Opinion issued November 18, 2021
In The
Court of Appeals For The
First District of Texas ———————————— NO. 01-19-00568-CV ——————————— BMLA, INC. D/B/A MURPHY’S DELI, Appellant V. KEZIAH JORDAN, Appellee
On Appeal from the 165th District Court Harris County, Texas Trial Court Case No. 2015-17893
DISSENTING OPINION
The majority holds that the evidence is legally insufficient to support the
verdict because Keziah Jordan did not introduce evidence that the kolache she
bought from Murphy’s Deli caused her food poisoning. I disagree with the
majority’s holding for two independent reasons. First, Murphy’s Deli did not preserve the issue of legal sufficiency for our review. Second, even if this issue had
been preserved for our review, Jordan introduced some evidence of causation, and
the evidence is therefore not legally insufficient to support the verdict.
Error Preservation
Applicable Law
Error preservation is not optional. Excepting fundamental error, a party cannot
present a complaint for the first time on appeal. TEX. R. APP. P. 33.1(a); Roberson v.
Collins, 221 S.W.3d 239, 241 (Tex. App.—Houston [1st Dist.] 2006, no pet.).
To challenge the legal sufficiency of the evidence on appeal from a judgment
entered on a jury verdict, a party must preserve the issue for review in the trial court
in one of several ways. The party can do so by moving for an instructed verdict,
moving for judgment notwithstanding the verdict, objecting to the submission of a
jury question, moving to disregard the jury findings, or moving for a new trial. T.O.
Stanley Boot Co. v. Bank of El Paso, 847 S.W.2d 218, 220 (Tex. 1992); Roberson,
221 S.W.3d at 241–42. The motion or objection must address legal sufficiency in
particular to preserve this issue. See Steves Sash & Door Co. v. Ceco Corp., 751
S.W.2d 473, 477 (Tex. 1988); U.S.A. Precision Machining Co. v. Marshall, 95
S.W.3d 407, 412 (Tex. App.—Houston [1st Dist.] 2002, pet. denied).
2 Nor can the party raise one legal-sufficiency complaint below and another on
appeal. The complaint made on appeal must conform to the one made below. In re
C.E.M., 64 S.W.3d 425, 428 (Tex. App.—Houston [1st Dist.] 2000, no pet.).
Analysis
It is undisputed that the first time Murphy’s Deli ostensibly raised the issue of
legal sufficiency was in its motion for new trial. In its new-trial motion, it argued:
The jury finding of liability against the Defendant for the incident made the basis of suit is against the great weight and preponderance of the evidence. There is no credible evidence the food product in question was tainted while in the possession of the Defendant. In support of this argument, Murphy’s Deli discussed the testimony of Jordan
and her treating doctor, asserting that her doctor’s testimony established the time
period in which food poisoning would appear and that Jordan’s testimony
established that her illness developed outside of this period. Murphy’s Deli then
reiterated that this evidence showed the “verdict finding liability is against the great
weight and preponderance of the evidence in this case.”
On its face, the new-trial motion solely challenges the factual sufficiency of
the evidence, not its legal sufficiency. The contention that a verdict is against the
great weight and preponderance of the evidence disputes the factual sufficiency of
the evidence. In re Bent, 487 S.W.3d 170, 179 (Tex. 2016); Transmission Exch. v.
Long, 821 S.W.2d 265, 274 (Tex. App.—Houston [1st Dist.] 1991, writ denied).
3 Notably, the new-trial motion does not include the terms “legal sufficiency”
or “no evidence.” The closest the motion comes to using either of these terms is “no
credible evidence.” But an assertion that the record lacks evidence that is credible,
or believable, is not the same thing as an assertion that there is no evidence
whatsoever. Instead, an attack on the credibility or believability of the evidence is
an attack on its persuasiveness or weight. See NEW OXFORD AMERICAN DICTIONARY
406 (3d ed. 2010) (defining “credible” as “able to be believed; convincing”). For this
reason, we construe challenges to the credibility or believability of the evidence as
factual-sufficiency challenges. See Hayes v. Wells Fargo Bank, No. 01-06-00720-
CV, 2007 WL 3038043, at *4 (Tex. App.—Houston [1st Dist.] Oct. 18, 2007, pet.
denied) (mem. op.); Lloyd v. Angel, No. 01-99-01070-CV, 2003 WL 22863081, at
*1 (Tex. App.—Houston [1st Dist.] Dec. 4, 2003, pet. denied) (mem. op.); see also
Crosstex N. Tex. Pipeline v. Gardiner, 505 S.W.3d 580, 615 (Tex. 2016) (courts set
aside finding for factual insufficiency if “credible evidence supporting the finding”
is too weak or too contrary to overwhelming weight of all evidence in record).
