106 P.3d 436
Idaho
2005

Lead Opinion

TROUT, Justice.

This is an appeal from a jury verdict convicting Richard Robinett (Robinett) of aggravated driving under the influence (DUI) and vehicular manslaughter.

I. FACTUAL AND PROCEDURAL BACKGROUND

On May 11, 2001, Robinett, his daughter Patricia Robinett (Patricia), and his brother-in-law Russell Lawrence (Lawrence), were drinking at a bar in Lewiston. At 2:25 a.m., they left in Robinett’s car with Robinett driving, Patricia in the front passenger seat and Lawrence in the back seat. At a curve in the road, the car went off the road and collided with several trees. Police were notified at approximately 3:00 a.m. and at the scene they found Robinett outside the vehicle. Patricia was trapped inside the car and Lawrence was found dead, lying across the front passenger seat. Robinett’s blood was drawn for a blood alcohol content (BAC) test approximately two hours after the accident and a second blood test was drawn about one-half hour later. The result of the first test was between .135 and .165 and the result of the second test was .12.

The State charged Robinett with aggravated DUI and vehicular manslaughter and elected to prosecute the DUI solely on the basis that Robinett was driving impaired and not as a per se violation of the statute based on the BAC results. Robinett’s defense was that he was not intoxicated at the time of the accident and that, even if intoxicated, his driving was not the cause of the accident. Robinett presented evidence that Lawrence became angry with Robinett, reached forward from the back seat and grabbed Robinett, causing him to lose control of the vehicle.

At trial, the district judge denied Robinett’s motion in limine to exclude evidence of the two BAC test results but granted Robinett’s motion to exclude evidence regarding the per se alcohol concentration standard of .08. Robinett was convicted of both charges and he appealed. The appeal was assigned to the Court of Appeals which remanded the case for a new trial. The State petitioned for review and this Court granted the petition.

II. STANDARD OF REVIEW

When considering a ease on review from the Court of Appeals, this Court gives serious consideration to the views of the *112Court of Appeals; however, this Court reviews the trial court’s decisions directly and acts as though it is hearing the matter on direct appeal from the decision of the trial court. State v. Statton, 136 Idaho 135, 136, 30 P.3d 290, 291 (2001); Leavitt v. Swain, 133 Idaho 624, 627, 991 P.2d 349, 352 (1999). The trial court has broad discretion in the admission and exclusion of evidence and its decision to admit evidence will be reversed only when there has been a clear abuse of that discretion. State v. Howard, 135 Idaho 727, 731, 24 P.3d 44, 48 (2001). On the question of whether evidence is relevant, this Court reviews the trial court’s ruling de novo. State v. Raudebaugh, 124 Idaho 758, 764, 864 P.2d 596, 602 (1993).

III. ANALYSIS

A. Admission of BAC test results

The offense of driving under the influence is codified in I.C. § 18-8004(l)(a) which provides:

It is unlawful for any person who is under the influence of alcohol, drugs or any other intoxicating substances, or any combination of alcohol, drugs and/or any other intoxicating substances, or who has an alcohol concentration of 0.08, as defined in subsection (4) of this section, or more, as shown by analysis of his blood, urine, or breath, to drive or be in actual physical control of a motor vehicle within this state, whether upon a highway, street or bridge, or upon public or private property open to the public.

This section has been interpreted as establishing one crime with two ways of proving a violation. State v. Garrett, 119 Idaho 878, 881-82, 811 P.2d 488, 491-92 (1991)(citing State v. Knoll, 110 Idaho 678, 718 P.2d 589 (Ct.App.1986)). As the Court of Appeals has indicated, the first way to prove a violation is to show under the totality of the evidence that the defendant was driving under the influence. The second way to prove a violation is to establish the defendant drove with an alcohol concentration of 0.08 percent or more. The State may elect to proceed against the defendant under either or both theories of proof. Evidence that is relevant under one theory of proof is not necessarily relevant under the other. State v. Edmondson, 125 Idaho 132, 133, 867 P.2d 1006, 1007 (Ct.App.1994); State v. Andrus, 118 Idaho 711, 713, 800 P.2d 107, 109 (Ct.App.1990); State v. Knoll, 110 Idaho 678, 682, 718 P.2d 589, 593 (Ct.App.1986).

