State v. Sercey

825 So. 2d 959 (2002)

STATE of Florida, Appellant,
v.
Rachel Louise SERCEY, Appellee.

No. 1D00-1334.

District Court of Appeal of Florida, First District.

June 17, 2002.

Robert A. Butterworth, Attorney General, and Charmaine M. Millsaps, Assistant Attorney General, Tallahassee, for Appellant.

Nancy A. Daniels, Public Defender, and Fred Parker Bingham, II, Assistant Public Defender, Tallahassee, for Appellee.

BARFIELD, J.

In this prosecution for DUI manslaughter and other related offenses, the State of Florida appeals the trial court's order which excludes all evidence of marijuana in the defendant's blood. Applying the principles enunciated in Frye v. United States, 293 F. 1013 (D.C.Cir.1923), and its Florida progeny, we reverse the order in part.

About noon on April 19, 1998, Rachel Sercey was driving a van carrying disabled persons on State Road 121 in Alachua County. The vehicle was traveling at a high rate of speed when the right rear tire began to come apart. According to witnesses, Sercey failed to brake and eventually lost control of the vehicle, which overturned. Three of the ejected disabled passengers died of their injuries. The injured Sercey was taken to a hospital, where it was noted that she smelled of alcohol and she admitted, "I had a couple beers before work" and "I smoke a little weed."

A medical specimen of Sercey's blood, drawn about one hour after the accident and tested in the hospital lab, indicated that her blood alcohol concentration (BAC) was "below detectable limits" (i.e., less than .02 grams/deciliter). A legal specimen of Sercey's blood, drawn almost three hours after the accident at the request of the Florida Highway Patrol (FHP) and tested by the Florida Department of Law Enforcement (FDLE) crime laboratory, indicated a BAC of .068 grams/deciliter.

The legal blood specimen was sent to the Chemical Toxicology Institute in Foster City, California, a private clinical and forensic toxicology lab directed by Dr. Randall Baselt, Ph.D. The lab received the blood specimen, and an accompanying letter from the FHP requesting analysis for marijuana, by express mail on May 27, 1998. The blood specimen was analyzed for marijuana on June 11 and 12, 1998. It was initially subjected to a preliminary radioimmunoassay (RIA) which indicated the presence of cannabinoids (chemical compounds found in marijuana) including tetrahydrocannabinol (THC), the psychoactive component of marijuana. The blood specimen was then subjected to GC/MS analysis to confirm and quantify the amount of THC and its more stable primary inactive metabolite, carboxytetrahydrocannabinol (referred to variously as cTHC, carboxy-THC, THCCOOH, or THC acid). Briefly stated, GC/MS analysis involves solvent extraction of the cannabinoids from the blood matrix, evaporation of the solvent and chemical reaction of the residue with a derivatizing agent to obtain a compound that can be directly injected into the GC/MS instrument, and separation of the individual cannabinoids by gas chromatography (GC), using a connecting mass spectrometer (MS) as the detector.^[1] This confirmation test indicated that the blood specimen contained 5.5 ng/ml THC and 91 ng/ml cTHC.

Sercey was charged with three counts of DUI manslaughter, three counts of vehicular homicide, and two counts of DUI with serious bodily injury. Her coworker, Evertice Cole, was charged with seven counts of abuse or neglect of disabled persons, and pled no contest to seven counts of culpable negligence. Cole was expected to testify at Sercey's trial that she was with Sercey the entire day of the accident, but did not see Sercey consume marijuana, and that Sercey did not appear to be impaired. The state apparently does not have an eyewitness willing to testify that Sercey consumed marijuana prior to the accident.

Sercey filed motions in limine seeking to exclude opinion testimony by the state's toxicology experts, Dr. Baselt and Dr. Bruce Goldberger, regarding the results of the RIA screen and GC/MS analysis, as well as their opinions regarding the recency of her use of marijuana prior to the blood specimen being drawn, the method of ingestion (i.e., that she had smoked marijuana as opposed to ingesting it by other means), and whether she was impaired by alcohol and marijuana at the time of the accident. She asserted that Dr. Baselt's testing procedures "are not reliable and not generally accepted in the relevant scientific field," that the general principle or theory underlying the expert opinions "is not sufficiently established to have gained general acceptance in the relevant scientific fields," and that "application of the scientific principles to the facts of this case and the conclusions drawn by the State's experts are not reliable and not generally accepted by the scientific community." She argued that no conclusions regarding impairment could be drawn from the THC/cTHC test results because there was only one blood specimen and because there is "no consensus in the scientific community as to any specific level of THC in blood that correlates with impairment." She asserted that impairment is an element of the state's proof of DUI manslaughter, an ultimate issue which is for the jury to decide by applying a legal standard to the facts, and that any opinion regarding impairment by the state's experts "is speculative and is not based on scientific principles" because the scientific tests upon which the state's experts rely "only show whether a driver consumed alcohol or drugs" and "do not measure impairment," and because there is "no consensus in the scientific community that the use of marijuana necessarily impairs a person's ability to drive" and "no consensus in the scientific community as to any specific level of THC or THC metabolites in the blood that correlates with impairment."

Sercey argued that "opinions as to the time of usage of marijuana" based on mathematical models developed in studies conducted by Dr. Huestis and others (the Huestis models) are not sufficiently established to have gained general acceptance in the relevant scientific community for several reasons. She asserted that the Huestis models were developed using "naive users of marijuana," that they were based on smoking marijuana (as opposed to other routes of ingestion), that they were based on "analysis of plasma concentrations of THC, as opposed to analysis of whole blood" (asserting that conversion between plasma and whole blood test results is more complex for THC than for alcohol), and that they "have not yet been widely peer reviewed for use in a forensic, legal setting." She argued that application of the Huestis models to the facts of this case is improper and unreliable because Dr. Baselt's calculations erroneously assumed that the partition coefficient for cTHC was the same as for THC, because there was evidence that she was a chronic marijuana user, but no evidence that she smoked marijuana on the date of the accident, and because the ranges of possible times of use based on the Huestis models "are too broad, and there is too much uncertainty as to the accuracy of those ranges in this case."

The state contended that because there was no serious controversy with regard to the relevance of the challenged testimony or the qualifications of the experts, the only issue was whether the expert testimony was based on a scientific principle or discovery that is "sufficiently established to have gained general acceptance in the particular field in which it belongs." It conceded that "the burden is on the proponent of the evidence to prove the general acceptance of both the underlying scientific principle and the testing procedures used to apply that principle to the facts at hand," but pointed out that a Frye hearing "is not necessary for areas of science well established over the years." The state did not dispute "that the scientific basis and methodology underlying the determination that an identifiable quantity of either alcohol or marijuana was in the defendant's blood is arguably subject to a Frye determination," but it challenged the suggestion by defense counsel "that once the ability of an expert to testify as to the presence of alcohol or marijuana in blood is established and meets the Frye test, an additional Frye determination must be made as to the opinion that flows from the data concerning the presence of alcohol or marijuana in the blood." It asserted that "the mere fact that the defense will produce testimony challenging the conclusions and methodologies employed by the State's experts does not require suppression when the Court nonetheless finds those conclusions to be generally accepted in the scientific community," citing Coppolino v. State, 223 So. 2d 68 (Fla. 2d DCA 1968), appeal dismissed, 234 So. 2d 120 (Fla.1969), cert. denied, Coppolino v. Florida, 399 U.S. 927, 90 S. Ct. 2242, 26 L. Ed. 2d 794 (1970).

