24CA0029 Peo v McDonald 03-05-2026 COLORADO COURT OF APPEALS
Court of Appeals No. 24CA0029 Douglas County District Court No. 22CR1211 Honorable Ryan J. Stuart, Judge
The People of the State of Colorado, Plaintiff-Appellee, v. James Paul McDonald, Defendant-Appellant.
JUDGMENT REVERSED AND CASE REMANDED WITH DIRECTIONS
Division II
Opinion by JUDGE SULLIVAN Fox and Kuhn, JJ., concur
NOT PUBLISHED PURSUANT TO C.A.R. 35(e)
Announced March 5, 2026
Philip J. Weiser, Attorney General, Austin R. Johnston, Assistant Attorney General, Denver, Colorado, for Plaintiff-Appellee
Megan A. Ring, Colorado State Public Defender, Chloe Sovinee-Dyroff, Deputy State Public Defender, Denver, Colorado, for Defendant-Appellant
¶1 Defendant, James Paul McDonald, appeals the judgment of conviction entered on a jury verdict finding him guilty of vehicular homicide. We reverse and remand for a new trial.
I. Background
¶2 In November 2022, McDonald crashed a vehicle after driving over 100 miles per hour. As a result of the crash, a female passenger was ejected from the vehicle. The passenger died at a hospital approximately seven hours later.
¶3 The prosecution charged McDonald with reckless vehicular homicide, § 18-3-106(1)(a), C.R.S. 2025, and speeding, § 42-4- 1101(1), C.R.S. 2025. McDonald didn’t deny that he was speeding or driving recklessly; rather, his theory of the case was that the car crash didn’t cause the passenger’s death.
¶4 At trial, the prosecution called as an expert witness the forensic pathologist who performed a postmortem examination of the passenger. The pathologist’s autopsy report listed twenty-two signs of external injury and seventeen signs of internal injury in the passenger. The pathologist testified, however, that none of these injuries were lethal when the passenger first arrived at the hospital.
2
¶5 The pathologist also testified that the passenger’s blood sample taken upon admission to the hospital contained 2,100 ng/mL of methamphetamine and 5 ng/mL of fentanyl. But the pathologist’s postmortem blood samples showed that these drug levels later increased, climbing to 4,800 ng/mL of methamphetamine and 21 ng/mL of fentanyl.1 According to the pathologist, the passenger could have died from the amount of drugs in her system upon admission to the hospital and at death, even if she hadn’t been in the car crash.
¶6 The pathologist offered three possible explanations for the increases in methamphetamine and fentanyl in the passenger’s system between her admission to the hospital and the autopsy. First, the passenger could have used drugs “during or around the time of the accident,” but her blood hadn’t yet reached a “steady state” when she arrived at the hospital. Second, postmortem redistribution could have caused the drug levels to become falsely
1 The pathologist performed the autopsy three days after the
passenger died, which she testified is normal in Douglas County, Colorado.
3
elevated.2 The pathologist clarified, however, that postmortem redistribution could only account for approximately one-half of the increase in methamphetamine (specifically, she estimated postmortem redistribution could have taken the methamphetamine level to 3,200 ng/mL). Thus, she opined that the passenger “could also have been potentially using while she was in the hospital at the time. I don’t know.” Finally, the pathologist explained that the hospital administered more fentanyl to the passenger in treating her. The pathologist added, however, that the hospital wouldn’t have given her methamphetamine, so her medical treatment couldn’t account for the increase in both drugs.
¶7 When asked about the passenger’s cause of death, the pathologist opined that she died from “[c]ombined drug intoxication and multiple injuries sustained in a motor vehicle collision.”
¶8 At the jury instruction conference, McDonald’s counsel requested that the trial court instruct the jury on independent intervening causation. While defense counsel focused primarily on the theory that the passenger’s body continued to metabolize drugs
2 Postmortem redistribution of drugs occurs when drugs seep out of organs and soft tissue and into the bloodstream after death.
4
she had consumed before the crash, counsel added that the parties presented evidence that the passenger may have used drugs while at the hospital.
¶9 The trial court initially said that “this seems exactly like the type of independent intervening cause that case law talks about as allowing.” But after hearing from the prosecution, the court denied McDonald’s request for an intervening cause instruction. The court first rejected the defense’s metabolization theory, explaining that the independent intervening act — here, the passenger’s alleged consumption of drugs before the crash — can’t occur before the criminal conduct. Next, the court reasoned that the parties presented “no evidence,” only “speculation,” that the passenger used drugs while at the hospital. While it denied the defense’s requested instruction, the court ruled that McDonald could still argue that the drugs, rather than the crash, caused the passenger’s death.