In sum, Murphy’s Deli did not raise the issue of legal sufficiency in its motion
for new trial. Thus, Murphy’s Deli did not preserve this issue for our review. See,
e.g., Sinegaure v. Bally Total Fitness Corp., No. 01-05-01070-CV, 2008 WL
5263235, at *2 (Tex. App.—Houston [1st Dist.] Dec. 18, 2008, no pet.) (mem. op.)
4 (new-trial motion challenging great weight and preponderance that did not suggest
a legal-sufficiency complaint preserved only factual-sufficiency challenge).
Moreover, even if one could somehow shoehorn a legal-sufficiency argument
into the new-trial motion’s language, Murphy’s Deli’s argument on appeal bears no
resemblance to the argument it made in its new-trial motion. On appeal, Murphy’s
Deli asserts that the evidence is legally insufficient because there is no “medical
expert testimony establishing that Jordan’s illness was caused by the consumption
of a kolache she purchased from Murphy’s Deli.” In contrast, its new-trial motion
said nothing about the absence of medical expert testimony as to causation. For this
additional reason, Murphy’s Deli did not preserve error. See, e.g., In re C.E.M., 64
S.W.3d at 428 (party could not challenge statutory predicate findings for termination
on appeal when her new-trial motion solely challenged best-interest finding).
Disregarding Murphy’s Deli’s repeated reference to “the great weight and
preponderance of the evidence,” the majority asserts that Texas courts consider the
phrase “no credible evidence” to be indistinguishable from “no evidence,” at least
when expert opinion and testimony about causation are at issue. In support of this
assertion, the majority cites two decisions: Jelinek v. Casas, 328 S.W.3d 526 (Tex.
2010); and Holt Atherton Indus. v. Heine, 797 S.W.2d 250 (Tex. App.—Corpus
Christi 1990), aff’d in part and rev’d in part, 835 S.W.2d 80 (Tex. 1992).
5 Inspection of these two decisions reveals that they do not support the
majority’s assertion. Neither decision faced the issue before us: distinguishing a
legal-sufficiency challenge from a factual one. Thus, neither one is apposite.
Jelinek was a medical malpractice case, in which the plaintiffs alleged that a
negligent lapse in treatment with an antibiotic during a hospital stay after abdominal
surgery caused the patient prolonged and needless suffering before she died. 328
S.W.3d at 529–31. The lone issue before the Court was whether the plaintiffs’
expert’s medical causation opinion was unsupported and thus legally insufficient.
Id. at 529. In addressing this issue, the Court stated and applied the traditional no-
evidence standard. Id. at 532. In articulating this standard, the Court did not make
any reference to credibility, the credibility of the evidence, or credible evidence. Id.
The Court ultimately held that the expert’s opinion was conjectural and therefore
legally insufficient. Id. at 538. In stating its holding, the Court once again did not
refer to credibility, the credibility of the evidence, or credible evidence. Id.
In its discussion, however, the Court made a single reference to “credible
evidence.” Id. at 538. The Court cautioned that courts “should not usurp the jury’s
role as fact finder” nor “question the jury’s right to believe one witness over
another.” Id. In the next sentence, the Court stated that appellate courts nonetheless
cannot “defer to the jury’s findings when those findings are not supported by credible
evidence.” Id. The Court then observed that if “the evidence compels the jury to
6 guess if a vital fact exists, a reviewing court does not undermine the jury’s role by
sustaining a no-evidence challenge.” Id. Such was the case in Jelinek because there
were several plausible causes of the patient’s suffering other than a negligence-
induced infection, none of which the plaintiffs’ expert could reasonably exclude. Id.
Thus, the Court concluded, “no evidence” supported the verdict. Id. at 536–38.
From the preceding discussion, the majority infers that “no credible evidence”
is the equivalent of “no evidence.” But Jelinek does not say so. Nor does Jelinek
imply that “no credible evidence” and “no evidence” are one and the same.