Citing Florida Power & Light Co. v. Tursi, 729 So. 2d 995 (Fla. 4th DCA 1999), the state asserted that "there is nothing new or novel about an expert who possesses the requisite background to do so rendering an opinion as to impairment based on the presence of intoxicants such as alcohol and marijuana," and that "under Section 90.704, the facts or data relied upon by an expert to support an opinion need not be admissible so long as those facts or opinions are of a type reasonably relied upon by experts in that field." Citing State v. McClain, 525 So. 2d 420 (Fla.1988), and other cases, it contended that "evidence of the presence of intoxicants is relevant and admissible in a prosecution of the sort involved in this case even if no quantitative analysis is possible because even a small amount of drugs can explain impaired conduct." It asserted that the expert opinions on impairment were scientific, not legal, and challenged the argument that the opinions "would invade the ultimate province of the jury," citing Gutierrez v. State, 739 So. 2d 1175 (Fla. 3d DCA 1999). It stated that it would establish "by the required burden of preponderance that the methodology employed by the State's experts in this case is generally accepted in the relevant scientific community, that the results obtained thereby are thus admissible, and that the State's expert witnesses should be allowed to render opinions interpreting such for the assistance of the jury."

THE FRYE HEARING

At the hearing on the motions in limine, the court qualified two experts for the state and three experts for the defense.^[2] The experts agreed on the pharmacokinetics of marijuana in the human body, explaining that marijuana is "a very lipophilic" chemical which is rapidly absorbed into the bloodstream when smoked, then is transferred into the fat tissues of the body, including the brain. The level of THC in the blood declines rapidly over the first few hours, then "levels out" and is dissipated over a period of days. The potency of the marijuana does not affect the absorption and dissipation calculations. A chronic user would have a residual or base-line level of THC in the blood as high as 2 ng/ml, and a baseline level of cTHC in the range of 10-20 ng/ml. All of the experts opined that Sercey's reported blood levels of THC and cTHC were significantly higher than the expected residual levels of chronic users.^[3]

The experts agreed that the generally accepted method of analyzing for THC and cTHC in blood or plasma employs an RIA screen to determine which drugs are present, then confirmation and quantification by GC/MS after isolation of the target compound from the matrix. The experts who were queried regarding the issue agreed that a single blood specimen is common in law enforcement, and that use of a single blood specimen does not affect the accuracy of the GC/MS analysis. None of the experts found fault with the techniques Dr. Baselt's lab used for preparing the calibration standards and internal standard solutions used in the analysis, nor with the liquid-liquid extraction technique used to isolate the THC/cTHC from the blood specimen and the standards, nor with the techniques for evaporating the solvent and chemically reacting the residues with the derivatizing agent, nor with the injection of the prepared specimen and standards into the GC/MS instrument, nor with the calibration and functioning of the GC/MS instrument.

The experts agreed that certain "quality control" procedures are necessary to insure the accuracy and precision of analytical test results. Analytical laboratories are accredited by various organizations and, as part of the accreditation, are regularly inspected and participate in "proficiency studies," in which identical portions of the same sample are analyzed by many labs at the same time in order to check the accuracy of each procedure as it is performed in the individual participating labs. Dr. Baselt's lab is accredited by the California Department of Health and the U.S. Department of Health and Human Services.^[4] The lab participates in proficiency testing for cTHC,^[5] but not THC because, as Dr. Baselt explained: "I don't think anybody has found a stable medium for THC that could be sent out to many different laboratories across the country and analyzed at different times." Dr. Baselt testified that his 1983 study of comparative testing of THC and cTHC in whole blood and in serum/plasma demonstrated that THC is not stable in blood, and the other experts agreed with this conclusion.

Another quality control requirement upon which the experts agreed is that a new analytical procedure must be validated before it is routinely used, usually by exhaustive and rigorous studies. The experts agreed that the GC/MS procedure has been properly validated and is considered the most accurate method of identifying and quantifying drugs and their constituent components, including THC and cTHC.

The experts also agreed that in order to quantify the GC/MS results, it is important to analyze, simultaneously with the "unknown sample" (in this case, Sercey's legal blood specimen), one or more of several types of "known samples." However, the testimony in this case is complicated by the various names and descriptions the experts used with regard to these "known samples," of which we have been able to delineate three categories.

The first category of known samples discussed by the experts is a set of "calibrators" or "standards," samples prepared in a matrix of blood, ethanol, or some other solvent, each of which contains a known amount of the target compound. A set of calibration standards includes a sample that contains none of the target compound (a "negative" standard), a sample that contains a known amount of the target compound which is more than the amount expected to be in the "unknown sample," and several samples that contain various known amounts of the target compound within the foregoing range. Dr. Baselt's lab extracted and analyzed, simultaneously with Sercey's blood sample, a set of six calibration standards prepared in a blood matrix, including a "negative" standard and five "positive" standards, each of which was "spiked" with a different measured amount of THC/cTHC. The concentrations of the standard samples ranged from no THC to 25 ng/ml THC, and from no cTHC to 250 ng/ml cTHC. The concentrations of THC and cTHC in Sercey's blood specimen fell within the range of concentrations of this "standard curve."

The second category of known samples, frequently used in GC analysis, is an "internal standard," a measured amount of a compound which is chemically similar to the target compound and is added to every sample, both known and unknown. Because the response time for the internal standard is slightly different from the response time for the target compound, the responses can be separately read and compared. The use of internal standards to compensate various types of potential problems in the analysis, including loss of minute amounts of sample during the extraction and injection procedures, fluctuating response, and signal suppression by the matrix, is the recommended method of assuring the highest precision for quantitative gas chromatography. Dr. Baselt explained:

[I]nternal standards are drugs that are supposed to resemble the substance that you are testing for, as closely as possible, but be differentiatable [sic] through your instrumental procedures. And the purpose of adding this is that you would put it into the sample very early in the analysis. And that this would [serve] as an internal marker of the carefulness with which you conducted your test. And, for instance, if you were to add this internal marker at the beginning of the test, at the end of the test and you did not find the internal marker, this would be an occasion that [indicates] that the test had failed. If you were to do the test, and you did find the internal marker at about the level that you expected, this would be an indication that the test had succeeded. That's, basically, the purpose of it.

Because the internal standard should have chemical properties similar to the target compound, a deuterated form of the target compound is often used as the internal standard.^[6] Dr. Baselt's lab used deuterated forms of THC and carboxy THC for the internal standards in this analysis. The mass spectrometer selectively monitors ions from THC and deuterated THC, the responses for which are near each other; there is a similar correspondence in responses from cTHC and deuterated cTHC. The quantitation calculation uses the "ion ratios,"^[7] with an acceptable variation from the equivalent standard ranging from ± 20% to ± 30%, depending upon the authority consulted. None of the experts found fault with the deuterated internal standards used by Dr. Baselt's lab, but they disagreed on the issue of whether Dr. Baselt's ion ratio calculations were acceptable.^[8]

The third category of known samples is an "external control," which is used to check the reproducibility of results of an analysis over time. An external control is a sample which is prepared in the same type of matrix as the unknown sample, but which has a known concentration of the target compound that is near (± 20%) the expected concentration of the "unknown sample," and which is stable over a significant period of time, at a minimum several months. Samples from proficiency studies, if they are sufficiently stable, are often used as external controls. The experts differed on the issue of whether Dr. Baselt's decision not to use external controls for THC and cTHC would invalidate the GC/MS analysis.^[9]

