Full opinion text
OPINION ON REHEARING TERRIE LIVINGSTON, Justice. Upon consideration of appellant General Motors Corporation’s motion for rehearing, we deny the motion; however, we withdraw our opinion and judgment of June 29, 2006 and substitute the following to make a nonsubstantive clarification. I. Introduction This is a classic battle of the experts case: a products liability design defect case arising out of a traffic accident in which Stacey Burry, a passenger in a 2001 Chevrolet Suburban, was permanently brain damaged. Appellant General Motors Corporation (GM), the manufacturer of the Suburban, appeals from a jury verdict finding that GM was hable for forty-nine percent of Stacey’s injuries because of a design defect in the Suburban and assessing approximately $38 million in damages against GM and Carol Reid, the driver of the Suburban, whom the jury found fifty-one percent responsible. In five issues, GM contends that (1) there is no evidence that a design defect in the Suburban caused Stacey’s injuries, (2) there is no evidence of any design defect in the Suburban, (3) several of the trial court’s evidentiary rulings constitute reversible error, (4) the jury was biased, unqualified, and erroneously instructed, and (5) the evidence is legally and factually insufficient to support the jury’s damage awards. We modify the judgment to delete the bystander damage awards and affirm it as modified. II. Background Facts On January 27, 2003, Stacey, her mother Carol, and Stacey’s three daughters, Rachel, Sarah, and Meghan, were traveling eastbound on Interstate 30 near Sulphur Springs, Texas. Carol was driving the 2001 Chevrolet Suburban, and Stacey was the front passenger. The three girls were in the middle seats, with Rachel directly behind Stacey. Carol exited the interstate and immediately attempted to turn right into a shopping center, crossing double white lines to do so. As she was moving across the access road, an eighteen-wheeler, which was traveling in the access road, struck the Suburban. The eighteen-wheeler’s left front fender initially hit the Suburban near its right rear wheel. The impact rotated the front right side of the Suburban into the front of the eighteen-wheeler so that the Suburban and eighteen-wheeler were perpendicular to each other and the eighteen-wheeler was pushing the Suburban sideways for a short time. Then the Suburban spun off of the eighteen-wheeler in a clockwise motion once or twice and landed in some dirt on the right side of the access road. Rachel, who was closest to the initial impact, suffered a contusion on her head and several breaks in her right leg. The other girls had no injuries. Carol was knocked unconscious but otherwise had no injuries. Stacey, however, struck the B-pillar — the part of the Suburban frame between the front passenger section window and middle section passenger window — as it intruded into the passenger compartment. The Suburban was equipped with front and side airbags, but no airbags deployed in the accident. Stacey was in a coma for ten weeks and suffered severe brain damage as a result of her injuries. Stacey, her husband Chris, and Chris as next friend for Rachel, Sarah, and Meghan, sued GM and Carol alleging that design defects in the Suburban and Carol’s negligence caused Stacey’s injuries. After a nine-day trial, a jury found Carol fifty-one percent liable, and GM forty-nine percent liable, for Stacey’s injuries. The jury also awarded appellees approximately $88 million in damages. III. No-Evidence Issues In its first two issues, GM contends that there is no evidence to support the jury’s findings that the 2001 Suburban had a design defect and that if there was a defect, it caused Stacey’s injuries. We will review these issues first. See Gross v. Burt, 149 S.W.3d 213, 229 n. 12 (Tex.App.-Fort Worth 2004, pet. denied) (op. on reh’g). A. Standard of Review A legal sufficiency challenge may only be sustained when: (1) the record discloses a complete absence of evidence of a vital fact; (2) the court is barred by rules of law or of evidence from giving weight to the only evidence offered to prove a vital fact; (3) the evidence offered to prove a vital fact is no more than a mere scintilla; or (4) the evidence establishes conclusively the opposite of a vital fact. Uniroyal Goodrich Tire Co. v. Martinez, 977 S.W.2d 328, 334 (Tex.1998), cert. denied, 526 U.S. 1040, 119 S.Ct. 1336, 143 L.Ed.2d 500 (1999); Robert W. Calvert, “No Evidence” and “Insufficient Evidence” Points of Error, 38 Tex. L. Rev. 361, 362-63 (1960). In determining whether there is legally sufficient evidence to support the finding under review, we must consider evidence favorable to the finding if a reasonable factfin-der could, and disregard evidence contrary to the finding unless a reasonable factfin-der could not. City of Keller v. Wilson, 168 S.W.3d 802, 827 (Tex.2005). Anything more than a scintilla of evidence is legally sufficient to support the finding. Cont’l Coffee Prods. Co. v. Cazarez, 937 S.W.2d 444, 450 (Tex.1996); Leitch v. Hornsby, 935 S.W.2d 114, 118 (Tex.1996). When the evidence offered to prove a vital fact is so weak as to do no more than create a mere surmise or suspicion of its existence, the evidence is no more than a scintilla and, in legal effect, is no evidence. Kindred v. Con/Chem, Inc., 650 S.W.2d 61, 63 (Tex.1983). More than a scintilla of evidence exists if the evidence furnishes some reasonable basis for differing conclusions by reasonable minds about the existence of a vital fact. Rocor Int’l, Inc. v. Nat’l Union Fire Ins. Co., 77 S.W.3d 253, 262 (Tex.2002). Any ultimate fact may be proved by circumstantial evidence. Russell v. Russell, 865 S.W.2d 929, 933 (Tex.1993). A fact is established by circumstantial evidence when the fact may be fairly and reasonably inferred from other facts proved in the case. Id. However, to withstand a legal sufficiency challenge, circumstantial evidence still must consist of more than a scintilla. Blount v. Bordens, Inc., 910 S.W.2d 931, 933 (Tex.1995). Absent an objection to the jury charge, the sufficiency of the evidence is reviewed in light of the charge submitted. Wal-Mart Stores, Inc. v. Sturges, 52 S.W.3d 711, 715 (Tex.2001). B. Existence of Design Defect GM contends that there is no evidence that the 2001 Suburban had any design defects because appellees “offered only unqualified speculation to support their theory that GM’s design was unreasonably dangerous.” Appellees’ theory at trial was that the Suburban should have had a dual-sensor system governing deployment of the side airbag, that it should have had a side airbag with head protection instead of just thorax protection, and that the B-pillar should have contained extra padding. 1. Qualifications of Appellees’ Experts As a threshold matter, GM contends that appellees’ experts were not qualified to opine about the design of GM’s side airbag system and any proposed alternatives to it; therefore, the trial court erred by allowing their testimony. A witness with the appropriate knowledge, skill, experience, training, or education is qualified to testify as an expert. Tex.R. Evid. 702; Roberts v. Williamson, 111 S.W.3d 113, 121 (Tex.2003); Toshiba Machine Co., Am. v. SPM Flow Control, Inc., 180 S.W.3d 761, 778 (Tex.App.