Case: 13-10922       Date Filed: 09/12/2014       Page: 1 of 31


                                                                                  [PUBLISH]

                  IN THE UNITED STATES COURT OF APPEALS

                            FOR THE ELEVENTH CIRCUIT
                              ________________________

                                     No. 13-10922
                               ________________________

                          D.C. Docket No. 1:11-cv-02733-JOF

PATRICIA HUGHES,
Mother and Administratrix of the
Estate of Allene J. Hughes

                                                                        Plaintiff - Appellant,

                                            versus

KIA MOTORS CORPORATION,
KIA MOTORS AMERICA, INC.

                                                                    Defendants - Appellees.
                               ________________________

                      Appeal from the United States District Court
                         for the Northern District of Georgia
                            ________________________

                                   (September 12, 2014)

Before TJOFLAT and WILSON, Circuit Judges, and BUCKLEW, ∗ District Judge.



       ∗
          Honorable Susan C. Bucklew, United States District Judge for the Middle District of
Florida, sitting by designation.
                Case: 13-10922        Date Filed: 09/12/2014        Page: 2 of 31


TJOFLAT, Circuit Judge:

       On May 2, 2005, Allene Hughes was involved in a vehicle collision with a

Mack truck in Chattanooga, Tennessee. The collision sent her Kia Optima

pinballing about, colliding with two parked cars, a fence, a tree, three metal posts

supporting a carport awning, and a flag pole before ultimately coming to rest

against a house. On May 3, Allene was declared dead from traumatic brain injury.

       Patricia Hughes, Allene’s mother and the administratrix of her estate, filed

suit against Kia Motors Corporation and Kia Motors America, Inc. (collectively,

“Kia”) in Georgia Superior Court, alleging that Kia’s failure to equip the Optima

with a fuel shut-off switch led to Allene’s death. Kia removed the case to federal

court,1 and discovery commenced. When discovery closed, Kia moved to exclude

the testimony of Hughes’s causation expert, Dr. Joseph L. Burton, and also moved

for summary judgment. The District Court granted both of Kia’s motions. Hughes

now appeals. Because the court did not abuse its discretion in excluding Burton’s

testimony and did not improperly grant summary judgment in Kia’s favor, we

affirm.



       1
          Kia asserted diversity jurisdiction under 28 U.S.C. § 1332, as the parties were citizens
of different states and the amount in controversy exceeded $75,000. Hughes is a resident of
Georgia, Kia Motors Corporation is a foreign national corporation organized under the laws of
the Republic of Korea and having its headquarters and principal place of business in Seoul,
South Korea, and Kia Motors America, Inc. is incorporated under California law with its
principal place of business in California.
                                                 2
              Case: 13-10922     Date Filed: 09/12/2014   Page: 3 of 31


                                          I.

      Allene Hughes died as a result of injuries she incurred from a vehicle

collision in Chattanooga, Tennessee, at 2:45 a.m. on May 2, 2005. That night,

Allene drove her 2004 Kia Optima out of a Waffle House parking lot onto Club

Drive, a one-way street. However, Allene exited the parking lot in the wrong

direction, and ended up traveling against the correct direction of traffic (no other

cars were on the Club Drive at the time). Upon realizing her mistake and reaching

the intersection of Moore and Brainerd Roads, Allene entered the through lanes of

Brainerd Road and was struck by the front of a westbound Mack truck pulling a

loaded fuel tanker trailer.

      After impact, the Mack truck came to a stop in the intersection. The Optima

traveled farther. The Optima initially traveled south, entering the property of a gas

station and convenience store. It then turned back toward the north, crossing all

the lanes of Brainerd Road and entering the parking lot of the Waffle House. In

the parking lot, the Optima struck a parked Geo Tracker and a parked Toyota

Corolla. It then ran through a fence to the north of the Waffle House, entering

Club Drive, and traveled east onto residential property. There, the Optima

sideswiped a tree in the front yard and struck three support posts of a carport

awning and a flagpole. The Optima came to a rest against the house.


                                          3
              Case: 13-10922    Date Filed: 09/12/2014   Page: 4 of 31


      The diagram below, which the police officers who arrived on the scene

created, indicates the Optima’s movement from the moment it left the Waffle

House until it came to a rest against the house. In the diagram, Allene’s vehicle is

indicated with the number 1, the Mack truck is number 2, and the Geo Tracker and

Toyota Corolla are indicated with numbers 3 and 4.




                                  DE 68-12, at 5.


                                          4
              Case: 13-10922    Date Filed: 09/12/2014   Page: 5 of 31


      When the Emergency Medical Service (“EMS”) arrived on the scene, Allene

had already been removed from the Optima. She was unresponsive, but was

breathing four to six times per minute with the assistance of a bag-valve mask.

EMS reported that her only obvious injury was an abrasion to the left knee.

      EMS transported Allene to Erlanger Medical Center, where doctors

discovered evidence of traumatic brain injury, left frontal subdural hematoma, and

epidural right pulmonary contusion. She also had suffered a pelvic fracture. An

emergency craniotomy and implantation of a Scott cannula to drain the blood from

subdural cavity was unsuccessful. Allene was brain dead. At 2:30 p.m. on May 3,

Allene was declared dead.

                                         II.