The experts also discussed their opinions regarding whether Sercey ate or smoked the marijuana,^[10] the likely time of her ingestion of marijuana (based on the experts' experience and based on the Huestis models),^[11] the conversion of the whole blood THC/cTHC values to serum or plasma THC/cTHC values (relevant to comparison with the Huestis models),^[12] Sercey's likely blood THC/cTHC levels at the time of the accident,^[13] and the correlation of degree of impairment with blood THC/cTHC levels and with blood alcohol/THC/cTHC levels.^[14]

TRIAL COURT ORDER

The order granting the motions in limine stated: "In Florida, an expert's opinion which is based upon a scientific principle is admissible only where the underlying principle, theory or methodology is generally accepted in the field to which it belongs, and the procedures followed to apply the technique are also generally accepted in the relevant scientific community. The burden of proof is on the proponent by a preponderance of the evidence." The trial judge found that "there is no generally accepted per se limit for impairment of THC and cTHC in blood" and that establishing a link between impairment and specific amounts of THC in blood is complicated by the fact that "chronic users of marijuana would likely have a baseline level of THC in their blood at all times," but that there was "no agreement as to how much THC would remain in the blood of chronic users as a baseline" and the "learned articles" supplied by both parties indicated that "the research to date has been conducted in laboratories involving volunteers rather than with chronic everyday users living in the community." He concluded that "there is no reliably established baseline level of THC in the blood of chronic marijuana users," and that chronic users "would require higher quantities than nonchronic users in order to become impaired." He found that the experts agreed "that the GC/MS is an appropriate scientific instrument for qualitatively identifying and quantifying THC and cTHC in liquids," and also "that THC and cTHC are not stable in whole blood, but degrade rapidly and at an unknown rate." He found that the experts did not agree, and the state had not established by a preponderance of the evidence, "that scientists using the GC/MS can reliably quantify the amount of THC and cTHC in whole blood at the earlier time of the blood draw, much less at some other hypothetical time in the past such as the time of the crash in this case."

The judge found that the state had failed to establish "that Dr. Baselt's testing procedures utilizing the GC/MS are reliable and generally accepted in the relevant scientific community." He specifically found that Dr. Baselt "failed to follow quality controls generally accepted in the relevant scientific community" and that the experts agreed that "quality control is critical to the integrity of GC/MS testing." He found that Dr. Baselt "tests no controls contemporaneous with testing for the unknown," and that "good scientific practices" require running both a "positive" and a "negative" control "at the same time and in the same run with the unknown." He stated that "[b]y comparing the GC/MS analytical results of the controls, the analyst can determine the accuracy of the analytical results of the unknown," and found that this is "particularly important with THC and cTHC because the scientists are working with low levels (nanograms per milliliter) of the controlled substance." He found that Dr. Baselt's testimony regarding the impossibility of obtaining or making reliable THC and cTHC controls in whole blood was "contradicted by the other experts" who testified that controls can be made by "spiking" whole blood that is free of controlled substances with known quantities of THC and cTHC, and that these controls are reliable "for a period of days (or weeks according to some testimony)." He found that these "spiked" samples "are stable enough for the purpose of running positive controls for a given analysis." He rejected Dr. Baselt's testimony "that his failure to run controls should be excused because he is periodically `proficiency tested'" and because "his laboratory from `time to time' compares results with other laboratories, by analyzing samples that other laboratories have analyzed." He stated that "Dr. Baselt's failure to follow quality controls generally accepted in the relevant scientific community undermines this Court's confidence in the results of his tests and the conclusions drawn therefrom." He found that these results "are further called into question" because, according to Dr. Bateh, "there is no consistency in the ion ratio range that he used to identify and quantify the THC."

The judge concluded that "the testing procedures utilized by Dr. Baselt in the case at bar do not pass the Frye test" and that the expert opinions of Dr. Baselt and Dr. Goldberger "that rely upon those tests will not be allowed." Applying section 90.403, Florida Statutes, he further found that "the probative value of admitting evidence of the presence of THC and cTHC in the Defendant's blood at the time of Dr. Baselt's June 11, 1998, analysis is substantially outweighed by the danger of unfair prejudice, confusion of the issues, and misleading the jury," and that "testimony and evidence of the presence of THC and cTHC in the Defendant's blood will not be allowed."

ARGUMENTS

The state contends that Frye "does not apply in this situation," because GC/MS analysis is well established and has been generally accepted in the relevant scientific community, citing Parker v. State, 777 So. 2d 937 (Ala.Crim.App.2000), and Jones v. State, 716 S.W.2d 142 (Tex.App.-Austin 1986). Alternatively, it asserts that courts have repeatedly found that GC/MS is generally accepted and satisfies the Frye standard, citing Jones and U.S. v. Distler, 671 F.2d 954 (6th Cir.), cert. denied, 454 U.S. 827, 102 S. Ct. 118, 70 L. Ed. 2d 102 (1981), and that federal courts have found GC/MS satisfies Daubert v. Merrell Dow Pharmaceuticals, Inc., 509 U.S. 579, 113 S. Ct. 2786, 125 L. Ed. 2d 469 (1993), citing U.S. v. Bynum, 3 F.3d 769 (4th Cir.1993), cert. denied, 510 U.S. 1132, 114 S. Ct. 1105, 127 L. Ed. 2d 416 (1994), and U.S. v. Vitek Supply Corp., 144 F.3d 476 (7th Cir.1998), cert. denied, 525 U.S. 1138, 119 S. Ct. 1026, 143 L. Ed. 2d 37 (1999). It argues that whether the correct quality control procedures were followed "are foundational considerations governed by ordinary evidentiary standards" which go to the weight of evidence, not its admissibility, unless "the testing procedures are so fundamentally flawed that the test results are rendered unreliable." The state asserts there is no dispute about the basic reliability of the test results, noting that the RIA screened Sercey's blood positive for THC prior to the GC/MS analysis, that "an external proficiency test was performed just prior to the test at issue," and that positive and negative standards prepared in whole blood "were run at the same time and in the same run as appellee's blood." It argues that a GC/MS analysis which uses internal standards but not external controls is admissible under the Frye standard, and that the perceived "confusion" in Dr. Baselt's testimony results from the fact that the trial judge does not understand the difference between internal standards and external controls. The state asserts that what Dr. Bateh was recommending was really a second set of standards from another source, not a "true control," which is a solution kept for an extended period of time and used to compare to all samples sent to the lab, while standards are prepared "every couple of weeks." It cites Vitek and State v. St. Pierre, 693 So. 2d 102 (Fla. 5th DCA 1997), as examples of cases in which the courts found GC/MS test results accurate and reliable even where not every possible quality control protocol was followed.

The state challenges the trial court's statement during the hearing that the federal and Florida administrative rules governing lab procedure constitute evidence of the requirements imposed by both governments on this particular analysis. It asserts that these rules do not clearly require both "controls" and "standards," but appear to treat them as interchangeable, that the rules do not on their face apply to drug testing of blood, and that in any case, use of standards rather than controls substantially complies with the rules. It contends that failure to follow administrative rules, manufacturer protocols, or protocols established by scientific groups does not render test results inadmissible under Frye, but only goes to the weight of the evidence and is for the jury to consider, citing Williams v. State, 710 So. 2d 24 (Fla. 3d DCA), rev. denied, 725 So. 2d 1111 (Fla. 1998), State v. Tankersley, 191 Ariz. 359, 956 P.2d 486 (1998), and State v. Brown, 949 S.W.2d 639 (Mo.App. E.D.1997). The state asserts that the experts' opinions that Sercey's blood THC level was higher at the time of the accident than when tested were based not upon GC/MS, but upon the fact that blood THC degrades rapidly. It argues that the levels of THC and cTHC in Sercey's blood are highly relevant to the proof of her impairment, notwithstanding that a statutory "per se limit for impairment of THC and cTHC" has not been established, noting that the prosecution may prove impairment even when the defendant's BAC is less than the.08 statutory level. It asserts that if chemical analyses performed by "one of the few labs that does this type of testing due to its complexity" are excluded from evidence based on the failure to use "controls" when no controls are available and when "blood standards" put through the same extraction process were used as the nearest possible equivalent, "there will be no prosecutions for drug impaired drivers because the State will never be able to perform drug testing that satisfy [sic] the trial court's requirements."

