ruling barring evidence that he consumed alcohol prior to the accident which resulted in his catastrophic injury.
The issue presented for this court's resolution is whether evidence of the plaintiff's consumption of alcohol prior to the accident in which he was injured may be introduced at trial, in light of New Jersey's evidentiary rule barring such evidence unless it is accompanied by independent "supplementary evidence" that the plaintiff was unfit to drive at the time of the accident. Specifically, this court must decide a previously unresolved question of New Jersey law: whether evidence that the plaintiff's blood alcohol level at the time of the accident was at or above .10%, constitutes sufficient "supplementary evidence" of intoxication. Because I conclude that evidence of a .10% or higher blood alcohol level is sufficient "supplementary evidence" of the plaintiff's intoxication in this case, and that the danger of unfair prejudice does not substantially outweigh the probative value of this evidence, plaintiff's in limine motion will be denied.
I. Facts and Procedural History
On July 28, 1991, Hulmes was severely injured as the result of a collision which occurred while he was riding a three-wheeled All-Terrain Vehicle ("ATV") designed and manufactured by Honda. Hulmes and his brother, Nicholas Hulmes, were riding their ATVs on the paved roadway of West Hunterdon Avenue in the City of Mantua, Gloucester County, New Jersey. There is some dispute as to how fast the ATVs were going at the time of the accident. Plaintiff's eyewitness testimony estimates their speed at about 17-20 m.p.h. See Dep. of Corey Thomas at 95. Defendants' accident reconstruction expert will testify that Hulmes's ATV was travelling at a speed of between 35-40 m.p.h. Report of David Mercaldi, Ph.D. at 8. It is undisputed that Hulmes's ATV came into contact with the ATV operated by his brother, although the force of this "collision" is an issue that is vigorously contested by the parties. Following the "collision," Hulmes's ATV rolled over, Hulmes was thrown from the vehicle, and both the vehicle and Hulmes came to rest some yards away. As a result of this accident, Hulmes suffered a spinal fracture at C3-C4 and spinal cord lesion, resulting in permanent quadriplegia.
Deborah Connelly, who lived nearby, heard the accident and went to the scene. Dep. of D. Connelly at 13, 19. Ms. Connelly testified that Nicholas Hulmes removed an unopened bottle of beer from the plaintiff's waistband shortly after the accident. Id. at 25-27. Ms. Connelly also testified that she did not smell alcohol on the plaintiff's breath, but, in view of the fact that the plaintiff was lying on the ground, she couldn't tell if the plaintiff was intoxicated. Id. at 39.
The Advanced Life Support Unit from Underwood Hospital arrived at the scene soon after the accident. Because Hulmes did not have a patent airway, the Advanced Life Support Unit paramedics inserted an endotracheal tube and ventilated Hulmes, using a bag and mask resuscitator. Dep. of C. McNulty at 98-99. Shortly thereafter, the trauma team from Cooper Hospital arrived by helicopter. The flight nurse, Catherine N. McNulty, R.N., testified that she continued ventilating Hulmes with the bag and mask resuscitator during transport to the trauma center, periodically suctioning from him an amber fluid that "smelled like alcohol." Id. at 105-108.
Plaintiff's brother, Nicholas Hulmes, testified at his deposition that neither he, nor his brother, had "consumed any alcoholic beverages" on the day of the accident. Dep. of N. Hulmes at 125. A member of the Sewell Fire Company Ambulance Squad and a witness at the scene, testified that Nicholas Hulmes was "belligerent" at the accident scene. Dep. of J. Neely at 24. Neely, however, also testified that he did not smell alcohol on Nicholas Hulmes's breath. Id. At oral argument, counsel represented that Hulmes himself had testified at his deposition that he could not remember whether he had consumed any alcoholic beverages prior to the accident.
The Third Circuit has noted that "in limine ruling on evidence issues is a procedure which should, in a trial court's discretion, be used in appropriate cases." In re Japanese Electronic Products Antitrust Litigation, 723 F.2d 238, 260 (3d Cir. 1983) (citing 21 C. Wright & K. Graham, Federal Practice and Procedure § 5037, at 193-99), rev'd on other grounds sub nom. Matsushita Elec. Indus. Co., Ltd. v. Zenith Radio Corp., 475 U.S. 574, 89 L. Ed. 2d 538, 106 S. Ct. 1348 (1986).