      On July 15, 2011, Patricia Hughes, Allene’s mother and the administratrix of

her estate, filed a wrongful death action against Kia in the Superior Court of

Gwinnet County, Georgia. The complaint included two theories of liability: strict

liability for the inherently dangerous design of the Optima and negligence based on

alleged defects in the Optima’s airbag deployment system, its lack of a fuel flow

shut-off device, Kia’s failure to warn of the unreasonable dangers inherent in the

design of the Optima, and Kia’s failure to remedy the defective condition of the

Optima.


                                          5
               Case: 13-10922       Date Filed: 09/12/2014     Page: 6 of 31


       Kia filed a timely notice of removal, pursuant to 28 U.S.C. § 1446(b), in the

United States District Court for the Northern District of Georgia. After Kia filed

its answer to the complaint’s allegations and affirmative defenses, the case

proceeded to discovery.

                                             A.

       Hughes hired Ralph Cunningham, an accident reconstructionist, 2 who

prepared a report on the collision between the Optima and the Mack truck.

According to the Mack truck’s Engine Control Module, which makes a permanent

record of a vehicle’s speed before and shortly after a hard breaking event, the

Mack truck approached the intersection where the collision occurred at 39 miles

per hour (mph). When the driver applied the brakes, the truck began to rapidly

decelerate at a steady speed until the truck came to a stop. Cunningham estimated

that the truck was traveling at a speed of not less than 10 mph and not more than

24 mph when it struck Allene’s Optima.

       Cunningham’s report indicates that the truck struck the Optima at 12 degrees

forward of perpendicular to the left side. For point of reference, a collision that is

perpendicular to the left side—a T-bone collision—would be 0 degrees. A head-on




       2
           Cunningham has a B.S.E. degree from Century University, an Engineer in Training
certification, and full accreditation as a Traffic Accident Reconstructionist.

                                              6
              Case: 13-10922     Date Filed: 09/12/2014    Page: 7 of 31


collision would be 90 degrees. Thus, the collision was closer to perpendicular than

head-on.

      When the vehicles collided, the Optima’s driver’s side airbag deployed.

Although the Optima was originally equipped with a driver’s front airbag, it had

deployed at some point prior to the May 2 collision and was not replaced; the

deployed airbag had been cut away from the steering wheel. The region of greatest

permanent deformation coincided with the left “A” pillar and front-door hinge

region, where the inward crush was 20 inches. The crush diminished toward the

rear, and scrapes on the left rear door and back of the Optima indicated sliding

contact, which Cunningham opined was likely from the vehicles separating from

each other. The brake pedal jammed such that it could not be applied, and it

deformed toward the right side of the vehicle, jamming the throttle pedal in an

applied position.

      Cunningham calculated the total delta-v—which represented the sudden

change in velocity that occurred during the contact phase of the collision with the

Mack truck—as 37.5 mph. Because the collision was neither head-on nor perfectly

perpendicular, the total delta-v could be separated into its longitudinal and lateral




                                           7
                 Case: 13-10922       Date Filed: 09/12/2014        Page: 8 of 31


components. Cunningham calculated the longitudinal delta-v using the total delta-

v, the 12 degree angle of impact, and simple trigonometry. 3

       From his calculations, Cunningham concluded that the longitudinal delta-v,

the sudden change in velocity in the longitudinal direction during the contact phase

of the collision, equaled approximately 8 mph. Cunningham did not calculate

lateral delta-v—the change in velocity along the perpendicular plane—because

Hughes’s attorneys did not request that he do so.

       Cunningham estimated the Optima’s probable post-impact travel if it had

been equipped with a fuel shut-off device by using two estimated speeds at which

the Optima could have been traveling when it collided with the Mack truck—38

mph and 26 mph. 4 Based on these two speeds, Cunningham predicted that the

Optima would have traveled between 69 and 98 feet if it had been traveling 38



       3
           The calculation can be expressed as follows:

                                                        = sin 12°
                                   longitudinal delta v


                                                        = sin 12°
                                       total delta v
                                   longitudinal delta v


                              longitudinal delta v = sin 12° (37.5 mph)
                                        37.5 mph


                                    longitudinal delta v ≈ 8 mph
       4
          Cunningham estimated using the maximum rate of acceleration possible for the Optima
and a lesser rate of acceleration of five feet per second per second. Cunningham believed that
even a speed of 26 using this lower acceleration rate was “somewhat improbably high,” DE 110,
at 9, but he did not calculate the Optima’s velocity using any other acceleration rate.

                                                 8
                Case: 13-10922       Date Filed: 09/12/2014        Page: 9 of 31


mph, and between 32 and 46 feet if it had been traveling at 26 mph. Had it

traveled the longer distance, the Optima would have stopped roughly on the curb

of the northwest corner of the gas station and convenience store. Had the Optima

been traveling at 26 mph, it would have come to a rest on Brainerd Road in the

eastbound lanes.

                                                B.

       Hughes also hired a medical expert, Dr. Joseph Burton, 5 to determine

Allene’s cause of death. His report concluded that Allene “died of blunt force

trauma that occurred predominantly to the left side of her head.” DE 63-2, at 4.

He further concluded, based extensively on Cunningham’s report, that

       the fact that the [Optima] did not stop operating in a timely manner
       after the initial impact with the Mack Truck was contributory or
       causative of [Allene’s] brain injuries. . . . [H]ad the [Optima] stopped
       in a timely manner and did not have the additional impact it is more to
       a reasonable degree of forensic, medical, and scientific probability
       that [Allene] would not have sustained a fatal brain injury.