Jelinek held that an expert’s opinion was no evidence, not because it was not
credible, but because it was speculative. Id. As the Court has always acknowledged,
there is a difference between the reliability of expert evidence, which can render it
no evidence, and the credibility of expert evidence, which continues to be an issue
for the jury. See E.I. du Pont de Nemours & Co. v. Robinson, 923 S.W.2d 549, 558
(Tex. 1995) (stating that even under gatekeeping standards applicable to expert
witnesses, “the jury will continue to assess the weight and credibility of the proffered
testimony”). The credibility of expert evidence is not a proper subject of no-evidence
review. See Keo v. Vu, 76 S.W.3d 725, 734–35 (Tex. App.—Houston [1st Dist.]
2002, pet. denied) (stating that “factual weaknesses underlying an expert’s causation
opinion generally go to the testimony’s weight, rather than its admissibility, and the
opinion is no evidence only if based completely upon speculation and surmise”).
7 Jelinek does not purport to alter this aspect of legal-sufficiency analysis and cannot
be read to suggest that “no credible evidence” and “no evidence” are synonyms.
Holt is still further afield. There, the Thirteenth Court used the phrase “no
credible evidence” in describing the appellant’s challenge of an award of $120,000
in lost profits, which the court of appeals then analyzed as a no-evidence challenge.
797 S.W.2d at 252–53. At first blush, this seems to support the majority’s position.
But Holt was an appeal from a no-answer default judgment. Id. at 252. In this
context, error preservation is not relevant because a party may challenge the legal
and factual sufficiency of awards for unliquidated damages in an appeal from a
default judgment. Id. at 252; see also Argyle Mech. v. Unigus Steel, 156 S.W.3d 685,
687 n.1 (Tex. App.—Dallas 2005, no pet.) (party need not preserve these issues). In
other words, the manner in which the appellant framed its evidentiary sufficiency
challenge—which is not quoted in the court’s opinion—was immaterial because it
could raise either legal or factual insufficiency issues for the first time on appeal.
In summary, neither Jelinek nor Holt support the majority’s position that “no
credible evidence” is the same thing as “no evidence.” Indeed, both decisions are so
readily distinguishable from the situation before us that the majority’s reliance on
them merely serves to underscore the untenability of the majority’s position.
To conclude, as the majority does, that “no credible evidence” and “no
evidence” are indistinguishable, the majority must discard “credible” as surplusage
8 and ignore Murphy’s Deli’s characterization of its complaint as being that the jury’s
liability finding is “against the great weight and preponderance of the evidence,” a
phrase that invariably signals a factual-sufficiency challenge. See Windrum v. Kareh,
581 S.W.3d 761, 781 (Tex. 2019) (explaining that factual-sufficiency review, unlike
legal-sufficiency review, weighs all evidence). The majority must also disregard the
longstanding recognition of our courts that “no credible evidence” means something
altogether different than “no evidence.” See Freeman v. Grashel, 145 S.W. 695, 698
(Tex. App.—San Antonio 1912, writ ref’d) (noting appellant did not claim there was
no evidence but rather that there was no credible evidence to support jury finding).
Murphy’s Deli did not preserve its legal-sufficiency challenge for our review.
The majority’s contrary holding cannot be reconciled with well-settled law.
Legal Sufficiency
Standard of Review
The test for legal sufficiency is whether the evidence at trial would allow
reasonable and fair-minded people to render the verdict under review. W & T
Offshore v. Fredieu, 610 S.W.3d 884, 898 (Tex. 2020). When reviewing a legal-
sufficiency challenge, we view the evidence in the light most favorable to the verdict,
crediting favorable evidence when reasonable jurors could do so and disregarding
contrary evidence unless reasonable jurors could not. Pike v. Tex. EMC Mgmt., 610
S.W.3d 763, 783 (Tex. 2020). Thus, conflicts in the evidence ordinarily present an
9 issue for the jury to resolve and play no part in legal-sufficiency analysis. See
McAllen Hosps. v. Lopez, 576 S.W.3d 389, 392 (Tex. 2019). We must indulge every
reasonable inference deducible from the evidence in favor of the verdict. JBS
Carriers v. Washington, 564 S.W.3d 830, 841–42 (Tex. 2018). We will sustain a
legal-sufficiency challenge if (1) there is a complete lack of evidence of a vital fact;
(2) rules of law or evidence bar us from considering the lone evidence of a vital fact;
(3) no more than a scintilla of evidence supports a vital fact; or (4) the evidence
conclusively proves the opposite of the vital fact. Pike, 610 S.W.3d at 783.
