Opinion
No. C-070563.
Decided June 27, 2008.
Appeal from the Hamilton County Municipal Court, No. C-06TRC-28350A.
Joseph T. Deters, Hamilton County Prosecuting Attorney, and Tanner B. McFall, Assistant Prosecuting Attorney, for appellee.
Laufman, Jensen Napolitano and Paul M. Laufman, for appellant.
{¶ 1} Defendant-appellant, Katie Grizovic, appeals her conviction following a jury trial for operating a motor vehicle while under the influence of alcohol in violation of R.C. 4511.19(A)(1)(a). Because the trial court allowed the admission of improper testimony concerning the statistical probability that if Grizovic's blood-alcohol level had been tested, it would have been higher than the legal limit, we reverse Grizovic's conviction.
Factual Background
{¶ 2} At trial, the state presented the testimony of Ohio State Highway Patrol Trooper Dexter Howard. Trooper Howard testified that he had initiated a traffic stop of Grizovic after he had, with the use of radar, detected her speeding. Howard followed Grizovic for a short distance before initiating the traffic stop but did not view any erratic driving as he followed her. Grizovic provided Howard with her driver's license, registration, and proof of insurance without event. Trooper Howard had Grizovic exit from her vehicle. He testified that as he questioned Grizovic, he smelled an odor of alcohol and noted that she had glassy, bloodshot eyes.
{¶ 3} Trooper Howard further testified that he was familiar with the rules for detecting impaired drivers provided by the National Highway Traffic Association and that he had attended a seminar on the advanced detection of alcohol impairment. Howard administered three field-sobriety tests to Grizovic, and he testified in detail about each test.
{¶ 4} He first administered the Horizontal Gaze Nystagmus ("HGN") test, which analyzes the involuntary jerking of the eye. According to Howard, on the HGN test there were six possible clues or signs of impairment that could be detected. Grizovic exhibited all six clues. Howard further testified that a detection of four out of the six possible clues indicated a 77% probability that the person being tested had a blood-alcohol level greater than .10. Howard explained that in Ohio, the legal, limit for driving was a blood-alcohol content of .08.
{¶ 5} Howard testified that he next had Grizovic perform the walk-and-turn test. According to Howard, a total of eight possible clues could be manifested on this test, and Grizovic had exhibited three clues. Howard further stated that the exhibition of two clues indicated a 68% likelihood that the person being tested had a blood-alcohol content greater than .10.
{¶ 6} The last test that Trooper Howard had Grizovic perform was the one-leg-stand test. Four possible clues could be detected on this test. According to Howard, Grizovic exhibited two clues, which indicated a 65% probability that she had a blood-alcohol content greater than .10.
{¶ 7} Trooper Howard testified that after observing Grizovic perform these field-sobriety tests, he determined that she was appreciably impaired and placed her under arrest. After being warned of the consequences that would follow, Grizovic declined to take a breath test to determine her blood-alcohol content.
{¶ 8} Grizovic was convicted of driving while under the influence of alcohol. In this appeal, she argues in her sole assignment of error that the trial court erred by admitting testimony concerning the mathematical likelihood that her blood-alcohol level had been in excess of the legal limit.
Impropriety of Testimony Concerning Statistical Probabilities
{¶ 9} Grizovic was convicted of driving while under the influence pursuant to R.C. 4511.19(A)(1)(a), which provides that "[n]o person shall operate any vehicle, streetcar, or trackless trolley within this state, if, at the time of the operation, * * * [t]he person is under the influence of alcohol * * *." Notably, Grizovic had been charged with, and convicted of, driving under the influence based on her conduct and appreciable impairment. Neither her blood nor her breath had been tested, and she was not charged with driving under the influence based upon the results of such a test.
{¶ 10} In State v. Bresson, the Ohio Supreme Court considered the admissibility of the results of HGN field-sobriety tests. After concluding that the results of an HGN test were admissible, the court further noted that "although results on an HGN test may be admissible at trial by a properly trained officer, such an officer may not testify as to what he or she believes a driver's actual or specific BAC level would be, based solely on the HGN test results."
State v. Bresson (1990), 51 Ohio St.3d 123, 554 N.E.2d 1330.
Id. at 128, 554 N.E.2d 1330.
Id. at 129, 554 N.E.2d 1330.
{¶ 11} The Fourth Appellate District interpreted this language in State v. Martin In Martin, the appellant had been convicted of driving while under the influence in violation of R.C. 4511.19(A)(1). At trial, the state elicited the following testimony from the arresting officer: "On the [HGN] test there's a total of six clues[;] however, [if] four or more clues are detected it's a 77% probability and at the time that this manual I was trained on came out that the person would test over the legal limit which at that point was .10."
State v. Martin, 4th Dist. No. 04CA24, 2005-Ohio-1732, 2005 WL 850863.