No doubt in an abundance of caution, plaintiff's counsel has identified over 1400 trial exhibits in plaintiff's portion of the Joint Final Pretrial Order. In the hope of narrowing the issues and expediting the trial of this case, this court granted the parties leave to file appropriate motions in limine. The parties subsequently filed a total of forty-four (44) in limine motions. On July 2, 1996, following a full day of oral argument on July 1st, this court entered an order deciding, or deferring until trial, forty-one (41) of the in limine motions. This opinion and the accompanying order decide one of the three motions taken under advisement following oral argument.
A. Plaintiffs' motion to exclude all references to alcohol consumption by Robert T. Hulmes.
Honda seeks to introduce evidence that the plaintiff was intoxicated at the time of the accident. Specifically, Honda proffers evidence of the presence of alcohol (ethanol) in Hulmes's blood, in an amount of 106.9 mg/dl, as determined by a blood serum test administered on a sample of Hulmes's blood drawn at Cooper Hospital approximately one hour after the accident,. Cooper Hospital Pathology Report at 8. Defendants' toxicology expert will testify that this reading extrapolates to a Blood Alcohol Content ("BAC") in whole blood of .11% at the time of the accident. Report of G. John DiGregorio, M.D. at 2.
A second Honda expert witness opines that the "most likely value for [Hulmes's] blood alcohol concentration" was "slightly above .10%. Plaintiff strenuously contests the accuracy of these extrapolations.
Moreover, plaintiff contends that the results of blood alcohol tests, unless supported by independent corroborating evidence of intoxication, not present in this case, are inadmissible under New Jersey law because the potential prejudicial effect of such evidence substantially outweighs its probative value.
B. The Prejudicial Effect of Evidence of Alcohol Consumption
The Third Circuit has held that, when confronted with the task of balancing the prejudicial effect of evidence of alcohol consumption with its probative value, a district court must look to state law to determine whether the prejudice to the party opposing the introduction of such evidence would be substantial. Rovegno v. Geppert Bros., Inc., 677 F.2d 327, 329 (1982) (relying upon Greiner v. Volkswagenwerk Aktiengeselleschaft, 540 F.2d. 85, 89 (3d Cir. 1976)).
Rovegno has proved to be a controversial decision. Judge Van Dusen, in a lengthy dissenting opinion, chastised the panel majority for its holding, declaring that Greiner may not so much have held, as assumed that the Erie Doctrine was applicable to a decision to admit evidence of drinking. See id. at 338 & n.21 (Van Dusen, J., dissenting) (discussing Erie R.R. v. Tompkins, 304 U.S. 64, 82 L. Ed. 1188, 58 S. Ct. 817 (1938)). There is an on-going and passionate debate about whether state law or federal law governs the admissibility of such evidence. Some authorities have suggested that the application of state law can be justified by simply rephrasing the question as one of the "materiality" under state law of evidence of alcohol consumption, rather than its probative value or prejudicial effect. See 5 Jack B. Weinstein et al., Weinstein's Evidence P 1101  (Matthew Bender 1995); see generally 22 Charles A. Wright & Kenneth W. Graham, Jr., Federal Practice and Procedure § 5201, at 234 & n.29 (1978 & Supp. 1996) (noting that the Third Circuit, in Rovegno, applied state law, "over vigorous dissent and without discussion of the effect of Rules 401 and 402").
Other courts have concluded that the Rovegno majority discounted the rule of Hanna v. Plumer, 380 U.S. 460, 14 L. Ed. 2d 8, 85 S. Ct. 1136 (1965), which held that federal law governed all "procedural" matters in federal courts, even in actions in which state law supplies the rule of decision. See McInnis v. A.M.F., Inc., 765 F.2d 240, 245 n.6 (1st Cir. 1985) (specifically criticizing Rovegno). The true difficulty, however, lies in "ascertaining whether a particular rule [of state law] is 'only evidentiary or is meant to prescribe rights and obligations of the parties not related to fact-finding.'" Weinstein, supra (citing Wellborn, The Federal Rules of Evidence and the Application of State Law in the Federal Courts, 55 Tex. L. Rev. 371, 396 (1977)). Cf. Gasperini v. Center for Humanities, 64 U.S.L.W. 4607, 4610 (U.S. June 24, 1996) ("Classification of a law as 'substantive' or 'procedural' for Erie purposes is sometimes a challenging endeavor.").