DE 63-2, at 10. Burton suggested that “[t]he dynamics of the left/frontal collision

with the Mack Truck combined with the deploying side airbag would have caused

[Allene’s] body to move forward towards the intrusion [at the A-pillar] and



       5
           Dr. Burton has an M.D. degree from Emory University Medical School. He is board
certified in forensic pathology. He has been chief medical examiner in several counties, a
clinical professor in forensic pathology and director of forensic pathology training program at the
Emory University Medical School.

                                                9
             Case: 13-10922     Date Filed: 09/12/2014    Page: 10 of 31


leftward interacting with the side impact airbag.” DE 63-2, at 9. Based on these

dynamics, Burton opined that “[i]t is possible that [Allene] sustained her left sided

brain injury in this impact due to her body moving forward past the deployed torso

bag and her head interacted with the intruding structures of the [Optima].” DE 63-

2, at 9. As the Optima impacted other objects and the various curbs, Allene’s body

“would have the opportunity to have interaction of her head with the steering

wheel (due to there being no airbag there to prevent interaction) and the instrument

panel in a manner which would either exacerbate or potentially cause her brain

injury.” DE 63-2, at 9.

      Thus, Burton summarized, Allene “could have sustained her brain injury in

the first impact and then due to subsequent impacts her head and body could have

interacted in a manner that would have exacerbated the brain injury she already

received.” DE 63-2, at 11. Alternatively, Allene “could have possibly sustained

the brain injury that caused her death as a result of her head’s contact with the

steering wheel and/or instrument panel during the additional impacts the [Optima]

experienced as a result of not stopping in a timely manner.” DE 63-2, at 11.

      Burton explained in his report that a specific error rate for clinical

impressions is difficult to calculate, but he held his opinion “to a reasonable degree

of forensic and scientific medical probability and certainty.” DE 63-2, at 11.

Burton explained that his conclusions were “supported by extensive references to
                                          10
             Case: 13-10922     Date Filed: 09/12/2014   Page: 11 of 31


the literature, to research and to 30 years of training, education and experience.”

DE 63-2, at 12. He further indicated that his opinions were “based on the specific

data [he] evaluated, the condition of the subject vehicle which [he] personally

inspected, [his] personal evaluation of various photographic evidence submitted

with the case file and on [his] education, background and training.” DE 63-2, at

13.

      Burton assured that the methodology he employed “comport[ed] to the

scientific method of inquiry.” DE 63-2, at 13. He explained that the scientific

method involves stating the problem, collecting data, formulating a hypothesis,

validating the hypothesis, and considering the potential for error.

                                          C.

      When Kia deposed Dr. Burton on May 3, 2012, he described his

methodology for reaching his opinion. Burton explained that he assessed the

forces that acted on the Optima, “[n]ot only how much they were, but where they

were coming from and how they were impacting in this case the package the

person was in, the [Optima] in this case.” DE 67-1, at 13. Burton opined that

“[w]e’re not dealing with some kind of high delta-v, high g-force crash.” DE 67-1,

at 22. Burton explained that, according to the National Highway Traffic Safety

Administration (the “NHTSA”), “if the delta-v is over 26, they consider it a severe

impact,” whereas a delta-v of 0 to 10 is “minor.” DE 67-1, at 22.
                                          11
              Case: 13-10922       Date Filed: 09/12/2014     Page: 12 of 31


       Kia asked Burton about his impression of the collision between the Optima

and the Mack truck, and he explained that the collision was 12 degrees forward of

a 90-degree side impact and “resulted in a delta-v for the [Optima] of about . . . 8

miles an hour.” DE 67-1, at 26. Burton described the pulse, or time of impact, as

“relatively short[,] like less than a hundred milliseconds.” DE 67-1, at 26. He

reasoned that “it is a big truck hitting a little car but it is not hitting it when it is

going very fast and it resulted in what would be, according to NHTSA, a low or

minor delta-v to the [Optima].” DE 67-1, at 26. Burton acknowledged that the 8

mph he referred to was actually the longitudinal delta-v, not the total or lateral

delta-v, but he opined that “delta-v is not the most important factor” in predicting

injury for a side impact. DE 67-1, at 26–27.

       Instead, Burton explained that the two most important factors for predicting

injury for a side impact are amount of intrusion into the interior of the Optima and

the velocity of the adjacent door panel. Burton said that he believed that

Cunningham’s report said that the Mack truck had slowed to approximately 12

mph, which meant “the front of the truck is moving at 12 miles an hour which

means at some point in time the door panel or part of the [Optima], its interior, will

be moving at approximately 12 miles an hour also.” DE 67-1, at 27–28. We pause

here to note two things. First, Cunningham’s report actually said the truck had

slowed to between 10 and 24 mph when it struck the Optima; the report never said
                                             12
               Case: 13-10922       Date Filed: 09/12/2014       Page: 13 of 31


the truck was traveling at 12 mph when it impacted the Optima. Second, and more

importantly, Burton’s explanation demonstrates that he measured the velocity of

the adjacent door panel as equal to the speed at which the truck was traveling at

impact.

