In the
United States Court of Appeals
For the Seventh Circuit
____________________
No. 17‐3390
GREGORY AREGOOD, JR., et al.,
Plaintiffs‐Appellants,
v.
GIVAUDAN FLAVORS CORPORATION,
et al.,
Defendants‐Appellees.
____________________
Appeal from the United States District Court for the
Southern District of Indiana, Indianapolis Division
No. 1:14‐cv‐00274‐SEB‐TAB — Sarah Evans Barker, Judge.
____________________
ARGUED JUNE 1, 2018 — DECIDED SEPTEMBER 13, 2018
____________________
Before RIPPLE, KANNE, and BRENNAN, Circuit Judges.
BRENNAN, Circuit Judge. More than twenty current and for‐
mer employees at the ConAgra microwave popcorn plant in
Rensselear, Indiana sued various manufacturers and suppli‐
ers of butter flavorings that contained the chemical diacetyl,
which if inhaled can cause a respiratory disease called “pop‐
corn lung.” All defendants were dismissed except Givaudan
Flavors Corporation (“Givaudan”), a long‐time supplier to
2 No. 17‐3390
the plant, which faced claims under Indiana product liability
law for strict liability, failure to warn, negligence, and design
defect.
Givaudan moved for and eventually received summary
judgment in full. The employee plaintiffs appeal, contending
that the district court erred in reviewing the evidence and ap‐
plying the law. Summary judgment for the flavor manufac‐
turer Givaudan is proper on many of plaintiffs’ claims, but
not that Givaudan failed to warn plaintiffs that its products
contained a dangerous substance. Whether an exception to
that duty to warn—the sophisticated intermediary doctrine—
applies to the employer ConAgra and exonerates Givaudan is
a fact question, so we remand for trial on that claim.
I. Background
As we review summary judgment in favor of the movant
Givaudan, we consider undisputed facts, all reasonable infer‐
ences from undisputed facts are drawn in favor of the non‐
movant employees, and we view disputed evidence in the
light most favorable to the employees. See Anderson v. Liberty
Lobby, Inc., 477 U.S. 242, 255 (1986); Weigle v. SPX Corp., 729
F.3d 724, 730 (7th Cir. 2013).
A. Factual
The employee plaintiffs are all current or former workers
at the Orville Redenbacher microwave popcorn plant in Rens‐
selaer, Indiana. Their employer is or was ConAgra (a division
of ConAgra Snack Food Group, not a party to this case). The
No. 17‐3390 3
employees worked at ConAgra when Givaudan1 manufac‐
tured and supplied to ConAgra butter flavorings containing
diacetyl, an organic additive with a buttery flavor. In the early
1990s, Givaudan began supplying these flavorings with diac‐
etyl to ConAgra for use in its plants, and by the mid‐2000s
Givaudan had sold these flavorings to ConAgra for use at its
Rensselear plant.2
Exposure to diacetyl, the employees allege, resulted in
their developing respiratory illnesses. When inhaled, diacetyl
can cause bronchiolitis obliterans—commonly referred to as
ʺpopcorn lungʺ—the inflammation and obstruction of the
smallest airways of the lungs. Symptoms of this disease in‐
clude a dry cough, shortness of breath, wheezing, fatigue, and
can lead to worse personal injuries.
In support of their claims, the employees have offered
opinions of various expert witnesses. An occupational physi‐
cian opined that the levels of diacetyl in the air at the Rens‐
selear plant when plaintiffs worked there likely caused bron‐
chiolitis obliterans. Another doctor examined the plaintiffs and
diagnosed them with flavoring‐induced bronchiolitis obliterans
caused by exposure to diacetyl at the Rensselear plant. An ep‐
idemiologist and occupational physician who had reviewed
the documentary and testimonial record opined Givaudan
1 Givaudan Flavors Corporation is the successor to several predecessors‐
in‐interest, including Tastemaker, Fries & Fries, Inc., Givaudan Roure Fla‐
vors Corporation, Givaudan Roure Corporation, Givaudan Corporation,
and Roure Corporation. We refer collectively to the current company and
its corporate predecessors as “Givaudan.”
2 Givaudan also makes diacetyl‐free butter flavorings, and has done so
since 1991.
4 No. 17‐3390
should have known that diacetyl caused lung disease, as well
as that Givaudan withheld from ConAgra the health risks of
its butter flavoring.
“Popcorn lung” has been an issue in the microwave pop‐
corn industry since at least the early 2000s when bronchiolitis
obliterans was discovered at a microwave popcorn plant in
Missouri (not owned or operated by ConAgra). In 2001, The
Wall Street Journal published an article on that outbreak, and
a federal health agency began a study to try to determine its
cause. After the Missouri occurrence and the national press,
ConAgra inspected its Rensselaer plant and concluded that its
processes and ventilation systems were different. ConAgra
wrote to its Rensselear employees about the differences be‐
tween the plants, concluding that ConAgra’s employees did
not face the same risks as in Missouri. Still, ConAgra devel‐
oped procedures for breathing tests for its employees, and
also cooperated with the federal health study, including im‐
plementing the study’s recommendation of ventilation im‐
provements.