Sercey contends that the notice of appeal should be treated as a petition for certiorari review because the evidence was not "suppressed." She argues that under the certiorari standard, this court must accept the trial court's findings of fact because they are supported by competent substantial evidence, and can look only at whether the trial court applied the correct law, not for "legal error." She asserts that the state's argument that Frye does not apply was not preserved below, that its suggestion that Daubert applies is contrary to Florida supreme court decisions, and that the cases the state cites are distinguishable. Finally, Sercey argues that even if the de novo appeal standard were applicable, the trial court applied the correct law and drew the correct legal conclusions that the procedure employed in this case was not generally accepted in the scientific community, and the court was not shown to have abused its discretion in ruling that the evidence would not be admitted because its probative value is outweighed by the danger of unfair prejudice, confusion of the issues, and misleading the jury.

The state replies that its appeal is authorized by Florida Rule of Appellate Procedure 9.140(c)(1)(B), because Sercey's blood THC/cTHC results are "evidence obtained by search and seizure," and that it may appeal this order because, unlike a defendant who may appeal a conviction, the state would have no remedy if the defendant were acquitted because it was not allowed to present evidence. It concedes that the argument that no Frye hearing was required was not presented to the trial court, but argues that Frye challenges are unique in that de novo appellate review is not limited to the record below, and that this court can take judicial notice for the first time on appeal that a scientific procedure is not novel, citing Florida Power & Light Co. v. Tursi and Hayes v. State, 660 So. 2d 257 (Fla.1995). It argues that the appellate court determines de novo the legal issues and decides whether competent substantial evidence supports the trial court's factual findings. It contends that the trial judge "simply did not understand the testimony regarding negative and positive standards," that he "improperly went beyond addressing the threshold question of admissibility of expert testimony under Frye by engaging in an analysis of the weight to be assigned to the expert testimony," and that he "improperly turned a Frye issue into a 403 balancing issue." The state asserts that it is not relying on Daubert, but that the two "gate-keeping" tests in Frye and Daubert (both "designed to remove junk science from the courtroom") are sufficiently similar that Daubert-based cases specifically involving gas chromatography are persuasive.

JURISDICTION

We review this case as a direct appeal of a nonfinal order under rule 9.140(c)(1)(B), which authorizes the state to appeal an order "suppressing before trial confessions, admissions, or evidence obtained by search and seizure," because the legal blood specimen upon which the expert opinion testimony sought to be excluded is based was "obtained by search and seizure."^[15]

RULES OF EVIDENCE

To determine whether the challenged evidence is admissible, we must apply to the facts the pertinent rules of evidence.^[16] Section 90.401, Florida Statutes, provides: "Relevant evidence is evidence tending to prove or disprove a material fact." Section 90.402 provides: "All relevant evidence is admissible, except as provided by law." Because there is no dispute that the testimony of the state's experts is relevant to some of the issues involved in this prosecution, the expert witness testimony is admissible unless otherwise excluded by law.

Section 90.702 provides: "If scientific, technical, or other specialized knowledge will assist the trier of fact in understanding the evidence or in determining a fact in issue, a witness qualified as a expert by knowledge, skill, experience, training, or education may testify about it in the form of an opinion; however, opinion is admissible only if it can be applied to the evidence at trial." There were no objections to the qualification of the state's witnesses as experts in the field of toxicology, nor was there any dispute that their scientific knowledge would assist the jury in understanding the evidence presented in this trial. Section 90.703 provides: "Testimony in the form of an opinion or inference otherwise admissible is not objectionable because it includes an ultimate issue to be decided by the trier of fact." The fact that some of the state's experts' opinions involved issues which the jury was called upon to determine (i.e., whether, at the time of the accident, Sercey's faculties were impaired by alcoholic beverages, or by marijuana, or by a combination of alcohol and marijuana) does not make their testimony inadmissible. Section 90.704 provides: "The facts or data upon which an expert bases an opinion or inference may be those perceived by, or made known to, the expert at or before the trial. If the facts or data are of a type reasonably relied upon by experts in the subject to support the opinion expressed, the facts or data need not be admissible in evidence." There is no dispute that RIA and GC/MS test results are the type of data reasonably relied upon by toxicologists to support their opinions about the presence and quantification of marijuana in a subject's blood, nor is there a dispute that textbooks and peer-reviewed published scientific studies in toxicology and pharmacokinetics are the type of data upon which toxicologists reasonably rely for opinions relating the presence and quantification of alcohol and marijuana in the subject's blood to other issues, including when the substances were ingested, the manner of ingestion, and whether it is probable that the subject was impaired at times related to the time the blood specimen was drawn.

Section 90.403 provides: "Relevant evidence is inadmissible if its probative value is substantially outweighed by the danger of unfair prejudice, confusion of issues, misleading the jury, or needless presentation of cumulative evidence." In excluding the expert witness testimony, the trial court found that "the testing procedures utilized by Dr. Baselt in the case at bar do not pass the Frye test" and that "the probative value of admitting evidence of the presence of THC and cTHC in the Defendant's blood at the time of Dr. Baselt's June 11, 1998, analysis is substantially outweighed by the danger of unfair prejudice, confusion of the issues, and misleading the jury." To determine whether this ruling is legally valid, we must examine the Frye reliability standard.

THE FRYE RELIABILITY STANDARD

In Frye, which involved the admissibility of evidence involving a precursor to the "lie detector," the court noted the "novel question" involved, and stated:

... Just when a scientific principle or discovery crosses the line between the experimental and demonstrable stages is difficult to define. Somewhere in this twilight zone the evidential force of the principle must be recognized, and while courts will go a long way in admitting expert testimony deduced from a well-recognized scientific principle or discovery, the thing from which the deduction is made must be sufficiently established to have gained general acceptance in the particular field in which it belongs.

We think the systolic blood pressure deception test has not yet gained such standing and scientific recognition among physiological and psychological authorities as would justify the courts in admitting expert testimony deduced from the discovery, development, and experiments thus far made.

293 F. at 1014. In Stokes v. State, 548 So. 2d 188, 193 (Fla.1989), the supreme court reaffirmed the applicability of the Frye test in Florida, notwithstanding adoption of the Florida Evidence Code:

The underlying theory for this rule is that a courtroom is not a laboratory, and as such it is not the place to conduct scientific experiments. If the scientific community considers a procedure or process unreliable for its own purposes, then the procedure must be considered less reliable for courtroom use.

In Flanagan v. State, 625 So. 2d 827, 828 (Fla.1993), the supreme court reiterated "the basic principle that novel scientific evidence is not admissible in Florida unless it meets the test established in [Frye]," under which, "in order to introduce expert testimony deduced from a scientific principle or discovery, the principle or discovery `must be sufficiently established to have gained general acceptance in the particular field in which it belongs.'" It noted that "pure opinion testimony" need not meet the Frye test "because this type of testimony is based on the expert's personal experience and training," but it found that the sex offender profile testimony challenged in that case "by its nature necessarily relies on some scientific principle or test, which implies an infallibility not found in pure opinion testimony," and therefore "must meet the Frye test, designed to ensure that the jury will not be misled by experimental scientific methods which may ultimately prove to be unsound." Id.