       Kia’s attorney then asked Burton to assume that Cunningham had calculated

that lateral delta-v during his deposition, which was taken earlier. 6 Burton was

asked to assume that lateral delta-v was approximately 37 mph. Then, Kia’s

attorney asked Burton for the velocity of the Optima’s interior door resulting from

the impact with the truck, to which Burton repeated his earlier calculation, which

he reached based on Cunningham’s report, of 12 mph. Burton then said, “If I

assume hypothetically that [the truck] hit [the Optima] at 36 miles an hour or 35,

then at some point in time part of the interior structure of the [Optima] was moving

at that speed.” DE 67-1, at 33. Notably, Burton assumed that the 35 or 36 mph

represented the truck’s velocity at the moment of impact, rather than lateral delta-


       6
          In fact Cunningham had been asked to do so. Using the angle of impact, the total delta-
v, and trigonometry, he calculated lateral delta-v as approximately 37 mph. The equation below


                                                    = cos 12°
demonstrates how Cunningham arrived at lateral delta-v.
                                    lateral delta v


                                                    = cos 12°
                                     total delta v
                                    lateral delta v


                               lateral delta v = cos 12° (37.5 mph)
                                      37.5 mph


                                    lateral delta v ≈ 37 mph

                                               13
              Case: 13-10922       Date Filed: 09/12/2014      Page: 14 of 31


v. 7 Then relying on that assumption about the Mack truck’s velocity—which was

different than the assumption Kia’s attorney asked Burton to make—he calculated

the interior velocity as equivalent to the truck’s assumed velocity.

       Kia’s attorney followed Burton’s response by asking, “[I]f the assumption

you’re making of Mr. Cunningham making that approximately 37-mile-per-hour

delta-v is in fact true, you would now agree with me that at some point in that

impact between the [Optima] and the Mack Truck part of that driver’s door was

traveling at least 35 miles per hour. Is that fair?” DE 67-1, at 33. Pause for a

moment to recognize that by asking about lateral delta-v—rather than the velocity

of the truck at impact—Kia’s attorney changed the thing being measured and asked

if it would support the same conclusion. In other words, Burton appears to have

assumed that the truck’s velocity would be equivalent to the Optima’s interior

velocity at the moment of impact, whereas Kia’s attorney asked if the Optima’s

lateral delta-v, its lateral change in velocity, would equal the Optima’s interior

velocity. But the truck’s velocity and the Optima’s lateral delta-v are different

measurements, and Burton likely should have recognized as much. Apparently

missing this sleight of hand, however, Burton agreed with Kia’s attorney that a




       7
         Cunningham never calculated the truck’s actual velocity at the moment of impact; he
estimated the truck’s velocity was between 10 and 24 mph.

                                              14
             Case: 13-10922     Date Filed: 09/12/2014    Page: 15 of 31


delta-v of approximately 37 mph would result in an interior velocity of at least 35

mph.

       During his deposition, Burton agreed that if lateral delta-v exceeded 35 mph,

it “could have caused a fatal brain injury.” DE 67-1, at 43. He further stated that

he would hold that opinion to a reasonable degree of medical certainty. Then, the

following colloquy occurred between Kia’s attorney and Burton:

       Q: But as I parse those words, you’re not of the opinion that the
       impact between the Mack truck and the [Optima] in fact was the
       impact that was the cause of her fatal injury in this case?
       A: My opinion is that I don’t know enough about the other impacts to
       exclude them. . . . So I know those are all possibilities. And did that
       happen here versus the Mack truck impact? I don’t know. I’m
       certainly not going to embarrass myself by arguing with somebody
       that the Mack truck impact couldn’t produce a fatal injury because I
       would look ridiculous and whatever credibility I might have to a jury
       or to a court I would lose if I tried to say that there’s no way that
       Mack truck impact could have killed this lady. I wouldn’t say that.

DE 67-1, at 43–44.

       Burton conceded that he lacked sufficient information to conclude that the

impact with the Toyota Corolla or the Geo Tracker—the two cars the Optima

collided with in the Waffle House parking lot—caused the fatal brain injury. He

also conceded that he could not conclude whether Allene received her injuries

from the side airbag, the A pillar, or the steering wheel rim, admitting that “[a]ll

three have the capacity to cause her brain to look the way it looks. But they all had


                                          15
             Case: 13-10922      Date Filed: 09/12/2014    Page: 16 of 31


the capacity to leave no telltale sign of interaction with the structure either.” DE

67-1, at 56. Kia’s attorney followed up:

       Q: So for all of the other impacts other than the Mack truck, you don’t
       have enough acceleration or deceleration information to make an
       analysis with respect to those impacts and the likelihood that during
       one of those impacts Ms. Hughes impacted some part of the [Optima];
       is that correct?
       A. Well, I know enough about curb impacts and things like that to
       know that it has the potential to cause her injury and to have her
       impact a structure from it. Did it and what structure was it, I don’t
       have enough information to answer that part of the equation.
DE 67-1, at 64–65.

       Nevertheless, Burton reaffirmed toward the end of his deposition that he

held the opinion that it was “more likely than not that” Hughes’s brain injury was

caused by some event after impact with the Mack truck because he believed the

side airbag “offered her significant protection in that Mack truck impact.” DE 67-

1, at 71.

                                           III.

                                           A.