Id. at ¶ 1.
Id. at ¶ 15.
{¶ 12} The Martin court questioned the propriety of this testimony, stating that "[i]n the case at bar, we note that the trooper did not testify that the HGN test results would show appellant's exact alcohol concentration. Instead, his testimony indicated to the jury that because the appellant exhibited more than four clues on the test, a 77% probability exists that the appellant would test over .10. We believe, however, that testimony to suggest a specific mathematical probability that the appellant would have tested over the statutory limit if she had taken a test is problematic. We believe that trial testimony (not testimony before a judge at a suppression hearing) of this nature is akin to testimony about the expected exact test result." But the Martin court concluded that in light of the other evidence that had been presented, the admission of the testimony was harmless error.
Id. at ¶ 37-38.
Id. at ¶ 39.
{¶ 13} We share the concerns noted by the Martin court. Although Trooper Howard did not give testimony in this case concerning Grizovic's exact blood-alcohol content, his testimony as to statistical probabilities essentially invited the jury to speculate that she would have tested over the legal limit.
{¶ 14} We conclude that testimony concerning the statistical probability that Grizovic would have tested over the legal limit based on the number of clues exhibited during field-sobriety tests was highly prejudicial and improperly admitted. Should the state seek to permissibly rely on such testimony, it must additionally provide expert testimony concerning the effect of a person's blood-alcohol content on his or her ability to drive.
{¶ 15} The Ohio Supreme Court has held that when a defendant is charged with driving while under the influence based on appreciable impairment and not on the results of a chemical test, the results of a chemical test are "not dispositive of guilt, but merely [constitute] some evidence to consider, if probative, in addition to all other evidence regarding the conduct of the defendant. Clearly, without expert testimony, prejudice could result from a jury giving too much weight to the test result itself rather than focusing on the critical issue of the defendant's conduct."
State v. French (1995), 72 Ohio St.3d 446, 452, 650 N.E.2d 887.
{¶ 16} The concerns noted by the Ohio Supreme Court are manifested in the present case. Here, expert testimony was necessary to explain how a blood-alcohol level of over .10 impaired Grizovic's ability to drive. As we have stated, Grizovic had not been charged with a per se violation for driving under the influence based on the results of a blood or breath test. Rather, she was charged with driving while under the influence of alcohol, which was to be proved by her visible and appreciable impairment. Expert testimony was necessary to link Grizovic's blood-alcohol conduct and impairment.
{¶ 17} The trial court erred by admitting testimony concerning the statistical probability that Grizovic would have tested over .10. We cannot determine what weight the jury gave to this prejudicial testimony. Nor can we conclude, based on the other evidence presented at trial, that the admission of such testimony amounted to harmless error, as in the Martin case. Consequently, we sustain Grizovic's assignment of error.
Conclusion
{¶ 18} The trial court erred by admitting testimony, in a prosecution for a violation of R.C. 4511.19(A)(1)(a), concerning the statistical probability that Grizovic would have tested over the legal limit based on the number of clues exhibited during field-sobriety tests, absent accompanying expert testimony linking blood-alcohol content to impairment. And because we cannot conclude that the admission of such testimony amounted to harmless error, we reverse Grizovic's conviction and remand this case for a new trial consistent with this decision.
Judgment reversed and cause remanded.
SUNDERMANN, P.J., Concurs.
PAINTER, J., concurs separately.
{¶ 19} A police officer — excuse me, trooper — was allowed to testify to a percentage prediction that the defendant would test over .10. This occurred in a case where the alcohol level was not an issue and inadmissible without expert testimony to explain to the jury what the level might mean. The defendant was not even charged with being over the prohibited alcohol level, which is of course .08, not .10. Because there was no expert testimony, the "predicted" alcohol level was irrelevant and inadmissible evidence. But not only was it admitted, it was discussed and made a basis of the prosecution's case.
{¶ 20} The trooper had attended some sort of training in predicting the probability that someone would test over .10 by counting the "clues" in a walk-and-turn test. We learn that the trooper could opine that there was a 68% chance that the defendant would test over .10.
{¶ 21} So perhaps the alcohol test is unnecessary if a trooper can predict the alcohol level just by the walking test. Though this "training" might be fine for establishing probable cause, these predictions cannot be admitted in the trial itself. Absent expert testimony, the alleged alcohol level is as inadmissible — without expert testimony — in a prosecution for driving while impaired as it is admissible in a prosecution for driving while over the prohibited level.
{¶ 22} And if the issue were whether the defendant was driving over the prohibited level, a prediction that there was a 68% likelihood would seem to be a smidge less than beyond a reasonable doubt. A 32% probability of testing under would surely be subject to a judgment of acquittal under Crim. R. 29 — if that were the issue, which it was not in this case.
{¶ 23} I am not making this up.