-Fort Worth 2005, pet. granted, judgm’t vacated w.r.m.). Whether an expert is qualified under rule 702 is a preliminary question to be decided by the trial court, and the party offering the expert’s testimony bears the burden of proving the witness’s qualifications. Gammill v. Jack Williams Chevrolet, Inc., 972 S.W.2d 713, 718 (Tex.1998); Broders v. Heise, 924 S.W.2d 148, 151 (Tex.1996). The decision to admit expert testimony is within the trial court’s discretion and will not be disturbed on appeal unless the trial court has abused its discretion. Gammill, 972 S.W.2d at 718-19; Toshiba, 180 S.W.3d at 778. In deciding whether an expert is qualified, the trial court “must ensure that those who purport to be experts truly have expertise concerning the actual subject about which they are offering an opinion.” Helena Chem. Co. v. Wilkins, 47 S.W.3d 486, 499 (Tex.2001). General experience in a specialized field is insufficient to qualify a witness as an expert. Pack v. Crossroads, Inc., 53 S.W.3d 492, 506 (Tex.App.-Fort Worth 2001, pet. denied). ‘What is required is that the offering party establish that the expert has ‘knowledge, skill, experience, training, or education’ regarding the specific issue before the court which would qualify the expert to give an opinion on that particular subject.” Broders, 924 S.W.2d at 153. GM challenges the qualifications of Don Friedman and Geoff Mahon, experts appel-lees proffered to testify about the design of the Suburban’s side impact protection system. Specifically, Friedman testified that the “occupant protection system” in the Suburban “did not provide the protection that would be expected in a side impact, and it didn’t perform in a way that one would expect in ... an accident of this severity.” Friedman’s testimony focused on whether a side airbag would have protected Stacey from the interior of the passenger compartment if it had deployed. Mahon, an airbag sensor engineer, testified regarding the deficiency of the sensing system for the passenger side airbag and opined that a safer alternative system would have included a second sensor to increase the likelihood of deployment of the side airbag. GM contends that Friedman was not qualified because he “last worked in the automotive industry over twenty years ago and has no experience with side airbags.” GM points out that Friedman “never ran a crash test with side impact airbags, never designed a side impact airbag, never designed a vehicle with side impact airbags, and never wrote any papers about side impact airbags.” Friedman testified that he had been involved in vehicle design for fifty-seven years and that he had worked for GM designing vehicles from 1960-68. He had designed fifteen cars and six “other vehicles.” After leaving GM, Friedman founded an automotive research and development company. Friedman testified that he had been investigating injuries that occur in real world accidents since 1970, about thirty-five years. He agreed that in his research he had investigated how an occupant moves once a vehicle is contacted. Friedman testified on voir dire that he had never run a crash test with side airbags, nor had he ever personally designed a side airbag system. Although he had written papers about side impacts, he had never written papers about side impact airbags. But Friedman also testified that he had designed frontal airbags before. He had also worked on designs to minimize side impact injuries in five vehicles. Friedman had also worked on an experimental research safety vehicle for the federal government that was never manufactured, but that could protect unbelted occupants in a forty mile per hour side impact collision. Friedman had built and tested airbags that protected unbelted occupants from fifty mile per hour impacts and authored numerous papers, reports, and government research on airbags, padding, and side impacts. He had investigated about thirty rollover cases with Suburbans and CK vehicles. In his deposition testimony, Friedman stated that most of his work at that time was research and was mostly related to rollovers and rollover occupant protection, which included “some relationship to the side impact.” In 1986 or 1988, he did computer modeling to analyze the performance of side impact airbags, but he had never done any statistical analysis of side impact airbags. But when asked whether he had “done statistical analysis of crash conditions and their relationship to injuries,” Friedman answered, “Certainly. That’s what I described as the basis for an appropriate design consideration.” Friedman had looked at GM crash tests when side airbags deployed and how they kept a dummy away from the side of the vehicle. Friedman testified at trial that if the installed thorax side airbag in the Suburban had timely deployed, it would have kept Stacey’s body, and thus, her head, away from the intruding B-pillar and Suburban interior. This opinion was based on Friedman’s study and reconstruction of the accident and his understanding of the way the human body interacts with intruding vehicle components during a collision. Friedman was offered as an expert with knowledge of the consequences to and behavior of vehicles and occupants in crashes, including side impact crashes, and how to design vehicles to minimize the impact to occupants in vehicle crashes, including side impact crashes. Based on the foregoing, we hold that the trial court did not abuse its discretion in determining that Friedman was qualified to testify on the actual subject on which he offered an opinion. GM further contends that Mahon, an airbag sensor engineer, was not qualified to testify regarding the sensing system on the Suburban because Mahon testified that his role as a sensing engineer was limited to developing a sensing system based on criteria provided by the vehicle manufacturer. He explained that his experience in the industry had been that the manufacturer decided how severe a crash must be before it wanted the airbag to deploy and then Mahon would design a system in which the airbag would deploy in crashes at least as severe as those dictated by the manufacturer. According to GM, this testimony shows that “Mahon deferred to manufacturers’ deployment criteria and had no experience establishing this criteria himself.” Mahon testified that he began working as an airbag sensor engineer in 1987 for a company that designed airbags and airbag sensors for automotive companies. That company at one point had “virtually 100 percent” of GM’s and Ford’s business. Mahon eventually became a vice president of the company, which has since gone bankrupt. Mahon expanded on his explanation about how he designed sensors based on criteria provided by the automotive companies by adding that because the testing that is done in the industry doesn’t cover every possible crash, my role is a little bit more than just picking a point [for deployment] and saying if I meet that point, that’s fine. You’ve got to look at meeting the severity requirements. Generally speaking, those severity requirements are based on injury criteria. In an affidavit presented by appellees to the trial court pretrial, Mahon averred that “[o]ver the years, [he had] personally reviewed literally thousands of crash test accelerometer plots and [had] a comprehensive understanding of all aspects of air bag sensor development crash testing.” He also averred that “[t]o ensure that our customers had at least a basic understanding of the principles of crash sensing,” he designed and taught a course known as Sensor 101 to engineers from various automotive companies, including GM. In the course, Mahon covered “the difference between