       Kia moved to exclude Burton’s testimony, arguing that Burton’s opinion

was not the product of reliable methodology and it did not fit the evidence in the

case. Specifically, Kia noted that Burton had admitted during his deposition that

the velocity of the interior of the door being impacted is a critical factor to


                                           16
             Case: 13-10922     Date Filed: 09/12/2014      Page: 17 of 31


analyzing the collision. But, Kia argued, Burton ignored lateral delta-v in forming

his opinion. According to Kia, because he ignored lateral delta-v, Burton

incorrectly assumed that the interior velocity of the door was 12 mph. Kia pointed

to Burton’s deposition, in which he agreed with Kia’s attorney that if lateral delta-v

were approximately 37 mph, the interior velocity of the door would have been at

least 35 mph. Kia also pointed to Burton’s deposition statement that if lateral

delta-v were 37 mph, it could cause a fatal brain injury.

      Kia also argued that Burton’s opinion should be excluded because he failed

to properly employ differential diagnosis as part of his methodology. According to

Kia, Burton could not exclude the impact by the Mack truck as the cause of the

fatal brain injury and did not have sufficient information about the other impacts to

know whether Allene would have collided with any of the interior structures of the

Optima. As such, he could not include any of the other impacts as part of his

analysis in trying to determine causation.

      Hughes opposed Kia’s motion, arguing that while considering lateral delta-v

may be relevant to the question of whether, as a general matter, a high lateral delta-

v can cause death, it does not establish whether Allene actually received her fatal

injury from the collision with the Mack truck. Hughes contended that Burton’s

methodology was sound because it considered the actual Optima damage and


                                         17
               Case: 13-10922       Date Filed: 09/12/2014        Page: 18 of 31


protection from the side airbag, just without a calculation of lateral delta-v. 8

Hughes then argued that, under Tennessee law, 9 she need only prove that the

Optima’s design defect was a substantial factor in causing Allene’s death. In other

words, Burton was not required to exclude other causes of Allene’s death. Hughes

pointed to Burton’s deposition testimony where he stated that he believed the side

airbag provided significant protection in the Mack truck impact, and she argued

that this demonstrated that Burton possessed sufficient information to determine

that later impacts caused Allene’s death.

       Kia replied that, under Tennessee law, causation refers to both cause in-fact

and proximate cause, and therefore Hughes was required prove that the collision

with the Mack truck was not the cause in-fact of her death. Because Burton could

not do so, Kia argued, his causation opinion should be excluded.




       8
          Hughes also relied on information contained in Burton’s supplement to his original
expert report and in his declaration, both of which were filed along with Hughes’s opposition to
Kia’s motion to exclude Burton’s testimony. Kia moved to strike those two filings, and the
District Court granted Kia’s motion. Hughes does not challenge the propriety of the District
Court’s striking Burton’s supplement and declaration. Therefore, we ignore those filings and the
contentions contained therein for purposes of this appeal. See United States v. Pilati, 

, 1364 (11th Cir. 2010) (“[W]hen on appeal a [party] fails to raise an issue when the
opportunity is presented, he waives that argument.”).
       9
          Georgia applies the choice-of-law principle lex loci delicti, which means apply the law
of the place where the tort was committed. See Dowis v. Mud Slingers, Inc., 

, 816,

, 419 (2005). Therefore, in this case we apply Tennessee law.

                                                18
                Case: 13-10922       Date Filed: 09/12/2014        Page: 19 of 31


      The District Court agreed with Kia and granted its motion to exclude

Burton’s testimony. Specifically, the court concluded that Tennessee law required

that Hughes demonstrate actual causation, and thus the lack of a shut-off switch

could not be a substantial factor in Allene’s fatal injury if she sustained that injury

on impact with the Mack truck. The court faulted Burton for failing to explain

how the scientific method assisted him in reaching his opinion; Burton simply

stated that he used the scientific method. The court also reasoned that in light of

Burton’s deposition admission that he would need to consider the forces acting on

the injured person, his failure to consider lateral delta-v demonstrated a failure to

follow his stated methodology. Finally, the District Court noted that Burton

admitted that a delta-v of 35 mph could have caused a fatal brain injury, and, after

he was informed that lateral delta-v in this case was 37 mph, that the interior

velocity of the driver’s side door would be at least 35 mph.

      The District Court also found fault in Burton’s failure to attempt to falsify

his conclusions. 10 Because Burton could not rule out the initial impact as the cause



      10
           The court expressed such failure in its dispositive order:
      [P]art of the scientific method is attempting to falsify one’s conclusions.
      Defendants correctly argue that Dr. Burton was neither able to rule out the initial
      impact with the tractor trailer as the cause of Ms. Hughes’s fatal brain injury, nor
      was he able to identify any particular impact after the tractor trailer which might
      have caused a fatal brain injury. Dr. Burton readily admitted that he did not know
      enough about the other impacts the [Optima] sustained after the initial impact
                                                 19
               Case: 13-10922         Date Filed: 09/12/2014     Page: 20 of 31


of the fatal injury and because he could not identify any particular impact after the

Mack truck collision that might have caused the injury, his opinion was unreliable.

Based on these problems with Burton’s opinion, the court ruled his opinion

inadmissible.

                                               B.

