to a FELA claim and BNSF has not argued that Havner does,

this appeal no longer involves a controlling question of law.

Baca’s argument is based entirely on the assumption that the controlling

question of law involved in the trial court’s order is the narrow question of

whether two specific Texas cases—Robinson and Havner—apply in a FELA

case. But that is not so. From the face of the trial court’s order and the record, it

is clear that the controlling question of law involved in the trial court’s order is not

the narrow question of whether Robinson and Havner, exclusively, are applicable

15
in determining the admissibility of an expert’s causation opinion in a FELA case.

Rather, the controlling question of law involved in the trial court’s order is the

broad question of whether FELA’s lower causation standard also lowers the

admissibility threshold of causation evidence in a FELA case such that the expert

admissibility standards expressed in cases like Robinson and Havner are

inapplicable in determining the admissibility of an expert’s causation opinion in a

FELA case.

1. The Plain Language of the Order

We start with the language of the trial court’s order. In relevant part, the

order states:

The controlling question of law as to which there is substantial
ground for difference of opinion is whether the [FELA’s] lower
causation standard—i.e. whether a railroad’s negligence played any
part, even the slightest, in bringing about the injury—makes
inapplicable the expert admissibility standards expressed in cases
like [Robinson] and [Havner]. The Court answered this question in
the affirmative, holding that FELA’s lower causation standard
contemplates a lower admissibility threshold for causation evidence.

The trial court’s order does not say that the controlling question of law involved in

its order was whether FELA’s causation standard makes Robinson and Havner

inapplicable in a FELA case. Nor does the order say that the controlling question

of law was whether FELA’s causation standard makes the admissibility standards

expressed in Robinson and Havner inapplicable in a FELA case. Both of these

formulations would have limited the controlling question of law involved in the trial

16
court’s order to a question of the applicability of Robinson and Havner,

specifically. Yet the trial court’s order used neither formulation.

Instead, the language in the trial court’s order states the controlling

question of law as “whether the [FELA’s] lower causation standard . . . makes

inapplicable the expert admissibility standards expressed in cases like

[Robinson] and [Havner].” (Emphasis added). The phrase “expressed in cases

like [Robinson] and [Havner]” is a modifier that clarifies the phrase preceding it—

“expert admissibility standards.” Examining the language of that modifying

phrase, the placement of the words “cases like” before “[Robinson] and [Havner]”

makes plain that the controlling question of law in the trial court’s order is not

limited to whether Robinson and Havner, specifically, are applicable. To read the

order as involving only the applicability of Robinson and Havner, as Baca does,

is to entirely erase the words “cases like” from the trial court’s order. So the

order involves the applicability of “the expert admissibility standards expressed in

cases like Robinson and Havner,” not the applicability of Robinson and Havner

exclusively.

This construction is confirmed by another portion of the trial court’s order

that Baca’s reading would also erase. After the portion stating the controlling

question of law, the very next sentence in the trial court’s order states, “The

Court answered this question in the affirmative, holding that FELA’s lower

causation standard contemplates a lower admissibility threshold for causation

evidence.” In this sentence, the trial court explains that it had decided that

17
[FELA’s] “lower causation standard . . . makes inapplicable the expert

admissibility standards expressed in cases like [Robinson] and [Havner].” And it

explained the sole reason why it had made that decision was its legal conclusion

that “FELA’s lower causation standard contemplates a lower admissibility

threshold for causation evidence.” That language confirms the trial court had

broadly ruled that FELA’s relaxed causation standard lowers Texas’s

admissibility threshold for causation evidence in a FELA case.

Thus, when all of the pertinent language in the trial court’s order is given

effect, it is clear that the controlling question of law on which the trial court ruled

is broader than the narrow question only of whether Robinson and Havner apply

in a FELA case. Rather, the trial court’s ruling was as follows: because FELA’s

lower causation standard contemplates a lower admissibility threshold for

causation evidence, FELA’s lower causation standard makes inapplicable the

expert admissibility standards expressed in cases like Robinson and Havner.

2. The Record

That the controlling question of law involved in the trial court’s order is not

limited to the narrow question of whether Robinson and Havner, specifically,

apply in a FELA case is also confirmed by the record.

In its motion to exclude, BNSF anticipated Baca would argue that “in light

of the causation standard under FELA, the standards for reliability and

admissibility of expert causation opinions should be relaxed.” BNSF maintained

18
that FELA’s relaxed causation standard does not relax the standards for

admissibility of expert causation opinions in a FELA case.