Givaudan knew about diacetyl and its harmful effects be‐
fore the Missouri outbreak, the attendant publicity, and the
federal health study. In the mid‐1980s, Givaudan learned
from its trade association that inhaling diacetyl was “harm‐
ful” and “capable of producing system toxicity.” In the 1990s,
three employees at Givaudan’s plant in Cincinnati were diag‐
nosed with bronchiolitis obliterans, and one died. In response,
Givaudan retained an occupational physician who confirmed
that the two surviving employees had contracted the disease.
That physician recommended further investigation into the
cause of the disease, but he was terminated. Givaudan was
sued twice for claims of lung injury from diacetyl exposure.
No. 17‐3390 5
Givaudan also responded to these diagnoses by promul‐
gating procedures designed to reduce the risk of personal in‐
juries from the use and handling of diacetyl. Those proce‐
dures required that “[w]henever liquid [d]iacetyl or a product
where liquid [d]iacetyl is present is to be used, a respirator
with chemical resistant gloves must be worn.” Further, “[a]ny
room containing [d]iacetyl in a liquid state must be labeled
respirator required.”
Givaudan likewise developed and implemented an em‐
ployee protection program. To do so, Givaudan hired three
physicians from the University of Cincinnati, who were re‐
quired to sign confidentiality agreements. A pulmonary toxi‐
cologist testified it was “somewhat fair” to say he was unable
to fully investigate the circumstances at Givaudan’s plant.
That doctor concluded Givaudan did not want to identify the
precise nature of the problem, and did not want him to put
anything in writing. An occupational physician stated he had
narrowed the chemicals suspected of causing the outbreak “to
a manageable list that could have been investigated,” and he
had offered the resources of the University of Cincinnati to
conduct the study, but Givaudan did not give him the green
light. Givaudan documents suggest the list of chemicals sus‐
pected as the cause of the bronchiolitis obliterans outbreak was
narrowed to three, including diacetyl. One of these physicians
testified that Givaudan limited the resources used to discover
the cause of the outbreak, and that he was under pressure to
not let employees know of any danger.
The parties dispute what information the flavor manufac‐
turer Givaudan gave to the employees’ company ConAgra
about diacetyl and how it should be handled. Federal Occu‐
6 No. 17‐3390
pational Safety and Health Administration regulations re‐
quire chemical manufacturers such as Givaudan to evaluate
and to classify each product sold and to provide with the
product a “material safety data sheet.” 29 C.F.R.
§ 1910.1200(g)(5) (2018). Data sheets accompanied Givau‐
dan’s shipments of butter flavoring containing diacetyl to
ConAgra. Those sheets did not disclose how much (or even
if) diacetyl was present in certain flavors, as Givaudan con‐
sidered that information a trade secret. The data sheets did
not include warnings that inhalation of fumes from butter fla‐
vorings made with diacetyl could cause permanent lung in‐
jury or bronchiolitis obliterans. While Givaudan informed
ConAgra generally that inhalation of the butter flavoring
could be “irritating to nose, throat, and lungs,” Givaudan also
stated on the sheet that the flavoring had “no known health
hazards.”
Givaudan supplemented the language in the material
safety data sheets over the time it sold butter flavorings to
ConAgra. Givaudan added that “ventilation meeting accepta‐
ble standards was recommended,” but it did not say employ‐
ees handling the flavoring with diacetyl should wear respira‐
tors, unless they were responding to spills or leaks. For “per‐
sonal protection,” the material safety data sheet stated, “res‐
piratory protection is not normally required” in well venti‐
lated areas, although in confined or poorly ventilated areas or
if material is toxic by inhalation, “the use of approved respir‐
atory protection is recommended.”
Givaudan’s material safety data sheets did not state that
the flavorings were hazardous. Givaudan’s toxicologist has
admitted that OSHA required Givaudan, for any flavor con‐
taining more than one percent of a hazardous chemical, to list
No. 17‐3390 7
in the material safety data sheet any health hazards and to
give safety instructions for that chemical. Givaudan sold
ConAgra five butter flavorings which contained more than
one percent diacetyl.
While diacetyl was not listed on the material safety data
sheets, ConAgra was aware that Givaudan’s butter flavorings
contained the additive. ConAgra was not aware, however, of
all the components of the butter flavorings and the levels of
diacetyl they contained. ConAgra’s representative testified
ConAgra understood the butter flavorings bought from
Givaudan were safe for its workers. ConAgra expected that
“flavor companies would disclose all the hazard information
that they possessed about flavors that were being provided to
ConAgra,” and ConAgra relied on “flavor suppliers to tell if
there was a particular hazard that could happen to” their em‐
ployees. ConAgra represented to its employees that it strictly
followed ingredient manufacturers’ instructions for the han‐
dling and use of all ingredients.