       At the same time it filed a motion to exclude Burton’s testimony, Kia filed a

motion for summary judgment, which the District Court granted. The court

determined that Hughes presented no evidence that Allene either did or did not

suffer her fatal brain injury at the time of the impact with the Mack truck. Even if

Burton’s testimony were admitted, the court said it would still find that Kia would

be entitled to summary judgment because Hughes could not demonstrate proximate

causation. The parties agreed that the front air bag did not deploy because it had

deployed in a previous accident and was never replaced. Thus, the court held,

there was no evidence in the record from which a reasonable juror could conclude

that the lack of a shut-off switch caused Allene’s fatal injuries.




       with the tractor trailer to determine whether Ms. Hughes suffered her fatal brain
       injury at any other point in the sequence of events.
DE 98, at 18–19 (citation omitted).
                                               20
             Case: 13-10922      Date Filed: 09/12/2014    Page: 21 of 31


                                          IV.

      We review the District Court’s exclusion of Burton’s expert testimony under

an abuse-of-discretion framework. Gen. Elec. Co. v. Joiner, 

, 142–43,

, 517, 

(1997). “[T]his standard of review requires

that we defer to the district court’s evidentiary ruling unless that ruling is

‘manifestly erroneous.’” Quiet Tech. DC-8, Inc. v. Hurel-Dubois UK Ltd., 

, 1340 (11th Cir. 2003) (quoting 

, 118 S. Ct. at

517). “Because the task of evaluating the reliability of expert testimony is

uniquely entrusted to the district court . . . we give the district court ‘considerable

leeway’ in the execution of its duty.” Rink v. Cheminova, Inc., 

,

1291 (11th Cir. 2005) (citation omitted) (quoting Kumho Tire Co. v. Carmichael,

, 152, 

, 1176, 

(1999)).

      The admission of expert evidence is governed by Federal Rule of Evidence

702, which provides:

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

      (a) the expert’s scientific, technical, or other specialized knowledge
      will help the trier of fact to understand the evidence or to determine a
      fact in issue;

      (b) the testimony is based on sufficient facts or data;



                                           21
             Case: 13-10922     Date Filed: 09/12/2014    Page: 22 of 31


      (c) the testimony is the product of reliable principles and methods;
      and

      (d) the expert has reliably applied the principles and methods to the
      facts of the case.

Rule 702 and the Supreme Court’s decisions in Daubert v. Merrell Dow Pharm.,

Inc., 

, 

, 

(1993), and its progeny

make clear that district courts act as gatekeepers, excluding evidence unless is it

reliable and relevant. “District courts are charged with this gatekeeping function

‘to ensure that speculative, unreliable expert testimony does not reach the jury’

under the mantle of reliability that accompanies the appellation ‘expert

testimony.’” 

(quoting McCorvey v. Baxter Healthcare

Corp., 

, 1256 (11th Cir. 2002)). Whether an expert’s testimony is

reliable depends on “the particular facts and circumstances of the particular case.”

Kumho Tire 

, 119 S. Ct. at 1179.

      This court has set out three requirements that an expert must meet before his

opinions may be admitted. First, the expert must be qualified on the matter about

which he intends to testify. City of Tuscaloosa v. Harcros Chemicals, Inc., 

, 562 (11th Cir. 1998). Second, he must employ reliable methodology.

expert’s testimony must be able to assist the trier of fact through the

application of expertise to understand the evidence or fact in issue. 

to whether an expert’s methodology is reliable, we look to a number of

                                          22
             Case: 13-10922     Date Filed: 09/12/2014    Page: 23 of 31


factors, including (1) whether the methodology can be and has been tested, (2)

whether the theory or technique has been subjected to peer review, (3) the known

or potential rate of error of the methodology employed, and (4) whether the

methodology is generally accepted. 

–94, 113 S. Ct. at

2797. This list of factors is not meant to be exhaustive, nor must each factor be

present in a given case. The proponent of the expert opinion must carry the burden

of establishing qualification, reliability, and helpfulness. United States v. Frazier,

, 1260 (11th Cir. 2004) (en banc).

      Our opinion in Frazier, is instructive, because there, as here, the expert

“offered precious little in the way of a reliable foundation or basis for his opinion.”

In that case, the defendant sought to introduce expert testimony on the

transfer of hairs and bodily fluids during sexual assault. 

The expert

was prepared to testify that no hair or bodily fluids were discovered at the crime

scene, and therefore it was unlikely the defendant had sexually assaulted the

alleged victim. 

moved to exclude the expert’s testimony, and

the District Court did so after holding a Daubert hearing. 

      Sitting as en banc, we affirmed the District Court’s exclusion of the expert’s

testimony, concluding that although the expert was qualified, his methodology was

not reliable and his opinion would not have been helpful to the jury. 

66. With respect to reliability, we explained that the expert had based his opinion
                                          23
             Case: 13-10922      Date Filed: 09/12/2014    Page: 24 of 31


on “his experience, and on various texts in forensic investigation.” 

problem, however, was that the expert never explained how his experience or the

relevant texts supported his opinion. 

basis for the opinion was left

unstated, “it would be very difficult indeed for the district court (or for that matter

the jury) to make even an informed assessment, let alone to verify” that the opinion

was reliable. 

      Since [the expert] was relying solely or primarily on his experience, it
      remained the burden of the proponent of this testimony to explain how
      that experience led to the conclusion he reached, why that experience
      was a sufficient basis for the opinion, and just how that experience
      was reliably applied to the facts of the case.

to do so, we concluded that the district court did not abuse its

discretion by excluding the expert’s testimony for being unreliable. 

here a similar set of circumstances, and thus we reach a similar

conclusion. Like the expert in Frazier, Burton asserts that he relied on the

evidence in the case, his experience, and relevant literature. He explained in his

report that he reached his conclusion based on the scientific method, without

further explaining how he tested his hypothesis to support his conclusions. During

his deposition, he explained a bit more—declaring that the amount of intrusion and

the velocity of the adjacent door were the most important factors to his




                                           24
               Case: 13-10922        Date Filed: 09/12/2014        Page: 25 of 31


evaluation 11—but even then, his explanation went no further. He did not explain

how those two variables were relevant, nor did he explain how he used those

factors to reach his conclusion.