In a foodborne illness case, the test for cause in fact is whether the plaintiff
would not have gotten ill but for eating the food in question. Smith v. Landry’s Crab
Shack, 183 S.W.3d 512, 514 (Tex. App.—Houston [14th Dist.] 2006, no pet.). A
finding of cause in fact may be based on appropriate circumstantial evidence. Id.
The majority states that Jordan’s treating physician did not determine that her
gastroenteritis was caused by food or that the kolache she ate in particular caused
her illness. The first statement is an indefensible mischaracterization of the record.
The second statement is accurate but immaterial because the treating physician’s
testimony, when considered in conjunction with other undisputed facts introduced
10 at trial, is legally sufficient to allow a reasonable jury to find that the kolache caused
Jordan’s illness.
Syed Ali Reza Hasan, M.D., who practices internal medicine, treated Jordan.
He diagnosed her as suffering from viral gastroenteritis and testified in no uncertain
terms that food more often than not is the source of viral gastroenteritis. In particular,
he stated that food is “the typical source of gastroenteritis.” There is little or no
ambiguity in Hasan’s statement, but the following exchange also took place:
Q. Is food the primary cause of gastroenteritis?
A. It can be. Q. In more cases than not is it true that food is the cause of gastroenteritis?
A. Correct.
Moreover, Hasan made it clear in his testimony that he effectively equated
viral gastroenteritis with food poisoning precisely because the latter more often than
not causes the former. This became evident when opposing counsel unsuccessfully
tried to get Hasan to concede that he could not diagnose Jordan with food poisoning:
Q. Based on reasonable medical probability, doctor, isn’t it true that you cannot make a medical determination that Ms. Jordan suffered food poisoning; am I correct?
A. An absolute determination? I can make a probable estimation of what triggered her symptoms, but I can’t say absolutely what triggered her symptoms. Q. Okay.
11 A. Can I just say that’s why in my documentation I wrote likely viral gastroenteritis. In other words, Hasan testified in his capacity as a doctor that if a patient, like
Jordan, has viral gastroenteritis, she more likely than not has it due to food. That is
a statement of reasonable medical probability. See Schronk v. Laerdal Med. Corp.,
440 S.W.3d 250, 258 (Tex. App.—Waco 2013, pet. denied) (reasonable medical
probability means more likely than not). To hold otherwise is to demand doctors use
magic words when testifying, instead of crediting the substance of their testimony,
as the law requires. See Stodgill v. Tex. Emp. Ins. Ass’n, 582 S.W.2d 102, 105 (Tex.
1979) (when testifying about cause, medical expert’s testimony is sufficient if
substance of what he says conveys reasonable medical probability). Moreover, when
asked to agree that he could not diagnose Jordan with food poisoning to a reasonable
degree of medical probability, Hasan did not agree.
The majority also asserts that Hasan did not testify that “viral gastroenteritis—
as opposed to gastroenteritis from all sources—is more often than not caused by
food.” But this assertion is an equally indefensible mischaracterization of the record.
Hasan testified that the predominant cause of gastroenteritis in general is food
and that about 80% of gastroenteritis is viral in nature. Taken together, a jury could
reasonably infer from this testimony that most cases of viral gastroenteritis are
caused by food. But Hasan went further and directly testified that this was so:
12 Q. Is it your opinion that the viral gastroenteritis was brought about because of food? A. That’s the typical source of gastroenteritis.
Moreover, the defense’s own expert repeatedly characterized viral
gastroenteritis as a type of food poisoning before the jury. Richard Fleming Stier, “a
food scientist with an emphasis in food safety, sanitation, food processing, and
training,” testified on behalf of Murphy’s Deli. In his testimony, Stier:
• stated without qualification that Hasan diagnosed Jordan with “viral gastroenteritis, which is a form of food poisoning”; and
• described “viral gastroenteritis” as one of multiple “foodborne illnesses” more than once and reiterated that it is “a form of food poisoning.”
Thus, the majority’s premise that there is no evidence that Jordan had food poisoning
is contradicted by the record, including Murphy’s Deli’s own expert evidence.