B. Procedural
The employees brought a multiple count complaint in In‐
diana state court asserting Givaudan (and other defendants
since dismissed) were liable for their injuries based on
(1) strict liability for providing a defective product, (2) failure
to warn about the butter flavoring causing respiratory dis‐
ease, (3) common law negligence, and (4) defective product
design, under both Indiana common law and the Indiana
Product Liability Act (“the Act”), IND. CODE § 34‐20‐1‐1
through § 34‐20‐9‐1. The defendants removed the case to
Indiana federal court under diversity jurisdiction; 28 U.S.C.
§§ 1332, 1441. Indiana law governs all of plaintiffs’ claims. Erie
8 No. 17‐3390
Railroad Co. v. Tompkins, 304 U.S. 64 (1938); C.W. ex rel. Wood
v. Textron, Inc., 807 F.3d 827, 831 (7th Cir. 2015).
First, the district court granted summary judgment to
Givaudan on plaintiffs’ claims for strict liability, negligence,
and failure to warn. Some months later,3 the district court
granted summary judgment to Givaudan on plaintiffs’ design
defect claim.4
II. Discussion
A. Standard of Review
We review de novo the district court’s grant of summary
judgment, which is proper only if there is no genuine issue of
material fact. Piltch v. Ford Motor Co., 778 F.3d 628, 631 (7th
Cir. 2015). A genuine issue of material fact exists whenever
“there is sufficient evidence favoring the non‐moving party
for a jury to return a verdict for that party.” Anderson, 477 U.S.
at 249. Summary judgment is proper only “if the admissible
evidence considered as a whole shows that ‘there is no genu‐
ine dispute as to any material fact and the movant is entitled
to judgment as a matter of law.’” Dynegy Marketing & Trade v.
Multiut Corp., 648 F.3d 506, 517 (7th Cir. 2011) (quoting FED.
3 The Honorable Larry McKinney presided over this case through the dis‐
trict court’s June 1, 2017 decision and order granting summary judgment
on the claims for strict liability, negligence, and failure to warn. After
Judge McKinney passed away, the Honorable Sarah Evans Barker pre‐
sided over the case and issued the October 18, 2017 decision and order on
the remaining claim of design defect.
4 In its October 18, 2017 decision and order the district court also denied
the plaintiffs’ motion to reconsider the district court’s June 1, 2017 sum‐
mary judgment order.
No. 17‐3390 9
R. CIV. P. 56(a)). The nonmovant bears the burden of demon‐
strating that such a genuine issue of material fact exists. See
Matsushita Elec. Indus. Co. v. Zenith Radio Corp., 475 U.S. 574,
586–87 (1986); Trask‐Morton v. Motel 6 Operating L.P., 534 F.3d
672, 677 (7th Cir. 2008).
B. Failure to Warn, and the Sophisticated Intermediary
Doctrine
In Indiana, when a user or consumer sues a manufacturer
for physical harm caused by a product, the Act governs the
case regardless of the legal theory. Piltch, 778 F.3d at 632 (cit‐
ing IND. CODE § 34‐20‐1‐1). Under the Act, a plaintiff must es‐
tablish that “(1) he or she was harmed by a product; (2) the
product was sold ‘in a defective condition unreasonably dan‐
gerous to any user or consumer; (3) the plaintiff was a fore‐
seeable user or consumer; (4) the defendant was in the busi‐
ness of selling the product; and (5) the product reached the
consumer or user in the condition it was sold.” Id. “The Act
provides that a plaintiff can satisfy the second element–that
the product was defective–by showing one of the following: a
design defect, a manufacturing defect, or a failure to warn.”
Ritchie v. Glidden Co., 242 F.3d 713, 720 (7th Cir. 2001) (citing
Natural Gas Odorizing, Inc. v. Downs, 685 N.E.2d 155, 161 (Ind.
Ct. App. 1997)).
“[I]n an action based on … an alleged failure to provide
adequate warnings or instructions regarding the use of the
product, the party making the claim must establish that the
manufacturer or seller failed to exercise reasonable care under
the circumstances in … providing the warnings or instruc‐
tions.” IND. CODE § 34‐20‐2‐2. Under Indiana law, there is a
duty to warn reasonably foreseeable users of all “latent dan‐
ger[s] inherent in the product’s use.” First Nat’l Bank & Tr.
10 No. 17‐3390
Corp v. Am. Eurocopter Corp., 378 F.3d 682, 690 (7th Cir. 2004)
(quoting Taylor v. Monsanto, 150 F.3d 806, 808 (7th Cir. 1998),
overruled on other grounds by Hill v. Tangherlini, 724 F.3d 965
(7th Cir. 2013)); Nat. Gas Odorizing, 685 N.E.2d at 161 (“[A] la‐
tent danger will, without more, cause the product to be un‐
reasonably dangerous as marketed.”) The duty is to warn of
the hidden danger itself, or the risks from a recognized dan‐
ger that far exceed that contemplated by the ordinary con‐
sumer. Am. Eurocopter Corp., 378 F.3d at 690.