crush and passenger compartment sensing methodologies, sensor system design, sensor calibration, and other matters of interest to restraint engineers.” According to Mahon, “[t]he course was designed to inspire discussions of ways to solve crash sensing problems ... admittedly with ... sensors” manufactured by his former company. Mahon further averred in the affidavit that “[a]s a result of the time spent in the automotive industry where [he] was involved in the design and manufacturing of air bag crash sensors, [he] gained considerable knowledge about how they perform under various conditions.” From May 1988 through the early 1990s, Mahon was “responsible for the sensor calibration and simulation activities of the corporation.” Mahon stated further that [c]rash sensing for side impact crashes is no different in principle than crash sensing for frontal impact crashes. The differences are in the details rather than the basic principles.... The significant difference between side impact events and frontal events is that there is much less time to gather data and make a decision in a side impact than in a frontal impact. Mahon’s testimony and affidavit show that in order to design a side airbag sensing system based on manufacturers’ criteria, he had to have comprehensive knowledge not only of how those sensors work, but when they must work based not only on the manufacturers’ criteria, but on injury criteria as determined by crash tests and industry knowledge. We hold that the trial court did not abuse its discretion in determining that Mahon was qualified to testify about the actual subject on which he offered an opinion. Thus, we conclude that the trial court did not abuse its discretion by determining that Friedman and Mahon were qualified to testify about the design of the Suburban’s airbag system and by denying GM’s motions to exclude their testimony. 2. Evidence of Unreasonably Dangerous Design and Safer Alternative Design GM also contends that there is no evidence that the design of the side airbag system in the Suburban was unreasonably dangerous or that a safer alternative design existed. a. Applicable Law When a claimant alleges a design defect, the burden is on the claimant to prove by a preponderance of the evidence that (1) there was a safer alternative design and (2) the defect was a producing cause of the personal injury, property damage, or death for which the claimant seeks recovery. Tex. Civ. PRAC. & Rem. Code Ann. § 82.005(a) (Vernon 2005); Davis v. Conveyor-Matic, Inc., 139 S.W.3d 423, 429 (Tex.App.-Fort Worth 2004, no pet.). A claimant must not only meet the proof requirements of the statute but must show, under the common law, that the product was defectively designed so as to be unreasonably dangerous, taking into consideration the utility of the product and the risks involved in its use. Hernandez v. Tokai Corp., 2 S.W.3d 251, 257 (Tex.1999); Honda of Am. Mfg., Inc. v. Norman, 104 S.W.3d 600, 604 (Tex.App.-Houston [1st Dist.] 2003, pet. denied). b. Side Airbag Design Appellees contended at trial that Stacey’s side airbag should have deployed in this type of accident; GM contended that the deployment of a side airbag was not necessary in this type of accident and that GM specifically designed the side airbag system not to deploy in this type of accident. 1. Unreasonably Dangerous GM first contends that it was entitled to a presumption — which appellees did not overcome — that the Suburban did not have a design defect because it had complied with standards promulgated by the National Highway Transportation Safety Administration (NHTSA) for the side structures of vehicles, which for 2001 vehicles did not require manufacturers to install any type of side airbag whatsoever. To support its contention, GM cites section 82.008(a) of the civil practice and remedies code and Harper. Tex. Civ. PRAC. & Rem. Code Ann. § 82.008(a) (Vernon 2005); Harper, 61 S.W.3d at 124. Section 82.008(a), which provides that a defendant in a products liability suit is entitled to a rebuttable presumption of no design defect when it shows it has complied with NHTSA minimum standards, applies only to suits filed on or after July 1, 2003. Tex. Civ. Peac. & Rem.Code Ann. § 82.008(a). This case was filed on May 28, 2003, so the rebuttable presumption in section 82.008(a) is not applicable here. Harper held that “[compliance with NHTSA regulations provides a presumption of no design defect.” Harper, 61 S.W.3d at 124 (citing Sims v. Washex Machinery Corp., 932 S.W.2d 559, 565 (Tex.App.-Houston [1st Dist.] 1995, no writ)) (emphasis added). But Sims stands only for the proposition that evidence of such compliance is strong and substantial — but not conclusive — evidence that a product was not defectively designed. Sims, 932 S.W.2d at 565 (citing Lorenz v. Celotex Corp., 896 F.2d 148, 150-51 (5th Cir.1990), which cites 59 Tex. Jur.3d Products Liability § 67 (1988), for the same proposition in a case applying Texas law); see also Coleman v. Cintas Sales Corp., 40 S.W.3d 544, 548 n. 1 (Tex.App.-San Antonio 2001, pet. denied). Neither Sims nor Lorenz describes this characterization of the evidence — as strong and substantial, but not conclusive — as a “presumption.” Thus, we disagree with the Eastland Court of Appeals’s characterization of such evidence as a presumption. Here, our review is for legal, not factual, sufficiency. Because evidence that GM complied with NHTSA standards is not conclusive, it cannot render appellees’ evidence insufficient. See Martinez, 977 S.W.2d at 334. Thus, regardless of GM’s strong, substantial evidence of no design defect, we must, under the no-evidence standard of review, determine whether ap-pellees’ design defect evidence is so weak as to be a mere scintilla or whether it is “some evidence,” i.e., more than a scintilla. Kindred, 650 S.W.2d at 63. GM also contends that appellees failed to show that GM’s design was unreasonable because they failed to take into account the risk of injuries caused by deployment of the airbag, which GM and appellees refer to as inflation-induced injuries, and because they failed to offer any evidence that deploying the airbag at the time of the second impact near Stacey would have prevented an occupant from hitting the side of the Suburban. See Coastal Transp. Co. v. Crown Cent. Petro. Corp., 136 S.W.3d 227, 232 (Tex.2004); Harper, 61 S.W.3d at 124. According to Dr. Thomas Perl, GM’s accident reconstructionist, the eighteen-wheeler initially impacted the Suburban at an approximately thirty-degree angle at a speed of forty-three miles per hour, but it was still traveling approximately thirty-nine miles per hour when it impacted the B-pillar “flush up against the Suburban,” which caused a change in velocity of the Suburban of thirty-one miles per hour. One of GM’s experts, Daniel Faust, testified by deposition that the side airbag should deploy when a 3,000 pound object crashed into the Suburban at a ninety-degree angle and a speed of thirteen to fourteen miles per hour or “if there’s a severe risk of injury to the occupant.” The evidence shows that one of GM’s performance goals was that the side airbag would have a deployment immunity (i.e., it would not deploy) in a ninety-degree side impact up to 17.5 miles per hour, and this result was achieved in tests presented to its Performance Assessment Committee (PAC). Thus, there is evidence that when the eighteen-wheeler hit the B-pillar of the Suburban, (1) the front of the