¶ 10 The jury found McDonald guilty of vehicular homicide and speeding. The trial court sentenced him to twelve years in the custody of the Department of Corrections.
5
¶ 11 McDonald appeals only the vehicular homicide conviction. He contends that the trial court reversibly erred by denying his request for an independent intervening cause instruction. He also asserts that the prosecutor committed multiple instances of misconduct that warrant reversal.
II. Independent Intervening Cause Instruction
¶ 12 We first address McDonald’s contention that he was entitled to an independent intervening cause instruction. Specifically, McDonald argues that at least some credible evidence suggested that either the passenger’s intervening consumption or the hospital’s administration of drugs constituted an independent intervening cause of the passenger’s death.
¶ 13 We conclude that the evidence suggesting that the passenger used methamphetamine at the hospital after the crash was sufficient to entitle McDonald to an independent intervening cause instruction. As a result, the trial court erred by denying McDonald’s requested jury instruction. We also conclude that the court’s error wasn’t harmless. We therefore reverse McDonald’s vehicular homicide conviction and remand the case for a new trial on that charge.
6
A. Applicable Law and Standard of Review
¶ 14 A person commits vehicular homicide if they operate or drive a motor vehicle “in a reckless manner, and such conduct is the proximate cause of the death of another.” § 18-3-106(1)(a). Proximate cause means a “cause which in natural and probable sequence produced the claimed injury . . . [and] without which the claimed injury would not have been sustained.” People v. Stewart, 55 P.3d 107, 116 (Colo. 2002) (quoting CJI-Crim. 9:10 (1983)). A person’s death may have more than one proximate cause. People v. Lopez, 97 P.3d 277, 280 (Colo. App. 2004).
¶ 15 But an independent intervening act may destroy the causal connection between the defendant’s act and the victim’s injury. People v. Saavedra-Rodriguez, 971 P.2d 223, 225-26 (Colo. 1998). If such an act occurs, the defendant is relieved from liability. Auman v. People, 109 P.3d 647, 662 (Colo. 2005).
¶ 16 To receive an independent intervening cause instruction, the defendant must establish three elements: (1) the ultimate harm wouldn’t have occurred absent the claimed intervening cause; (2) the intervening cause wasn’t reasonably foreseeable by the
7
defendant; and (3) the defendant didn’t participate in the intervening cause. Id.
¶ 17 An intervening cause defense is treated like an affirmative defense “for the purpose of determining the quantum of evidence necessary to submit the issue to the jury.” Saavedra-Rodriguez, 971 P.2d at 228. Thus, a defendant need only offer “[s]ome credible evidence” on each element to receive the instruction.3 Id. (citation omitted); see § 18-1-407(1), C.R.S. 2025. This standard requires only a “small quantum of evidence,” and the evidence may come from any source, even from the prosecution. People v. Newell, 2017 COA 27, ¶ 21. The court must view the evidence in the light most favorable to the defendant and draw all reasonable inferences from the evidence. See id.; Saavedra-Rodriguez, 971 P.2d at 228.
¶ 18 Whether a defendant has met this evidentiary burden is a question of law that we review de novo. People v. Reynolds, 252 P.3d 1128, 1131 (Colo. App. 2010).
3 Colorado appellate courts “have understood ‘some credible
evidence’ to be interchangeable with ‘some evidence,’ ‘any credible [even if highly improbable] evidence,’ ‘a scintilla of evidence,’ a ‘small quantum of evidence,’ and ‘any evidence.’” Galvan v. People, 2020 CO 82, ¶ 24 (citations omitted). For consistency, we adhere to the “some credible evidence” formulation throughout this opinion.
8
B. Some Credible Evidence Supported McDonald’s Intervening Causation Theory
¶ 19 We start by assessing whether some credible evidence supported McDonald’s claimed independent intervening cause. The trial court concluded that, “[a]t most,” the evidence suggested that the passenger had the “opportunity” to use drugs while at the hospital, and “no evidence” suggested she actually used drugs after arriving at the hospital. We disagree and conclude that McDonald elicited some credible evidence to support his theory that the passenger used methamphetamine after the crash.4
¶ 20 While neither party presented direct evidence that the passenger used methamphetamine while at the hospital, sufficient circumstantial evidence supported McDonald’s theory. Specifically, the pathologist testified that (1) the amount of methamphetamine in the passenger’s blood increased significantly while she was at the hospital; (2) after accounting for postmortem redistribution, only
4 We don’t address McDonald’s separate arguments that he was
also entitled to an independent intervening cause instruction based on either the passenger’s alleged consumption of fentanyl after the crash or the hospital’s alleged gross negligence in administering additional fentanyl. We also offer no opinion on McDonald’s metabolization theory.