For the employees to show that the butter flavorings were
defective under the Act and to prevail on their failure to warn
claim, they must show Givaudan had a duty to adequately
warn about a latent dangerous characteristic, and Givaudan
breached that duty by failing to so warn. See Nat. Gas Odoriz‐
ing, 685 N.E.2d at 161. Rather than Givaudan breaching the
duty, the district court ruled that Givaudan discharged its
duty to warn by reasonably relying on ConAgra, the plain‐
tiffs’ employer, to warn them about any dangers from diace‐
tyl.
The duty to warn usually cannot be delegated, although
Indiana law recognizes an exception to this general rule,
called the sophisticated intermediary doctrine. Id. at 163 (cit‐
ing RESTATEMENT (SECOND) OF TORTS, § 388). The doctrine has
three requirements: “(1) the product is sold to an intermedi‐
ary with knowledge or sophistication equal to that of the
manufacturer; (2) the manufacturer adequately warns this in‐
termediary; and (3) the manufacturer can reasonably rely on
the intermediary to warn the ultimate consumer.” Am. Eu‐
rocopter Corp., 378 F.3d at 691. Courts interpreting Indiana law
have enumerated additional factors to be considered when
deciding if the sophisticated intermediary doctrine applies:
No. 17‐3390 11
[T]he likelihood or unlikelihood that harm will
occur if the [intermediary] does not pass on the
warning to the ultimate user, the trivial nature
of the probable harm, the probability or improb‐
ability that the particular [intermediary] will
not pass on the warning and the ease or burden
of the giving of the warning by the manufac‐
turer to the ultimate user.
Nat. Gas Odorizing, 685 N.E.2d at 163 (brackets in original)
(quoting Dole Food v. North Carolina Foam Indus., Inc., 188 Ariz.
298, 303 (App. 1996)); see also Ritchie, 242 F.3d at 724; Am.
Eurocopter, 378 F.3d at 692.
Whether a manufacturer has discharged its duty under the
sophisticated intermediary doctrine is “almost always a ques‐
tion for the trier of fact.” Ritchie, 242 F.3d at 724; Am. Eurocop‐
ter, 378 F.3d at 692 (“In general, summary judgment should
not be granted in favor of a manufacturer based on the sophis‐
ticated intermediary doctrine.”) (citation omitted); see also
Hathaway v. Cintas Corp. Svcs., Inc., 903 F. Supp. 2d 669, 676
(N.D. Ind. 2012) (citation omitted); Nat. Gas Odorizing, 685
N.E.2d at 164 (“Whether a manufacturer has discharged its
duty under the sophisticated intermediary doctrine is almost
always a question for the trier of fact.”) This admonition fol‐
lows from the fact‐bound inquiries the parties litigate: What
was the knowledge or sophistication of the intermediary?
How did that knowledge or sophistication compare with that
of the manufacturer? Were warnings given, and if so were
12 No. 17‐3390
they adequate?5 What harm could occur if warnings were not
relayed, and how much? What is the ease or burden of the
giving of the warning? Was it reasonable for the manufacturer
to rely on the intermediary to warn the ultimate users of dan‐
ger? A jury, the archetypal determiner of “reasonableness,”
usually makes those decisions,6 although this rule has excep‐
tions. See Am. Eurocopter Corp. 378 F.3d at 691–93; Taylor, 150
F.3d at 808–09; Hathaway, 903 F. Supp. 2d at 675–78.
5 “In general, the adequacy of the warning is a question of fact for the
jury.” Am. Eurocopter Corp., 378 F.3d at 691 (citing Nat. Gas Odorizing, 685
N.E.2d at 161).
6 See RESTATEMENT (SECOND) OF TORTS § 388, cmt. n, which notes the rea‐
sonableness question proper for the jury:
Giving to the third person through whom the chattel is
supplied all the information necessary to its safe use is not
in all cases sufficient to relieve the supplier from liability.
It is merely a means by which this information is to be
conveyed to those who are to use the chattel. The question
remains whether this method gives a reasonable assurance that
the information will reach those whose safety depends upon
their having it. All sorts of chattels may be supplied for the
use of others, through all sorts of third persons under an
infinite variety of circumstances. This being true, it is ob‐
viously impossible to state in advance any set of rules
which will automatically determine in all cases whether
one supplying a chattel for the use of others through a
third person has satisfied his duty to those who are to use
the chattel by informing the third person of the dangerous
character of the chattel, or of the precautions which must
be exercised in using it in order to make its use safe.
(Emphasis added.)