eighteen-wheeler was “flush” with the side of the Suburban and, therefore, the body of the eighteen-wheeler was at a right angle to the Suburban, (2) that it hit the B-pillar area of the Suburban at a speed greater than that at which the airbag was designed not to deploy, and (3) that it hit the B-pillar area of the Suburban at a speed greater than the thirteen or fourteen miles per hour at which GM’s expert testified the side airbag should deploy. Friedman, Mahon, and Robert Caldwell, another of appellees’ experts who reconstructed the accident, testified that the impact at the B-pillar was sufficiently severe to require deployment of the side airbag. The interior of the Suburban intruded into the passenger compartment about eight to ten inches at the top and about twelve inches at the level of the window sill. Rescue personnel had to use the Jaws of Life® to get Stacey out of the Suburban after the accident. The foregoing evidence is sufficient to support the conclusion that the side impact in this case was severe enough that the side airbag should have deployed and that a side airbag design that did not allow for the side airbag to deploy in this type of crash was unreasonably dangerous based on the severity of the crash. GM contends, however, that it had designed the airbag not to deploy in an accident like Stacey’s because it was concerned about inflation-induced injuries and unnecessary deployments. In May 1999, just a few months before the 2001 Suburban was to go into production, the PAC reviewed the airbag design and expressed concerns. Specifically, the airbag did not timely deploy in a crash test involving a steel pole that intruded directly into the passenger compartment. The PAC was also concerned that there was no “data, analysis, judgment or rational[e]” addressing the deployment of side airbags in “vehicle angle impacts” or “side impacts from light trucks.” According to the PAC report, such impacts occur at high frequency in the field. The PAC was also concerned about the potential for injury due to inflation of the side airbags, particularly to children. Ultimately, the PAC concluded that “[l]ack of necessary information is responsible for the inability to concur on several of the evaluation considerations. However, performance is the issue in others. In particular, the PAC felt the SIAB [side impact airbag] presented additional risk to small children and from potentially late deployment without any demonstrate^ benefit.” Appellees’ expert, Mahon, testified that at the time of the May 1999 PAC report, GM had two choices with respect to the side airbag problems — change the sensing system or re-evaluate its deployment goals. But Mahon also testified that if GM’s goals were realistic, its only choice was to change the sensing system. According to Mahon, GM’s goals were realistic, but its engineers declined to change the sensing system because it would have created a costly delay to the program. In an October 1999 PAC report, which was issued after GM had already put the 2001 Suburban into production, GM ultimately accepted the slower inflation time in the pole test because the sensing system could not be adjusted to timely deploy while simultaneously being immune to a door slam. In other words, when GM experimented with making the airbag more sensitive to deployment, the airbag would deploy in unnecessary circumstances, such as when a door was slammed too hard. The PAC report states that “[s]uch a potential increase in unnecessary deployments would be inappropriate because it would increase possible exposure to inflation[-]induced injuries.” According to Mahon, when GM “drove up the calibrations to avoid the door slam, [it] ended up with late deployments on the pole crash.” However, Faust explained that the crash in the pole test was unlike the crash here because the pole, being extremely narrow, slices into the vehicle and causes much more intrusion. The October 1999 PAC report also shows that in response to the PAC’s earlier concern about side impact crashes between the Suburban and light trucks, “[a]n explanation was provided.... This crash sensing system evaluation was judged from the harder to sense FMVSS 214 barrier impacts.” According to Mahon, the FMVSS 214 test is a “moving to formidable barrier” test, which was designed to be a test in which the sensor does not deploy. Mahon could not understand how running a test in which the airbag specifically does not deploy would address the PAC’s concerns about the timing of deployment in a collision in which the airbag was supposed to deploy. According to Mahon, the test presented to the PAC in October “doesn’t matter.” The October 1999 report also shows that GM’s engineers declined to perform an angled impact crash test because it was only required in Europe. Mahon testified that the crash in this case was an angled impact and that data says that angled impacts account for “something like 30 percent” of side impacts. The trial court admitted evidence that there was only a minimal risk of inflation-induced injury due to deployment of side airbags; specifically, appellees offered an October 2001 NHTSA study, which identified no fatalities and only three reports of injuries, two minor, resulting from the deployments of side airbags as of the date of the study. A GM internal document calculated this risk as “0.80 serious inflation[-]induced injury event per 12 million car years.” The document also noted that this projection was “for unrestrained or improperly restrained children and is not adjusted for the positive effects of the warnings and increased public awareness.” Both Friedman and Dr. Carley Ward, appellees’ biomechanical engineering expert, testified that they had not seen any inflation-induced injuries from side airbags. Mitchell Scherba, a GM safety director, also acknowledged that side airbags can be more “benign” in larger vehicles because there is more space between the occupant and the location where the airbag is mounted. Moreover, Mahon testified that “you could still have inflation-induced injuries” in a wanted deployment. Jurors are the sole judges of the credibility of the witnesses and the weight to give their testimony. City of Keller, 168 S.W.3d at 819. We therefore conclude that in determining whether GM’s design was unreasonable, the jury was entitled to weigh appellees’ evidence of the benefit of occupant protection from side airbag deployment in a crash of this type against GM’s evidence of the slight risk of inflation-induced injury and unwanted deployment and to determine based on that evidence that a design which favored the reduction of the slight risk of inflation-induced injury and unwanted deployment over occupant protection was unreasonable. Based on the foregoing evidence, we conclude that appellees introduced more than a scintilla of evidence that the 2001 Suburban had a design defect that was unreasonably dangerous. 2. Safer Alternative Design GM also contends that there is no evidence of a viable safer alternative design. In addition to proving that a product is unreasonably dangerous, a plaintiff must prove that (1) there was a safer alternative design, (2) the safer alternative design would have prevented or significantly reduced the risk of injury, without substantially impairing the product’s utility, and (3) the safer alternative design was both technologically and economically feasible when it left the manufacturer’s control. Tex. Civ. PRAC. & Rem.Code Ann. § 82.005(a)-(b); Gen. Motors Corp. v. Sanchez, 997 S.W.2d 584, 588 (Tex.1999); Bic Pen Corp. v. Carter, 171 S.W.3d 657, 671 (Tex.App.