The majority notes that Hasan did not order laboratory tests to identify the
virus that Jordan had. The majority implies that without this testing and similar
testing of the kolaches at Murphy’s Deli for pathogens, Jordan cannot show that the
kolache she ate was a cause in fact of her illness. But insistence on laboratory testing
is a demand for medical certainty, which is not required. See Hospadales v. McCoy,
513 S.W.3d 724, 737 (Tex. App.—Houston [1st Dist.] 2017, no pet.).
Hasan testified that viral gastroenteritis is typically diagnosed by clinical
observation of symptoms, not laboratory testing, for two reasons. First, the
pathogens that usually cause gastroenteritis can be hard to isolate in testing. Second,
13 laboratory testing for the specific cause of gastroenteritis is not cost-effective. Thus,
the imposition of a laboratory-testing requirement to show cause in fact as to food
poisoning would be tantamount to abolishing any legal remedy for it.
Contrary to the majority, however, courts generally do not insist on laboratory
testing to prove cause in fact in suits involving foodborne illnesses. As the Virginia
Supreme Court noted in Bussey v. E.S.C. Restaurants, suits “involving food
poisoning present unique circumstances because the primary source of evidence is
usually consumed and transmuted in the ordinary course of its use. As a result, most
cases will necessarily rely on circumstantial evidence.” 620 S.E.2d 764, 768 (Va.
2005). In Bussey, the court held the plaintiff’s proof sufficient based on evidence she
had eaten one meal in the last 36 hours, consisting of beef that smelled bad, and then
suffered a sudden illness, which was diagnosed as food poisoning. Id.
Nor is Virginia peculiar in this regard. Other jurisdictions have likewise
approved of reliance on circumstantial evidence to prove causation in food-
poisoning cases. E.g., Patterson v. Kevon, 818 S.E.2d 575, 579–81 (Ga. 2018); San
Francisco v. Wendy’s Int’l, 656 S.E.2d 485, 501 (W. Va. 2007); Wilson v. Circus
Hotels, 710 P.2d 77, 79–80 (Nev. 1985); Miller v. Atl. Bottling Corp., 191 S.E.2d
518, 519–20 (S.C. 1972); see David Polin, Proof of Liability for Food Poisoning, 47
AM. JUR. PROOF OF FACTS 3d 47 § 18 (1998) (“Circumstantial evidence may suffice
14 to prove food poisoning, or at least to prove a reasonable inference of it, in the
absence of definitive medical proof.”).
Texas courts have explicitly adopted this position as well. In Hebert v.
Loveless, the Ninth Court of Appeals held that a “causal connection” between
unwholesome food and illness “may be inferred by a balance of probabilities.” 474
S.W2d 732, 738 (Tex. App.—Beaumont 1971, writ ref’d n.r.e.). As the Hebert court
rightly recognized, the absence of “any direct evidence of the unwholesome nature
of the food or a condition therein which rendered it unfit for human consumption is
not dispositive.” Id. at 737. Our own court has reiterated this rule on more than one
occasion. See Evans v. MIPTT, L.L.C., No. 01-06-00394-CV, 2007 WL 1716443, at
*3 (Tex. App.—Houston [1st Dist.] June 14, 2007, no pet.) (mem. op.) (“This Court
has long recognized that a party may rely on circumstantial evidence to prove a claim
that a food product is defective.”); Jim Dandy Fast Foods v. Carpenter, 535 S.W.2d
786, 789 (Tex. App.—Houston [1st Dist.] 1976, no writ) (“The fact that food served
was unfit for human consumption can be shown by circumstantial evidence.”).
This is not to say that the consumption of food followed by sudden illness,
standing alone, is legally sufficient evidence of food poisoning. It is not. Sarti v. Salt
Creek, 85 Cal. Rptr. 3d 506, 512 (Cal. Ct. App. 2008) (observing that it is logically
fallacious to assume one event caused another merely because it preceded the other
and that this logical fallacy cannot be deployed “to win a food poisoning case”);
15 Valenti v. Great Atl. & Pac. Tea Co., 615 N.Y.S.2d 84, 85 (N.Y. App. Div. 1994)
(same). But in conjunction with the relatively rapid onset of illness, courts have
credited other circumstances as being probative of a food’s unwholesomeness. These
circumstances include an unusual appearance, flavor, odor, or texture. E.g., Williams
v. O’Charley’s, Inc., 728 S.E.2d 19, 22 (N.C. Ct. App. 2012) (chicken had bad
aftertaste and was dry); Bussey, 620 S.E.2d at 768 (beef had bad smell); Miller, 191
S.E.2d at 519 (soda included unexpected substance that was “repulsive in
appearance and had a vile smell and taste”); see also F.W. Woolworth Co. v. Garza,
390 S.W.2d 90, 92 (Tex. App.—San Antonio 1965, writ ref’d n.r.e.) (noting there
was no evidence of unwholesomeness of hamburger that “had no unusual odor, taste
and appearance” in suit alleging allergic reaction from food). Another probative
circumstance may be the plaintiff’s ingestion of little or no other food in proximity
to the emergence of food poisoning. E.g., Bussey, 620 S.E.2d at 768 (plaintiff had
only eaten one meal in last day-and-a-half). In addition, a factfinder may reasonably
infer unwholesomeness from evidence that others who ate the same food became
sick. E.g., Patterson, 818 S.E.2d at 579 (large numbers of people who ate same food
became ill); Turner v. Wilson, 86 S.E.2d 867, 869 (S.C. 1955) (several who ate same
food fell ill after eating it). Or, alternatively, a factfinder may draw the opposite
inference when those who eat the same food do not fall ill, though this circumstance
is evidentiary rather than conclusive. E.g., Wilson, 710 P.2d at 79 (jury could have
16 discounted bologna eaten by plaintiff almost three days earlier as possible alternative
source of illness given that sister ate same bologna and did not become ill).