Baca indeed made the argument BNSF anticipated he would. In Section

V, Paragraph D of his response to BNSF’s motion, Baca argued that “[i]t is well

established that FELA’s featherweight causation standard should significantly

influence a determination of the admissibility of an expert’s causation testimony.”7

In Section V, Paragraph E of his response, Baca argued that “[t]he remedial

nature of FELA has a significant effect on the admissibility of expert testimony.”

He pointed to Hines as “the leading case on point,” stating that the Hines court

had held that “FELA[’s] relaxed standard of causation also relaxes the threshold

of admissibility for the reception of expert testimony.” Baca argued that because

of this relaxed admissibility standard, (1) Havner is inapplicable in a FELA case

and (2) Robinson “should be applied with a more lenient standard” in a FELA

case.8

For this “well established” proposition, Baca cited only one case: Hines v.
7

Consolidated Rail Corp., 926 F.2d 262, 269 (3d Cir. 1991).
8
As part of his argument for why this case no longer involves a controlling
question of law, Baca argues that he agrees that Robinson applies to his FELA
claim and that, therefore, there is no substantial disagreement between the
parties as to the applicability of Robinson in this case. Not so. It is patently clear
that the parties did disagree as to the applicability of Robinson to Dr. Schonfeld’s
causation opinion. As we have noted, BNSF argued that Robinson’s
admissibility standards apply the same in a FELA case as they do in a non-FELA
case, whereas Baca argued that those standards apply differently—that is, “more
lenient[ly]”—in a FELA case than in a non-FELA case.

19
 The trial court considered BNSF’s motion by submission. Before signing

the order denying the motion, the trial court first sent a letter ruling to the parties

informing them of its decision. In the letter, the trial court explained that it had

decided to deny the motion and explained its reason for doing so, stating that it

“went with [Baca] on the issue of the application of FELA causation standards

to . . . Dr. Schonfeld’s opinion on causation.” The trial court referred the parties

to specific portions of Baca’s response to BNSF’s motion, telling them to “[s]ee

generally paragraph[s] D & E” of the response. The trial court additionally stated

that it “believe[d] this to be a ‘controlling question of law as to which there is

substantial ground for difference of opinion.’”

Thus, the record confirms that the parties disagreed over the broad

question of whether FELA’s lower causation standard lowers the threshold of

admissibility for expert causation testimony; that the trial court ruled in the

affirmative on that broad question; and that the trial court based its decision to

deny BNSF’s motion solely upon its resolution of that broad question.

Given the foregoing, we disagree with Baca’s argument that this appeal no

longer involves a controlling question of law, as well as with his argument that

BNSF’s appellant’s brief presents a controlling question of law that differs from

the one the parties presented to the trial court and which the trial court resolved.

C. BACA’S RELIANCE ON BLAKENERGY

Baca also notes that BNSF has argued only that Robinson applies to his

FELA claim; it has not argued that Dr. Schonfeld’s opinion fails to meet

20
Robinson’s admissibility standards. But he argues that, to the extent BNSF’s

appeal can be construed as presenting the issue of whether Dr. Schonfeld’s

opinion satisfies Robinson’s admissibility standards, this court has already

determined it is not appropriate to review by permissive appeal a “plain vanilla

expert challenge under Robinson”—that is, a trial court’s determination of

whether an expert’s testimony satisfies Robinson’s admissibility standards. See

Blakenergy, 2014 WL 4771736, at *1. He asks us to follow our precedent in

Blakenergy and dismiss this appeal.

Assuming Baca’s characterization of our decision in Blakenergy is correct,

we need not address this argument because this appeal does not present the

“plain vanilla expert challenge” of whether Dr. Schonfeld’s opinion satisfies

Robinson’s admissibility requirements.9 Rather, the trial court denied BNSF’s

motion to exclude based on its legal conclusion that because FELA’s

featherweight standard of causation “contemplates a lower admissibility threshold

for causation evidence,” state procedural standards governing the admissibility of

expert testimony “expressed in cases like [Robinson] and [Havner]” are

inapplicable in a FELA case. Thus, the sole issue in this appeal is a pure

question of law: whether FELA’s causation standard lowers the admissibility

threshold for causation evidence in FELA cases.

9
Because we need not address this argument, we express no opinion as to
the merit of Baca’s characterization of this court’s denial of the petition for
permissive appeal in Blakenergy. See id.