No. 17‐3390 13
The employees’ failure to warn claim focuses on Givau‐
dan’s manufacturing and supply of butter flavorings, their ac‐
companying material safety data sheets, and the lack of warn‐
ings that the flavorings contained diacetyl which can cause
respiratory disease. As a result of Givaudan’s failure to ade‐
quately instruct and warn of the dangers of diacetyl, the em‐
ployees contend Givaudan’s butter flavorings were defective
and unreasonably dangerous when put to their reasonably
anticipated use. Givaudan responds that the district court cor‐
rectly granted it summary judgment because it relied on the
employees’ sophisticated and knowledgeable employer,
ConAgra, to warn its employees of any dangers associated
with Givaudan’s butter flavorings. Givaudan asserts its mate‐
rial safety data sheets sufficiently acquitted its responsibilities
about the use and handling of diacetyl. Givaudan maintains
ConAgra knew everything Givaudan did.
The district court concluded on summary judgment that
the duty to warn had effectively passed from Givaudan the
flavor manufacturer to ConAgra the employer for two rea‐
sons: (1) “neither Givaudan, nor its hired professionals, nor
its trade association discovered any connection between res‐
piratory problems at flavoring plants and diacetyl in the
1990s” and (2) “ConAgra knew as much if not more than
Givaudan with respect to the danger of butter flavors.”7 But
in arriving at these reasons, the district court did not always
adhere to the rule that in considering a summary judgment
motion, “[t]he evidence of the non‐movant is to be believed,
and all justifiable inferences are to be drawn in [the non‐mo‐
vant’s] favor.” Anderson, 477 U.S. at 255; Sojka v. Bovis Lend
7 2017 WL 2378258 at *10–*11.
14 No. 17‐3390
Lease, Inc., 686 F.3d 394, 397 (7th Cir. 2012). Factual inferences
were made contrary to the employees’ evidence.
In its first reason—that Givaudan did not discover any
connection in the 1990’s between employees’ respiratory
problems at its plants and diacetyl—the district court failed to
properly account for the employees’ evidence. This included
the occurrences of bronchiolitis obliterans at Givaudan’s plant,
the resulting injuries to Givaudan employees, the investiga‐
tion of the outbreak, and the development and implementa‐
tion of an employee protection program at Givaudan related
to the outbreak.
Givaudan responds that its 1990s investigation related
specifically to diacetyl, not butter flavorings. That characteri‐
zation does not give full weight, though, to what transpired
at Givaudan’s predecessor companies, and the policies and
procedures Givaudan adopted as a result of the bronchiolitis
obliterans outbreak. Those policies and procedures governed
the use and handling of diacetyl, and were designed to reduce
the risk of personal injuries. The district court saw no differ‐
ence in these policies for liquid diacetyl, and diacetyl in a mix‐
ture, such as included in butter flavoring. Givaudan bets
heavily on this distinction as relieving it from having to pass
on its knowledge from the 1980s and 1990s to ConAgra. But
while a jury could find that the policies Givaudan instituted
applied to only pure liquid diacetyl, a jury also could read the
policy to apply to mixtures (“or a product where liquid diac‐
etyl is present”). That the policies could be so read renders
this a fact question not for summary judgment. Givaudan
pushes back, seeing a distinction in the procedures for raw
ingredients versus for a mixture, and proclaims a sophisti‐
No. 17‐3390 15
cated intermediary would know the difference. Such an inter‐
mediary cannot know such a difference, though, if Givaudan
does not relay the quantity of the substance to ConAgra.
Givaudan maintains this information is a trade secret and it
chose not to include that information in the material data
safety sheets.
Also on this first reason, the district court’s conclusion
does not fully account for Givaudan cutting short the work
done by the physicians whom it retained to study the bronchi‐
olitis obliterans outbreak in the 1990s, as well as Givaudan
placing restrictions on these physicians’ work and its docu‐
mentation. A jury could reasonably conclude that since then
Givaudan has known of diacetyl’s dangers, and the need to
protect its own employees and the users of its products.
The district court’s second reason for applying the sophis‐
ticated intermediary doctrine was that “ConAgra knew as
much if not more than Givaudan with respect to the danger
of butter flavors.” The parties pull and tug on this question,
Givaudan citing information it believes ConAgra knew or
should have known, and ConAgra contrasting what it knew
from Givaudan with what ConAgra found out later. The rec‐
ord, viewed through the proper lens, reveals factual disputes
and discrepancies between what Givaudan knew and
ConAgra knew.
This second reason incorrectly relied on Givaudan’s view
of the evidence, rather than the employees’. The only source
cited for some of the critical information related to ConAgra’s
alleged knowledge of flavoring‐related problems was Givau‐
dan’s interrogatory answers. Some information, such as
ConAgra’s size and sophistication, does not even relate to
ConAgra’s knowledge of the dangers of diacetyl. By failing to
16 No. 17‐3390
credit evidence that contradicted its key factual conclusions,
the district court improperly “weigh[ed] the evidence” and
resolved disputed issues in favor of the moving party, Ander‐
son, 477 U.S. at 249, inappropriate at the summary judgment
stage.