Notwithstanding the U.S. Supreme Court's ruling in Daubert that the Federal Evidence Code had superceded the Frye test in federal court proceedings, Florida has continued to adhere to Frye. In Ramirez v. State, 651 So. 2d 1164 (Fla.1995), the supreme court stated:

The admission into evidence of expert opinion testimony concerning a new or novel scientific principle is a four-step process. First, the trial judge must determine whether such expert testimony will assist the jury in understanding the evidence or in determining a fact in issue. Second, the trial judge must decide whether the expert's testimony is based on a scientific principle or discovery that is "sufficiently established to have gained general acceptance in the particular field in which it belongs."... The third step in the process is for the trial judge to determine whether a particular witness is qualified as an expert to present opinion testimony on the subject in issue. Fourth, the judge may then allow the expert to render an opinion on the subject of his or her expertise, and it is then up to the jury to determine the credibility of the expert's opinion, which it may either accept or reject.

Id. at 1166-67 (citations omitted). The court noted that the second step of this process embodies the Frye test, and observed that "[t]he principle inquiry under the Frye test is whether the scientific theory or discovery from which an expert derives an opinion is reliable." Id. at 1167. It stated that in utilizing the test, "the burden is on the proponent of the evidence to prove the general acceptance of both the underlying scientific principle and the testing procedures used to apply that principle to the facts of the case at hand," that the trial judge "has the sole responsibility to determine this question," and that "general acceptance under the Frye test must be established by a preponderance of the evidence." Id. at 1168.

In Hadden v. State, 690 So. 2d 573, 578 (Fla.1997), the supreme court reaffirmed its "intent to use the Frye test as the proper standard for admitting novel scientific evidence in Florida, even though the Frye test is not set forth in the evidence code," observing that "[r]eliability is fundamental to issues involved in the admissibility of evidence" and that, like hearsay evidence, "[n]ovel scientific evidence must also be shown to be reliable on some basis other than simply that it is the opinion of the witness who seeks to offer the opinion." It noted that the appropriate standard of review is de novo, that an appellate court "should consider the issue of general acceptance at the time of the appeal rather than at the time of trial," and that the appellate court "may examine expert testimony, scientific and legal writings, and judicial opinions in making its determination." Id. at 579. It observed that "the Frye standard for admissibility of scientific evidence is not applicable to an expert's pure opinion testimony which is based solely on the expert's training and experience." Id. at 579-80.

In Berry v. CSX Transp., Inc., 709 So. 2d 552 (Fla. 1st DCA), rev. denied, 718 So. 2d 167 (Fla.1998), this court addressed the Frye test in the context of toxic tort litigation. It noted that the challenged expert testimony "was grounded upon numerous peer-reviewed and published epidemiological studies demonstrating an association between exposure to organic solvents and toxic encephalopathy," but found that "the trial court went beyond addressing the threshold question of admissibility of expert testimony under Frye, which was the issue before it, and in effect engaged in an analysis of the weight to be assigned to the expert testimony or the sufficiency of the evidence." Id. at 554. Declining to interpret language in Hadden "as meaning that in all cases expert opinion testimony, not otherwise developed through clinical experience, must be Frye tested," the court held that "under Frye and its Florida progeny, when the expert's opinion is well-founded and based upon generally accepted scientific principles and methodology, it is not necessary that the expert's opinion be generally accepted as well." Id. at 566-67. It found that the claims of bias, lack of plausibility, and other alleged defects in the studies upon which the expert relies "go to the weight, rather than the admissibility, of the studies," which is to be considered by the fact finder. Id. at 571. It held: "The fact that the experts have all derived their opinions from the same generally-accepted methodology, the epidemiological studies contained in the record, but simply disagree upon how to interpret the scientifically (and legally) reliable data, is not a valid reason for excluding the plaintiffs' experts' opinions altogether." Id.

In U.S. Sugar Corp. v. Henson, 787 So. 2d 3, 5 (Fla. 1st DCA 2000), approved, 823 So. 2d 104 (Fla.2002), this court held that the Frye standard applies in workers' compensation cases, and affirmed the lower tribunal's reliance on the claimant's experts, finding that "there is general acceptance in the relevant scientific community both (i) for claimant's general causation theory that certain pesticides to which he was repeatedly exposed over a long period of time can cause peripheral neuropathy, and (ii) for the differential diagnosis methodology employed by claimant's physicians, which they used to exclude the other factors that might cause his condition and to determine that his pesticide exposure specifically caused his injury." It stated:

We share Judge Altenbernd's concerns with an appellate court undertaking a de novo review not only based upon the scientific literature considered by the lower court or submitted by the parties on appeal, but also based upon literature outside of the record. See Brim, 779 So.2d at 435 ("Both due process and the limited technical competence of the judiciary require that this review take place with certain safeguards that we have not yet provided."). Nevertheless, we read the Florida Supreme Courts's opinion in Brim v. State, 695 So. 2d 268, 274 (Fla.1997), as requiring such an undertaking for this court to reach a de novo determination of whether there is general acceptance within the relevant scientific community of the novel science before us. Further, the issue of general acceptance is to be made as of the time of appeal, rather than the trial. See Hadden, 690 So.2d at 579; E.I. Du-Pont, 748 So.2d at 1115.

787 So.2d at 15. The court noted that the expert's role "is to apply the generally accepted principles to the facts relating to the claimant's condition" and emphasized that "the weight to be given to stated scientific theories, and the resolution of legitimate but competing scientific views, are matters appropriately entrusted to the trier of fact." Id. at 20.

ANALYSIS

It is clear from the case law that the Frye reliability standard applies only to expert opinions which are based on new or novel scientific principles or procedures. It is clear from this record that neither the RIA screen for THC/cTHC, nor the GC/MS analysis for THC/cTHC can be considered "new or novel," and that both tests are generally accepted in the scientific community. Under Ramirez, once the judge found that the scientific principles and procedures underlying the experts' opinions regarding the presence of THC and cTHC in Sercey's blood (i.e., the RIA and GC/MS analyses, and the numerous chemical, biochemical, and physical principles upon which those procedures are based) are generally accepted in the relevant scientific community (i.e., chemistry/toxicology), the threshold determination of admissibility of those opinions was met, and the judge was required to allow the experts to render opinions on the presence of THC and cTHC in Sercey's blood, leaving for the jury to determine the credibility of those expert opinions.^[17] The issue of whether the particular analysis of Sercey's blood by Dr. Baselt's lab was performed in the proper manner, including whether sufficient quality control procedures were used to ensure the accuracy of the test results, was a question of credibility and probative value for the jury to determine, after hearing the testimony of all the experts.^[18]

Sercey sought to exclude both the results of the RIA and GC/MS analyses and the experts' opinions deduced from those results with regard to how the marijuana was ingested, how recently it was ingested prior to the blood draw, and whether Sercey was impaired at the time of the accident. To the extent these opinions are based solely upon the experts' knowledge and experience, they constitute "pure opinion" and are not subject to Frye analysis. To the extent they are based on generally accepted scientific principles, they are not subject to Frye analysis. To the extent they are based upon new or novel scientific theories or studies, they would be subject to Frye scrutiny.