21
 Because we are unpersuaded by Baca’s arguments that this appeal no

longer involves a controlling question of law, we deny his motion to dismiss.

IV. THE TRIAL COURT’S ERRONEOUS LEGAL CONCLUSION WAS
AN ABUSE OF DISCRETION

In its sole issue, BNSF argues the trial court abused its discretion by

concluding that in a FELA case, FELA’s featherweight causation standard lowers

our state’s standards governing the admissibility of expert testimony.

A. STANDARD OF REVIEW

Even in a FELA case, a trial court’s evidentiary rulings, including its

decision whether to admit expert testimony, is subject to an abuse-of-discretion

standard of review. See Abraham, 233 S.W.3d at 17; Mo. Pac. R.R. v. Navarro,

90 S.W.3d 747, 750 (Tex. App.—San Antonio 2002, no pet.). Here, the trial court

based its decision to deny BNSF’s motion to exclude solely on its determination

of a pure question of law. A trial court has no discretion to make an erroneous

legal conclusion even in an unsettled area of law. In re United Scaffolding, Inc.,

301 S.W.3d 661, 663 (Tex. 2010) (orig. proceeding); In re Mo. Pac. R.R.,

998 S.W.2d 212, 216 (Tex. 1999). We review a trial court’s determination of a

pure question of law de novo, affording no deference to the trial court’s

conclusion. See RSI Int’l, Inc. v. CTC Transp., Inc., 291 S.W.3d 104, 107 (Tex.

App.—Fort Worth 2009, no pet.).

22
 B. APPLICABLE LAW

As a general rule, when a plaintiff brings a FELA claim in state court, the

court applies federal substantive law and state procedural law. See Dickerson,

470 U.S. at 411; Oney, 380 S.W.3d at 800. But a state procedural rule may not

interfere with, lessen, or destroy any of the substantive rights afforded to the

plaintiff under FELA. See Brown, 338 U.S. at 298–99; Scott v. Atchison, Topeka

& Santa Fe Ry., 572 S.W.2d 273, 281 (Tex. 1978) (op. on reh’g); Oney,

380 S.W.3d at 800. Any state procedural rule that does so must yield. See

Brown, 338 U.S. at 298–99; Scott, 572 S.W.2d at 281; Oney, 380 S.W.3d at

799–800.

C. TEXAS’S PROCEDURAL STANDARDS APPLY

In accordance with the general rule that state courts apply state procedural

law in cases brought under a federal statute in state court, Texas courts apply

state rules of evidence when determining the admissibility of expert testimony in

FELA cases. See Nichols, 379 S.W.3d at 382 (“Although [appellees’] claim is

pursuant to [FELA], the trial court must follow state procedure in determining the

reliability of expert testimony.”); Abraham, 233 S.W.3d at 18 (“Despite the fact

that appellants assert a claim under the [FELA], the trial court must follow state

procedure in determining the reliability of expert testimony.”); Navarro, 90 S.W.3d

at 754–59 (applying state procedural rules in determining the admissibility of

expert testimony in a FELA case). Accordingly, as relevant here, in a FELA case

brought in state court, expert testimony must be reliable to be admissible. See

23
Nichols, 379 S.W.3d at 382 (citing Robinson, 923 S.W.2d at 556). This general

proposition is not in dispute here.

In the trial court, Baca argued that FELA’s featherweight causation

standard relaxes Texas procedural standards for determining the admissibility of

expert testimony regarding causation in a FELA case. The trial court agreed and

based its decision to deny BNSF’s motion to exclude solely on that legal

conclusion. BNSF argues that this conclusion is contrary to what the courts in

this state, the federal courts, and the courts in many other states have

recognized: the standard of causation under FELA and the standards for

admission of expert testimony under the applicable rules of evidence are distinct

issues that do not affect one another. Thus, BNSF argues, the trial court abused

its discretion by concluding that FELA’s featherweight causation standard relaxes

our state’s procedural standards for admissibility of expert testimony in a FELA

case.

