State v. Shadden

235 P.3d 436, 290 Kan. 803, 2010 Kan. LEXIS 433

• Court: Supreme Court of Kansas
• Decided: July 9, 2010
• Docket: 97,457
• Status: Published

Opinion

The opinion of the court was delivered by

Luckert, J.:

Richard Shadden appeals his conviction for driving under the influence of alcohol (DUI) to an extent that it rendered him incapable of safely driving a vehicle, in violation of K.S.A. 2009 Supp. 8-1567(a)(3). During the trial, a law enforcement officer testified Shadden failed the National Highway Traffic Safety Administration’s (NHTSA) standardized walk-and-tum test and his failure meant there was a 68 percent chance that his blood alcohol content (BAC) was more than .10. On appeal, Shadden argues this testimony presents scientific opinion evidence that is not admissible without the State laying the foundation required in Frye v. United States, 293 F. 1013 (D.C. Cir. 1923), which was adopted in Kansas in State v. Lowry, 163 Kan. 622, 629, 185 P.2d 147 (1947). Shadden further argues that words like “ ‘tests’ or other related terms such as pass, fail, or points” should not be allowed in an officer’s testimony or the State’s arguments because the words add scientific credibility to the officer’s opinion.

We agree that the Frye test must be met before admitting evidence establishing a relationship between a NHTSA test failure and a specific measurement of a driver’s BAC. In this case, the State did not lay the necessary foundation, and the district court erred in admitting the officer’s opinion that 68 percent of the time a person exhibiting two clues has a BAC of more than .10. Nevertheless, we disagree that it was error to allow the State and its witnesses to use words like “tests,” “pass,” “fail,” or “points” when referring to Shadden’s performance on the NHTSA test. These words are commonly used by lay and expert witnesses to describe evidence that is not scientific in nature. Therefore, it is not nec-essaiy to meet the Frye test before these words are used. We also reject other issues raised by Shadden and find the Frye error harmless. Therefore, we affirm Shadden’s conviction.

Facts and Procedural Background

On December 27, 2005, Officers Nick Weiler and Shannon Goodnight observed the driver of a pickup truck run a stop sign and fail to yield the right of way to an oncoming car, causing the driver of the car to stop quickly to avoid an accident. The officers activated emergency fights, and the pickup’s driver turned onto another street. Rather tifian pull to the curb, the driver stopped in the middle of a lane of traffic.

Officer Weiler approached the driver, later identified as Shad-den, and asked for his driver’s license and proof of insurance. Officer Weiler detected a strong odor of alcohol from Shadden and asked him to step out of the truck. The smell of alcohol persisted after Shadden emerged from the truck. Officer Weiler observed Shadden sway as he walked and noted that Shadden also slurred some of his words and had difficulty communicating, frequently pausing and asking to have questions repeated. Officer Weiler also noted that Shadden’s face appeared flushed and his eyes were bloodshot and watery.

Officer Weiler decided to perform some NHTSA standardized field sobriety tests. Because of the grade of the street, he did not conduct a one-leg-stand test. He asked Shadden to perform the walk-and-tum test, however. After the officer instructed Shadden and demonstrated the test, Shadden attempted to perform it. Pursuant to the NHTSA standards, Officer Weiler was trained to look for eight possible clues of intoxication based on an individual’s performance of the walk-and-tum test. Under NHTSA protocols, if an individual demonstrates two or more clues, the individual is deemed to have failed the test.

Officer Weiler noted that Shadden failed to maintain his balance while listening to the test instructions and he started to take steps before he was instmcted to begin. During the first nine steps, Shadden stopped once, stepped sideways once, raised his arms twice, and failed to place the heel of one foot against the toe of the other foot on four occasions. While turning, Shadden stepped outside the acceptable range of motion. On the final nine steps, Shadden stopped twice, stepped sideways twice, raised his arms five times, and failed to place his heel against his toe five times. Based on these errors, Officer Weiler identified all eight clues of intoxication.

Then, the other officer, Officer Goodnight, conducted three nonstandardized sobriety tests: the alphabet test, a counting test, and the finger-to-nose test. Shadden was unable to recite file alphabet from A to Z without a mistake. He counted to 15 correctly but repeated a few numbers when counting back down to 1. In six attempts, Shadden failed to touch his nose correctly during the finger-to-nose test.

Officer Weiler arrested Shadden for DUI. At the police station, Officer Weiler provided Shadden with the implied consent advisory form (DC-27), which included a warning that a test refusal may be used against the individual in a trial for DUI. When Officer Weiler asked if Shadden would submit to a breath test on the Intoxilyzer 5000, Shadden refused. After being Mirandized, Shad-den waived his rights and spoke with Officer Weiler. When the officer asked how much Shadden had to drink that evening, Shad-den indicated that he had three or four beers. He also volunteered that he had smoked marijuana.

The State charged Shadden with operating or attempting to operate a vehicle while under the influence of alcohol to an extent that it rendered him incapable of safely driving a vehicle, in violation of K.S.A. 2009 Supp. 8-1567(a)(3). Before trial, Shadden filed a motion in limine seeking to prevent the State or witnesses from referring to the field sobriety exercises as “ ‘tests’ or other related terms such as pass, fail, or points.” In addition, Shadden sought an order requiring the State to refrain from attaching any scientific significance to Shadden’s performance on the NHTSA test.

At the hearing regarding the motion, defense counsel argued that such tests “don’t pass the Frye test or Daubert test, they have never been tested for reliability or validity.” See Daubert v. Merrell Dow Pharmaceuticals, Inc., 509 U.S. 579, 125 L. Ed. 2d 469, 113 S. Ct. 2786 (1993); Frye, 293 F. 1013. Defense counsel argued that under that premise, it could be problematic to present the testimony of an arresting officer as an “expert” using “scientific tests” and cited rationale from Florida courts as persuasive authority. The district court denied the request, stating:

“Well, the NHTSA field sobriety test that has been standardized [is] recognized in Kansas courts. They are based on statistical research. They have their limitation.

“Counsel is fully permitted to, in the interrogation of the officer, to cover the limitations of the test, but they are tests and can be so indicated to the jury. They do score these, based upon clues of impairment, to what is a failure. So the walk- and-tum, one-leg-stand test, they are valid in Kansas.

“The officer may testify in his opinion whether they were ‘pass’ or ‘fail.’ It is all subject to a scathing cross-examination by the defense counsel as to how much weight that the jury ought to give these tests in this particular case.”

During the jury trial, Officer Weiler testified he believed Shad-den “was driving under the influence of alcohol or drugs due to his admissions” and also due to his “sobriety tests, how he failed all those, and the odor of alcohol, [and] bloodshot, wateiy eyes.” In addition, Officer Weiler indicated he had been trained regarding the NHTSA and the tests’ indication of impairment. On direct examination, Officer Weiler gave the following testimony regarding the NHTSA:

“Q. Now, you’re trained with regard to percentages, and if a person has this many clues, it indicates a percentage of whether or not they are .10 over the legal limit, correct?

“A. Yes.

“Q. If a person exhibits two clues, which you have said is a failure, what is the percentage that the person is over .10?

“A. Sixty-eight percent.

“Q. And that is if they exhibit two of eight clues?

“A. Yes.

“Q. And what is the legal limit in the state of Kansas?

“A. Point zero eight.

“Q. So it is — the legal limit is even lower than .10 which is recognized by the NHTSA training?

“A. Yes.”

On cross-examination, the defense attorney questioned Officer Weiler about the NHTSA standards, making the point that the test results are not reliable if the test is not administered in the prescribed manner. Further, in attempting to attack the officer’s testimony regarding the NHTSA-percentage testimony, the defense attorney asked Officer Weiler whether “the converse of 68 percent accuracy is 32 percent inaccuracy.” Officer Weiler admitted that was “correct,” as did Officer Goodnight when she subsequently testified and was cross-examined. The defense attorney also asked extensive questions regarding how many points were available for scoring and how many points should have been deducted for mistakes made by Shadden. In response to cross-examination questions, Officer Weiler indicated Shadden correctly performed 76 percent of the steps he took during the test.

Following a 2-day trial, a jury found Shadden guilty of DUI. For purposes of sentencing, the district court found that Shadden had three prior DUI convictions. The court imposed a sentence of 1 year in jail and a fine of $2,500. The court also imposed court costs, Board of Indigents’ Defense Services (BIDS) attorney fees, and a requirement of 12 months of substance abuse treatment upon release from jail.