On this second reason, the district court also did not view
the employees’ evidence in a light most favorable to them.
Givaudan had received information from its trade association
about diacetyl’s harmful effects. Givaudan had suffered its
own outbreak of bronchiolitis obliterans. Givaudan had hired
scientists to investigate who narrowed the cause of the out‐
break to three suspected chemical compounds, one of which
was diacetyl. Injuries and a death among its employees re‐
sulted in Givaudan implementing its own policies aimed spe‐
cifically at diacetyl. From these facts, a jury could conclude
that Givaudan knew more than ConAgra about diacetyl and
its harmful effects, the decision point as to whether or not the
sophisticated intermediary doctrine applies.
Another example of differing knowledge between Givau‐
dan and ConAgra is when employees should, or need not, use
a respirator when handling diacetyl or products containing
diacetyl. Givaudan contends “there is no evidence to suggest
that the respirator requirement was implemented to prevent
lung disease.” To Givaudan, “the procedures were imple‐
mented months before Tastemaker first discovered that it had
any employees with breathing issues and only to prevent eye
irritation, not lung disease.” But the evidence on which
Givaudan relies is from an individual who cannot place the
implementation of the policy before the bronchiolitis obliterans
outbreak, and who had no first‐hand knowledge of the reason
for the policy.
No. 17‐3390 17
An additional reasonable inference for the employees re‐
jected by the district court was in the memoranda ConAgra
sent its employees following the bronchiolitis obliterans out‐
break in Missouri. ConAgra attempted to reassure its employ‐
ees that Rensselaer had different handling procedures and
ventilation, so such an outbreak would not be repeated in In‐
diana. A reasonable inference from the memoranda is that
ConAgra concluded there was a ventilation problem at the
other company’s Missouri plant, not with Givaudan’s butter
flavoring. The district court rejected this inference as implau‐
sible, but a jury makes that call.
Given this view of the evidence, the sophisticated interme‐
diary doctrine does not fit. ConAgra is not bound by the same
regulations as Givaudan, as an employer can rely on the ma‐
terial safety data sheets provided by the manufacturer.
ConAgra relied on the warnings and safe handling instruc‐
tions in Givaudan’s material safety data sheets. Givaudan
knew this. Givaudan did not share with ConAgra the possi‐
bility that diacetyl could be the source of bronchiolitis obliterans
among Givaudan’s workers.
These examples rebut Givaudan’s contention that while
Givaudan supplied ConAgra’s Rensselear plant with butter
flavorings containing diacetyl, ConAgra had the same infor‐
mation as Givaudan. A jury, rather than the district court by
summary judgment, should resolve these factual disputes
about what the manufacturer knew versus what the employer
knew, and thus whether the employer was a sophisticated in‐
termediary.
18 No. 17‐3390
Now summary judgment is not precluded in all cases with
a defense that the sophisticated intermediary doctrine re‐
leases a manufacturer. But Givaudan’s attempts to analogize
this case to those precedents fail.
Like the helicopter manufacturer and the helicopter pas‐
sengers in American Eurocopter, Givaudan contends it and
ConAgra had the same level of knowledge of the possible
harm. Yet as noted, genuine and material factual disputes ex‐
ist about the parties’ relative knowledge of diacetyl and the
butter flavoring. American Eurocopter also involved unique
facts: business executives had been warned explicitly and re‐
peatedly from exiting a helicopter while the rotor blades were
in motion, yet they insisted on so disembarking. 378 F.3d at
693. Here, ConAgra did not warn its employees that inhaling
diacetyl could cause permanent lung damage. Unlike in
American Eurocopter, the ConAgra employees did not refuse
to wear a respirator or other items of personal protection fol‐
lowing an explicit or repeated warning, as they could not re‐
ject a warning they were never provided.
Another decision Givaudan relies on, Taylor, also differs
materially. In Taylor, employees of a corporate user of chemi‐
cals sued their manufacturer alleging failure to warn (and
other theories). The manufacturer raised the sophisticated in‐
termediary doctrine. We noted “Indiana courts have in the
past taken a broad, multi‐factor view of the level of sophisti‐
cation required under the ‘sophisticated intermediary doc‐
trine.’” 150 F.3d at 809 (citations omitted); Unlike the differ‐
ences in knowledge between ConAgra and Givaudan, in Tay‐
lor the user of chemicals gave their manufacturer individual‐
ized specifications, and the user had vast in‐house expertise
and had participated in federal and industry task forces and
No. 17‐3390 19
committees on the chemicals. Id. at 808–09. Other cases in
which summary judgment was granted under the sophisti‐
cated intermediary doctrine are also factually distinguishable
from this case. E.g., Hathaway, 903 F. Supp. 2d at 676–78
(parties’ written rental agreement expressly discharged sup‐
pliers’ duty to warn).