Returning to our case, Jordan testified that she ate a kolache she bought from
Murphy’s Deli sometime between 9:00 and 9:30 on a Monday morning after having
blood drawn at a doctor’s appointment. She further testified that the kolache “tasted
a little off” when she ate it. This general description of the kolache’s taste also
appears in her medical records, indicating that she had told Hasan the same thing.
After Jordan ate the kolache, she soon became “nauseated” and “got really
bad stomach cramps,” followed by “several episodes of diarrhea.”
Jordan testified that she had not eaten anything other than the kolache since
dinner Sunday evening due to the early Monday morning blood draw, which
required her to be fasting. She had Sunday dinner with her husband and son. She did
not become ill afterward and neither did they. Before that, her last meal was Sunday
breakfast, which she ate with her husband. Again, neither of them became ill
afterward. Before that, Jordan’s last meal was Saturday dinner, which she ate with
her husband and son. None of them became ill after eating this meal either.
All of these Saturday and Sunday meals apparently were homecooked. Jordan
is the only one of the three who ate these meals who eventually became ill. The lone
difference in their diets was the suspect kolache she ate Monday morning.
17 Taken as a whole, these circumstances—the kolache’s ill taste, rapid onset of
the illness, limited number of other meals in the previous day-and-a-half, and
continued good health of those who shared these other meals—are some evidence,
when joined with Hasan’s diagnosis of viral gastroenteritis, that Jordan had food
poisoning and the kolache was a cause in fact of her food poisoning. Some evidence,
of course, is legally sufficient evidence. The majority errs in holding otherwise.
The majority effectively reads this evidence out of the record. For example,
the majority argues that there is “no expert testimony that would permit the jury to
find medical causation based on the kolache’s taste.” But the majority is mistaken.
First, the majority is mistaken about the record. The defense’s own expert
testified that the taste of the kolache was important. Stier’s testimony is clear:
Q. Isn’t it important to you that the kolache’s taste was, as she said “dubious”? A. Yeah. And the reason I think it’s important is if something tastes dubious, you know, it shouldn’t be consumed.
Thus, there is in fact expert testimony from which the jury could have reasonably
found that the kolache’s reported taste was an indicator that it was unwholesome or
unfit for consumption in some way and hence the cause of Jordan’s sudden illness.
Second, to the extent the majority contends otherwise, it is mistaken that
expert evidence is required to prove that food with an untoward taste is
unwholesome. Expert evidence is not required to prove causation when the common
18 understanding of laypersons would enable them to find from the evidence, with
reasonable probability, a relationship between an alleged cause and injury. Mack
Trucks v. Tamez, 206 S.W.3d 572, 583 (Tex. 2006). For example, in Jelinek, the
Court discussed a prior decision, in which a previously healthy woman suffered
watering eyes, blurred vision, headaches, and swelling of the breathing passages
after exposure to leaking chemicals. 328 S.W.3d at 533–34 (discussing Morgan v.