Court of Appeals" Decision

On direct appeal, the Court of Appeals reversed Shadden’s conviction and remanded the case for a new trial, finding the district court erred in admitting some, although not all, of Officer Weiler’s testimony about NHTSA tests. State v. Shadden, 40 Kan. App. 2d 1103, 199 P.3d 167 (2009). The Court of Appeals concluded it was not error to admit the officer’s testimony that a NHTSA field sobriety test was administered and that, based upon the officer’s training and experience, the driver failed that test. It was error, however, to admit testimony that took “the additional step of equating a level of certainty or probability to the officer’s opinion or to correlate a driver’s performance with a specific BAC level.” Shadden, 40 Kan. App. 2d at 1114. The Court of Appeals stated that Officer Weiler was “clearly not qualified to testify about the reliability of the NHTSA standards, and no expert testimony was provided to qualify the NHTSA standards under Frye.” Shadden, 40 Kan. App. 2d at 1114. Because Shadden had no effective means of cross-examining the reliability of the NHTSA standards, “[t]he result is the officer’s opinion that the criminal defendant is intoxicated is given an undeserved scientific credibility.” Shadden, 40 Kan. App. 2d at 1114. The Court of Appeals found the error was not harmless due to the possibility that the jury placed undue weight on the field sobriety test results. Shadden, 40 Kan. App. 2d at 1115. Consequently, the Court of Appeals reversed and remanded for a new trial.

In addition, the Court of Appeals addressed a prosecutorial misconduct issue raised by Shadden in which he argued the State violated the district court’s order in limine when it asked the officers for their personal opinion on whether Shadden was intoxicated. The Court of Appeals concluded the questions were not a violation of the order and the evidence was admissible. The Court of Appeals further found that two remaining issues had not been preserved before the district court. Those issues were that the district court erred in admitting testimony that Shadden had refused to take a breath test and that the officers created an unconstitutional condition by asking Shadden to submit to a breath test. As to this unconstitutional condition issue, Shadden argued the request to submit to a breath test required him to choose between a waiver of his Fourth Amendment rights in consenting to a breath test and a waiver of his Fifth Amendment rights by refusing to take a breath test.

The Court of Appeals addressed one final issue, holding the district court erred by imposing BIDS attorney fees without first considering Shadden’s financial resources and the burden that payments would impose. Shadden, 40 Kan. App. 2d at 1121-22 (citing State v. Robinson, 281 Kan. 538, Syl. ¶ 1, 132 P.3d 934 [2006]).

Following the Court of Appeals’ decision, the State filed a petition for review and Shadden filed a cross-petition. We granted both the petition and cross-petition, and our jurisdiction arises from K.S.A. 22-3602(e). In those petitions, neither party requested review of the BIDS attorney fee issue; therefore, that issue is not before this court. However, all other issues are raised either in the State’s petition for review or Shadden’s cross-petition.

The State’s petition for review raises two arguments. First, the State argues it was inappropriate for the Court of Appeals to sua sponte consider the admissibility of testimony regarding the relationship between failing a NHTSA test and a probability of having a BAC above .10. The State contends the issue statements in Shad-den’s appellate brief did not specifically mention Officer Weiler’s testimony in which he indicated that, under the NHTSA standards, missing two clues in the wallc-and-tum test means there is a 68 percent probability that the defendant’s BAC was .10 or higher. Second, the State argues the Court of Appeals erred when it concluded the admission of the testimony regarding the relationship between failing a NHTSA test and a probability of having a BAC above .10 required reversal of the conviction.

Shadden in Inis cross-petition argues the Court of Appeals erred in ruling the officers could use words like “test” and “fail.” In ad dition, he raises the prosecutorial misconduct issue that the Court of Appeals rejected and the constitutional issues that the Court of Appeals concluded had not been preserved.

Preservation of BAC Issue

First, we consider the State’s complaint that Shadden did not preserve or present on appeal an issue that the Court of Appeals considered — i.e., the admissibility of the officer’s testimony that Shadden’s failure on the walk-and-turn test meant there was a 68 percent probability he had a BAC above .10. As the State contends, ordinarily an appellate court will not consider an issue on appeal not raised in the district court. Graham v. Dokter Trucking Group, 284 Kan. 547, 555, 161 P.3d 695 (2007); In re Tax Appeal of Alsop Sand Co., Inc., 265 Kan. 510, 521, 962 P.2d 435 (1998). This rule does not present an impediment to consideration of the issue in this case, however, because the issue of scientific validity or credibility, including the relationship between test failures and BAC, was raised in tire district court and in Shadden’s brief before the Court of Appeals.

More specifically, before both the district court and the Court of Appeals, Shadden argued drat field sobriety tests do not satisfy the Frye standard and that tire accuracy of such tests is outside the common knowledge of laypersons. In his motion in limine, Shad-den argued that “[t]he testimony of Officer Weiler and Officer Goodnight shall be limited to only the description of their lay observations as to signs of impairment.” Shadden made two related requests for an order. First, he sought to limit the use of terminology like “ Tests’ or other related terms such as pass, fail, or points.” Second, he sought to limit the officers’ testimony to lay opinions. Shadden supported these requests by citing State v. Meador, 674 So. 2d 826 (Fla. Dist. App.), rev. denied 686 So. 2d 580 (Fla. 1996). Meador supported Shadden’s second request to limit testimony to lay opinions because the Florida Court of Appeals concluded the State could not admit any evidence of a scientific opinion that stated a relationship between a driver’s performance during field sobriety tests and a certain level of intoxication. In addition, Shadden cited the following authorities and made the following parenthetical statements:

“See also State v. Ross, 938 P.2d 797 (Or. App. 1997) (arresting officer prohibited from rendering opinion that defendant’s behavior was consistent with a blood alcohol level above the legal limit); State v. Lummus, 1997 Ariz. App. LEXIS 189 (officer’s statement that ‘on a scale of 1 to 10 the defendant rated a 10 plus for intoxication’ is inadmissible opinion testimony that implies the defendant was over the blood alcohol limit); State v. Ferrer, 2001 Hawaii App. LEXIS 116 (officer can’t testify that defendant ‘failed’ field sobriety exercises, he can only testify to personal observations).”

This discussion clearly raised the issue of whether the officers could testify about the relationship between a test failure and BAC. In addition, Shadden had a standing objection to the evidence at trial. See K.S.A. 60-404 (providing that a verdict shall not be set aside “by reason of the erroneous admission of evidence unless there appears of record objection to the evidence timely interposed and so stated as to malee clear the specific ground of objection”).

Then in his appellate brief submitted to the Court of Appeals, Shadden asserted that “it is inappropriate for the officer to inform the jury that the NHTSA tests — or any other tests for that matter— are rehable scientifically.” Although Shadden’s appellate brief focused on the terminology argument, the arguments clearly reflected the overarching theme that the officers should have been prohibited from venturing beyond a recitation of their personal observations and should not have been allowed to state a scientific opinion.

Hence, each of the issues addressed in the Court of Appeals’ decision was raised and preserved before the district court and pursued in Shadden’s brief before the Court of Appeals. The fact that the Court of Appeals’ interpretation of die law led it to a conclusion the State had not foreseen does not make the panel’s action inappropriate. See Graham, 284 Kan. at 556 (finding the Court of Appeals’ interpretation of the workers compensation statute, leading it to a conclusion unforeseen by the claimant, did not render the panel’s action inappropriate).

Motion in Limine Ruling

Consequently, we will discuss the Court of Appeals’ holdings that the officer’s opinion that 68 percent of the time a person exhibiting two clues has a BAC of more than .10 was inadmissible and that it was not error to admit testimony referring to the field sobriety exercises as “tests,” indicating that the defendant’s total “points” meant the driver “passed” or “failed” such exercises, or lending scientific credibility to the results of the exercises.

Standard of Review

In recent years, this court has applied an abuse of discretion standard when reviewing a district court’s decision regarding a motion in limine. E.g., Unruh v. Purina Mills, 289 Kan. 1185, 1193, 221 P.3d 1130 (2009). Although this statement of the standard is generally utilized, it is a departure from the original decisions of this court that separately analyzed two factors that underlie a district court’s motion in limine. Those two factors were explained in the first Kansas case to discuss motions in limine in which this court stated:

“The purpose of a motion in limine is to assure all parties a fair and impartial trial by prohibiting inadmissible evidence, prejudicial statements, and improper questions by counsel. It is generally agreed a protective order issued on a motion in limine should be granted only when the trial court finds two factors are present: (1) The material or evidence in question will be inadmissible at a trial under the rules of evidence; and (2) The mere offer of or statements made during trial concerning the material will tend to prejudice the jury. [Citations omitted.] The material to which the motion in limine is addressed may be either inadmissible under an established rule of evidence, such as the hearsay rule, or it may be excludable under a statute, such as K.S.A. 60-445, because its probative value is substantially outweighed by its tendency to prejudice.” State v. Quick, 226 Kan. 308, 311, 597 P.2d 1108 (1979), overruled on other grounds State v. Jackson, 244 Kan. 621, 772 P.2d 747 (1989).