The expert opinion testimony that Sercey smoked marijuana (as opposed to orally ingesting it) appears to be based on the experts' knowledge and experience, and on the pharmacokinetics of THC and cTHC, about which there is no dispute. The expert opinion testimony regarding the recency of Sercey's use of marijuana prior to the blood specimen being drawn appears to be based upon the experts' knowledge and experience, independent of the Huestis mathematical models, and is therefore not subject to Frye analysis. But any opinions as to recency of use of marijuana which were based upon the Huestis models would be subject to Frye scrutiny, to determine whether the Huestis models are generally accepted in the toxicology community.

The results of the initial study on which the Huestis models are based, and the other articles referring to the Huestis models, have been peer-reviewed and published in respected scientific journals, an indication that they may be generally accepted among toxicologists, within the announced limits of their application. However, both Dr. Goldberg and Dr. Baselt testified that the models do not apply to "excessive" levels of THC/cTHC like those involved in this case, that they are of limited usefulness with chronic users, and that the time of marijuana usage predicted by Model I in this case (1-1.5 hours prior to the blood draw, which was three hours after the accident) would not be possible. On this record, we find that the trial court could reasonably find that the probative value of application of the Huestis models to the particular facts in this case is substantially outweighed by the danger of misleading the jury, and that opinions based on the Huestis models should be excluded under section 90.403, Florida Statutes.

The experts' opinions that Sercey was impaired by alcohol and marijuana at the time of the accident appear to be based on the experts' knowledge and experience, as well as peer-reviewed and published studies indicating that the combined use of moderate amounts of alcohol and marijuana results in significant impairment.^[19] Sercey did not challenge these scientific findings, but instead focused on the fact that there is no consensus among toxicologists on a minimum per se level of THC upon which courts could base a presumption of impairment, as there is with alcohol. In Florida, the state may prove impairment by alcohol even when the BAC is less than the .08 "legal" limit, using evidence of the presence of alcohol in the defendant's blood and other evidence of impairment, including erratic driving and the fact that an accident occurred. In the same way, the state may also prove impairment by a controlled substance, or impairment by a combination of alcohol and a controlled substance. Evidence of the presence of THC/cTHC in Sercey's blood is therefore admissible even if it cannot be quantitatively related to impairment, unless it is excluded under section 90.403, Florida Statutes.^[20]

The trial court's ruling that "the probative value of admitting evidence of the presence of THC and cTHC in the Defendant's blood at the time of Dr. Baselt's June 11, 1998, analysis is substantially outweighed by the danger of unfair prejudice, confusion of the issues, and misleading the jury" seems inextricably tied to his findings that the expert witness testimony did not meet the Frye reliability standard, in particular that "the testing procedures utilized by Dr. Baselt in the case at bar do not pass the Frye test." In light of the fact that Sercey did not challenge GC/MS analysis and the scientific principles underlying GC/MS, which has been considered for many years the "gold standard" for drug analysis, and considering that a challenge to the way in which a particular test was administered pertains to the weight and credibility of the evidence rather than to its admissibility, this ruling under section 90.403 must be reversed.^[21]

The trial court's determination that Sercey's particular test results were not reliable because Dr. Baselt's lab did not use "controls" appears to be the result of semantic confusion. What Dr. Bateh refers to as the required negative and positive THC/cTHC "controls" (made by spiking "clean" blood samples with known amounts of THC/cTHC and running the spiked samples through the same extraction procedure and GC/MS analysis as the "unknown" sample of Sercey's blood) are precisely what Dr. Baselt refers to as the negative and positive THC/cTHC "standards" his lab actually used in this analysis. All the experts agree that, because of the instability of THC in blood, what Dr. Baselt correctly refers to as a THC/cTHC "external control" (a blood sample with known THC/cTHC concentrations that can be run with each analysis over an extended period of time, for the purpose of determining whether the analytical procedure gives the same result over time) is not feasible.

Our extensive research of the scientific principles underlying the issues, required in order to fulfill our obligation to review this case de novo, highlights two areas of inquiry that in our opinion were not adequately explored by the attorneys at the pre-trial hearing, and might be helpful to the jury in understanding the accuracy of the particular GC/MS results in this case. These are the explanation of the "deuterated internal standards" which were added to all the blood samples, including the calibration standards, and the question of whether Dr. Baselt's preparation and use of calibration standards in a blood matrix, combined with the aforementioned internal standards, proficiency studies, and duplicate analyses of samples by another lab, ensured the accuracy of the analysis in this case.^[22] It is our opinion that Dr. Baselt's lab followed the protocols necessary to ensure the accuracy of the test results, given the unstable nature of THC in a blood matrix. However, this determination is for the jury to make, after considering fully the quality control aspects of the GC/MS analysis.

The trial court's exclusion of any expert witness opinion testimony regarding the recency of Sercey's use of marijuana which is based upon the Huestis models is AFFIRMED. The remainder of the order is REVERSED and the case is REMANDED for further proceedings consistent with this opinion.

WOLF and LEWIS, JJ., CONCUR.

NOTES

[1] The McGraw-Hill Encyclopedia of Science and Technology describes gas chromatography as "a method for the separation and analysis of complex mixtures of volatile organic and inorganic compounds" which is "based on the fact that different compounds move at different rates through a column containing a solvent liquid or solid." It notes that "[a]fter separation of the mixture with gas chromatography, each component can be analyzed further with a mass spectrometer, which determines molecular structure based on the fragmentation of ions formed when the molecules are ionized." GC/MS is also briefly described in U.S. v. Distler, 671 F.2d 954 (6th Cir.), cert. denied, 454 U.S. 827, 102 S. Ct. 118, 70 L. Ed. 2d 102 (1981), and in Andrews v. Webb, 685 F. Supp. 579 (E.D.Va.1988).

The scientific principles underlying solvent extraction, gas chromatography, and mass spectrometry were discovered and developed in the early 1900s, and the particular combined method of analysis referred to as GC/MS has been in use for over forty years in a wide variety of applications, including drug testing. The experts agreed that GC/MS analysis is generally accepted in the scientific community as the best method for determining the presence and quantity of THC and cTHC in blood. The principal reasons for preferring GC/MS analysis for drug analysis are "its sensitivity in detecting minute amounts of a given substance and its selectivity in distinguishing various compounds," Jones v. State, 716 S.W.2d 142, 149 (Tex.App.-Austin 1986). GC/MS has been described as the "gold standard" for drug confirmation testing. See Goebel v. Warner Transp., 612 N.W.2d 18, 22 (S.D.2000). See also Andrews v. Webb (RIA and GC/MS analysis for THC in urine); Williams v. Secretary of Navy, 787 F.2d 552 (Fed.Cir.1986)(RIA, GC, and GC/MS analysis for THC in urine); Parker v. State, 777 So. 2d 937 (Ala.Crim.App.2000) (GC/MS analysis of cocaine); U.S. v. Bynum, 3 F.3d 769 (4th Cir.1993), cert. denied, 510 U.S. 1132, 114 S. Ct. 1105, 127 L. Ed. 2d 416 (1994) (GC analysis of cocaine impurities). And see State v. Price, 731 S.W.2d 287 (Mo.App. W.D. 1987) ("The reported decisions and authorities confirm that the mass spectrometry and gas chromatography technique is an extremely reliable specific test for the measurement, separation and identification of particular organic compounds-drugs among them").

[2] Dr. Bruce Goldberger, Ph.D., Director of Toxicology at the University of Florida College of Medicine and a federal inspector of laboratories, has been qualified about 30 times in the last three years in federal, state, and military courts as an expert in the field of forensic toxicology. He regularly gives opinions regarding impairment from drugs or alcohol.