Compugraphic Corp., 675 S.W.2d 729 (Tex. 1984)). Under these circumstances, the
Court noted, “lay testimony sufficed to connect the specific injury to the negligence
with no evidence of causation beyond the leaking chemicals.” Id. at 534. That an
untoward taste indicates unwholesomeness is likewise well within the ken of
laymen. Thus, in conjunction with the suddenness of Jordan’s illness and Hasan’s
diagnosis, the kolache’s taste is evidence that a jury reasonably could credit in
deciding that the kolache caused her illness and resulting injuries. See id. at 533
(“When lay testimony is credited as evidence of causation, it usually highlights a
connection between two events that is apparent to a casual observer.”).
The majority says its holding is logically compelled by our sister court’s
decision in Smith v. Landry’s Crab Shack. But Smith is distinguishable on more than
one ground. First, the plaintiff in that suit did not testify that the suspect food she ate
seemed unwholesome in any way. 183 S.W.3d at 514. Second, the plaintiff’s expert
testified only that “none of the circumstances” of her illness were “inconsistent with
19 a diagnosis of gastroenteritis.” Id. He did not actually diagnose her as having this
illness to a degree of reasonable medical probability. Id. at 515. In contrast, Jordan
testified that the kolache she ate “tasted a little off,” and Hasan testified that she had
viral gastroenteritis to a degree of reasonable medical probability. Thus, the majority
is quite mistaken when it describes the evidence in Smith as being “even stronger
than the evidence presented here.” On the contrary, the opposite is the case.
Smith is also inapposite in that the lone issue before the court in that summary-
judgment case was the legal sufficiency of the causation testimony of the plaintiff’s
expert. See id. at 514. The parties’ dispute focused on the expert’s testimony, so the
court’s opinion did too. But the court did not hold, as the majority seems to think,
that expert testimony is invariably required to establish a causal connection between
a food and a medically diagnosed foodborne illness. As discussed above, a plaintiff
can prove cause in fact with circumstantial evidence. Hebert, 474 S.W.2d at 737–
38. Smith recognizes as much when it states that “the test for cause in fact is whether
the illness would not have occurred if Smith had not eaten at the restaurant” in one
breath, and in the next notes that a “finding of cause in fact may be based on
circumstantial evidence.” 183 S.W.3d at 514. The majority effectively overreads
Smith’s resolution of the parties’ dispute about the sufficiency of expert testimony
as foreclosing circumstantial proof even though Smith explicitly says otherwise.
20 The majority’s reliance on our decision in Farroux v. Denny’s Restaurants is
even more strained. In that case, we affirmed summary judgment for the defendant
because it had conclusively disproved causation. 962 S.W.2d 108, 110 (Tex. App.—
Houston [1st Dist.] 1997, no pet.). There, among other things, the plaintiff’s treating
physician “said there were too many possibilities to determine what made the
plaintiff ill.” Id. at 111. The lone evidence to the contrary introduced by the plaintiff
was his own affidavit, in which he swore that his doctor had told him otherwise. Id.
The plaintiff’s affidavit, however, “was directly contrary to the plaintiff’s deposition
testimony” on the subject, so we disregarded the affidavit as a sham. Id.
The majority argues that Farroux is like our case inasmuch as Jordan’s doctor,
Hasan, testified that viral gastroenteritis has numerous causes. He did. But when he
did so he was discussing the specific pathogens that can cause gastroenteritis. He
was not, as the majority implies, testifying about non-foodborne causes of the illness.
At any rate, Hasan never disavowed his diagnosis of viral gastroenteritis or his
testimony that food more often than not is the cause of viral gastroenteritis. Unlike
the doctor in Farroux, Hasan simply did not testify that the possible causes of
Jordan’s illness were too numerous to allow him to ascertain its cause.
Moreover, to the extent there are non-foodborne causes of viral
gastroenteritis, they are irrelevant unless evidence shows one of them to be a
plausible alternative cause in fact of Jordan’s illness. See Transcontinental Ins. Co.
21 v. Crump, 330 S.W.3d 211, 218 (Tex. 2010). An expert is not required to disprove
or discredit every conceivable cause other than the one he espouses. Id. He need only
reasonably exclude another cause if evidence shows it is a plausible alternative that
could be negated. Id.; see also Havner v. E–Z Mart Stores, 825 S.W.2d 456, 460
(Tex. 1992) (plaintiff not required to prove case “to an absolute degree of certainty”
and does not have to conclusively disprove possibility that defendant did not cause
his injuries). The majority does not refer to any evidence plausibly suggesting that
something other than food may have caused Jordan’s viral gastroenteritis.