In subsequent cases, both criteria have been broadened somewhat. As to the first criteria, cases have recognized that common-law doctrines as well as rules of evidence may require the exclusion of evidence. See Martinez v. Milburn Enterprises, Inc., 290 Kan. 572, 233 P.3d 205 (2010) (applying collateral source rule). As to the second criteria, at least implicitly it has been recognized that an order in limine is permissible under a district court’s inherent authority to manage the course of trials and thus the purpose of a motion in limine can be broader than guarding against prejudice. See Luce v. United States, 469 U.S. 38, 41 n.4, 83 L. Ed. 2d 443, 105 S. Ct. 460 (1984). More specifically, a limine order may be appropriate because the introduction or mention of the evidence may cause unfair prejudice, confuse the issues, or mislead the jury; tire consideration of the issue during the trial might unduly interrupt and delay the trial and inconvenience the jury; or a ruling in advance of trial may limit issues and save the parties time, effort, and cost in trial preparation. See Palmieri v. Defaria, 88 F.3d 136, 141 (2d Cir.1996); Sperberg v. Goodyear Tire & Rubber Co., 519 F.2d 708, 712 (6th Cir. 1975), cert. denied, 423 U.S. 987 (1975); see generally K.S.A. 60-216(c)(7) (allowing pretrial consideration of issues that “may aid in the disposition of the action”); U.S.D. No. 232 v. CWD Investments, 288 Kan. 536, 205 P.3d 1245 (2009) (affirming order in limine limiting damage claims on the basis that they were not timely or adequately disclosed). However, these advantages to a pretrial ruling must be balanced against the reality that a district court is usually in a better position during trial to assess the value and utility of evidence and its potential prejudice. See Quick, 226 Kan. at 312.

Hence, to restate the Quick factors by incorporating these considerations, a motion in limine may be granted when a district court finds two factors are present: (1) The material or evidence in question will be inadmissible at a trial; and (2) the pretrial ruling is justified as opposed to a ruling during trial because the mere offer or mention of the evidence during trial may cause unfair prejudice, confuse tire issues, or mislead the jury; the consideration of the issue during the trial might unduly interrupt and delay the trial and inconvenience tire juiy; or a ruling in advance of trial may limit issues and save the parties time, effort, and cost in trial preparation. In determining if a pretrial ruling is justified a district court should weigh whether the court will be in a better position during trial to assess the value and utility of evidence and its potential prejudice.

On appeal, an appellate court considers the same factors. Although the Quick court did not discuss the standard for reviewing a district court decision regarding these factors, this court later adopted a separate standard of review for each factor. See Board of Educ., U.S.D. No. 464 v. Porter, 234 Kan. 690, 694, 676 P.2d 84 (1984). Eventually, the court conflated the two standards, probably because it routinely applied an abuse of discretion standard to all issues regarding the admissibility of evidence and that same standard routinely applied to decisions regarding whether the circumstances justified a ruling in advance of trial. See State v. Rowell, 256 Kan. 200, 208, 883 P.2d 1184 (1994).

Recently, this court refined the standard for reviewing eviden-tiary rulings, separately identifying a standard for each analytical step that is inherent in a decision to admit or exclude evidence. An abuse of discretion standard applies to the review of some steps of the analysis; a de novo standard applies to others. This refinement and modification of the evidentiary standard of review means that it is no longer appropriate to apply an abuse of discretion standard to the first motion in limine factor — i.e., the admissibility of evidence. Instead, the multistep evidentiary standard should be applied when decisions regarding that factor are reviewed.

Under the multistep evidentiary analysis, the first question is relevance. K.S.A. 60-401(b) defines relevant evidence as evidence that is probative and material. On appeal, the question of whether evidence is probative is judged under an abuse of discretion standard; materiality is judged under a de novo standard. See State v. Reid, 286 Kan. 494, 507-09, 186 P.3d 713 (2008). The second step is to determine which rules of evidence or other legal principles apply. On appeal, this conclusion is reviewed de novo. Boldridge v. State, 289 Kan. 618, Syl. ¶ 10, 215 P.3d 585 (2009). In the third step of the analysis, a district court must apply the applicable rule or principle. The appellate court’s standard of review of this third step varies depending on the rule or principle that is being applied. Some rules and principles grant the district court discretion, while others raise matters of law. State v. Riojas, 288 Kan. 379, 383, 204 P.3d 578 (2009). Finally, an analysis under K.S.A. 60-445 may be required, depending on the issue and parties’ arguments. Under that statute, a district judge “may in his or her discretion exclude evidence if he or she finds that its probative value is substantially outweighed by the risk that its admission will unfairly and harmfully surprise a party who has not had reasonable opportunity to anticipate that such evidence will be offered.” This analysis is reviewed under an abuse of discretion standard. See Reid, 286 Kan. at 509.

Like the final step of the multistep admission of evidence standard, the second prong of the motion in limine test — whether a pretrial ruling is justified instead of a ruling during the trial — rests in the discretion of the district court. Hence, an abuse of discretion standard applies. Luce, 469 U.S. at 40.

Relevance and Applicable Evidentiary Rules

In this appeal, the focus is on the first motion in limine factor of whether the evidence was admissible under the applicable rules of evidence. Applying the multistep evidentiary standard, the first step — relevance—is not in issue. Shadden concedes that Officer Weiler’s testimony regarding the field sobriety tests was relevant to the DUI charge. Indeed, a driver’s performance on tests designed to test motor and cognitive skills for signs of impairment is probative and material to the question of whether the driver was operating a vehicle while under the influence of alcohol.

As to the second step of the evidentiary analysis, the parties agree that the applicable rules are K.S.A. 60-456 and the foundation rule that is known as the Frye test. K.S.A. 60-456 generally governs the admissibility of all opinion testimony, regardless of the subject matter of the testimony or of the categorization of the witness as lay or expert. The Frye test relates to the admission of scientific opinion evidence.

Under K.S.A. 60-456(a), a layperson is allowed to offer opinions or inferences “as the judge finds (a) may be rationally based on the perception of the witness and (b) are helpful to a clearer understanding of his or her testimony.” Different criteria apply to the determination of whether an expert’s opinion is admissible. Those criteria, found in K.S.A. 60-456(b), require that the expert’s opinion be “(1) based on facts or data perceived by or personally known or made known to the witness at the hearing and (2) within the scope of the special knowledge, skill, experience or training possessed by the witness.” Whether offered by a lay or expert witness, “[tjestimony in the form of opinions or inferences otherwise admissible under this article is not objectionable because it embraces the ultimate issue or issues to be decided by the trier of the fact.” K.S.A. 60-456(d).

On appeal, this portion of the third step of the evidentiary anal ysis — i.e., the district court’s application of K.S.A. 60-456 — is reviewed under an abuse of discretion standard. Kuhn v. Sandoz Pharmaceuticals Corp., 270 Kan. 443, 455, 14 P.3d 1170 (2000).

In addition to considering K.S.A. 60-456, a district court must determine whether the Frye test has been met if an opinion is based on scientific methods or procedures and is offered for admission. See Frye, 293 F. at 1014; Lowry, 163 Kan. at 629. The Frye test requires a showing that the basis of a scientific opinion is generally accepted as rehable within the expert’s particular scientific field. See Kuhn, 270 Kan. at 454; State v. Witte, 251 Kan. 313, 323, 836 P.2d 1110 (1992).

The Frye test applies only to testimony based on a scientific method or procedure. It does not apply to pure opinion testimony, which is an expert opinion developed from inductive reasoning based on the expert’s own experiences, observations, or research. Rather than being subject to a Frye analysis, the validity of pure opinion is tested by cross-examination of the witness. See Kuhn, 270 Kan. at 457. “The distinction between pure opinion testimony and testimony based on a scientific method or procedure is rooted in a concept that seeks to limit application of the Frye test to situations where there is the greatest potential for juror confusion.” Kuhn, 270 Kan. at 460.

On appeal, while the admission of expert testimony is generally subject to an abuse of discretion standard, the determination of whether the Frye test was correctly applied is subject to de novo review. Kuhn, 270 Kan. at 456; see State v. Elnicki, 279 Kan. 47, 51, 105 P.3d 1222 (2005).