9
two explanations could account for the large increase in methamphetamine — one being that the passenger used methamphetamine while at the hospital; and (3) the passenger had a history of methamphetamine use.5 This evidence, taken together and viewed in the light most favorable to McDonald, was sufficient to satisfy the low “some credible evidence” standard. See People v. Gallegos, 2023 COA 47, ¶ 47 (circumstantial evidence can satisfy the exceedingly low some credible evidence standard), aff’d, 2025 CO 41M.
¶ 21 For two reasons, we aren’t convinced by the People’s arguments that the pathologist’s testimony was too speculative to satisfy the some credible evidence standard.
¶ 22 First, divisions of this court have rejected similar attempts to cast circumstantial evidence as mere speculation. In Newell, ¶ 29, for example, the division rejected the People’s argument that the evidence supporting a self-defense instruction amounted to “‘mere
5 An investigator also testified that three syringes and an arm
tourniquet were found in the passenger’s purse at the hospital. But the trial court sustained the prosecution’s objection to this line of questioning for lack of foundation, so we don’t consider it in our analysis.
10
speculation’ equivalent to ‘no evidence.’” The division explained that (1) a witness saw the victim holding a pair of scissors; (2) the defendant had a cut on his face that could be consistent with a cut from a pair of scissors; (3) no witness explained how the defendant was injured; and (4) the victim scuffled with and threw a chair at the defendant. Id. at ¶ 26. This circumstantial evidence, the division concluded, constituted sufficient evidence to entitle the defendant to a self-defense instruction. See id. at ¶ 29.
¶ 23 Newell’s reasoning applies here. Although no one witnessed the passenger use methamphetamine at the hospital, the pathologist testified that the drug’s increased level in the passenger’s blood was consistent with her using methamphetamine after arriving at the hospital. Moreover, the evidence suggested that the passenger had a history of methamphetamine use, and no witness offered a definitive alternative explanation for the increase. Even if other explanations remained possible, this circumstantial evidence was sufficient to permit the jury to infer that the increase in methamphetamine was attributable to the passenger’s drug use after the crash. See id. at ¶¶ 27-28 (explaining “slight, unreasonable, or improbable” circumstantial evidence may support
11
an affirmative defense instruction, even if conflicting); cf. Kulik v. Pub. Serv. Co. of Colo., 605 P.2d 475, 478-79 (Colo. App. 1979) (intervening cause instruction was proper when the evidence “raised a number of potential intervening causes, including the possibility” that someone broke a safety valve that eventually failed), aff’d sub nom., Metro. Gas Repair Serv., Inc. v. Kulik, 621 P.2d 313 (Colo. 1980).
¶ 24 Second, the pathologist’s testimony that the passenger may have used methamphetamine at the hospital doesn’t fit the definition of speculative testimony under People v. Ramirez, 155 P.3d 371, 378 (Colo. 2007). In Ramirez, the supreme court explained that an expert’s opinion testimony is speculative if it “has no analytically sound basis.” Id. On the flip side, an expert’s testimony isn’t speculative if the method employed by the expert is scientifically sound, and the opinion is based on facts that sufficiently satisfy CRE 702’s reliability requirements. Id.
¶ 25 In this case, the pathologist based her expert testimony on the reliable fact that the level of methamphetamine in the passenger’s blood increased while at the hospital. And we perceive no basis for concluding — and the People don’t develop any argument to
12
suggest — that the pathologist used scientifically unsound methodologies in reaching her opinion. While the pathologist relayed her opinion about the two possible explanations for the increased level of methamphetamine with less than absolute certainty (adding, “I don’t know,” after providing her opinion), such confidence isn’t required. See id. (expert’s statements signaling “less than certainty, i.e., ‘I think’ or ‘it is possible’” don’t render the testimony speculative).
¶ 26 Finally, we aren’t persuaded by the People’s argument that other evidence established that it would have been “very difficult” for the passenger to use drugs while at the hospital. Our task isn’t to weigh the contrary evidence that may rebut the defendant’s affirmative defense; that job falls to the jury. Rather, we need only determine whether some credible evidence supported McDonald’s requested intervening cause instruction. See Gallegos, ¶ 52 (“[W]e do not consider the persuasiveness of the evidence in reviewing whether a defendant was entitled to an affirmative defense.”).
