Opinion by:
SARAH B. DUNCAN, Justice.Hossein Bagheri was convicted of driving while intoxicated. On appeal he con*659tends the trial court erred in admitting testimony regarding a retrograde extrapolation of the intoxilyzer test result. The State has confessed error, admitting that the facts here are no better than they were in Mata v. State, 46 S.W.3d 902 (Tex.Crim.App.2001), in which the Texas Court of Criminal Appeals concluded the trial court abused its discretion in admitting a retrograde extrapolation by the same witness who testified in this case. Id. at 917.
In keeping with our prior opinions in Hartman v. State, 2 S.W.3d 490, 494 (Tex.App.-San Antonio 1999, pet. ref'd), and Mata v. State, 75 S.W.3d 499, 500-01 (Tex.App.-San Antonio 2002, pet. filed) (not yet published), in which this court held that the error was harmless because the evidence was sufficient to support conviction under the alternate impairment theory, the State urges that the error was harmless. On our own initiative, we have reconsidered the harm issue en bane and now hold that the sufficiency of the evidence under the alternate impairment theory is irrelevant; it does not establish that the error in admitting the flawed retrograde extrapolation analysis was harmless. We therefore overrule Mata and Hartman to the extent of the conflict and reverse the trial court’s judgment and remand the cause for a new trial.
Factual and Procedural Background
Bagheri was arrested at approximately 2:30 a.m. by Officer Randall McCumbers after McCumbers observed Bagheri’s vehicle traveling at approximately 70 m.p.h., veer onto the inside shoulder of the road on two occasions, and, increasing his speed to approximately 85 m.p.h., move across three lanes of traffic without signaling. After Bagheri was stopped, McCumbers noticed that his speech was slurred, his breath smelled of alcohol, and his eyes were red and glassy. Bagheri seemed confused and stumbled when he exited his car; and he appeared unsteady as he walked to the rear of his car. Field sobriety tests indicated to McCumbers that Bagheri was driving while intoxicated. In-toxilyzer tests revealed Bagheri’s blood alcohol level to be 0.113 at 3:34 a.m. and 0.107 at 3:37 a.m. At trial George Allen McDougall, Jr., Bexar County’s breath test technical supervisor, testified that these intoxilyzer test results established that Bagheri’s blood alcohol level was 0.10 or more at the time of the stop using a retrograde extrapolation.
The court’s charge asked the jury to find whether Bagheri was guilty of the offense of driving while intoxicated and instructed the jury that a person is deemed to be intoxicated while driving within the meaning of the law if: (1) he does not have the normal use of his mental or physical faculties by reason of the introduction of alcohol into the body; or (2) he has an alcohol concentration of 0.10 percent or more. The jury found Bagheri guilty; and he was sentenced to eight months in jail, probated for 120 days, and fined $1500.
Applicable Law
Because the error of which Bagheri complains is nonconstitutional, we must disregard it unless it affected his “substantial rights.” Tex.R.App. P. 44.2(b). “[S]ubstantial rights are not affected by the erroneous admission of evidence ‘if the appellate court, after examining the record as a whole, has fair assurance that the error did not influence the jury, or had but a slight effect.’ ” Solomon v. State, 49 S.W.3d 356, 365 (Tex.Crim.App.2001). Stated another way, “[t]he accused’s substantial rights are not affected unless there is a reasonable possibility that the errors are prejudicial; i.e., that they ‘might have contributed’ to the defendant’s conviction.” Hinds v. State, 970 S.W.2d *66033, 35 (Tex.App.-Dallas 1998, no pet.) (quoting United States v. Brown, 897 F.2d 162, 163 (5th Cir.1990)). Among the facts to be considered are whether the evidence was cumulative and whether it was elicited from an expert. See Solomon, 49 S.W.3d at 365.