Viewing the evidence in the employees’ favor, and draw‐
ing justifiable inferences for them, a reasonable jury could
conclude that Givaudan failed to discharge its duty to warn
the plaintiff employees on the dangers of diacetyl. Thus, sum‐
mary judgment should not have been granted to Givaudan on
plaintiffs’ failure to warn claim.8
C. Common Law Negligence Claim
In its first summary judgment decision, the district court
concluded that the employees’ claim for common law negli‐
gence against Givaudan was subsumed by the Act, which ap‐
plies to all claims “brought by a user or consumer.” IND. CODE
§ 34‐20‐1‐1. The employees submitted they were not “users or
consumers” under the Act, relying on cases of distributors or
“middle men” of products. Although Indiana courts have
held that individuals who fell within these groupings were
not “users or consumers,” the district court concluded that the
employees did not act as “middle men” or distributors.
8 Both parties reference Stults v. International Flavors & Fragrances, 31 F.
Supp. 3d 1015 (N.D. Iowa 2014), in which the district court determined
that a reasonable jury could conclude that flavor manufacturing defend‐
ants withheld pertinent safety information from ConAgra. Because Givau‐
dan was not a party to Stults, in which a consumer plaintiff sued, rather
than an employee, and the case involved Michigan rather than Indiana
law, the case was not persuasive.
20 No. 17‐3390
According to the district court, because the employees were
users of the butter flavoring, any claim against Givaudan, a
manufacturer, must be brought under the Act.
On appeal, the employees raise the same argument, con‐
tending that “[w]orkers who come into contact with the prod‐
uct before it is sold to the public in general are usually not
considered consumers.” Again, the employees assert they are
not end users, but work for an intermediary, and that their
common law negligence claim survives.
The Indiana law the employees rely upon does not sup‐
port their position, however. In Thiele v. Faygo Beverage, Inc.,
489 N.E.2d 562 (Ind. Ct. App. 1986), a warehouse worker was
injured while moving a case of beverages. The Indiana Court
of Appeals noted cases had established that the phrase “user
or consumer” in the Act “does not include intermediaries in
the distributive chain.” Id. at 588. The plaintiff in Thiele, there‐
fore, was not entitled to the Act’s benefits because “the legis‐
lature has required a ‘sale’ to a ‘first consuming entity’ before
the protection afforded by the Act is triggered, and Robert
Thiele’s injury occurred before such a transaction took place.”
Id.
The other case employees note, Keen v. Nestle Waters North
America, Inc., 2012 WL 1365444 (S.D. Ind. Apr. 19, 2012), also
involved an employee injured while moving beverages at a
supermarket. Id. at *1. In Keen, the district court determined
that the supermarket’s employees were “middle men” em‐
ployees, like the plaintiff in Thiele. Id. at *5; Thiele, 489 N.E.2d
at 585.
Unlike in Thiele and Keen, ConAgra and its employees
used the butter flavorings as an ingredient in ConAgra’s
No. 17‐3390 21
microwave popcorn. ConAgra was the first consuming en‐
tity—not a “middle man”—for further sale of butter flavor‐
ings to the public. ConAgra incorporated Givaudan’s butter
flavoring into its microwave popcorn which was sold to the
public. ConAgra was a user or consumer within the meaning
of the Act, as were ConAgra’s employees, who fall within the
Act and its definition of “user or consumer” as recognized in
Thiele.
The Indiana Supreme Court has ruled that under the Indi‐
ana Product Liability Act “an employee of a ‘consuming en‐
tity,’ … falls under the definition of ‘user or consumer’ estab‐
lished in Thiele.” Butler v. City of Peru, 733 N.E.2d 912, 919 (Ind.
2000). So the district court did not err by granting Givaudan
summary judgment on plaintiffs’ common law negligence
claim which is preempted by the Indiana Product Liability
Act.
D. Design Defect Claim
Under the Act, a plaintiff also can prove that a product
was sold “in a defective condition unreasonably dangerous to
any user or consumer” (IND. CODE § 34‐20‐1‐1) by showing a
design defect. See Ritchie, 242 F.3d at 720 (citing Nat. Gas Odor‐
izing, Inc. 685 N.E.2d at 161). “[A] defective‐design plaintiff
must establish that the defective condition rendered the prod‐
uct ‘unreasonably dangerous.’” Weigle v. SPX Corp., 729 F.3d
724, 735 (7th Cir. 2013) (quoting IND. CODE § 34‐20‐2‐1).
“[D]efective‐design claims sound in negligence, so a party al‐
leging a design defect ‘must establish that the manufacturer
or seller failed to exercise reasonable care under the circum‐
stances in designing the product.’” Id. at 734 (quoting IND.