       11
          The parties devote considerable attention in their briefs to the significance, or lack
thereof, of Cunningham’s calculation of lateral delta-v. We decline to address the point, in part
because we need not do so to resolve this appeal, but also because the attorneys in this case
appear to conflate multiple measurements to make their arguments.
        As alluded to earlier, Burton calculated the interior velocity of the adjacent door based on
the actual velocity of the Mack truck at the time of impact, which he assumed to be 12 mph.
Cunningham estimated the Mack truck’s velocity to be somewhere between 10 and 24 mph at
impact. But Cunningham also calculated lateral delta-v as 37 mph. While Hughes claims this
number is hypothetical, it is based on trigonometry. See supra note 6. When Kia asked Burton
during his deposition to assume that the lateral delta-v was 37 mph, this assumption was based
on evidence that could have, and likely would have, been established at trial.
       Burton then testified that if the Mack truck were traveling at 37 mph when it impacted the
Optima, the interior velocity of the adjacent door would be at least 35 mph. But there is no
evidence in the record to assume that the Mack truck was traveling at 37 mph when it impacted
the Optima. Cunningham never retreated from his estimate that the Mack truck was traveling
between 10 and 24 mph at impact; his calculation of lateral delta-v is an entirely different
measurement.
         To be sure, Burton agreed with Kia’s attorney that a lateral delta-v of 37 mph would
mean the interior velocity of the adjacent door was at least 35 mph. But Burton never equated
the Mack truck’s velocity to lateral delta-v, and he never expressly based the interior velocity on
lateral delta-v. In short, Burton appears not to have recognized that Kia’s attorney asked about
interior velocity using a different measurement, lateral delta-v, instead of velocity of the Mack
truck.
        However, Burton never indicated that he made a mistake when he said a 37 mph lateral
delta-v would lead to an interior velocity of at least 35 mph, and Hughes’s attorneys have not
made any argument about the apparent discrepancy we describe here. Thus, we cannot say for
sure that Burton misspoke. And we would therefore be left to accept Burton’s testimony that a
37 mph lateral delta-v would lead to an interior velocity of at least 35 mph, which means that
Burton did not employ his stated methodology—relying on intrusion and interior velocity—
correctly.
        All of this is to say that we cannot conclude that the District Court committed clear error
by relying on Burton’s admission, even though it appears to be based on a faulty assumption.
                                                25
               Case: 13-10922       Date Filed: 09/12/2014      Page: 26 of 31


       Moreover, Burton was unable to rule out the Mack truck impact as the cause

of Allene’s fatal injury. He admitted that he lacked sufficient information about

the Optima’s impacts with the Toyota Corolla and Geo Tracker to determine

whether these impacts caused or did not cause the injury. And although he

concluded that the impacts with the curbs had the potential to cause Allene to

impact a structure inside the car and thereby cause her fatal injury, he could not

opine that the curb impacts actually caused a collision with an interior structure

and the fatal injury. Nor could Burton say that the impact with the house was

sufficient to cause the fatal injury. Nevertheless, he asserted that Allene would not

have sustained the fatal injury had the Optima been equipped with a shut-off

switch.

        “[S]omething doesn’t become scientific knowledge just because it’s uttered

by a scientist; nor can an expert’s self-serving assertion that his conclusions were

derived by the scientific method be deemed conclusive.” McDowell v. Brown, 

, 1299 (11th Cir. 2004) (alteration in original) (internal quotation marks




And thus, based on this factual finding, the court’s legal conclusion—that Burton did not follow
his methodology and thus his opinion is unreliable—is not clearly erroneous. However, we need
not rely exclusively on Burton’s statement because we may affirm the District Court on any basis
supported by the record, Thomas v. Cooper Lighting, Inc., 

, 1364 (11th Cir. 2007),
and, as described elsewhere, Hughes failed to carry her burden to demonstrate that Burton’s
testimony was reliable.

                                              26
             Case: 13-10922     Date Filed: 09/12/2014    Page: 27 of 31


omitted). “The trial court’s gatekeeping function requires more than simply taking

the expert’s word for it.” Fed. R. Evid. 702, advisory committee’s note, 2000

amendment (internal quotation marks omitted). As the Supreme Court explained

in Joiner,

      Trained experts commonly extrapolate from existing data. But
      nothing in either Daubert or the Federal Rules of Evidence requires a
      district court to admit opinion evidence that is connected to existing
      data only by the ipse dixit of the expert. A court may conclude that
      there is simply too great an analytical gap between the data and the
      opinion 

, 118 S. Ct. at 519.