Common Knowledge v. Scientific Opinion

As we apply these rules to the evidence in this case, both the lay and expert opinion provisions of K.S.A. 60-456 are implicated because both lay and expert witnesses are permitted to testify as to their observations of a driver’s acts, conduct, and appearance and, based on those observations, to give opinions of the driver’s state of impairment. See, e.g., State v. Kendall, 274 Kan. 1003, 1013, 58 P.3d 660 (2002) (it does not matter if the officer’s opinion that defendant was intoxicated is given as expert or lay witness, since either iype of testimony is permitted even though it embraces the ultimate issue); State v. Townsend, 146 Kan. 982, 986, 73 P.2d 1124 (1937) (permitting lay witnesses’ opinions as to intoxication). In addressing the categorization of a law enforcement officer’s testimony as lay or expert, this court in City of Dodge City v. Hadley, 262 Kan. 234, 241, 936 P.2d 1347 (1997), recognized the testimony can fall into both categories.

The common component of lay and expert opinions regarding impairment or intoxication is the common knowledge that excessive alcohol consumption can cause problems with coordination, balance, and mental acuity. Each of these psychomotor skills is tested by the NHTSA one-leg-stand test and walk-and-tum test, assuring an opportunity to observe and judge coordination, balance, and mental acuity. While a layperson may be able to observe some or all of these same skills, a law enforcement officer’s opinion — while ultimately based on common knowledge — also draws from his or her training in administering these psychomotor tests and his or her experience in observing drivers’ performances on these tests. Nevertheless, because the opinion rests on the common knowledge of the effects of excessive alcohol consumption, the officer provides pure opinion, not scientific, testimony. See State v. Slater, 267 Kan. 694, 705, 986 P.2d 1038 (1999) (“ The objective signs of intoxication are matters of common knowledge and experience.’ ”); see also, e.g., Cumbie v. City of Montgomery, 703 So. 2d 423, 425 n.1 (Ala. Crim. App. 1997) (explaining that the battery of field sobriety tests typically includes the one-leg-stand test, the walk-and-tum test, and the finger-to-nose test, are designed to disclose “physical manifestations of intoxication,” and do not “require the evidentiary foundation for the admission of expert scientific testimony”); State v. Superior Court, 149 Ariz. 269, 276, 718 P.2d 171 (1986) (distinguishing the horizontal gaze nystagmus [HGN] test, which rests upon an assertion of scientific legitimacy, from the psychomotor tests, which rely upon a basis of common knowledge, and holding that “[different rules therefore apply to determine” the admissibility of HGN test results); People v. Williams, 3 Cal. App. 4th 1326, 1332, 5 Cal. Rptr. 2d 130 (1992) (holding that psychomotor tests, unlike the HGN test, rest on a basis of common knowledge, and that lay witnesses may opine as to “another’s state of intoxication when based on the witness’s personal observations of such commonly recognizable signs as an odor of alcohol, slurring of speech, unsteadiness, and the like”); Hawkins v. State, 223 Ga. App. 34, 36, 476 S.E.2d 803 (1996) (holding that expert testimony was not required as a foundation for admission of testimony regarding the results of psychomotor tests because these types of exercises are not based on a scientific principle or technique, but instead are “physical dexterity exercises that common sense, common experience, and the laws of nature’ show are performed less well after drinking alcohol”); People v. DiNonno, 171 Misc. 2d 335, 336, 659 N.Y.S.2d 390 (1997) (explaining that since psychomotor tests are “not truly scientific in nature” but “are based upon the indisputable fact that intoxication affects physical coordination and mental acuity[,]” “proof of their acceptance in the scientific community is not required”); State v. Murphy, 953 S.W.2d 200, 202 (Tenn. 1997) (“[T]he HGN test does differ fundamentally from other field sobriety tests because the witness must necessarily explain the underlying scientific basis of the test in order for the testimony to be meaningful to a jury.”); Plouff v. State, 192 S.W.3d 213, 223 (Tex. App. 2006) (coordination, balance, and mental agility problems exhibited during field sobriety tests are observations grounded in common knowledge).

Yet, implicit in these cases is the notion that evidence regarding psychomotor field sobriety tests must not go beyond the common knowledge of laypersons, unless the evidence is shown to be reliable under Frye. This conclusion is also implicit in three Kansas cases cited by the Court of Appeals — Witte, 251 Kan. 313, State v. Chastain, 265 Kan. 16, 960 P.2d 756 (1998), and State v. McHenry, No. 93,872, unpublished opinion, filed June 30, 2006. A discussion of these cases reveals the line between pure opinion and scientific opinion relating to intoxication.

In Witte, 251 Kan. 313, this court provided an extensive analysis of the positions taken by other jurisdictions regarding the admission of the horizontal gaze nystagmus (HGN) test, which is a NHTSA-approved test. The Witte court first considered whether testimony regarding the HGN test involved scientific evidence requiring expert testimony that qualified under Frye. Witte, 251 Kan. at 318-22. This court summarized the reasons other jurisdictions had concluded that the HGN test is scientific in nature by stating:

“These courts have given various reasons for holding that I-IGN evidence is scientific in nature: The HGN test is distinguished from other field sobriety tests in that science, rather than common knowledge, provides the legitimacy for HGN testing. [Citations omitted.] Certain reactions to alcohol are so common that judicial notice will be taken of them; however, HGN testing does not fall into this category. [Citation omitted.] HGN test results are ‘scientific evidence based on the scientific principle that consumption of alcohol causes the type of nystagmus measured by the HGN test.’ [Citation omitted.] HGN evidence could have a disproportionate impact on the jury’s decisionmaking process because of the test’s scientific nature and because the jury may not understand the nature of the test or the methodology of its procedure. [Citations omitted.]” (Emphasis added.) Witte, 251 Kan. at 321.

The Witte court then concluded:

“Alcohol’s effect on a person’s sense of balance is common knowledge. The same cannot be said for HGN. The HGN test is based upon scientific principles and exceeds common knowledge. We hold that the HGN test results are scientific evidence. As such, the Frye foundation requirements for admissibility must be satisfied.” Witte, 251 Kan. at 322.

After reviewing cases from other jurisdictions, the Witte court ultimately concluded the HGN test results were not sufficiently rehable to meet the requirements of Frye. Witte, 251 Kan. at 329-30. While Witte did not address the admissibility of other field sobriety tests, the decision implies that the results of field sobriety tests based on common knowledge of the effects of alcohol, such as poor balance, would be admissible irrespective of Frye.

Six years later, in Chastain, 265 Kan. 16, this court was asked to reexamine Witte in light of subsequent decisions in four states that allowed the use of the HGN test. Reaffirming Witte, this court stated: “[W]e are not satisfied that such testing has achieved general acceptance within the relevant scientific community.” Chas- tain, 265 Kan. at 23. We also considered the evidence presented in the trial, including that of an expert witness called to testify regarding the reliability of the HGN test and concluded the expert was not qualified to address all of the reliability issues raised in Witte and had not done so. Chastain, 265 Kan. at 23.

More recently, in the unpublished case of State v. McHenry, No. 93,872, filed June 30, 2006, the Court of Appeals considered whether a 12-step drug recognition examiner (DRE) protocol required demonstration of reliability under Frye. The McHenry court specifically noted that the district court restricted the arresting officer s testimony to discussions of the kinds of symptoms different drugs produce and the symptoms the officer observed in the defendant. In rejecting McHenry’s argument that the officer’s testimony implied scientific evidence that was not demonstrably reliable under Frye, the Court of Appeals quoted Williams v. State, 710 So. 2d 24 (Fla. Dist. App. 1998), at length. The McHenry court essentially concluded that the testimony of the officer who conducted the DRE protocol was admissible outside of Frye because the testimony related to physiological conditions within the common knowledge of the jurors:

“ ‘Police officers and lay witnesses have long been permitted to testify as to their observations of a defendant’s acts, conduct, and appearance, and also to give an opinion on the defendant’s state of impairment based on those observations. [Citations omitted.] Objective observations based on observable signs and conditions are not classified as “scientific” and thus constitute admissible testimony.’ ” McHenry, slip op. at 17 (quoting Williams, 710 So. 2d at 28-29).