Robinett argues the numerical BAC test results should not have been admitted because the State elected to prosecute solely under a driving under the influence theory. This Court has never decided whether a numerical BAC test result can be admitted where a defendant’s BAC level is actually tested but the defendant is prosecuted solely under an impairment theory. The Court of Appeals has stated the impairment method of proof considers the totality of the evidence, meaning “circumstantial evidence of impaired driving ability or other observable symptoms of intoxication.” State v. Barker, 123 Idaho 162, 163, 845 P.2d 580, 581 (Ct.App.1992)(quoting State v. Knoll, 110 Idaho 678, 682, 718 P.2d 589, 593 (Ct.App.1986)). We hold today that a numerical BAC test result is relevant to a prosecution for driving under the influence (as opposed to a per se violation) only if a proper foundation is laid to, assure the validity of the test result, including evidence extrapolating the result back to the time of the alleged offense. .

This case is distinguishable from State v. Sutliff, 97 Idaho 523, 547 P.2d 1128 (1976), where we held BAC test results need not be related back to the time of the offense as a foundational prerequisite to admissibility. In Sutliff the defendant was charged with driving under the influence after he was involved in an accident which killed another person. Breath and blood samples were taken from the defendant fifty to sixty minutes after the accident. The defendant was prosecuted under a statute that provided certain presumptions depending on the percentage of alcohol in the body. We held “the statute does not require extrapolation back but establishes that the percentage of blood alcohol as shown by chemical analysis relates back to the time of the alleged offense for purposes of applying the statutory presumption.” Sutliff, 97 Idaho at 525, 547 P.2d at 1130. Although the statute in Sutliff was worded differently than the statute under which Robinett was prose*113euted, it is clear the prosecutor in Sutliff was in effect prosecuting the case in a manner analogous to the current per se method of proof. Where the prosecution elects to use the per se method, the question is what the alcohol level was at the time the sample was taken. “The lapse of time prior to the extraction of samples goes to the weight to be afforded the test results and not to their admissibility.” Sutliff, 97 Idaho at 524, 547 P.2d at 1129. For that reason, it is appropriate to admit results drawn an hour or more after the alleged offense without having to actually extrapolate the evidence back to the time of the alleged offense.

Unlike proceeding on a per se theory, admission of a numerical BAC test result for purposes of demonstrating impairment must be extrapolated back to the time of the alleged offense to be relevant. The whole purpose of admitting the BAC test results is to show there was alcohol in the defendant’s blood and that the level of alcohol in his system would have impaired his ability to drive in some identifiable way. This requires there be evidence connecting the test result to the time of the alleged offense and evidence showing how the numerical result relates to the issue of impairment. In other words, numerical test results can be admitted in a driving under the influence prosecution if there is a connection shown between the numerical test result and driving impairment at the time the motor vehicle was operated. Therefore, the numerical results of Robinett’s BAC tests are not relevant to a prosecution for driving under the influence because the results were not accompanied by any evidence that correlated the results to the time of the accident and how it would have impacted Robinett’s ability to drive.

Having concluded the numerical BAC test results were erroneously admitted, it must be determined if the error was harmless. “The test for harmless error ... is whether a reviewing court can find beyond a reasonable doubt that the jury would have reached the same result without the admission of the challenged evidence.” State v. Moore, 131 Idaho 814, 821, 965 P.2d 174, 181 (1998) (quoting Giles v. State, 125 Idaho 921, 925, 877 P.2d 365, 369 (1994)). After reviewing the record, this Court cannot say the jury would have reached the same result without the numerical BAC test results and, therefore, the error in admitting the results without further foundation was not harmless.