      The decision to exclude expert testimony is committed to the sound

discretion of the District Court, see 

, and here the court

reasonably could have questioned the reliability of Burton’s ultimate opinion given

the vague manner in which Burton described his methodology coupled with his

inability to express an opinion about how the various impacts would have affected

Allene. Cf. Kumho Tire 

, 119 S. Ct. at 1177 (“The court could

reasonably have wondered about the reliability of a method of visual and tactile

inspection sufficiently precise to ascertain with some certainty the abuse-related

significance of minute shoulder/center relative tread wear differences, but

insufficiently precise to tell with any certainty from the tread wear whether a tire

had traveled less than 10,000 or more than 50,000 miles.” (internal quotation

                                          27
             Case: 13-10922     Date Filed: 09/12/2014    Page: 28 of 31


marks omitted)). “[T]he deference that is the hallmark of abuse-of-discretion

review requires that we not reverse an evidentiary decision of a district court unless

the ruling is manifestly erroneous.” 

(citations omitted)

(internal quotation marks omitted). Here, the District Court concluded that the

leap from data to opinion was too great, and therefore excluded Burton’s

testimony. Because the court properly applied the law and did not make clearly

erroneous factual findings in reaching its decision to exclude Burton’s opinion, we

hold that the court did not abuse its discretion in excluding his testimony.

                                          V.

      Having concluded that the District Court did not abuse its discretion by

excluding Burton’s testimony, we must consider whether summary judgment was

proper. When reviewing a district court’s grant of summary judgment, we exercise

de novo review. Fils v. City of Aventura, 

, 1287 (11th Cir. 2011).

Summary judgment is appropriate where the moving party—here, Kia—“shows

that there is no genuine dispute as to any material fact and the movant is entitled to

judgment as a matter of law.” Fed. R. Civ. P. 56(a). In reviewing the material

facts, we draw all inferences in favor of the nonmoving party—here, Hughes. 

.




                                          28
              Case: 13-10922      Date Filed: 09/12/2014      Page: 29 of 31


       In enhanced injury cases such as this one, general principles of causation

still apply under Tennessee law. 12 That is, “[a]ny claim for ‘enhanced injuries’ is

nothing more than a claim for injuries that were actually and proximately caused

by the defective product.” Whitehead v. Toyota Motor Corp., 

,

694 (Tenn. 1995). Thus, Hughes has the burden of proof on causation, which

means she must present evidence to create a genuine issue of material fact on both

cause in-fact and proximate causation. Hale v. Ostrow, 

, 718

(Tenn. 2005). As the Supreme Court of Tennessee has explained:

       The defendant’s conduct is the cause in fact of the plaintiff’s injury if,
       as a factual matter, it directly contributed to the plaintiff’s injury. In a
       case such as this one, we must ask whether the plaintiff’s injury would
       have happened ‘but for’ the defendants’ act. If not, then the
       defendants’ conduct is a cause in fact of the plaintiff's injury. It is not
       necessary that the defendants’ act be the sole cause of the plaintiff’s
       injury, only that it be a cause.

“[T]he mere occurrence of an injury does not prove

negligence . . . . Even when it is shown that the defendant breached a duty of care

owed to the plaintiff, the plaintiff must still establish the requisite causal

connection between the defendant’s conduct and the plaintiff’s injury.” Kilpatrick

v. Bryant, 

, 599 (Tenn. 1993) (citation omitted).




       12
          As previously mentioned, we apply Tennessee law because Georgia applies the choice-
of-laws rule of lex loci delicti. See supra note 9.

                                             29
                 Case: 13-10922         Date Filed: 09/12/2014     Page: 30 of 31


        In the context of this case, the Optima’s lack of a shut-off switch must have

been an actual cause of Allene’s enhanced injury—the fatal injury—before we

consider whether it was also a proximate cause. Thus, the question of actual

causation depends then on whether Allene sustained her fatal injury as a result of

the impact with the Mack truck. If Allene sustained her fatal injury from the Mack

truck impact, then the additional impacts were not a but-for cause of the fatal

injury. 13 Without Burton’s opinion that the lack of a shut-off switch caused

Allene’s fatal injury, Hughes has presented no evidence of actual causation. 14 She

has not shown that the Mack truck impact did not cause Allene’s fatal injury, nor

has she shown that the lack of a shut-off switch (as opposed to some other cause,

such as the removed driver’s front airbag) was an actual cause of the fatal injury.

As such, Kia is entitled to judgment as a matter of law because Hughes failed to

create a genuine issue of material fact with respect to actual causation.


       13
            As one treatise explains:
       If one of the two fires burns the plaintiff’s property to the ground before the other
       spreads to the scene, the second fire is not a factual cause at all, even though it
       would have burned the plaintiff’s property in the same way. Your acts today
       cannot in any practical sense cause something that happened in 1939, or even
       something that happened one second before you acted.
Dan B. Dobbs, Paul T. Hayden & Ellen M. Bublick, The Law of Torts § 189 (2d ed. 2011).
       14
         Even with Dr. Burton’s testimony, Hughes would have a problem getting past
summary judgment because Burton could not exclude the Mack truck impact as a cause of death.
Moreover, Burton opined to a reasonable degree of medical certainty that a lateral delta-v of 37
could have caused Allene’s fatal injury. If Allene sustained her fatal injury as a result of the
Mack truck impact, then the subsequent impacts would not have led to an enhanced injury.

                                                 30
            Case: 13-10922     Date Filed: 09/12/2014   Page: 31 of 31


                                        VI.

      The District Court did not abuse its discretion in excluding Burton’s

testimony. Nor did the court err in granting summary judgment for Kia. We

therefore AFFIRM the court’s judgment.

      SO ORDERED.




                                        31