1. Texas Authorities

BNSF is correct that two of our sister courts have squarely held that

FELA’s featherweight causation standard does not lower the threshold for

admissibility of expert testimony in a FELA case. The first to do so was the

Fourth Court of Appeals in Navarro. See 90 S.W.3d at 751. That case involved

a FELA claim in which the plaintiff alleged that her exposure to diesel exhaust

during her employment with the defendant railroad caused her to develop bone

marrow cancer. Id. at 749. On appeal, the railroad argued that the testimony of

24
the plaintiff’s causation experts should have been excluded because the

testimony was scientifically unreliable. Id. at 749. In addressing that argument,

the court acknowledged FELA’s lower burden of causation but noted that lower

burden had “not been generally applied to the admissibility of expert testimony” in

a FELA case and that “the Daubert standard of admissibility of expert evidence

[in a FELA case] extends to each step in an expert’s analysis all the way through

the step that connects the work of the expert to the particular case.”10 Id. at 751

(internal quotation omitted). And in analyzing whether the challenged expert

causation testimony was unreliable, the court applied Texas’s standards. See id.

at 754–59.

The Fourteenth Court of Appeals followed suit in Abraham. See

233 S.W.3d at 17–20. That was a FELA case in which the plaintiffs alleged that

their exposure to creosote during their employment with the defendant railroad

caused them to suffer diseases of the throat, lungs, and skin, including cancer.

Id. at 16. The defendant railroad challenged the plaintiff’s expert’s causation

opinion on the ground that it was not scientifically reliable. Id. The court applied

state procedural standards in determining the reliability of the challenged expert

testimony. See id. at 18 (citing Robinson, 923 S.W.3d at 565). It agreed that the

standard of causation under FELA and the standards of expert testimony under

the rules of evidence are distinct issues and do not affect one another. See id. at

10
The court noted that our supreme court adopted Daubert in Robinson.
See id. at 750.

25
19–20. And, consequently, it agreed that FELA’s featherweight causation

standard did not relax our state’s procedural standards governing the

admissibility of expert testimony such that courts must admit expert testimony in

FELA cases that would be inadmissible in other contexts. See id.

We also find the supreme court’s decision in In re GlobalSantaFe Corp.,

275 S.W.3d 477 (Tex. 2009) (orig. proceeding), instructive to our analysis. That

case involved a claim under the Jones Act, which provides for a cause of action

against the employer of a seaman who sustains an injury or death in the course

of his employment. See 46 U.S.C.A. § 30104 (West 2007); GlobalSantaFe,

275 S.W.3d at 479–80, 480 n.1. The Jones Act expressly incorporates FELA

and the case law developing that statute. See Ellis, 971 S.W.2d at 406; see also

46 U.S.C.A. § 30104. Thus, like FELA, the Jones Act establishes a

featherweight standard of causation. See Ellis, 971 S.W.2d at 406. And courts

therefore often look to caselaw applying FELA when analyzing claims under the

Jones Act. See Wills v. Amerada Hess Corp., 379 F.3d 32, 46–47, 47 n.9 (2d

Cir. 2004) (looking to FELA cases when analyzing whether the Jones Act’s

relaxed standard of causation lowered the standards for admissibility of expert

testimony under the rules of evidence); see also Lies v. Farrell Lines, Inc.,

641 F.2d 765, 770 (9th Cir. 1981) (noting that because the Jones Act expressly

incorporates FELA and case law developing that statute, the court could

“appropriately look to FELA cases to test the sufficiency of the allegations and

proof in this Jones Act claim”).

26
 In GlobalSantaFe, the supreme court considered whether the Jones Act

preempted certain provisions of Chapter 90 of the civil practice and remedies

code, including section 90.004, which requires a plaintiff in silica cases to serve a

detailed expert report on each defendant. See 275 S.W.3d at 479–80; see also

Tex. Civ. Prac. & Rem. Code Ann. § 90.004 (West 2017). This provision, the

court said, “endeavors to assure that claims are not brought and pursued unless

they are supported by reliable expert evaluations of the claimant.”

GlobalSantaFe, 275 S.W.3d at 482. After discussing the general principles of

preemption in Jones Act cases, the court concluded that “[t]he requirements

embedded in Chapter 90 to assure reliable expert confirmation of silica-related

diseases are not preempted by the Jones Act.”11 See id. at 486.

In explaining its conclusion, the court discussed the standards governing

the admissibility of expert testimony. See id. at 486–87. In particular, it noted

that federal caselaw has developed standards for admission of expert testimony

that focus on the trial court’s role in determining the reliability of such testimony

and that Texas caselaw, drawing heavily from federal jurisprudence, had

developed similar standards. See id. It then discussed the issue of whether the

Jones Act’s featherweight causation standard affected the application of state

procedural rules governing the admission of expert testimony. The court said,

11
The court concluded, however, that the Jones Act preempts section
90.004(b)(2) because that provision “requires claimants in some cases to
establish a minimal level of impairment,” which conflicts with the Jones Act. See
id. at 483, 489.