Several Circuit Courts of Appeals have disagreed with Rovegno, holding that a determination of the possible prejudicial effect of the introduction of evidence of alcohol consumption is within the trial court's discretion under Rule 403 of the Federal Rules of Evidence, and is not controlled by state law. See Romine v. Parman, 831 F.2d 944 (10th Cir. 1987) (relevancy and prejudice determinations are made under Fed. R. Evid. 403); McInnis, 765 F.2d at 244-45 (1st Cir. 1985) (a panel, which included then-Judge Breyer, ruled that the Federal Rules of Evidence, and not state law, apply to a determination whether to admit evidence of intoxication); Ballou v. Henri Studios, Inc., 656 F.2d 1147, 1153 (5th Cir. 1981); Levitt v. H.J. Jeffries, Inc., 517 F.2d 523, 525 (7th Cir. 1975). Indeed, the Rovegno majority does not explain how the application of Fed. R. Evid. 403, rather than state law, could "rationally affect private ordering or encourage forum shopping." McInnis, 765 F.2d at 246 (citing Erie). The weight of authority seems to support the conclusion reached in McInnis, that, in deciding whether to admit evidence of alcohol consumption, the application of Fed. R. Evid. 403 by a federal court cannot violate the "twin aims" of Erie. Hanna v. Plumer, 380 U.S. at 468.
Honda devotes a significant portion of its brief in opposition to plaintiff's motion to a discussion of the controversy engendered by Rovegno to demonstrate that Honda would prevail on this evidentiary issue if I were to apply Fed. R. Evid. 403, rather than New Jersey law. Rovegno, however, bars me from taking that path.
In Straley v. United States, 887 F. Supp. 728, 732-39 (D.N.J. 1995), a case brought under the Federal Tort Claims Act, in which, as in diversity cases, the rule of decision is supplied by state law, Judge Debevoise conducted a comprehensive analysis of the current state of the law in light of the debate over the Rovegno decision. Judge Debevoise finally concluded, as I do, that the Third Circuit meant what it said in Rovegno, and that, "on the question of alcohol consumption, [Rovegno ] requires the application of state law rules of admissibility." Id. at 737.
Happily, Rovegno itself points to one way out of the quagmire it has created. In its analysis of Fisher v. Dye, 386 Pa. 141, 125 A.2d 472, 476 (Pa. 1956), the leading Pennsylvania decision interpreting the admissibility of evidence of alcohol consumption, the Rovegno majority concluded "that the Pennsylvania decision implicitly requires the same discretionary weighing required by Rule 403. Thus, in interpreting Fisher v. Dye and its progeny, we may draw on our own decisions dealing with review of Rule 403 exercises." Rovegno, 677 F.2d at 329. Rovegno's invitation to district courts in this Circuit to inform the application of state law by reference to Rule 403, is reinforced by the happy coincidence that on July 1, 1993, New Jersey amended its rules of evidence, closely modeling its new rules on the Federal Rules of Evidence. Indeed, New Jersey's Rule of Evidence 403 closely parallels Fed. R. Evid. 403.
Accordingly, while Rovegno instructs me to apply New Jersey law to decide this issue, I may also look to the decisions of federal courts applying Rule 403 in balancing the probative value of evidence of alcohol consumption against the potential prejudicial effect of such evidence.
C. The "Balancing Test"
Under either New Jersey Evidence Rule 403 or its federal equivalent, a trial court may exclude otherwise relevant evidence "if its probative value is substantially outweighed" by the danger of unfair prejudice. Thus, as a threshold matter, I must decide whether the evidence of Hulmes's alcohol consumption is probative. Under either the New Jersey Evidence Rules or the Federal Rules of Evidence, relevant evidence is that which tends to prove or disprove any fact that is of consequence to the outcome of the action. Fed. R. Evid. 401; N.J. R. Evid. 401. Plaintiff, in this in limine motion, does not challenge the relevance of the evidence of Hulmes's alcohol consumption, but only the prejudicial effect of such evidence.
Nevertheless, the probity of the evidence of alcohol consumption bears analysis because, in a balancing test, evidence which is probative on two or more issues might arguably require a greater degree of prejudice before it could be said that the danger of unfair prejudice "substantially" outweighs the probative value of the evidence.