B. Jury instructions on superseding cause

Robinett raises a number of other issues on appeal, most of which need not be addressed since this matter is being remanded for a new trial. “Where an appellate court reverses or vacates a judgment upon an issue properly raised, and remands for further proceedings, it may give guidance for other issues on remand.” Smith v. Idaho Com’n on Redistricting, 136 Idaho 542, 545, 38 P.3d 121, 124 (2001) (citation omitted). Robinett asserts the district judge erred in refusing several of his requested jury instructions including one concerning superseding cause. He argues these instructions were necessary to allow the jury to properly consider his evidence that Lawrence caused the accident by attacking Robinett while Robinett was driving. Based on the record, we conclude the jury was properly instructed and this is not a case involving a superseding cause. Robinett sought to show that he committed no act that could have had a causal relationship to the car accident and that the sole cause of the accident was Lawrence’s attack. Robinett’s theory required the jury to choose one of two alternative causes; not to assess whether a second action occurred subsequent to-the first which intervened or superseded to actually result in the accident. The jury was appropriately called upon to decide which of the two causes it believed actually caused the accident. Therefore, the district court did not err by refusing Robinett’s proposed jury instructions on superseding cause.

IV. CONCLUSION

The numerical result of Robinett’s BAC tests should not have been admitted without evidence connecting the test results to the time of the alleged offense and evidence showing how the test results related to the issue of impairment. Because this error was *114not harmless, we reverse and remand for a new trial.

Chief Justice SCHROEDER and Justices KIDWELL and EISMANN concur.



Concurrence Opinion

Justice BURDICK

Specially Concurring.

I concur with the majority opinion and write only to address defendant’s assertion of prosecutorial misconduct. I believe that the issue was correctly addressed by Judge Lansing in State v. Robinett, Docket No. 28564, 2004 WL 32949:

Robinett contends that there was misconduct in several components of the prosecutor’s closing argument. We find it necessary to address only one of these for guidance on remand.
While discussing the proof required to establish Robinett’s guilt of aggravated DUI, the prosecutor said: “What they do, because this is essentially a DUI statute, is they make it strict liability. If you are intoxicated and you’re driving, [and] an injury occurs, then you are guilty of this.” This comment misstated to the jury the causation element of aggravated DUI. The offense is defined in I.C. § 18-8006 as follows:
Any person causing great bodily harm, permanent disability or permanent disfigurement to any person other than himself in committing a violation of the provisions of section 18-8004(l)(a) or (l)(e), Idaho Code, is guilty of a felony....
(Emphasis added.) Although this statute does not require that the State prove any negligent act while driving under the influence, it does require that there be “some causal connection between the defendant’s driving while under the influence and the victim’s injuries.” State v. Johnson, 126 Idaho 892, 895, 894 P.2d 125, 128 (1995).1 See also State v. Frank, 51 Idaho 21, 28, 1 P.2d 181, 184 (1931) (holding that, under an analogous statute establishing the offense of involuntary manslaughter, the State was required to prove not only that the defendant was driving under the influence but also that his unlawful driving was the proximate cause of the victim’s death).

The prosecutor’s argument here was improper because it suggested to the jury that the State need not prove any causal relationship between Robinett’s driving under the influence and the accident that caused Patricia’s injuries. Indeed, it indicated that even if the wreck occurred in the manner asserted by Robinett, solely as a result of Lawrence suddenly grabbing Robinett from behind, Robinett would be criminally liable. This assertion in the prosecutor’s argument that a defendant is “strictly liable” for any injury that occurs while he is driving under the influence eliminated the causation element and misled the jury regarding the State’s burden of proof.

Notes

. In holding that I.C. § 18-8006 does not require a negligent act, Johnson overruled, sub silentio, the court of Appeals decision in State v. Nelson, 119 Idaho 444, 446-47, 807 P.2d 1282, 1284-85 (Ct.App., 1991), where we held that the statute required proof that the driver, while intoxicated, committed a negligent act or omission.