27
 To the extent that Jones Act jurisprudence recognizes a special
standard for proving causation, federal cases have held that this
causation standard does not exempt Jones Act cases from the
general rules for admission of expert testimony. We see no basis for
holding that Texas law generally governing the admission of expert
testimony, which draws so heavily from federal law, is preempted by
the Jones Act.

Id. (citations omitted). Thus, in GlobalSantaFe, the supreme court expressed

disapproval of the notion that a featherweight causation standard alters the

applicability of Texas law generally governing the admission of expert testimony.

2. Federal and State Authorities

The decisions in Navarro, Abraham, and GlobalSantaFe are in line with the

overwhelming weight of authority in the federal courts, as well as the courts in

other states. As BNSF has pointed out in its brief, several federal authorities

have held that a featherweight causation standard—whether under FELA or the

Jones Act—does not relax the threshold of admissibility for expert testimony.

See, e.g., Wills, 379 F.3d at 47 (noting that “the standards for determining the

reliability and credibility of expert testimony are not altered merely because the

burden of proof [under the Jones Act] is relaxed”); Claar v. Burlington N. R.R.,

29 F.3d 499, 503 (9th Cir. 1994) (“The standard of causation under FELA and the

standards for admission of expert testimony under the Federal Rules of Evidence

are distinct issues and do not affect one another.”); see also Taylor v. Consol.

Rail Corp., No. 96-3579, 1997 WL 321142, at *6–7 (6th Cir. June 11, 1997)

(unpublished table decision) (holding that the typical standards for determining

the admissibility of expert testimony “are not modified by the relaxed standard of

28
proof in FELA cases”); Cantrell v. BNSF Ry. Co., No. Civ 12-0129, 2013 WL

8632378, at *4 (D.N.M. June 28, 2013) (noting that the relaxed causation

standard under FELA and the standard for admission of expert testimony are

distinct issues that do not affect one another); Clements v. CSX Transp., Inc.,

No. 3:09-cv-122-TCB, 2011 WL 13136959, at *3 (N.D. Ga. Dec. 13, 2011)

(noting that if a plaintiff in a FELA case chooses to introduce expert testimony,

the trial court must apply the same admissibility analysis that applies in any other

case), amended in part on other grounds, 2012 WL 12897089 (N.D. Ga. Jan. 9,

2012). In addition, as BNSF observed in its brief, the courts in several states

have concluded that FELA’s relaxed burden of proof does not alter the

applicability of their state’s procedural standards governing the admissibility of

expert testimony. See, e.g., Smith v. CSX Transp., Inc., 806 S.E.2d 890, 893–94

(Ga. Ct. App. 2017); Smart v. BNSF Ry. Co., 369 P.3d 966, 971–72 (Kan. Ct.

App. 2016); see also Russell v. Ill. Cent. R.R., No. W2013-02453-COA-R3-CV,

2015 WL 4039982, at *2–3 (Tenn. Ct. App. June 30, 2015) (applying state

procedural rules in determining admissibility of expert testimony in FELA case);

McNeel v. Union Pac. R.R., 753 N.W.2d 321, 328–30 (Neb. 2008) (applying state

procedural standards governing admissibility of expert testimony in a FELA case

in the same manner as in other cases.).

Thus, in addition to the fact that several Texas courts have concluded that

a relaxed standard of causation does not relax our state’s procedural standards

governing the admissibility of expert testimony, there is a broad range of

29
authority from the federal courts and courts in other states holding that FELA’s

lower causation burden does not alter the applicable procedural standards

governing the admissibility of expert testimony.

3. Baca’s Authorities

In his appellee’s brief, Baca did not address the merits of the question

presented in this permissive appeal or the significant number of authorities BNSF

presented in support of its position on that question. Rather, he reurged some of

the arguments he made in his motion to dismiss this permissive appeal.12 Our

analysis above concerning those arguments is equally applicable here, and we

need not restate it. We note, however, that in his response to BNSF’s motion in

the trial court, Baca cited two cases that he argued supported his position that

FELA’s relaxed causation standard relaxed the threshold for admissibility of

expert testimony in this FELA case. We briefly address those cases.