The plaintiff in a products liability case must demonstrate that the product was not misused, or that the misuse was foreseeable to the manufacturer. Jurado v Western Gear Works, 131 N.J. 375, 619 A.2d 1312 (1993). In a case based upon a failure to warn, evidence of misuse might "include a showing that the plaintiff either disregarded an adequate warning or would have done so had the warning been given." London v. Lederle Labs, 290 N.J. Super. 318, 327, 675 A.2d 1133 (App. Div. 1996) (citing Coffman v Keene Corp., 133 N.J. 581, 604, 628 A.2d 710 (1993), and Calderon v. Machinenfabriek Bollegraaf Appingedam, B.V., 285 N.J. Super. 623, 667 A.2d 1111 (App. Div. 1995)). Thus, evidence of Hulmes's consumption of alcohol is relevant to rebut the presumption that Hulmes's would have followed additional warnings had they been given. See Coffman v Keene Corp., 133 N.J. 581, 628 A.2d 710 (1993) (the plaintiff in a failure to warn case benefits from a rebuttable presumption that he or she would have followed an adequate warning had one been given).
Furthermore, evidence of misuse of a product "is relevant to the issue of proximate cause because it may establish that the plaintiff's conduct was the cause-in-fact of the injury." Id. (citing Johansen v. Makita U.S.A., Inc., 128 N.J. 86, 92, 607 A.2d 637 (1992)). Evidence of Hulmes's alcohol consumption is also relevant, therefore, to the issue of proximate cause. In some cases, the plaintiff's conduct may be so remotely related to the accident, if it is related at all, that whether the plaintiff had consumed alcohol prior to the accident is of little probative value on the issue of causation. See, e.g., Lewis v. Horace Mann Ins. Co., 442 So. 2d 526 (La. Ct. App. 1983) (overturning finding that alcohol was a substantial contributing cause when the plaintiff testified that the defendant was not using his turn signal and was within 20 or 30 feet when he suddenly turned into the plaintiff). By contrast, in this case, both parties agree that the accident was precipitated by the contact between plaintiff's Honda ATC250R and the ATV driven by his brother, while the plaintiff was attempting to "pass" his brother's ATV.
The leading New Jersey case on the admissibility of evidence of alcohol consumption by a party prior to an accident in which that party is involved is Gustavson v. Gaynor, 206 N.J. Super. 540, 503 A.2d 340 (App. Div. 1985). In Gustavson, the defendant admitted to having had "two or three bottles of beer" some five to six hours before the accident. Id. at 543. There was no blood test available and no eyewitness testimony corroborating the defendant's intoxication. The Gustavson court held that:
The mere fact that a driver had consumed some alcoholic beverages is by itself insufficient to warrant an inference that the driver was intoxicated and that the intoxication was of such a degree as to render him unfit to drive at the time of the accident. The admission of such evidence without supporting evidence is unduly prejudicial in view of its capacity to inflame the jury.
Id. at 545.
Other "supplementary evidence from which the [jury] . . . may reasonably conclude that the drinking affected the safe operation of the vehicle" includes, without limitation, "excessive drinking, driving at an excessive speed, recklessness or erratic driving, drunken behavior at the accident scene, or similar acts suggestive of an unfitness to drive." Gustavson, 206 N.J. Super. at 544-45 (citing Rovegno, 677 F.2d at 330-31); cf. Straley, 887 F. Supp. at 737 (citations omitted).
The rule announced in Gustavson has also been applied in Guzzi v. Clarke, 252 N.J. Super. 361, 599 A.2d 956 (L. Div. 1991), and in opinions of the United States District Court for the District of New Jersey in Clement v. Consolidated Rail Corp., 130 F.R.D. 530 (Wolfson, M.J.), aff'd, NO. CIV. 88-3793, 1990 WL 49978 (D.N.J. April 17, 1990) (Fisher, J.). The Gustavson rule, like the Pennsylvania rule of Fisher v. Dye, and similar rules in other states, is designed to exclude potentially prejudicial evidence unless that evidence reliably supports a conclusion that the party against whom it will be introduced was "unfit to drive" at the time of the accident. See Clement, 130 F.R.D. at 537.