As these cases recognize, courts generally agree that there is a dividing line between admitting held sobriety test results as circumstantial evidence of intoxication, which is admissible, and the use of such results to assert or imply a specific level of intoxication, which is not admissible unless an appropriate scientific opinion foundation has been laid. See, e.g., Ballard v. State, 955 P.2d 931, 940 (Alaska App. 1998), overruled on other grounds by State v. Coon, 974 P.2d 386 (Alaska 1999) (concluding that HGN test results are admissible as circumstantial evidence of intoxication but inadmissible to establish a particular BAC); State v. Campoy, 214 Ariz. 132, 134-35, 149 P.3d 756 (Ct. App. 2006) (same); People v. Rose, 268 Ill. App. 3d 174, 181, 643 N.E.2d 865 (1994) (distinguishing between the admission of field sobriety test results and preliminary breath test results because the breath test registered body chemistry rather than recording behavioral characteristics); Schmidt v. State, 816 N.E.2d 925, 946 (Ind. App. 2004) (suggesting that admission of evidence regarding statistical probability that an individual who failed a field sobriety test would have a BAC over .10 is improper in the State’s case-in-chief); State v. Murphy, 451 N.W.2d 154, 157-58 (Iowa 1990) (quoting a statement from State v. Nagel, 30 Ohio App. 3d 80, 80, 506 N.E.2d 285[1986], that “ ‘[objective manifestations of insobriety, personally observed by the officer, are always relevant where, as here, the defendant’s physical condition is in issue’ ”); Wilson v. State, 124 Md. App. 543, 553, 723 A.2d 494 (1999) (permitting testimony regarding HGN test results but finding error when testimony included an opinion of defendant’s BAC); State v. Rose, 86 S.W.3d 90, 100 (Mo. App. 2002) (allowing evidence of HGN test results as circumstantial evidence of intoxication but not as evidence of specific BAC); State v. Baue, 258 Neb. 968, 985-87, 607 N.W.2d 191 (2000) (same); State v. Dahood, 148 N.H. 723, 734, 814 A.2d 159 (2002) (same); Brewer v. Ziegler, 743 N.W.2d 391, 400 (N.D. 2007) (same); State v. Sullivan, 310 S.C. 311, 315-16, 426 S.E.2d 766 (1993) (same); Emerson v. State, 880 S.W.2d 759, 769 (Tex. Crim.), cert. denied 513 U.S. 931 (1994) (same).

At oral argument before this court, the State acknowledged this long line of authority and conceded that no foundation was laid during the trial that would allow evidence of a relationship between fading the NHTSA tests and a specific BAC. Such a concession is well-advised considering this testimony is necessarily based on studies measuring the BAC and creating statistical probabilities of intoxication based on a driver’s NHTSA test failure. There was no evidence during the trial of this case establishing that the NHTSA tests have any enhanced scientific reliability not readily observable by the average layperson who sees a driver walk, step off a curb, or engage in a conversation — i.e., eveiyday activities that exhibit the same psychomotor skills as are tested by the NHTSA exercises. Nor was there evidence establishing the basis for the conclusion that two observed clues during the performance on NHTSA tests means there is a 68 percent chance that the BAC is greater than .10.

In addition, as the Court of Appeals aptly observed, there was no evidence that the officers who testified in this case had the mathematical and scientific expertise to lay the necessary foundation. “Therefore, Shadden had no effective means of cross-examining the reliability of the NHTSA standards because Weiler was merely relating information promulgated by the NHTSA.” Shadden, 40 Kan. App. 2d at 1114.

We agree with the Court of Appeals that the admission of the evidence regarding the relationship between the test and a specific BAC without laying a Frye foundation was error.

Field Sobriety Terminology

Shadden argues this reasoning suggests that the State and its witnesses should be prohibited from using terms that imply scientific credibility to the test. In his cross-petition for review, Shad-den suggests that this conclusion can be implied from the Court of Appeals’ reasoning, although not its conclusion. In other words, in his view, it is inconsistent to prohibit the tests as scientific evidence but to allow scientific terminology to be used when talking about the tests. Where we disagree with Shadden is in the suggestion that words like “points,” “clues, “test,” “pass,” or “fail” indicate a scientific opinion.

Kansas courts have consistently referred to field sobriety exercises as “tests” and have described an individual’s performances on such tests as “passing” or “failing.” See, e.g., State v. Stevens, 285 Kan. 307, 319, 172 P.3d 570 (2007) (finding DUI conviction was supported by evidence that defendant “was unable to satisfactorily complete the field sobriety tests”); Bruch v. Kansas Dept. of Revenue, 282 Kan. 764, 765, 148 P.3d 538 (2006) (noting that driver technically “passed the field sobriety tests”); State v. Martinez, 268 Kan. 21, 24, 988 P.2d 735 (1999) (discussing defendant’s failure on “field sobriety tests”); State v. Neuman, 266 Kan. 319, 320, 970 P.2d 988 (1998) (“Field sobriety tests were administered and defendant failed the tests. He also failed a breath test.”); City of Dodge City v. Norton, 262 Kan. 199, 204-05, 936 P.2d 1356 (1997) (discussing the validity of “field sobriety test” as clue to physical impairment); State v. Shaw, 37 Kan. App. 2d 485, 487, 154 P.3d 524, rev. denied 284 Kan. 950 (2007) (summarizing officer’s testimony that defendant failed the walk-and-tum test by exhibiting four of eight “clues of intoxication”); City of Dodge City v. Ingram, 33 Kan. App. 2d 829, 831, 109 P.3d 1272 (2005) (discussing defendant’s failure on “alphabet test,” the “walk-and-tum test,” and the “one-leg-balance test”).

None of the cited cases specifically addressed the issue raised by Shadden — that it is inappropriate for a testifying officer to imply that the NHTSA tests are scientifically reliable by using the allegedly prejudicial terminology. Further, we are not aware of, and Shadden does not cite to, any Kansas case holding that, when an officer uses terms like “points,” “clues, “test,” “pass,” or “fail,” the officer is no longer testifying as a lay witness and begins to testify as a scientific expert, who therefore must be so qualified.

Courts in other jurisdictions have addressed the question and are divided on this issue. A number of jurisdictions conclude that, depending on the circumstances, the use of certain words may give the officer’s lay witness testimony an “aura of scientific validity”— implying reliability and transforming the testimony into expert testimony. State v. Meador, 674 So. 2d 826, 833 (Fla. App.), rev. denied 686 So. 2d 580 (Fla. 1996); see United States v. Horn, 185 F. Supp. 2d 530, 559-61 (D. Md. 2002); State v. Ferrer, 95 Hawaii 409, 427, 23 P.3d 744 (2001); State v. O’Key, 321 Or. 285, 291, 899 P.2d 663 (1995).

Other jurisdictions do not give such weight to ordinary words used to describe an officer’s field observations. In the present appeal, the Court of Appeals pointed to the Arizona Court of Appeals in Campoy, 214 Ariz. 132, where the State appealed the trial court’s decision to prohibit the arresting officer from using terminology such as “sobriety,” “test,” “field sobriety test,” “impairment,” “pass,” “fail,” or “marginal” in describing the defendant’s performance of field sobriety exercises because such language added “unwarranted scientific credibility” to the State’s evidence. Campoy, 214 Ariz. at 134.

Reversing the trial court, the Arizona Court of Appeals first observed that the results of field sobriety tests are admissible as evidence of a criminal defendant’s general impairment “so long as no correlation is made between performance and BAC and no scientific validity is assigned to the tests themselves as accurate measures of BAC.” Campoy, 214 Ariz. at 135. The Campoy court further noted that performance on field sobriety tests has repeatedly been found to be relevant evidence of a defendant’s impairment, “thus, we disagree with the respondent’s implicit conclusion to the contrary.” Campoy, 214 Ariz. at 135.

Additionally, the Campoy court pointed out that the purportedly prejudicial terminology is “pervasive throughout the case law concerning [field sobriety tests] and have not been found, or even suggested to be, inadmissible.” Campoy, 214 Ariz. at 135. The trial court had ordered the restrictions on terminology based on a finding that there is “no scientific correlation between impairment and performance” on field sobriety tests, a finding, in turn, based on expert testimony that several factors other than alcohol impairment can lead to a “cue of impairment” on a field sobriety test. Campoy, 214 Ariz. at 135. This rationale was rejected by the Campoy court, which stated that expert testimony goes to the weight to be given to evidence related to field sobriety testing, not admissibility or relevancy. Moreover,

“[t]he mere self-evident fact that circumstances other than alcohol impairment can be responsible for cues of impairment on [field sobriety tests] does not establish that such tests are necessarily uncorrelated with impairment. Indeed, our courts have repeatedly found [field sobriety tests] are tests of impairment, albeit not definitive indicators of such, and police officers should be permitted to testify accordingly. [Citation omitted.]” Campoy, 214 Ariz. at 135.

The Campoy court found that the purportedly prejudicial words did not, by themselves, suggest a scientific basis for the tests or lend the tests unwarranted scientific credibility. “Rather,” said the Campoy court, “they make plain the tests’ purpose as indicators of impairment and enable the State to demonstrate their probative value.” Campoy, 214 Ariz. at 136. Testimony that a defendant exhibited “ Tour cues of impairment’ on a Tield sobriety test’ does not improperly assert or imply the defendant has been scientifically proven to have been impaired." Campoy, 214 Ariz. at 136.