C. The Trial Court Erred by Denying the Instruction
¶ 27 Having concluded that some credible evidence supported McDonald’s theory that the passenger used methamphetamine
13
while at the hospital, we must now determine whether that conduct satisfied Auman’s three elements and therefore entitled McDonald to an intervening cause instruction. See Auman, 109 P.3d at 662.
¶ 28 All agree that McDonald didn’t participate in the passenger’s drug use. And the People don’t dispute that it would be unforeseeable that the passenger would use more methamphetamine at the hospital after the car crash and after having already used a potentially lethal amount of methamphetamine earlier in the day. Indeed, no evidence suggested that McDonald knew the passenger had used drugs that day or that she may use drugs again at the hospital. Our analysis therefore zeros in on Auman’s first element: whether some credible evidence established that the passenger’s death wouldn’t have occurred if she hadn’t used methamphetamine at the hospital. See id.
¶ 29 The pathologist testified that the passenger died due to “[c]ombined drug intoxication and multiple injuries sustained in a motor vehicle collision.” After the prosecutor asked whether she was “saying that the cause of death was not one or the other, but a combination of both,” the pathologist responded, “That is correct.”
14
The pathologist also explained that the passenger had enough methamphetamine in her system at the time of her death that she “easily” could have overdosed from that alone. And on cross- examination, the pathologist confirmed that the passenger “had no lethal injuries upon her arrival at the hospital.”
¶ 30 This testimony, if believed by the jury, established that the car crash wasn’t the sole cause of the passenger’s death, that her death was at least partially attributable to the methamphetamine in her system, and that the level of methamphetamine in her blood increased while at the hospital. Viewed together and in the light most favorable to McDonald, this constitutes some credible evidence that the passenger would have survived if the level of methamphetamine in her body hadn’t increased while at the hospital. And as we’ve already concluded, at least some credible evidence suggested the passenger’s methamphetamine level increased due to her own intervening drug use at the hospital.
¶ 31 We recognize, as the People emphasize, that the pathologist’s statement that the passenger could have died from the drugs alone doesn’t necessarily mean that the passenger would have survived had she refrained from using more drugs after the crash. But the
15
pathologist also testified that, in her opinion, (1) the passenger’s injuries when she arrived at the hospital weren’t lethal; and (2) the passenger died from a combination of the drugs and the crash, not from either one individually. Thus, the pathologist made clear that the crash alone likely wasn’t enough to cause the passenger’s death and that, but for the drugs in her system, she would have survived.6 See Saavedra-Rodriguez, 971 P.2d at 227 (If the defendant caused an injury that “would probably not have been fatal, but the victim dies as a result of [an intervening act, that intervening act] relieves the defendant of criminal liability for the death.”).
¶ 32 Accordingly, because McDonald satisfied the required elements, the trial court erred by denying his request for an independent intervening cause instruction.
D. The Error Wasn’t Harmless
¶ 33 The parties disagree over whether we should apply the constitutional or nonconstitutional harmless error standard to
6 We also reject the People’s argument that “the record contains no
evidence that [the passenger] died from a drug overdose.” As discussed, the pathologist testified that “drug intoxication” was one of two causes that collectively led to the passenger’s death.
16
assess whether the trial court’s error requires reversal. We need not decide this question because we conclude that the court’s error is reversible under either standard. We therefore assume, for the purpose of our analysis only, that the more demanding nonconstitutional harmless error standard applies. See Hagos v. People, 2012 CO 63, ¶ 12 (explaining that “[r]eversal is more difficult to obtain under [the nonconstitutional harmless error] standard than under the constitutional harmless error standard”).
¶ 34 Under the nonconstitutional harmless error standard, we must reverse if “the error affects the substantial rights of the parties.” Id. (citing Crim. P. 52(a)). Stated differently, we reverse if “the error ‘substantially influenced the verdict or affected the fairness of the trial proceedings.’” Id. (quoting Tevlin v. People, 715 P.2d 338, 342 (Colo. 1986)). But an error is harmless if “there is no reasonable possibility that it contributed to the defendant’s conviction.” Pernell v. People, 2018 CO 13, ¶ 22.
¶ 35 In urging us to find the trial court’s error harmless, the People point out that the court still permitted McDonald to argue that the passenger’s drug use, not the crash, proximately caused her death. According to the People, providing the jury with an intervening
17
cause instruction “would have changed nothing,” and its omission didn’t mislead the jury. We disagree.