In Guzzi, the Law Division admitted expert testimony analyzing the probable results of combining a relatively small amount of alcohol with a prescription medication. The facts of Guzzi, which involved expert testimony on the effects of mixing alcohol and a prescription medicine, distinguish it from other decisions weighing the probative value of evidence of alcohol consumption against its prejudicial effect. Guzzi, therefore, is not dispositive of the issue presented in this case. Nevertheless, it is instructive that the Guzzi court admitted evidence of intoxication, despite the absence of testimony of "excessive drinking, driving at an excessive speed, recklessness or erratic driving [or] drunken behavior at the scene." Guzzi, 252 N.J. at 366. Significantly, the court concluded that the list of "supplementary evidence of intoxication," set forth above, was not exhaustive. The court in Guzzi found the evidence of intoxication "highly probative" of why the defendant did not react more quickly with a "550-foot clear line of vision." Id. At oral argument on this motion, counsel for Honda produced photographs, exchanged by the parties during discovery in this case, showing that West Hunterdon Avenue was dry and the weather clear at or about the time of Hulmes's accident.
In Clement, Magistrate Judge Wolfson excluded evidence of alcohol consumption based upon a finding that a blood alcohol content of .051%, without any other "supplementary evidence," did not meet the Gustavson standard. Clement, 130 F.R.D. at 535. Magistrate Judge Wolfson's decision was affirmed, in part, on the basis of New Jersey's Motor Vehicle Law then in effect, under which blood alcohol levels ranging between .05% and .09%, instead of giving rise to a statutory presumption of intoxication, constituted only one type of "competent evidence to be used in determining the guilt or innocence of the defendant." Clement v. Consolidated Rail Corp., 1990 U.S. Dist. LEXIS 4598, NO. CIV. 88-3793, 1990 WL 49978 *2 (D.N.J. April 17, 1990) (citing N.J.S.A. § 39:4-50.1 (repealed)).
It is not clear, however, what effect the current New Jersey Motor Vehicle Code's irrebutable presumption of unfitness to drive with a BAC of .10% or higher has upon the admissibility of such evidence in a civil case.
Unlike some states, New Jersey does not specifically limit the use of evidence of a probable violation of section 39:4-50 of its Motor Vehicle Code to criminal trials. See 1 James F. Mosher, Liquor Liability Law § 14.02 (Matthew Bender 1995) (identifying Arkansas, Connecticut, Kentucky, Minnesota and Louisiana as states that prohibit the use of the statutory presumption of intoxication in civil cases, and Arizona, Florida, Georgia, Montana, Oregon, Vermont and Washington as states in which the statute specifically provides for a rebuttable presumption of intoxication in civil cases based on the statutory blood alcohol level).
In Straley v. United States, Judge Debevoise excluded evidence of the plaintiff's blood alcohol level, which was, by one calculation, as high as .12% at the time of the accident. Judge Debevoise noted that there was eyewitness testimony that Straley did not appear to be intoxicated at the time of the accident and no "on-the-scene evidence that Straley was actually impaired." Straley, 887 F. Supp. at 739.
In Straley, however, even the defense experts agreed that the .147 serum blood alcohol level taken at the hospital was probably an artificially high reading, due to Straley's considerable loss of blood following the accident. More importantly, all the experts in Straley apparently agreed that "it was likely Straley's blood alcohol was below .10% at the time of the accident." Id.
In this case, Hulmes's blood alcohol level can be determined only by extrapolation from the serum test performed at Cooper Hospital upon his admission. Honda's toxicology expert opines that Hulmes's BAC at the time of the accident would have been .11%. In his brief in opposition, plaintiff attaches a letter from Dr. William L. Mannion, which contests the findings of defendants' experts.
Dr. Mannion opines that plaintiff's blood alcohol "could have been significantly below the .10% level of legal intoxication at the time of the accident." Hulmes's Brief. exhibit C at 2. Finally, however, Dr. Mannion suggests only that the extrapolations made by the defendants' experts could be in error by plus or minus ten percent. This conclusion, however, does not cast significant doubt on the evidence that Hulmes's BAC was at or above the .10% level which constitutes a per se violation of N.J.S.A. § 39:4-50. Any doubt raised by Dr. Mannion's opinion goes to the weight of the testimony of the defendants' experts, rather than to its admissibility. At least one other court which has considered this issue has reached the same conclusion.