In addition, the court expressed a concern that permitting restrictions on vocabulary in DUI cases would open the door to “creative wordsmithing” and would invite perpetual and unnecessary litigation. Such restrictions would also place an unnecessary burden on the parties, could cause testimony to take on an unnatural tone, and would be transparent to the jury. Campoy, 214 Ariz. at 136; see State v. Askren, 147 Ariz. 436, 437, 710 P.2d 1091 (Ct. App. 1985) (“Any juror would know that the purpose of giving the field tests was to try to determine if appellant was under the influence of alcohol.”).

Like Arizona, Connecticut has rejected the argument that the State must lay a scientific foundation for evidence regarding walk- and-tum and one-leg-stand tests before using testing terminology. In State v. Kelley, 95 Conn. App. 423, 896 A.2d 129, rev. denied 279 Conn. 906 (2006), the defendant specifically argued that the State should have been prohibited from using words like “tests,” “results,” “pass,” “fail,” and “points” when referring to the walk- and-tum and one-leg-stand tests because “ "such words wrongly [implied] that the matters had scientific validity and, thus, [prejudiced] the defendant.’ ” Kelley, 95 Conn. App. at 432. The Kelley court found no merit to this notion, stating: “Although there may be situations when language imbues unscientific evidence with scientific significance, using testing language to describe field sobriety tests is not one of them.” Kelley, 95 Conn. App. at 433. The court observed that the purportedly prejudicial words are commonly used by the average person to describe unscientific topics. “In this context,” said the Kelley court, “the language is nothing more than descriptive and does not automatically imply that the topic is scientific in nature.” Kelley, 95 Conn. App. at 433.

Other courts have also followed this type of rationale. See Com. v. Lizardo, 2009 WL 4667583, at *2 (Mass. App. 2009), (unpublished opinion) rev. denied 455 Mass. 1109 (2010) (stating that the field sobriety tests “were of the ordinary variety that measure balance, coordination, and mental acuity, and that are well within the comprehension of lay jurors” and rejecting the notion that the jury would have misunderstood their significance or endowed them with undue scientific authority merely because the words “test” and “pass” were used at trial); McRae v. State, 152 S.W.3d 739, 746 (Tex. App. 2004) (stating that the words “clues,” “test,” and “divided attention” merely refer to observations by the officer, based on common knowledge observations and do not convert the lay witness testimony into expert testimony).

We agree with the Court of Appeals’ determination that the reasoning of the Arizona court, and other courts of like mind, is more persuasive than the reasoning of courts requiring the avoidance of specific words. As the Court of Appeals stated:

“[W]here officer testimony does not link test performance with a specific level of intoxication, the mere use of the term ‘test’ or an indication by the officer that the defendant failed to perform the tests adequately and, therefore, ‘failed’ the test does not lend scientific credibility to the test results. There is only a semantic difference between ‘field sobriety test’ and ‘field sobriety exercise’ or between ‘failing a test’ and ‘being unable to perform an exercise adequately.’ An officer must be permitted to relate the activities a suspected drunk driver was asked to perform and to indicate that certain deficiencies in the performance of these activities indicated that the driver was intoxicated. A juror is not likely to mistake the purpose of a driver standing in the street on one foot while counting to 30 or walking heel-to-toe for 18 steps on a straight fine after being stopped by a law enforcement officer. To this end, it is appropriate for the officer to testify that field sobriety tests were administered and that, based upon the officer’s training and experience, the driver failed those tests. It is impermissible to take the additional step of equating a level of certainty or probability to the officer’s opinion or to correlate a drivers performance with a specific BAC level.” (Emphasis added.) Shadden, 40 Kan. App. 2d at 1114.

Harmless Error

Applying this holding to this case, much of Officer Weiler’s testimony was admissible. On the other hand, his testimony that implied a level of scientific certainty in the conclusion that there was a relationship between performance on the walk-and-tum test and an illegal BAC should not have been admitted because a Frye foundation had not been laid. This leads to the next question: Was the admission of the improper testimony harmless? The Court of Appeals concluded it was not harmless, and the State urges this court to reverse that holding. The State argues there was overwhelming evidence of Shadden s guilt, including the near collision observed by the officers, Shadden’s demeanor during the traffic stop, Shad-den’s admission to the use of alcohol and marijuana, and Shadden’s failure on the nonstandardized field sobriety tests.

The standard of review for evaluating these arguments, which do not implicate the constitutional harmless error standard, is statutorily defined by K.S.A. 60-261, which provides:

“No error in either the admission or tire exclusion of evidence ... is ground for vacating, modifying or otherwise disturbing a judgment or order, unless refusal to take such action appears to the court inconsistent with substantial justice. The court at every stage of the proceeding must disregard any error or defect in the proceeding which does not affect the substantial rights of the parties.” K.S.A. 60-261.

The determination of whether substantial justice has been done and whether an error affirmatively causes prejudice to tire substantial rights of a party requires examination of the trial record as a whole, not just examination of the error in isolation. State v. Warledo, 286 Kan. 927, 943, 948, 190 P.3d 937 (2008).

In this appeal, the Court of Appeals cited the statutory harmless error standard and concluded Shadden’s conviction must be reversed because there was a “possibility” that the jury placed weight on the erroneously admitted testimony. As the United States Supreme Court recently concluded, a finding of reversible error because of a “possibility” of a different outcome is an overly rigorous criterion of whether the substantial rights of a party have been denied because of error. United States v. Marcus, 560 U.S. 258, _, 176 L. Ed. 2d 1012, 130 S.Ct. 2159, 2164 (2010) (concluding circuit court’s “any possibility” standard could not be reconciled with federal plain error standard that requires error to affect appellant’s “substantial rights”).

Nevertheless, the Court of Appeals is not alone in its conclusion; several courts have reversed convictions when evidence similar to that admitted in this case was presented to a jury. See, e.g., Robinson v. State, 982 So. 2d 1260, 1262-63 (Fla. App. 2008); Wilson v. State, 124 Md. App. 543, 559, 723 A.2d 494 (1999); State v. Michaud, 342 Mont. 244, 254-55, 180 P.3d 636 (2008); State v. Downey, 145 N.M. 232, 242, 195 P.3d 1244 (2008). The rationale of these decisions reflects the underlying purpose of the Frye test, which is to prevent juries from being prejudiced by scientific evidence that is not deemed rehable. See Kuhn v. Sandoz Pharmaceuticals Corp., 270 Kan. 443, 460, 14 P.3d 1170 (2000).

Yet other courts have determined improperly admitted evidence regarding a NHTSA test would not have reasonably affected or contributed to the verdict in light of the remaining evidence of impairment. See, e.g., People v. Kirk, 289 Ill. App. 3d 326, 334, 681 N.E.2d 1073 (1997); Com. v. Dumais, 60 Mass. App. 70, 72-73, 799 N.E.2d 125 (2003), rev. denied 441 Mass. 1101 (2004); State v. Poole, 216 S.W.3d 271, 275-77 (Mo. App. 2007). This split in authority is not surprising given that a determination of harmlessness is necessarily fact intensive and case specific.

In this case, the Court of Appeals’ conclusion that the error was not harmless was primarily based on State v. Witte, 251 Kan. 313, 330-31, 836 P.2d 1110 (1992), in which a divided court held that the admission of HGN test results was not harmless. The court in Witte did not specify the standard it applied in reaching this conclusion, i.e., whether it was applying the statutory harmless error standard or some other standard. Regardless, there are significant distinctions between this case and the facts in Witte that lead us to the opposite conclusion.

In Witte, a police officer observed Witte swerving as he drove. Witte explained to the jury that he had never driven the vehicle before and he found the steering to be loose and noticed it pulled to the left. The car’s owner, who was a passenger at the time of the stop, verified “there was a little play in the steering.” Witte, 251 Kan. at 314. Both Witte and his passenger testified that the officer performed field sobriety tests and then told them they would be free to leave after he checked their licenses, which they understood meant Witte had passed the field tests. However, a driver’s license check revealed that Witte had a suspended license. Consequently, he was arrested and taken to the police station where a breath test was administered. Witte’s BAG registered .003 above the legal limit. There was considerable evidence regarding the breath test’s margin of error, some of which would support a conclusion that Witte’s BAG could have been below the legal limit. Given the possibility jurors could view the evidence in this light, the Witte court concluded that “the admission of the HGN test results and the officer’s testimony that the jerking of defendant’s eyes was with certainty’ because Witte was under the influence of alcohol may have influenced the juiy by leading the jury to believe one scientific test supported and gave credibility to the other.” Witte, 251 Kan. at 331. The majority reached this conclusion despite other evidence of intoxication, including a strong odor of alcohol coming from inside the car, bloodshot and watery eyes, the defendant’s admission that he had drunk two beers, and the defendant’s failure to satisfactorily complete the walk-and-tum test or the one-leg-stand test. Witte, 251 Kan. at 331-32 (Six, J., dissenting).