-Corpus Christi 2005, pet. filed). To establish a safer alternative design, a plaintiff must show that the alternative design would not, under other circumstances, impose an equal or greater risk of harm. Martinez, 977 S.W.2d at 337. The safety benefits from the proposed design must be “foreseeably greater than the resulting costs, including any diminished usefulness or diminished safety.” Id.; Costilla v. Crown Equip. Corp., 148 S.W.3d 736, 739 (Tex.App.-Dallas 2004, no pet.). In support of their theory that a safer alternative design was available for the system, appellees “proposed the addition of a second side airbag sensor to be used alone or, optimally, in combination with inflatable head protection.” GM contends that there is no evidence that the alternative system proposed by appellees would have “commanded deployment in time to cushion [Stacey’s] collision with the B-pillar.” GM also contends that the testimony of Friedman and Mahon with regard to Mahon’s proposed two-sensor system was conclusory and speculative and not based on any tests, studies, data, or other scientific support. See Earle v. Ratliff, 998 S.W.2d 882, 890 (Tex.1999) (“An expert’s simple ipse dixit is insufficient to establish a matter; rather, the expert must explain the basis of his statements to link his conclusions to the facts.”); Gammill, 972 S.W.2d at 726-27; see also Coastal Transp. Co., 136 S.W.3d at 232. Specifically, GM complains that the evidence shows that the two-sensor system had never been implemented on an actual vehicle and no manufacturers had ever purchased the two-sensor system developed by Ma-hon’s former employer. Friedman, Mahon, and Dr. Ward all testified that a timely deployed side airbag would have prevented or reduced the risk of Stacey’s injuries. Mahon opined that the airbag did not deploy here because GM chose a system with only one sensor to pick a calibration that would achieve door slam immunity even though it would not meet the requirements for a lot of potential crashes. Mahon theorized that a second sensor located in the B-pillar would “make the sensing system twice as smart” by accurately and timely detecting events calling for “wanted deployments.” Mahon thus concluded that a two-sensor system would have timely deployed in Stacey’s accident. Mahon testified that “something that we knew about in the late 1980s from frontal crashes, is that when you don’t get enough information in a crash ... if you add another sensor, you now have two devices gathering information, two devices able to think about what’s going on, so you’re twice as smart.” Mahon testified that when another sensor is added to the system, there is better deployment on the “must deploy” crashes and better immunity to the “must not deploy” crashes. To support his theory that a timely deployed airbag would have prevented Stacey from hitting the B-pillar, Mahon testified in conjunction with a video of a GM crash test in which a fourteen inch steel pole struck the door of a vehicle at eighteen miles per hour. In that test, the airbag was able to timely inflate and protect the crash test dummy from the pole because the sensor was properly calibrated to timely deploy. According to appellees, if a single sensor can sense and timely deploy in that test, then it is logical that a second sensor located in the B-pillar would also “command timely deployment” when being impacted by an eighteen wheeler going thirty-nine miles per hour. The record supports appellees’ contention. According to the reconstruction performed by Caldwell — GM’s accident reconstruction expert — Stacey’s head was approximately five and one-half inches from the B-pillar at sixty milliseconds after the initial impact between the eighteen-wheeler and the Suburban, and it struck the B-pillar at the earliest at ninety milliseconds after impact. Caldwell stated that these were conservative estimates and that a head attached to a body would travel toward the B-pillar a little slower than the free particle upon which he based his conclusion. The eighteen wheeler impacted the B-pillar at thirty to fifty .milliseconds. According to Caldwell, when the Suburban rotated around into the front of the eighteen-wheeler, the B-pillar was directly impacted. Using Caldwell’s measurements, Mahon opined that a properly designed sensor in the B-pillar would have detected the impact, commanded deployment, and caused the airbag to inflate in approximately ten milliseconds. Thus, at sixty milliseconds into the accident, Stacey’s head would have been about five inches from the B-pillar, enough clearance for the airbag to inflate between her body and the side of the Suburban. Mahon testified in detail about a two-sensor system and stated that it provides better deployment in crashes in which deployment is necessary and also better immunity from unwanted deployments. He described how having a sensor in the B-pillar works together with the more forward sensor in the door to provide “better decision-making, [and] better discrimination.” According to Mahon, “door slam immunity is trivial” with a two-sensor system because the two sensors detect opposite impacts and “have to come together” on a deployment decision. Thus, you would have a much better chance of deployment in a crash like this one in which the initial impact was a little bit off (i.e., not at a ninety-degree angle). We believe that Mahon sufficiently explained the basis for his testimony about how a two-sensor system would work based on his experience in developing frontal side airbag sensors as well as side airbag sensors. That the two-sensor system had not been used on other vehicles in production does not render Mahon’s testimony speculative or conclusory. Cf. Norman, 104 S.W.3d at 606 (holding that to prove safer alternative design, technology must be in existence or scientific knowledge to produce it must be reasonably achievable). Friedman explained that the purpose of the sensor is to detect “the acceleration ... of the surface to which it’s attached.” The change in acceleration indicates a crash is occuiTing, and the sensor then sends a signal to the airbag inflator to deploy. Thus, even though the initial impact was not near Stacey, if a sensor had been located in the B-pillar, it would have recognized that the vehicle was being pushed and detected the acceleration that caused Stacey to move toward the B-pillar. Thus, Friedman’s testimony is consistent with Mahon’s. Mahon compared videotapes from two GM crash tests: one in which an airbag timely deployed and one in which an airbag did not timely deploy. Mahon testified that these tests show that a timely deployed side airbag can hold an occupant away from intrusion and dramatically reduce head injuries. Dr. Ward also used videos from GM crash tests to show how a properly inflated side airbag will keep a person’s head away from the interior of the vehicle. Based on these test results and their experiences, both Mahon and Dr. Ward concluded that a timely deployed side airbag would have been out, inflated, and available for the protection of Stacey’s head. And Dr. Ward stated that if the airbag had had head protection, it would have kept Stacey’s head even further away from the B-pillar. Dr. Ward testified that if the existing thorax bag had inflated, it would have tended to keep Stacey away from the