¶ 36 The trial court instructed the jury that, to find McDonald guilty of vehicular homicide, it must conclude that his reckless driving was “a proximate cause of the death of another.” The court further instructed the jury on proximate causation as follows:
The phrase “proximate cause” means an act or failure to act which in natural and probable sequence produced the death of [the passenger]. It is a cause without which the death of [the passenger] would not have occurred. There can be more than one proximate cause of [the passenger’s] death, and Mr. McDonald’s conduct does not have to be the only, nearest, or last cause of death, so long as it is a cause but for which the death would not have occurred.
¶ 37 By contrast, an intervening cause instruction would have informed the jury that McDonald’s conduct couldn’t be the cause of the passenger’s injuries if, “in order to bring about such injuries, it was necessary that his or her conduct combine or join with an [unforeseeable] intervening cause that also contributed to cause the injuries.” CJI-Civ. 9:20 (2025); see also COLJI-Crim. ch. H, cmt. 8 (2024) (referencing CJI-Civ. 9:20 when discussing instructional language for an intervening cause defense). In other words, an
18
intervening cause instruction would have told the jury that “conduct that is broken by an independent intervening cause cannot be the proximate cause of injury to another.” Martinez v. People, 2024 CO 6M, ¶ 13 (quoting Stewart, 55 P.3d at 121).
¶ 38 By finding McDonald guilty of vehicular homicide, the jury necessarily concluded that McDonald’s reckless driving was a proximate cause of the passenger’s death. See § 18-3-106(1)(a). But we have no way of knowing whether the jury also believed McDonald’s theory that the passenger used methamphetamine after the crash and would have survived had she not done so. Had the court provided an intervening cause instruction, the jury wouldn’t have been able to find McDonald guilty if it also believed that the passenger’s use of methamphetamine at the hospital constituted an intervening cause that broke the causal connection between his driving and the passenger’s death. But under the instructions provided by the court, no such restriction was placed on the jury. To the contrary, the jury could have believed McDonald’s theory and still found him guilty because the given instructions (1) said nothing about an intervening cause breaking the causal connection; and (2) expressly contemplated a conviction even if McDonald’s
19
conduct wasn’t “the only, nearest, or last” cause of the passenger’s death.
¶ 39 Because we don’t know whether the jury credited McDonald’s theory, we can’t agree with the People that instructing the jury on intervening causation “would have changed nothing.” Cf. People v. Harris, 2015 COA 53, ¶ 36 (erroneous admission of evidence wasn’t harmless under the nonconstitutional harmless error standard because “it is impossible for us to know what evidence the jury relied on in making [its] findings”); People v. Anderson, 991 P.2d 319, 321 (Colo. App. 1999) (error not harmless when the prosecutor’s argument “improperly permitted the jury to adopt a version of the law pursuant to which it could find defendant guilty, regardless of when he formed the [requisite] intent”).
¶ 40 We also aren’t convinced by the People’s argument that other explanations for the passenger’s elevated drug levels at the hospital were “far more likely.” As discussed above, the task of weighing the evidence and resolving conflicts falls to the jury, not us. See People v. Vanderpauye, 2021 COA 121, ¶ 49 (rejecting the People’s argument that the jury wouldn’t have credited certain statements
20
because “[w]e cannot say what weight the jury would have given the evidence”), aff’d, 2023 CO 42.
¶ 41 Given all this, we conclude that there is a reasonable possibility that the trial court’s error contributed to McDonald’s conviction. See Pernell, ¶ 22. We therefore reverse the judgment of conviction for vehicular homicide and remand for a new trial on that charge.
III. Remaining Contentions
¶ 42 McDonald also contends that the prosecutor committed multiple acts of misconduct during closing argument that require reversal. We decline to address this contention because we can’t be certain that it will arise again on remand. See People v. Gillespie, 2024 COA 98, ¶ 56.
¶ 43 Further, because we have concluded that the trial court’s decision denying the intervening cause instruction requires reversal, we need not address McDonald’s contention that the cumulative impact of the trial court’s other alleged errors requires reversal. See Vanderpauye, ¶ 71.
21
IV. Disposition
¶ 44 We reverse the judgment of conviction for vehicular homicide and remand the case to the trial court for a new trial on that charge. We don’t disturb the portions of the judgment not challenged on appeal.
JUDGE FOX and JUDGE KUHN concur.