Discussion
The State argues the error in admitting the retrograde extrapolation was harmless because the court’s charge submitted both the “per se” and “impairment” theories of intoxication; and the jury may have convicted Bagheri based on the impairment theory, which is supported by McCumber’s testimony and the intoxilyzer test results. The State’s argument is supported by our previous opinions in Mata v. State, 75 S.W.3d 499, 500-01 (Tex.App.San Antonio 2002, pet. filed), and Hartman v. State, 2 S.W.3d 490, 494 (Tex.App.San Antonio 1999, pet. ref'd).
Even if we assume the intoxilyzer test results would be admissible in the absence of a retrograde extrapolation tying the results to the time of the stop,1 or that McCumber’s testimony alone would be sufficient to establish intoxication under the impairment theory, the fact remains that the court’s charge submitted both theories in a single question. As a result, we do not and cannot know which theory persuaded the jury beyond a reasonable doubt. Cf. Crown Life Ins. Co. v. Casteel, 22 S.W.3d 378, 388 (Tex.2000). Without knowing that the jury found Bagheri guilty under the impairment theory, the State’s assertion that the evidence is sufficient to support conviction under that theory is simply irrelevant; it in no way leads us to a “fair assurance that the error [in admitting the retrograde extrapolation evidence] did not influence the jury, or had but a slight effect.” See Solomon, 49 S.W.3d at 365. We therefore hold the sufficiency of the evidence to support conviction under an impairment theory does not alone establish that the error in admitting a flawed *661retrograde extrapolation analysis was harmless and overrule Hartman and Mata to the extent of the conflict. We will instead adhere to the harm test enunciated by the Texas Court of Criminal Appeals. See id.
After reviewing the record, we cannot conclude that the retrograde extrapolation evidence “did not influence the jury, or had but a slight effect.” The retrograde extrapolation evidence was placed before the jury by an expert witness, whom the State touted in closing arguments as having “studied with the leading experts in the field in the United States”; and it was not cumulative of other evidence. We therefore hold the error was not harmless, reverse the trial court’s judgment, and remand the cause for a new trial consistent with this opinion.
PHIL HARDBERGER, C.J., concurs.
. Whether the intoxilyzer test results are admissible in the absence of a retrograde extrapolation is a question the Texas Court of Criminal Appeals has expressly left open. See Mata, 46 S.W.3d at 910 n. 42 & accompanying text. In its post-submission brief, the State indicates it "frankly ... doesn't see how [the breath test result] is not relevant.” Perhaps counsel is not aware that courts in other jurisdictions have recognized that an intoxi-lyzer test administered an hour or more after the stop does not tend to establish a defendant’s blood alcohol level at the time of the stop and is therefore no evidence of the essential element of "per se” intoxication. See Commonwealth v. Jarman, 529 Pa. 92, 601 A.2d 1229, 1231 (1992) (A blood test performed approximately one hour after stop and indicating a blood alcohol content of 0.114 "was no evidence upon which the expert could offer an opinion as to whether appellant's blood alcohol level was in fact greater than or equal to 0.10% at the time of driving."); Commonwealth v. Modaffare, 529 Pa. 101, 601 A.2d 1233, 1235 (1992) (A blood alcohol test performed one hour and fifty minutes after an accident and indicating a blood alcohol content of 0.108 was "no evidence ... appellant’s blood alcohol level was greater than or equal to 0.10% at the time of the accident.”).
In support of its argument that the test result would be admissible and relevant absent a retrograde extrapolation to establish intoxication under an impairment theory, the State cites Míreles v. Texas Dep’t of Pub. Safety, 9 S.W.3d 128, 131 (Tex.1999), and Price v. State, 59 S.W.3d 297 (Tex.App.-Fort Worth, 2001, pet. ref’d.). However, neither case analyzes relevance in the context of the established science; both opinions simply rely upon Forte v. State, 707 S.W.2d 89, 95 (Tex. Crim.App.1986). But Forte predates the entire Daubert "revolution” and its application to DWI cases such as Mata.
In short, the admissibility and relevance of the intoxilyzer test result in the absence of a retrograde extrapolation has not been authoritatively decided by the Texas Court of Criminal Appeals. We have therefore adopted a harm analysis that renders the issue moot.