Instead, the majority simply posits that Hasan’s failure to rule out other
possible causes of Jordan’s illness is “problematic given Jordan’s purchase of food
on hospital grounds, where Jordan could have encountered any number of germ
sources.” But the majority does not cite evidence suggesting that Jordan was exposed
to other germ sources at the hospital that could have possibly caused her viral
gastroenteritis. The majority does not do so because the record lacks this evidence.
On the contrary, the evidence on this particular subject refutes the majority.
Hasan was asked about staphylococcus, which he agreed can cause rapid-onset
gastroenteritis. Hasan likewise agreed that staphylococcus is frequently found in
hospitals. However, he unequivocally testified that staphylococcus is bacterial, not
viral. Thus, Hasan’s diagnosis that Jordan more likely than not had viral
22 gastroenteritis was inconsistent with staphylococcus causing Jordan’s illness. After
Hasan was questioned about the role of staphylococcus, the exchange concluded:
Q. So if staphylococcus is not viral, then staphylococcus can’t be the cause of her viral gastroenteritis. Am I correct?
A. Could you say that again? I’m sorry. Q. Okay. If staphylococcus is not viral, which we’ve established?
Q. All right. Then the staphylococcus can be—cannot be the cause of her viral gastroenteritis? A. Right. The record does not contain evidence of any other possible cause of viral
gastroenteritis that Jordan could have possibly encountered at the hospital. Thus,
when the majority writes of “any number of germ sources,” it is speculating. If
experts had to rule out purely speculative alternatives in order for their testimony to
constitute reliable evidence of causation, their testimony seldom would pass muster.
See Crump, 330 S.W.3d at 218. But Texas law does not require such rigor. Id.; see
also Bustamante v. Ponte, 529 S.W.3d 447, 457 (Tex. 2017) (plaintiff need not
speculate about other possible unknown causes and then disprove speculation); JLG
Trucking v. Garza, 466 S.W.3d 157, 162 (Tex. 2015) (citing with approval court of
appeals decision rejecting argument that plaintiff had to negate other causes of her
injury due to lack of supporting evidence in the record for these alternatives).
23 In sum, neither Smith nor Farroux supports the majority’s legal-sufficiency
analysis. Further, the majority’s holding is contrary to Texas’s longstanding rule that
food poisoning may be proved by circumstantial evidence. The majority gives short
shrift to multiple circumstances that together support the jury’s verdict, including:
• Jordan’s testimony that the kolache she ate tasted dubious;
• Stier’s testimony that a dubious taste signifies food should not be eaten;
• Jordan’s testimony that she fell ill soon after eating the kolache; • Hasan’s diagnosis of Jordan’s illness as viral gastroenteritis; • Hasan’s testimony that viral gastroenteritis is typically foodborne;
• Stier’s testimony that viral gastroenteritis is a foodborne illness; and • Jordan’s testimony about the other food she ate with her family during the 36 hours or so before her illness and her family’s lack of illness. These circumstances are legally sufficient evidence of cause in fact. Indeed, these
circumstances are factually sufficient evidence of cause in fact. The majority reaches
its contrary conclusion by adopting a divide-and-conquer approach, in which it
considers each circumstance in isolation from the remainder of the evidence, rather
than considering whether these circumstances in combination are sufficient. In doing
so, the majority disobeys our law’s directive that we consider circumstantial
evidence in its complete context. See Jelinek, 328 S.W.3d at 532 (when considering
circumstantial evidence, we must view each piece, not in isolation, but in light of all
known circumstances).
24 As our Supreme Court has observed, “the common law has applied more
stringent rules to sales of food than to sales of other merchandise” due to the danger
posed to the public by unwholesome food. Jacob E. Decker & Sons v. Capps, 164
S.W.2d 828, 829 (Tex. 1942). The majority’s holding, which sets a high bar for
proving liability in food-poisoning cases, conflicts with Texas’s stringent public
policy as to food sales. This stringent public policy is embodied by the well-settled
principle that a plaintiff may prove causation with circumstantial evidence in food-
poisoning cases. The circumstances of this case amply support the jury’s verdict.
Conclusion
A jury rendered a verdict in Jordan’s favor. The majority upends the verdict
on the unpreserved ground that the jury did not hear legally sufficient evidence of
cause in fact, but the majority misunderstands what the law requires in terms of
proof. As there is sufficient evidence of cause in fact, I respectfully dissent.
Gordon Goodman Justice
Panel consists of Chief Justice Radack and Justices Goodman and Farris.
Justice Goodman, dissenting.