Here, unlike Witte, Shadden was not charged with driving while his BAC was above the legal limit. See K.S.A. 8-1567(a)(l). As a result, unlike a presumptive intoxication case that is based on a measurement of BAC, the jury was not instructed that it could presume intoxication based on Shadden’s BAC. Instead, the jury had to find beyond a reasonable doubt that Shadden was intoxicated to a point where he could not safely operate a vehicle. In that regard, the evidence in this case was substantial, especially when compared to the Witte circumstances where the driver swerved and the swerving was explained by loose steering. Here, the jury heard both officers testify that they observed Shadden ran a stop sign and fail to yield the right of way to an oncoming vehicle, nearly causing an accident that was avoided by the other driver’s actions. Then, when emergency lights were activated, Shadden stopped his car in the middle of a busy road and in the lane of traffic rather than pulling to the side of the road. Shadden explained these behaviors by testifying that he was not familiar with the area and was uncertain of his route of travel from one work site to another.

The jury also heard evidence regarding the officers’ observations of Shadden and his behavior. While the officer in Witte observed bloodshot eyes, smelled alcohol, and noted some psychomotor impairment, in this case the evidence was much stronger. Both officers described Shadden’s flushed complexion and his bloodshot and watery eyes. In addition, the officers reported that they observed impaired motor skills even before asking Shadden to perform any NHTSA or other field sobriety tests. As Officer Goodnight described, “when he exited the vehicle and was walking towards us, he was swaying excessively.” Officer Goodnight added that Shadden “swayed excessively the whole time we had contact with him.” Officer Weiler explained that Shadden slurred some of his words and he frequently paused and asked the officers to repeat questions. Both officers testified to the strong odor of alcohol, which was so strong Officer Goodnight could smell it when standing 5 or 6 feet from Shadden.

In addition to the standardized test that is the focus of this appeal, Officer Goodnight asked Shadden to perform three nonstan-dardized field sobriety tests, i.e., the alphabet test, the counting test, and the finger-to-nose test. Officer Goodnight explained that when asked to say the alphabet, Shadden “did not say the letter E. There were errors during the K, L, M sequence. He said N, O, P, Q, O, R, and . . . then went back to C, S, T, W, S, Y, Z.” On the counting test, he was asked to count 1 to 15 and then count backwards to 1 without repeating the number 15. When counting backwards, Shadden got to 3 and then went back to 6 and started counting down again. In six attempts, Shadden was unable to touch his finger to his nose. Based on diese observations, Officer Goodnight formed the opinion that Shadden was “highly intoxicated.”

Shadden explained his poor performance on his need to shift his weight because his “foot was throbbing.” Shadden testified that he had stepped on a nail earlier in the day, puncturing his boot and injuring his left foot under his smallest toe. In addition, Shadden testified he was distracted by the flow of traffic, was “freaked out,” and was cold. According to the officers, Shadden did not mention being cold at the scene or having an injury.

Further, Shadden admitted to Officer Weiler that he had consumed drugs and three or four beers before driving. His testimony at trial was somewhat different in that he testified he drank three-quarters of a 32-ounce bottle of beer about an hour before the stop. He explained the difference by saying that he told Officer Weiler he drank “equal to” three to four beers, “I’m thinking ounce wise.”

These facts provide substantial evidence of guilt. In addition, as we have discussed, the officers were subjected to rigorous cross-examination regarding the NHTSA test. Defense counsel effectively made the point that the field testing was subject to error, that Shadden successfully completed a larger percentage of the tasks than he failed, and that there were alternative explanations for the mistakes. We find these facts to be vastly different from those in Witte. As we consider the record as a whole in this case, we conclude the evidence of the relationship between a test failure and a specific BAC did not prejudice the substantial rights of Shad-den, and the error of admitting this testimony does not require reversal.

Prosecutorial Misconduct

Shadden argues that his trial was prejudiced by prosecutorial misconduct. Specifically, he contends the State violated the district court’s ruling on Shadden’s motion in limine, prohibiting the officers from rendering their opinions regarding Shadden’s intoxication. Shadden focuses on the direct examination of Officer Weiler. The Court of Appeals correctly found no merit to Shadden’s contentions.

Preservation of Issue

First, the State contends that Shadden failed to preserve this issue for appeal in that Shadden did not contemporaneously object on the specific grounds of “violation of the order in limine” or “prosecutorial misconduct.” To support this argument, the State relies on State v. Aikins, 261 Kan. 346, 932 P.2d 408 (1997).

In Aikins, the defendant argued that the district court abused its discretion in refusing to grant a mistrial because the State violated the court’s order in limine precluding gang evidence at trial. The defendant took issue with the State’s reference to an AK-47 weapon during the testimony of one witness. Because the defendant’s objection at the time of the trial was not that the reference violated the order in limine, but that the line of questioning was irrelevant, this court held the issue was not preserved for appeal. Aikins, 261 Kan. at 377.

The State is correct that a contemporaneous objection must be made to all evidentiary claims — including those alleging prosecu-torial misconduct — to preserve the issue for appellate review. State v. King, 288 Kan. 333, 349, 204 P.3d 585 (2009) (noting that K.S.A. 60-404 “dictates that evidentiary errors shall not be reviewed on appeal unless a party has lodged a timely and specific objection to the alleged error at trial”)- Nevertheless, this case is distinguishable from Aikins in that Shadderis objections revolved around the very purpose of the motion in limine. Shadden made contemporaneous objections on the bases of foundation or ultimate conclusion (province of the jury), both of which were discussed at the hearing on the motion in limine and addressed by the judge; moreover, Shad-den made a standing objection at trial. See King, 288 Kan. at 349.

Hence, we find the issue was preserved for appeal.

When an appellate court reviews a claim of prosecutorial misconduct involving the improper elicitation of testimony, the court must first consider whether the questions posed were impermissible. If the questions were impermissible, the reviewing court then determines whether the misconduct constituted plain error; that is, whether the evidence prejudiced the jury against the defendant and denied the defendant a fair trial. State v. Hunt, 285 Kan. 855, 871-72, 176 P.3d 183 (2008).

In making the assessment of whether a prosecutor s misconduct in introducing evidence is plain error, an appellate court must consider: (1) whether the misconduct was gross and flagrant; (2) whether the misconduct exhibited ill will toward the defendant by the prosecutor; and (3) whether the misconduct may be deemed harmless in light of the evidence of guilt presented at trial. None of these three factors is individually controlling. Where the first two factors weigh against the prosecutor, a reviewing court may find the misconduct harmless only when both the statutory and constitutional harmless error tests are satisfied. State v. McRey- nolds, 288 Kan. 318, 323, 202 P.3d 658 (2009); Hunt, 285 Kan. at 872.

Analysis

Applying this standard, we begin with the question of whether the prosecutor impermissibly elicited testimony in violation of the district court’s order in limine. The order was entered during the hearing on the motion. In discussing Shadden’s argument that the arresting officers should be prohibited from rendering an opinion regarding Shadden’s impairment, the district court initially commented that the State should address tire officers’ opinions in terms of their reasons for arresting Shadden. The State then noted that City of Dodge City v. Hadley, 262 Kan. 234, 936 P.2d 1347 (1997), and State v. Carr, 230 Kan. 322, 634 P.2d 1104 (1981), overruled on other grounds hy State v. Cantrell, 234 Kan. 426, 673 P.2d 1147 (1983), specifically permitted officers to provide opinion testimony concerning tire state of intoxication of a criminal defendant. Defense counsel then conceded that the State could elicit testimony from the officers regarding their opinions that Shadden was intoxicated to a point that established probable cause to arrest. Defense counsel clarified that the officers should not be permitted to invade the province of the jury by offering testimony such as, “We are absolutely of the opinion that he was intoxicated to the point of not being able to safely drive a car.”

The district court then stated that the officers may “reach their opinions based on conduct, field sobriety results, what have you, to arrest him, and to say that, 'Okay. Based on what we saw, what we observed, the results of these tests, we arrested him for DUI.’ ” After further discussion, the district court added:

“And I suppose that it is really a question of how you frame your question. If you say, ‘Based upon his conduct as well, did that cause you any pause as to whether he was under the influence at the time that you arrested him?’ I think that you can reach the same result. But I think that it is a little bit less than just turning to the juiy and saying, ‘In my expert opinion he was under the influence, members of the jury,’ period.