In his response in the trial court, Baca argued that “[i]t is well established

that FELA’s featherweight causation standard should significantly influence a

determination of the admissibility of an expert’s causation testimony.” For this

proposition, he cited only one case: Hines v. Consolidated Rail Corp., 926 F.2d

262 (3rd Cir. 1991). Indeed, the court in that case concluded that the causation

12
That is, in his appellee’s brief, Baca argues that because he agrees that
Robinson applies to his claim, and because BNSF has not specifically argued
that Havner and Borg-Warner apply, “there is nothing further for this [c]ourt to
consider[,] and this [c]ourt should affirm the trial court’s ruling on BNSF’s motion
to exclude the expert opinions of Dr. Schonfeld.”

30
standard under FELA can significantly influence a determination of the

admissibility of expert testimony. Id. at 269. However, as our sister court

observed in Abraham, the Hines decision predates the Supreme Court’s seminal

decision in Daubert, which articulated standards for assessing the reliability of

expert testimony. Abraham, 233 S.W.3d at 18. And of course, our supreme

court subsequently adopted Daubert. See Robinson, 923 S.W.2d at 556–57. As

the cases we have discussed above demonstrate, since Daubert, Texas, federal,

and sister-state courts have consistently applied in FELA cases the same expert-

testimony admissibility standards they apply in non-FELA cases. We thus find

Hines unpersuasive. See Wills, 379 F.3d at 47 (declining to follow Hines and

holding that “Daubert’s standards for determining the admissibility of expert

testimony apply regardless of whether the plaintiff’s burden to prove causation is

reduced”); Abraham, 233 S.W.3d at 20 (concluding Hines was unpersuasive

because it was decided before Daubert and Robinson).

In his response in the trial court, Baca also argued that the Third Court’s

decision in Navarro and the Fourteenth Court’s decision in Abraham, both of

which concluded that FELA’s relaxed causation burden does not relax the

standards for admissibility of expert testimony, are “no longer the law in a FELA

case” after the Supreme Court’s decision in McBride. See McBride, 564 U.S.

685; Abraham, 233 S.W.3d at 19–20; Navarro, 90 S.W.3d 750–51. But McBride

was a jury-charge case concerning whether the trial court had erred by failing to

include a proximate-cause instruction in the jury charge. See 564 U.S. at 689–

31
90. The Court concluded it had not, holding that in FELA cases, juries “are

properly instructed that a defendant railroad ‘caused or contributed to’ a railroad

worker’s injury ‘if [the railroad’s] negligence played a part—no matter how

small—in bringing about the injury.’” Id. at 705. Thus, McBride did not deal with

the admissibility of expert testimony in FELA cases. And even after McBride,

courts have continued to recognize a distinction between FELA’s causation

standard and the procedural standards governing the admissibility of expert

testimony. See Whalen v. CSX Transp., Inc., No. 13 Civ. 3784, 2016 WL

5723877, at *17 n.17 (S.D.N.Y. Sept. 29, 2016); Cantrell, 2013 WL 8632378, at

*4; Clements, 2011 WL 13136959, at *3; Smith, 806 S.E.2d at 893–94; Farley v.

BNSF Ry. Co., No. 112,872, 2016 WL 1169426, at *7 (Kan. Ct. App. Mar. 25,

2016). We are therefore unpersuaded that McBride upended this well-

established distinction.

4. Texas’s Admissibility Standards Apply

We agree with the reasoning of the courts that have concluded that the

standard of causation under FELA and the standards for admission of expert

testimony under the applicable rules of evidence are distinct issues that do not

affect one another and that, consequently, FELA’s featherweight causation

standard does not require a Texas trial court to admit expert testimony in a FELA

case that would be inadmissible in a non-FELA case. See Claar, 29 F.3d at

503–04; Abraham, 233 S.W.3d at 19–20. We therefore hold that FELA’s

featherweight causation standard does not render inapplicable in FELA cases

32
our state procedural law governing the admissibility of expert testimony. The trial

court abused its discretion by concluding to the contrary. See United Scaffolding,

301 S.W.3d at 663 (“[A]n erroneous legal conclusion is an abuse of discretion,

even if it may not have been clearly erroneous when made.”). We sustain

BNSF’s sole issue.

V. CONCLUSION

Having concluded that this appeal continues to involve a controlling

question of law as to which there is a substantial ground for difference of opinion,

we deny Baca’s motion to dismiss. And having sustained BNSF’s sole issue, we

reverse the trial court’s order denying BNSF’s motion to exclude and remand this

case for further proceedings. See Tex. R. App. P. 43.2(d).

/s/ Lee Gabriel

LEE GABRIEL
JUSTICE

PANEL: WALKER, GABRIEL, and KERR, JJ.

DELIVERED: March 29, 2018

33