Dr. Randall Baselt, Ph.D., has a B.S. in chemistry and a Ph.D. in pharmacology, and has worked as a toxicologist since 1965, including two years of toxicology research for the City of San Francisco, a ten-year professorship of toxicology, and positions as director of four toxicology labs, including director of his current lab since 1984. He has been qualified as an expert in toxicology more than 2200 times in 25 states (including Florida), the federal courts, and three foreign countries. In each of the Florida cases, he testified regarding recent usage of marijuana and impairment based on THC in blood. In 1983, he authored a study of comparative testing for THC/cTHC in whole blood and in serum/plasma.

Dr. Ricky Bateh, Ph.D., is an attorney, chemist, former clinical lab director, consultant, and adjunct professor at Florida Community College in Jacksonville, where he teaches GC/MS, toxicology, and environmental clinical chemistry. He has testified 75-80 times in state, federal, and military courts as an expert in forensic chemistry and toxicology. He has done some research, but he has not published anything on blood THC.

Dr. Edward Willey, M.D., a former hospital pathologist, testified that he does not consider himself a toxicologist because he is not a chemist, but he considers himself qualified to testify regarding the results of toxicological tests. He admitted that he had been hired to interpret the blood alcohol data and give a possible explanation for the two different test results, and that he did not challenge the GC/MS results or the RIA results.

Thomas Burr has a B.S. in chemistry and biology with 32 years experience in areas relating to forensic testing and forensic science, including 21 years in a Minnesota crime lab and "quite a few" years running GC/MS analyses. He has testified as an expert on alcohol and drugs in 22 states, in federal court, and in Canada, and he has testified as an expert in Florida cases "on numerous occasions." He has written a number of review articles in the area of alcohol testing and has given a dozen or more seminars on the effects of alcohol and drugs on human performance and the forensic aspects of alcohol and drug testing. He has not published any articles on THC, has done no research, and has done no THC analysis in blood, only in urine specimens.

[3] Dr. Baselt opined that the levels of THC and cTHC his lab found in Sercey's blood

... are in our experience very significant concentrations. They each exceed the typical numbers that we see in forensic situations involving marijuana. And together, they would indicate recent usage of marijuana by the individual. And they would also be consistent with the person being under the influence of marijuana at the time of the blood collection.

Dr. Goldberger testified that Sercey's cTHC level was "actually in the [higher] range that you'd find in urine." Dr. Bateh testified that there have been studies in which baseline THC levels for chronic users have been as high as 4 ng/ml, but he admitted that a cTHC level of 91 ng/ml is "excessive." Dr. Willey testified that 5.5 ng/ml at the terminal phase is "a reasonably high level." Burr testified that he would not expect 5.5 ng/ml to be a residual level of THC, and that he did not see anything in the scientific literature to suggest that baselines for chronic users "could be that high."

[4] Dr. Goldberger testified that Dr. Baselt "is highly regarded in the field of forensic toxicology," and observed that "in California the rules about a forensic laboratory are very strict, probably more strict than they are in the State of Florida." Dr. Bateh testified that Dr. Baselt is well-respected in the field of toxicology, but that he is not certified by any forensic organization.

[5] The most recent cTHC proficiency test prior to the June 11, 1998, analysis of Sercey's blood specimen was performed on April 14, 1998. Dr. Baselt testified that in that proficiency test, his lab obtained a 94 ng/ml reading for a sample with a target value of 101 ng/ml, which "tells us that we are doing reasonably well."

[6] A deuterated compound is one in which one or more hydrogen atoms have been replaced by deuterium atoms. Deuterium is an isotope of hydrogen, which contains an extra neutron, and therefore has double the atomic weight of hydrogen.

[7] Dr. Baselt explained:

The ion ratio is simply the ratio between the abundance of two different ions within the same chemical, or two ions representing different chemicals, such as THC in a person's bloodstream, and the internal standard deuterated form of THC. So these can be used either qualitatively or quantitatively; meaning, you would either employ them in identifying the substance, or in measuring the amount of the substance.

[8] Dr. Baselt testified that the ion ratios his lab used for quantitation of the GC/MS results were 410/413 for THC and 640/643 for cTHC, that the ion ratios were within the required ± 25% of the equivalent standard, and that the ± 25% requirement was "developed in-house as being reasonable, based on our validation studies when we were first setting this method up."

Dr. Bateh testified that Dr. Baselt's data has "no consistency" in the ion ratio range used to identify THC, which he believed "calls into question" the 5.5 ng/ml result. He stated that for forensic quantitative analysis using selective ion monitoring, three ions are usually monitored (two to identify the THC, and the third to quantify it). He stated that Dr. Baselt used the 367/410 ion ratio to qualitatively identify the unknown as THC, and the 410/413 ratio to quantify it, but that his 367/410 ratios range from .625 to 1.051, which is "totally unacceptable" because there is more than ± 20% variation. He noted that Dr. Baselt used the 489/640 ion ratio to qualitatively identify the unknown as cTHC, and the 640/643 ratio to quantify it, and that his 489/640 ratios range from .565 to .589, which is well within the ± 20% range.

Dr. Goldberger explained that because ion ratios may be concentration dependent, most GC/MS labs, including his lab and Dr. Baselt's lab, use a technique called "concentration matching" for ion ratios (i.e., matching the ion ratios for a concentration that is close to the sample) which allows "some variation in these ranges" beyond the accepted ± 20% range, noting that the range would be within ± 20% of the standard near the same concentration as the sample.

[9] Dr. Baselt testified that his lab did not use external controls for THC and cTHC, and explained:

We don't actually have a true control for the blood THC GC/MS testing ... we have never found a material that is stable over a long period of time for THC and carboxy THC content. That is the definition of a control. We can do it in urine for carboxy THC but not in blood ... [which] degrades in terms of both THC and carboxy THC content significantly over time .... the purpose of a control is to be able to compare the result of the test that we did last week with a test that we are doing this week and with tests that we will be doing for the foreseeable future. And simply to guarantee that our results are reproducible, not accurate but reproducible. And, preferably, a control should be stable for up to a year, and you would use the same material over that period of time. Therefore, you could say that every test that we did, every week, for that period of one year was the same; there were no significant differences. You may be 100 percent inaccurate in those results. But, at least, you were reproducible. That's the point of a control.

He testified that the lack of external controls does not affect the reliability of the test, which is determined by comparison of the results of the unknown sample with the results of the standards that were analyzed with it. He stated that the accuracy of the testing procedures are verified about once a month by rerunning blood samples that have been tested by other labs across the U.S., and "we find that we typically agree very well with those folks." He testified that, as a general rule, it "would certainly be advisable" to use standards from one source and external controls from another source, but "for this particular procedure we have not found that as a possibility." To his knowledge, there are no blood THC controls, and they have not been able to find "a matrix that duplicates blood that is stable in terms of its THC content."

Dr. Bateh testified that calibration standards and external controls should come from different sources, and "while there is no blood sample that you can buy that contains THC, what's very commonly done in this field is you would go to your blood bank for those pints of blood or units of blood that are no longer available for transfusion" and "spike into blood samples," then run the spiked samples through the same extraction procedure and GC/MS analysis as the "unknown" sample. He stated that a standard is used to calibrate the instrument "every time you run the test," while a control is used to "evaluate whether or not the calibration of the instrument was done properly." He opined that running controls is "the only way you can monitor your extraction process for your unknowns" and that without the use of controls, there is no way to know if the results generated by the standard curve are accurate:

Because for the most part the standard curve materials do not always undergo the same extraction process as your unknowns do. Some spike them in blood and others just run them from the methanol national vial that you get them in.