“It needs to be related to the jury in the context of what they were seeing and observing, ‘Based upon the arrest of this individual, he was arrested for this behavior and conduct/ that that is why they arrested him. They can relate to that. But I think that everything in the way of their opinions needs to be related to the reasons why they arrested him.”

When the prosecutor requested clarification of die court’s ruling, die district court added:

“My point simply is this ... if you are simply going to put an officer on the stand and say, ‘You observed him all that night?’ ‘Yeah, I did.’ ‘Do you have an opinion for the jury?’

“It has to be ... in the context of what the officers did, why they arrested him . . . they needed to get him off the street, this is why they did it, this, this, this, and this.

“I think that you can get in what you are trying to get in if you phrase your question correctly. I’m asking that, if you have any questions ahead of time, you can bounce those off the Court. If you don’t want to, you don’t have to.

“But we may have to approach, we may have to do something outside the presence of the jury. I’m just anticipating that issue. I’ve had it come up in other cases and it is, I think, better practice for the State to have the officers give their opinions in relation to what they did. That is, they arrested the fellow, they detained him, gave him an Intoxilyzer, whatever they did, to explain what they are doing. That is all fair for the jury to hear and understand.

“But I think that when you just bring in an officer and say, ‘Okay. You observed everything; do you have a[n] opinion for the jury?’ period, that is a little bit different.”

During the State’s direct examination of Officer Weiler, the prosecutor reviewed the officer’s observations of Shadden’s performance during the traffic stop and subsequent field sobriety tests before eliciting the officer’s opinion of whether Shadden was intoxicated. Defense counsel objected several times, arguing that the State failed to lay a sufficient foundation to qualify Officer Weiler as an expert on DUI or that the question invaded the province of the jury. The district court did not sustain or overrule the objections, but it directed the prosecutor to ask the officer why he arrested Shadden. The prosecutor then asked Officer Weiler why Shadden was arrested. When the officer testified that Shadden was arrested for DUI, the prosecutor inquired of the officer’s basis for determining that Shadden had committed DUI and asked whether the officer believed Shadden was intoxicated. Defense counsel again objected, but the district court overruled the objection.

The Court of Appeals concluded that the State “clearly attempted to comply” with the district court’s order in limine and, ultimately, the State did not violate the order. The order appeared to require the State to lay a foundation for Officer Weiler’s opinion that Shadden was intoxicated before eliciting that opinion, and the opinion should be given in the context of why Shadden was arrested. Shadden, 40 Kan. App. 2d at 1118. This is precisely what the State tried to do.

Regardless, even if tire district court’s ruling could be interpreted to prohibit opinion testimony by the officers regarding Shadden’s intoxication, we cannot conclude that plain error occurred, as Shadden was not substantially prejudiced by the State’s eliciting such opinion testimony because the evidence is generally admissible if an opinion foundation was laid. K.S.A. 60-456(d) provides: “Testimony in the form of opinions or inferences otherwise admissible under this article is not objectionable because it embraces the ultimate issue or issues to be decided by the trier of the fact.”

Further, as the Court of Appeals observed and we have discussed, it is well established in Kansas law that an officer who has observed a criminal defendant during a traffic stop may provide an opinion on whether the defendant was intoxicated at the time of die stop. Shadden, 40 Kan. App. 2d at 1119 (citing Hadley, 262 Kan. at 241-42); see State v. Kendall, 274 Kan. 1003, 1013, 58 P.3d 660 (2002).

It appears the district court was attempting to comply with these authorities and assure a foundation was properly laid for the admission of the opinion evidence. Further, the prosecutor attempted to comply with this ruling. Consequently, we cannot conclude that the prosecutor committed gross and flagrant misconduct or exhibited ill will toward the defendant in eliciting Officer Weiler’s opinion regarding Shadden’s impairment. Of course, the admission of the officer’s testimony concerning the 68 percent likelihood that Shadden was under the influence was error, but die Court of Appeals correctly stated that the State’s elicitation of that testimony was not a violation of the order in limine.

Shadden’s contention that the State committed misconduct has no merit.

Breath Test, Miranda, and Unconstitutional Conditions

In addition, Shadden raises two issues regarding the admission of evidence that he refused to take a breath test. First, he contends that the district court improperly permitted the State to elicit testimony regarding Shadden’s refusal to take a breath test in violation of his rights under the Fifth, Sixth, and Fourteenth Amendments to the United States Constitution. Specifically, Shadden claims that the statement he made in refusing to take the breath test was inadmissible because Officer Weiler had not read his Miranda rights before requesting that he take the test. He argues that this refusal was the equivalent of an incriminating statement given during a custodial interrogation. Second, Shadden contends that he was forced to choose between constitutional rights, prohibited by the doctrine of unconstitutional conditions, because he was forced to assert either his right against self-incrimination or his right to be free from an unreasonable search. He argues that the breath test is an unreasonable search prohibited by the Fourth Amendment to the United States Constitution and that his refusal to take the breath test then cannot be used against him, because to do so would violate the Fifth Amendment’s protection against self-incrimination.

The Court of Appeals refused to consider these issues because Shadden did not object to the admission of this evidence. Shadden acknowledges that he makes these arguments for the first time on appeal and recognizes that Kansas law generally requires a contemporaneous objection to the admission of evidence to be lodged in order to preserve the evidentiary issue for appeal. K.S.A. 60-404. He nevertheless argues that the statutory requirement should be waived because consideration of the issue is necessary to serve the ends of justice or to prevent denial of fundamental rights. See State v. Hawkins, 285 Kan. 842, 845, 176 P.3d 174 (2008) (recognizing three categories when appellate courts can consider new issues on appeal, one of which is to serve the ends of justice or to present denial of fundamental rights); Pierce v. Board of County Commissioners, 200 Kan. 74, 80-81, 434 P.2d 858 (1967) (same).

Shadden’s argument is contrary to this court’s recent decisions that emphasized “the importance of th[e] legislative mandate” contained in K.S.A. 60-404, which “dictates that evidentiary errors shall not be reviewed on appeal unless a party has lodged a timely and specific objection to the alleged error at trial.” King, 288 Kan. at 349. In King, the prosecutor asked the defendant on cross-examination at trial about his continued silence after receiving Miranda warnings. Although there was no objection lodged at trial, the defendant subsequently appealed on the grounds of prosecu-torial misconduct, arguing the cross-examination violated his rights under Doyle v. Ohio, 426 U.S. 610, 49 L. Ed. 2d 91, 96 S. Ct. 2240 (1976) (prosecutor’s use of defendant’s post-arrest silence to impeach credibility violates Fifth and Fourteenth Amendments to the United States Constitution). King, 288 Kan. at 339-40.

Refusing to consider the issues presented in King’s appeal on the merits, this court found that the defendant’s challenge to the testimony at issue, while perhaps impheating constitutional rights, was an evidentiaiy-based claim of prosecutorial misconduct intended by the legislature to be included within the scope of the contemporaneous objection rule set forth in K.S.A. 60-404. As such, and in the absence of an objection as required by the statute, we concluded that King had failed to preserve his Doyle claim for appellate review. King, 288 Kan. at 347-49.

Notwithstanding past decisions, which may have relaxed the statutory obligation to object in the prosecutorial misconduct context, this court expressly clarified that “[fjrom today forward, in accordance with the plain language of K.S.A. 60-404, evidentiary claims— including questions posed by a prosecutor and responses to those questions during trial — must be preserved by way of a contemporaneous objection for those claims to be reviewed on appeal.” King, 288 Kan. at 349.

Since King, this court has continued to apply this rationale and has reiterated that the contemporary objection requirement must be followed even where constitutional concerns are at stake. State v. Hollingsworth, 289 Kan. 1250, 1257, 221 P.3d 1122 (2009) (stating that “the fact that an evidentiary claim may have a federal constitutional — rather than a state statutory — basis does not alone excuse the lack of compliance with K.S.A. 60-404”); State v. Richmond, 289 Kan. 419, 429-30, 212 P.3d 165 (2009) (stating “if we were to overlook the lack of this particular objection and consider the issue because it is necessary to serve the ends of justice or to prevent the denial of Richmond’s right to a fair trial, these and other case law exceptions would soon swallow the general statutory rule”).

Hence, we conclude that without a timely and specific objection, the constitutional implications concerning the testimony regarding Shadden’s refusal to take a breath test have not been properly preserved for appellate consideration.

The decision of the Court of Appeals on the issues subject to review is reversed. The judgment of the district court on the issues subject to review is affirmed.

Davis, C.J., not participating. Larry T. Solomon, District Judge, assigned.