B-pillar and when she reached it, she would have hit it with less force. Even an inflated airbag with only thorax protection keeps the head further inboard than if an airbag had not been there. According to Dr. Ward, the existing thorax bag may not have kept Stacey’s head from hitting the B-pillar altogether, but it would have prevented her from having irreversible brain injury because she would have hit it with less force. Additionally, Dr. Ward explained that an airbag absorbs energy. Dr. Ward also testified that in all of her experience, she had never seen, and was not aware of, any problems with inflation-induced injuries from side airbags. She also testified that her research on alternatively designed airbags, i.e., side curtain airbags and other airbags with added head protection, shows that there is no inflation-induced injury risk with those designs, including with alternative airbags designed by GM. See Martinez, 977 S.W.2d at 338 (holding that defendant’s evidence of a greater risk of harm was not conclusive; thus plaintiffs evidence created a fact issue for the jury to resolve). Based on the foregoing, we conclude that there is at least some evidence, more than a scintilla, of the existence of a safer alternative design. Having determined that the evidence is legally sufficient to support the jury’s finding that the Suburban possessed an unreasonably dangerous design defect and that a safer alternative design existed, we overrule GM’s second issue. C. Causation GM contends in its first issue that even if there is sufficient evidence to support the jury’s determination that the Suburban suffered from a design defect, there is no evidence that the defect proximately caused Stacey’s injuries. Specifically, GM contends that the testimony of appellees’ biomechanical expert, Dr. Ward, is conelu-sory, speculative, lacks probative value on its face, and is not grounded in the methods and procedures of science. GM’s first three contentions are determinable on the face of the record. 1. Probative Value of Dr. Ward’s Testimony Dr. Eugene George, Stacey’s doctor while she was at the hospital after the accident, testified that as a result of the accident, Stacey had a right occipital con-dyle fracture, multiple facial fractures on her left side underneath her eye and on her nasal bone, a dilated, fixed (or “blown”) right pupil, a diffuse axonal injury to her brain stem, damage to the frontal lobes of her brain, and a traumatic sub-arachnoid hemorrhage. According to Dr. George, Stacey’s right side injury was consistent with a forceful blow to the right, back side of the head. When asked to explain what had to happen for Stacey to receive those injuries, Dr. George testified as follows: This is a complex area when you try to work out how injuries occur. In this case you have an injury to the base of the skull on the right side, and you have fractures on the left face. And then you have some diffuse injuries within the brain in between. It would appear that she had an injury on one side and — and a counter injury that — that caused her to go forward or backward on the other. I think that’s probably safe to ... say that ... it occurred. The unlikely, [sic] without some sort of a rotational and flection [sic] injury that she would have that kind of a occipital skull — condyle fracture. I guess my only caveat is after seeing a few decades of these injuries, the brain not only is ... complex, but it’s in these complex structures and dura sacs within the brain that it’s hard to categorize injuries, the rotation, the translational forces and so forth. As we try to set up a scenario for certain injuries, as we try to investigate how we can improve the treatment of them, it’s hard even to classify them together.... Having said that, I would say that this patient had had some sort of ... a lateral rotation injury, whether she was struck on this side and went forward and hit her head on that side — on the left side or the reverse, it’d be a little difficult to say beyond that. [Emphasis added.] Dr. Ward testified that Stacey received her primary head injury from contacting the B-pillar as she moved back and to the right towards the B-pillar as the Suburban moved out from under her. Dr. Ward based her determination by looking at the Suburban; Stacey’s X-rays, CAT scans, and medical records; Caldwell’s accident reconstruction; and the movements of a human volunteer in an exemplar vehicle. She also saw slight marks on the B-pillar and deformation on the door window on Stacey’s side. Dr. Ward testified that the primary forces in the accident would have moved Stacey to the right and to the back and that Stacey’s injuries were consistent with a downward head injury. She also testified that she could tell the initial impact with the B-pillar was the primary force in the accident because Stacey’s CAT scans showed swelling on the right side of her brain. In addition, Dr. Ward explained that Stacey could have incurred the left-side injuries much more easily because the eye socket area is very easy to fracture and this type of impact is energy-absorbing, i.e., the face breaks to absorb energy and protect the brain. Dr. Ward explained that people can get a lot of facial fractures with no brain injury because the face absorbs the shock of the blow. She said that there was nothing in the event to pull Stacey hard to the left and that GM’s theory that the primary forceful blow was from the left was inaccurate because human bodies don’t rebound well; a person can’t have a rebound and subsequent impact with more energy than the first impact. Dr. Ward admitted that a dummy’s head will rebound in such a circumstance, but that is because the dummy’s head is made of rubber. Dr. Ward theorized that the left side of Stacey’s face turned toward the door when the Suburban rolled off the eighteen-wheeler and spun around and that that caused the injuries to the left side of her face. She stated that the only thing happening after the impact near Stacey was spinning, which would keep Stacey to the right. According to Dr. Ward, if Stacey had rebounded to the left side and hit Carol’s head, Carol would have a corresponding head injury. Carol was knocked unconscious, but she did not have any swelling, bruising, or other evidence that she had received a blow to the right side of her head. Dr. Ward agreed that a diffuse axonal head injury is caused by movement of the brain relative to its attachments. GM contends that Dr. Ward’s testimony fails to account for the diffuse axonal injury diagnosed by Stacey’s own physician because that type of injury is not caused by a single blow, but by a rotational movement of the head. GM also contends that Dr. Ward’s theory is based on speculation about how Stacey moved during the accident, rather than testing, studies, or data. See Volkswagen of Am., Inc. v. Ramirez, 159 S.W.3d 897, 912 (Tex.2004). According to GM, “[i]t is undisputed that a diffuse axonal injury results when a force causes a person’s head to rapidly rotate and whip around. Ward never explained what caused [Stacey’s] head to move in that way.” Appellees contend, however, that this evidence is not undisputed. Rather than confirming that a diffuse ax-onal injury occurs only from rapid rotational movement from side to side, Dr. George testified that a diffuse axonal injury occurs when an impact moves the brain from side to side or from front to back or any other way. He also agreed that Stacey’s brain injury was consistent with a “heavy blow to the right side and back of [her] head.” And the excerpt from Dr. George’s testimony that we previously quoted specifically contemplates Stacey’s head hitting something and then moving forward. GM’s position also fails to take into account the fact that appellees’ experts testified that the Suburban “rapidly rotated” off of the eighteen-wheeler, which forced Stacey’s head down to the left and onto the window. GM’s own expert based his disagreement with Dr. Ward’s theory on his view that a “very violent motion” pitched Stacey to the left. But the jury was entitled to believe Dr. Ward’s testimony over the testimony of GM’s expert. See Morrell v. Finke, 184 S.W.3d 257, 282 (Tex.App.