CODE § 34‐20‐2‐2). To demonstrate a design defect under In‐
diana law, “the plaintiff must compare the costs and benefits
22 No. 17‐3390
of alternative designs” and “show that another design not
only could have prevented the injury but also was cost‐effec‐
tive under general negligence principles.” Piltch, 778 F.3d at
632 (quoting Pries v. Honda Motor Co., 31 F.3d 543, 545–46 (7th
Cir. 1994)). This demonstration requires expert testimony
because a lay jury would not be able to establish and compare
the costs and benefits of a diacetyl‐free butter flavor design
with Givaudan’s butter design containing diacetyl, or that an
alternative diacetyl‐free design was cost‐effective.
In its second dispositive order, the district court granted
Givaudan summary judgment on the employees’ design
defect claim. The district court noted how expert testimony is
required on the topic of costs and benefits of alternative de‐
signs, and to show that another design could have prevented
the injury and was also cost‐effective. Because the employees
provided “no expert testimony on the costs and benefits of a
diacetyl‐free butter flavor or that the diacetyl‐free butter fla‐
vors that existed prior to 2007 were cost‐effective alternatives
under general negligence principles,” the district court con‐
cluded that the employees had not met their burden on their
design defect claim.
On appeal, the employees contend they have presented
evidence to survive summary judgment on this claim. They
assert the butter flavorings Givaudan sold were designed
defectively because they contained diacetyl, and refer to their
expert witnesses who opined that butter flavorings contain‐
ing diacetyl caused this lung disease from which plaintiffs
suffer. The employees also refer to Givaudan’s expert witness
who testified to the availability and use of diacetyl‐free butter
flavors since before Givaudan supplied products to ConAgra.
No. 17‐3390 23
None of the evidence the employees point to shows that
Givaudan’s butter flavorings were defective. See IND. CODE
§§ 34‐20‐2‐1, 34‐20‐2‐2. Instead, on this point the employees
offer conclusory factual assertions, without explanation or
supporting data. See McMahon v. Bunn‐O‐Matic Corp., 150 F.3d
651, 658 (7th Cir. 1998) (“[a]n expert who supplies nothing but
a bottom line supplies nothing of value to the judicial pro‐
cess”) (citations omitted). The employees’ evidence that alter‐
natives to diacetyl existed before 2007 also does not prove
Givaudan negligently designed the flavorings. See Bourne v.
Marty Gilman, Inc., 452 F.3d 632, 638 (7th Cir. 2006) (“mere ex‐
istence of a safer product is not sufficient to establish liabil‐
ity”). Nor does their evidence show viable alternatives to di‐
acetyl before 2007. The employees designated as an expert a
senior flavor chemist at Givaudan who has worked at Givau‐
dan making artificial butter flavors since the early 1990s who
testified there was one butter flavoring without diacetyl. But
the chemist did not testify that the flavoring was used for pop‐
corn, he acknowledged diacetyl had uniquely potent butter
flavor, and he said when Givaudan started reformulating fla‐
vors to eliminate diacetyl, the new products were “not … the
same and [did] not taste the same.”
Statistical evidence would have been helpful on a design
defect claim like this: How many employees worked in the
plant? How many worked with butter flavorings? How many
have contracted the claimed disease, within the plant and
throughout the country? The employees presented no such
evidence.
The employees also have not come forward with evidence
that at the time ConAgra was purchasing butter flavorings
from Givaudan for the Rensselaer plant, diacetyl‐free butter
24 No. 17‐3390
flavorings, if used, would have prevented the employees’ in‐
juries. Cf. id. at 637–38. In Bourne, the plaintiffs’ expert failed
to provide how a finder of fact could evaluate the frequency
of injuries caused by the product, or calculate the extent to
which risk would actually be reduced by the alternative
designs, or justify the cost of those alternatives relative to the
benefits of the alleged defective product. Id.
Here, the evidence the employees offer on frequency of
harm is even less detailed than in Bourne. The senior Givau‐
dan flavor chemist who testified Givaudan made a diacetyl‐
free butter flavor in the early 1990s did not testify about the
costs or benefits of that product. The employees did not pro‐
vide evidence of an alternative, cost‐effective butter flavoring
design that would have prevented bronchiolitis obliterans. In
fact, the employees weakened their position in the district
court when they admitted many diacetyl substitutes have
been linked to lung disease. Once they noted “diacetyl‐substi‐
tutes may be just as harmful as diacetyl,” and cited various
studies. In the absence of such evidence, the employees’
design defect theory fails, and the district court did not err in
granting summary judgment to Givaudan on this claim.
III. Conclusion
For these reasons, we AFFIRM the grant of summary judg‐
ment to Givaudan on the employees’ claims other than the
failure to warn, REVERSE the grant of summary judgment on
the employees’ duty to warn claim based on the sophisticated
intermediary doctrine, and REMAND for further proceedings
on that claim.