Interpreting an Illinois rule similar to the Gustavson rule, the court in Cuellar v. Hout, 168 Ill. App. 3d 416, 522 N.E.2d 322, 118 Ill. Dec. 867 (Ill. Ct. App. 1988), admitted disputed expert opinion testimony that the plaintiff's BAC was .104 at the time of the accident, holding that the expert opinion itself was "supporting evidence which 'either directly or by reasonable inference, [shows] that the conduct of [plaintiff] at and before the accident was or may have been affected by the use of alcoholic beverages.'" Id. at 327 (citations omitted). The Cuellar court, addressing facts similar to those before this court, concluded that the plaintiff's attacks on the methodology of the expert's extrapolation went to the weight, and not the admissibility of such testimony. Id. at 326.
The question whether a BAC of .10% or higher may, in and of itself, constitute Gustavson-type "supplementary evidence" of intoxication is unsettled in New Jersey. In Clement, Magistrate Judge Wolfson declared her unwillingness "to infer that one is unfit to drive based solely on one's blood alcohol level -- even when that level is extremely high." Clement, 130 F.R.D. at 535. Clement's BAC, however, was only .051%. More importantly, the view taken by Magistrate Judge Wolfson ignores the fact that a BAC of .10% or higher is, without more, irrebutable evidence of unfitness to drive.
In Straley, on the other hand, Judge Debevoise opined that "a clear showing that Straley's blood alcohol content at the time of the accident was above the .10% standard for legal intoxication might satisfy Gustavson's corroboration requirement." Straley, 887 F. Supp. at 739. Notably, Straley was riding upon, but not driving, a vehicle at the time he was injured. Arguably, the application of a statutory presumption of unfitness to operate a vehicle is somewhat less compelling in a civil case when the allegedly intoxicated plaintiff was not driving.
In this case, Hulmes was driving a motor vehicle within the meaning on the New Jersey Motor Vehicle Code.
Cf. Locke v. Claypool, 426 Pa. Super. 528, 627 A.2d 801, 803 (Pa. Super. Ct. 1993) (holding that statutory standards of intoxication apply equally to drivers of automobiles and to bicyclists, because, a bicycle is a "vehicle" under the Pennsylvania Motor Vehicle Code, which proscribes driving "vehicles" while under the influence of alcohol).
The question left unanswered in Straley and Clement is precisely the one confronting this court on this motion in limine. Can evidence of a blood alcohol content of .10%, or higher, with little or no additional corroborative evidence of intoxication, satisfy the Gustavson requirement for other "supporting evidence" of unfitness to drive?
Courts grappling with similar evidentiary rules in other states have come to inconsistent conclusions. For example, in Smith v. Deere & Co., CIV. A. No. 87-7998, 1989 WL 24903 (E.D. Pa. March 17, 1989), a suit against the product manufacturer, the court admitted evidence of a BAC of .1483%, which was accompanied only by the plaintiff's admission that he had been drinking prior to the accident, as evidence of unfitness to drive a snowmobile. It is unclear, however, whether the court in Smith applied Fed. R. Evid. 403, or Pennsylvania law.
In Kempe v. Dometic Corp., 866 F. Supp. 817 (D. Del. 1994), the court excluded evidence of plaintiff's blood alcohol level in a products liability suit. After discussing the difficulty of establishing the alcohol content of whole blood by extrapolation from a serum level, the court concluded that "the probative value of the serum alcohol is outweighed by its potential for unfair prejudice." Id. at 821 (citing Rovegno, 677 F.2d at 331). Curiously, however, the court admitted evidence that the plaintiff drank two cocktails prior to the accident because it concluded that such evidence "would not confuse or distract the jury." Id. Ultimately, it is unclear whether Kempe was decided on the basis of a Maryland rule of law which operates to exclude unsupported evidence of intoxication because of its prejudicial effect, or because the evidence was thought to be too "confusing" or because it might "mislead" the jury.
There is little "supplementary evidence" in this case of Hulmes's unfitness to drive. Applying Gustavson, this court must decide which, if any, of the following proffered evidence, provides support for the admission of evidence intended to show that Hulmes's consumed alcoholic beverages prior to the accident:
1. Expert testimony that Hulmes's ATV was traveling at 35-40 m.p.h. at the time of the accident;
2. Testimony from the flight nurse that she suctioned an "amber liquid that smelled like alcohol" out of Hulmes's stomach shortly after the accident;