-Fort Worth 2005, pet. filed). Dr. Ward’s conclusions were based on her extensive knowledge about how human bodies, which she distinguished from crash test dummies in simulated accidents, move when forces are applied to them. She explained how she came to those conclusions based on her examination of the forces applied in the accident, which she determined in part by examining the damage to the vehicle and the results of an accident reconstruction — a common method of determining after the fact what actually happened during an accident. Dr. Ward even explained how it is impossible to know exactly what happened to a person in an accident because we cannot perform tests with real people. We therefore conclude and hold that her testimony had probative value and was not speculative. 2. Reliability of Dr. Ward’s Testimony GM also contends that Dr. Ward’s testimony is unreliable because it fails to account for Stacey’s left-side injuries, the inward forces applied to the B-pillar (i.e., the fact that the B-pillar was intruding in on the structure as Stacey was moving toward it) as evidenced by the leftward (toward the driver’s side) tilting of Stacey’s seat, and Carol’s unconsciousness at the scene. a. Standard of Review Expert testimony must be relevant to the issues in a case and be based on a reliable foundation. Tex.R. Evid. 702; E.I. du Pont de Nemours & Co. v. Robinson, 923 S.W.2d 549, 553 (Tex.1995); Gross, 149 S.W.3d at 237. The trial court must make an initial determination of whether the expert’s testimony is relevant and reliable so as to be admissible. Robinson, 923 S.W.2d at 557; Gross, 149 S.W.3d at 237. However, even when challenged expert testimony is admitted by the trial court, a party may later complain on appeal that the expert testimony is legally insufficient to support the judgment because it is unreliable. Mar. Overseas Corp. v. Ellis, 971 S.W.2d 402, 409 (Tex.), cert. denied, 525 U.S. 1017, 119 S.Ct. 541, 142 L.Ed.2d 450 (1998); Gross, 149 S.W.3d at 237. Unreliable expert testimony is not evidence. Merrell Dow Pharms., Inc. v. Havner, 953 S.W.2d 706, 714 (Tex.1997), cert. denied, 523 U.S. 1119, 118 S.Ct. 1799, 140 L.Ed.2d 939 (1998). In determining whether expert testimony is reliable and, therefore, some evidence supporting the judgment, the appellate court must employ “an almost de novo-like review and, like the trial court, look beyond the expert’s bare testimony to determine the reliability of the theory underlying it.” Austin v. Kerr-McGee Refining Corp., 25 S.W.3d 280, 285 (Tex.App.-Texarkana 2000, no pet.); see Guadalupe-Bianco River Auth. v. Kraft, 77 S.W.3d 805, 808 (Tex.2002); Havner, 953 S.W.2d at 713. “An expert’s simple ipse dixit is insufficient to establish a matter; rather, the expert must explain the basis of his statements to link his conclusions to the facts.” Earle, 998 S.W.2d at 890. The court does not focus on the correctness of the expert’s opinion, but on the reliability of the analysis the expert used in reaching his or her conclusions. Gross, 149 S.W.3d at 237; State Farm Fire & Cas. Co. v. Rodriguez, 88 S.W.3d 313, 319 (Tex.App.-San Antonio 2002, pet. denied) (op. on reh’g). The supreme court has articulated six nonexclusive factors appellate courts should consider in determining whether scientific testimony is reliable: (1) the extent to which the theory has been or can be tested; (2) the extent to which the technique relies upon the subjective interpretation of the expert; (3) whether the theory has been subjected to peer review and publication; (4) the technique’s potential rate of error; (5) whether the underlying theory or technique has been generally accepted as valid by the relevant scientific community; and (6) the nonjudicial uses that have been made of the theory or technique. Havner, 953 S.W.2d at 714; Robinson, 923 S.W.2d at 557. Scientific evidence that is not based on “the methods and procedures of science” is no more than a “subjective belief or unsupported speculation.” Robinson, 923 S.W.2d at 557. “If the foundational data underlying opinion testimony are unreliable, an expert will not be permitted to base an opinion on that data because any opinion drawn from that data is likewise unreliable.” Havner, 953 S.W.2d at 714. In addition, if the expert’s conclusions from data are based upon flawed methodology, the testimony is unreliable even if the data is sound. Id.; Gross, 149 S.W.3d at 238. The Texas Supreme Court has recognized that the Robinson analysis may not apply to certain types of expert testimony. Helena Chem. Co., 47 S.W.3d at 499 (citing Gammill, 972 S.W.2d at 726); Gross, 149 S.W.3d at 238. In these types of cases, there must still be a rehable basis for the expert’s opinion. Helena Chem. Co., 47 S.W.3d at 499; Gross, 149 S.W.3d at 238. “Experience alone may provide a sufficient basis for an expert’s testimony” unless there is “too great an analytical gap between the data and the opinion proffered.” Gammill, 972 S.W.2d at 726-27. Thus, regardless of whether the Robinson factors are applied, the proponent of the expert testimony must prove that it is based upon a rehable foundation. Gross, 149 S.W.3d at 238. b. Analysis GM contends that Dr. Ward’s testimony about how Stacey’s body and head moved during the accident is based on mere speculation about how Stacey moved in the accident and fails to close the analytical gap between Dr. Ward’s opinion and the evidence. Dr. Ward testified about her qualifications and background at trial. A biome-chanical engineer, Dr. Ward had been working and determining how people got hurt in crashes for thirty-eight years. She had a bachelor’s degree in mechanical engineering, a master’s degree in engineering mechanics, and a Ph.D. in biome-chanics and dynamics. She also attended medical school at UCLA for “a number of years” to study the structure of the human body. Dr. Ward testified that after she received her master’s degree, she “began to specialize in the study of dynamics, how things respond on impact.” She worked in the auto industry after that and “utilized the dynamics of impact of structures at that time.” Dr. Ward had also taught engineering and taught medical students how to investigate how injuries occur. Dr. Ward had been on a national committee for the Department of Transportation “looking at how people are injured in car crashes” and had served on other similar committees: “[o]n identifying injuries, on doing autopsies, [and] how to identify the injuries in the autopsy.” Dr. Ward had also served for ten years as the deputy coroner at USC medical school. Dr. Ward testified that she had published peer-reviewed literature about how to investigate how head injuries occur and that she had written about fifty papers. Moreover, she answered ‘Tes” when asked if she had developed “scientific models to absolutely determine reasonably what happened to people and how them head injuries occurred” and whether such models were used by engineers in the field. GM first claims that Dr. Ward’s testimony about Stacey forcefully hitting the B-pillar is suspect. GM claims that Dr. Ward testified that Stacey’s head moved downward and the B-pillar hit her on the top of the head. But Dr. Ward testified that the right side of Stacey’s head was impacted toward the back and just above her right ear. GM also criticizes Dr. Ward’s testimony that swelling on the right side of Stacey’s head was an indication that the B-pillar hit Stacey on the right side of her head because the swelling was not identified until the day after the accident. Dr. Ward explained that such swelling does not always occur at the point of impact on the same day; it can occur up to twelve to twenty-four hours after impact. In addition, the EMT report taken on the day of the accident indicates external head injuries on Stacey’s right side. GM also contends that Dr. Ward failed to account for the left-side fractures to Stacey’s face and that her testimony that after Stacey’s head hit the B-pillar, she went forward and hit the left side of her face on the windowsill is mere speculation. Both Dr. Ward and Friedman testified that the forces applied when the Suburban started spinning to the right and off the eighteen-wheeler would have caused Stacey to pull hard to the right against the interior of the Suburban. Dr. Ward also found a one-eighth inch deformation in the trim piece on the top of the windowsill next to where Stacey had been sitting. GM additionally appears to contend that because its expert concluded that the intrusion of the B-pillar caused Stacey’s entire body to move violently to the left (inward toward the center of the vehicle) and because Dr. Ward did not explain that conclusion, Dr. Ward’s testimony about how Stacey received her left-side injuries is somehow deficient. But Dr. Ward, Caldwell, and Friedman all explained how Stacey’s seat would have moved under her instead of with her. The jury was entitled to believe appellees’ experts over GM’s. Morrell, 184 S.W.3d at 282. GM also claims that Dr. Ward failed to account for Carol’s unconsciousness immediately after the accident. But Dr. Ward testified that it is not uncommon for a person to lose consciousness after such an accident even if that person did not receive a blow to the head. Further, GM’s implicit assertion that if Carol’s unconsciousness was due to a blow to the head, then that blow could have only come from Stacey’s head, does not negate Dr. Ward’s testimony. Indeed, Dr. Ward’s testimony that Stacey’s head did not impact Carol’s is supported by the evidence that Carol had no external marks,, bruises, or swelling on her head or any other indication that she had been hit in the head during the accident. GM further contends that Dr. Ward ignored the fact that Stacey’s seat was tilted inward as a result of the impact. But rather than ignoring the tilting, as we have observed above, Dr. Ward testified that Stacey would have moved independently of the seat because it was moving out from under her and, thus, Stacey would still have moved the way the forces on her body were moving. Friedman also agreed that what happened to the seat had nothing to do with what happened to Stacey even though she was belted in. Finally, GM argues that Dr. Ward failed to take into account the absence of any head injury to Rachel. According to GM, Stacey’s injuries are inconsistent with Rachel’s, yet Rachel was on the same side of the Suburban as Stacey, seated directly behind her. But Friedman distinguished their positions. Rachel was injured more on the rest of her body because she was seated at the place of the greatest crush. Rachel was not seated where her head could be impacted by, or could impact, the C-pillar. It is clear from most of the experts’ testimony, both appellees’ and GM’s, that the forces applied to Stacey and Rachel were different because of the two-impact scenario. Dr. Ward did not find Rachel’s injuries to be a “useful tool” in determining how Stacey received her injuries; thus, her failure to account for Rachel’s injuries does not show that her testimony was unreliable. As we have explained above, Dr. Ward did not fail to account for Stacey’s left-side injuries, the inward forces applied to the B-pillar, or Carol’s unconsciousness at the scene, as GM contends. Based on Dr. Ward’s qualifications and experience in the study of how human bodies move and react when forces are applied to them, we conclude and hold that Dr. Ward’s causation testimony was reliable. 3. GM’s Theory of Causation In support of its contention that the evidence is legally insufficient to support the jury’s findings on causation, GM focuses heavily on its own theory of how Stacey received her injuries: that Stacey’s brain injury was caused by impact with Carol’s head rather than any part of the vehicle. GM’s expert testified at trial that although Stacey initially hit the right side of her head on the B-pillar, her brain injury was caused when she rebounded off the B-pillar and hit the left side of her head against Carol’s head. But as we have already explained, Dr. Ward also testified as to her theory of how Stacey received her left-side injuries — that after Stacey hit her head on the B-pillar, the Suburban started spinning, which forced Stacey’s head to spin down and hard to the right against the windowsill — and we have determined that Dr. Ward’s testimony was rehable. It is the province of the jury to resolve conflicts in the evidence. City of Keller, 168 S.W.3d at 820. In addition, as we have already observed, the jury was entitled to believe the testimony of appel-lees’ expert as to causation over the testimony of GM’s. See Morrell, 184 S.W.3d at 282. Thus, GM’s causation evidence does not render appellees’ causation evidence legally insufficient. Based on our review of the entire record, we conclude that there is more than a scintilla of probative evidence that a design defect in the Suburban was a proximate cause of Stacey’s injuries. We overrule GM’s first issue. IV. Evidentiary Rulings In its third issue, GM contends that the trial court “repeatedly made evidentiary rulings that paved the way for a plea to the jury to render a verdict based on emotion, not law or fact. Independently and cumulatively, these rulings were reversible error.” Specifically, GM contends that the trial court abused its discretion by allowing appellees to introduce the Burry children to the jury in person, by excluding evidence of Carol’s negligence in causing the accident, by admitting appellees’ statistics and excluding GM’s statistics, and by admitting evidence more properly designated as rebuttal evidence during ap-pellees’ case-in-chief. A. Standard of Review A trial court’s rulings in admitting or excluding evidence are reviewable under an abuse of discretion standard. Nat’l Liab. & Fire Ins. Co. v. Allen, 15 S.W.3d 525, 527-28 (Tex.2000). An appellate court must uphold the trial court’s evidentiary ruling if there is any legitimate basis in the record for the ruling. Owens-Corning Fiberglas Corp. v. Malone, 972 S.W.2d 35, 43 (Tex.1998). To determine whether a trial court abused its discretion, we must decide whether the trial court acted without reference to any guiding rules or principles; in other words, we mu