He testified that he was "shocked knowing Dr. Baselt's reputation" when he discovered that Dr. Baselt "failed to take a very common procedure of making his own controls and running them." He opined that monthly proficiency testing would not cover an "every-day situation." He conceded that the National Institute on Drug Abuse (NIDA) and the Substance Abuse and Mental Health Services Administration (SAMHSA) mandatory standards for labs reference urine testing and blood alcohol testing, but not blood THC testing. He stated that chapter 440, Florida Statutes, requires controls for Drug Free Workplace testing, and opined that "[t]he analytical principles behind urine testing is the same, and should be the same, as those analytical principles applied to blood testing."

Dr. Baselt and Dr. Goldberger testified that drug testing of blood and drug testing of urine are "just two different systems," and that NIDA, SAMSHA, and the Florida Administrative Code rule for clinical labs do not address analysis of THC in blood. Dr. Baselt explained that the quality control requirements are different for the two systems because when testing urine for marijuana use, "you do not look at THC," but "only test for carboxy THC," and because "[u]rine tends to be more stable for almost anything, than blood."

Dr. Goldberger testified that the Society of Forensic Toxicologists (SOFT) recommends one positive control and one negative control with each sample batch. He stated that in his lab, they usually spike blood the same day to make a "laboratory control sample" for blood analysis, but that this would not be acceptable in a urine testing laboratory. He acknowledged that spiking a blood sample the same day is "probably not the best way, because it doesn't give you any feedback on the stability problem," but he testified that a THC blood or serum control that is stored for some period of time "wouldn't be stable" and "is going to degrade very quickly," so "I think the only solution would be to spike that day." He testified that Dr. Baselt's lab work met the scientific requirements and procedures generally accepted within the scientific community, and that although Dr. Baselt did not run positive controls, he ran external proficiency samples and he does "inter-laboratory comparisons, which are very useful and part of quality control," so "those are considered control samples." He testified that peer evaluation by proficiency testing is an excellent method of control, perhaps even better than spiked controls, noting that if the lab control is improperly made, "my results could be wrong every day, even with a control in every batch."

Dr. Baselt testified that spiking blood samples in the laboratory is not a "control":

That is a standard. A control is something that you use over a minimum of three months, and you do it repetitively. And you start off by analyzing it 20 times, over a period of a couple of weeks. So it has to be very stable to be able to work as a control.

He acknowledged that external controls are recommended, but stated that "it's something that has to be looked at on an individual basis, depending on what you are testing for. And there are simply things that cannot be controlled through the usual mechanisms."

Burr testified that Dr. Baselt's procedures in this case are not scientifically reliable because "no quality controls, positive or negative," were run at the time of the analysis of Sercey's blood. He described a "control" as a sample with a known concentration, which is handled "just as if it were an unknown sample" and is used to check the calibration of the instrument. He stated that it is a "very basic tenet of toxicological testing" that controls must be run with evidence samples, and "you cannot judge the accuracy of results without control tests." He testified that in this case a "negative control" would be a blood sample with no THC or cTHC, and a "positive control" would be a blood sample with a known concentration of THC and cTHC, both of which would be processed with the unknown samples through the extraction procedure and the GC/MS analysis, "to validate both your instrumental measurements and your calibration of your GC/MS procedure, and to validate your analytical procedures of getting the sample to the gas chromatograph and mass spectrometer." He stated that if no commercially prepared controls are available, it is a common practice in this field to "manufacture your own control materials" using blood determined to be drug-free, but he noted that because "the blood will degrade after you make up your control," there is "a limited time that you can use these controls" and "you have to make them up on a regular basis" and "determine in your laboratory how long you can reliably use them after you make them up." He testified that proficiency testing is not an acceptable substitute for controls unless the proficiency sample were run at the same time as the standard curve samples.

[10] Dr. Goldberger testified that the method of ingestion of marijuana affects the blood levels. He explained that orally ingested marijuana has a delayed and reduced absorption which peaks after 2-3 hours, that when marijuana is eaten, "the levels in the blood are much lower compared to the much higher levels in the blood when you smoke," and that in his "scientifically accepted opinion," the marijuana in Sercey's blood was smoked. He testified that it is "possible, but highly unlikely," that the THC levels in Sercey's blood resulted from oral consumption, explaining that to reach the high levels of THC/cTHC found in the GC/MS analysis, Sercey would have had to eat "a huge amount."

Dr. Baselt testified that he was not able to determine "just from our test results" how the marijuana in Sercey's sample was ingested, but that based on published studies, "[o]ne would have to consume approximately three to four times as much THC by the oral ingestion route to obtain the same blood THC concentration relative to smoking marijuana." He stated that the peak blood levels of THC are 50 ng/ml (if the marijuana is smoked) and 4-11 ng/ml (if the marijuana is orally ingested).

Burr testified that the results of the GC/Ms analysis do not indicate the method of consumption of marijuana, but that a blood THC level of 5.5 ng/ml is "not a likely level that you could get from secondhand smoke," and that the marijuana was likely ingested through some direct source.

[11] When Dr. Goldberger was asked for the time of Sercey's consumption of marijuana, he opined that in view of the high levels of THC/ cTHC, "it was recent exposure." He testified that he uses the Huestis models only "as guidance to confirm my gut feeling or my intuition on roughly when the marijuana was used," which is based on the facts of the case. He stated that based on his experience, he can give an opinion as to time of usage even if the subject is a repetitive user and even without the Huestis models. He described the Huestis models, which he testified have been "peer reviewed" and are used by many law enforcement agencies, but he admitted that the models were developed from studies which did not involve chronic marijuana users and that there was an indication the models are not useful if the cTHC level is "excessive." Using Model I (based on serum THC levels) and multiplying the GC/MS whole blood test results by a conversion factor of two, he estimated that Sercey smoked marijuana 1-1.5 hours before her blood was drawn, which he admitted would be "virtually impossible" because she was in custody. Using Model II (based on both THC and cTHC serum levels), he estimated that she smoked marijuana 3-3.5 hours before her blood was drawn.

Dr. Baselt testified that Sercey's "very significant concentrations" of THC/cTHC, taken together, "would indicate recent usage of marijuana by the individual." He stated that he has been testifying to this type of thing for at least fifteen years, and that this conclusion is generally accepted within the scientific community of toxicologists. He explained that cTHC is a metabolite of THC which "may take an hour or two or sometimes even three hours to reach its peak [level in the blood]," and "generally is an indication that the usage actually occurred several hours prior to the actual blood draw." He conceded that "a certain portion of what we see in this case could be a result of chronic accumulation, but certainly that would represent a minority of what is present at the time of the blood collection in this case." He stated that he uses the Huestis models as predictors, "simply to estimate the time of use of marijuana," but he noted that Model I, which predicts the marijuana would have been ingested approximately one hour prior to blood collection, "typically does not work" when the THC/cTHC levels are high, as they are in this case. He found that Model II "gives us a predicted time of use of three hours prior to the blood collection," but he admitted that the Huestis models are of limited usefulness for chronic users and that the models assume ingestion of a single average-sized marijuana cigarette. He explained that in many cases the ingestion "has occurred over hours, sometimes even days, prior to blood collection," and that "a much more reliable method would be to compare the concentrations of the THC and the metabolite as a ratio so you could remove a number of variables, one of which would be the amount of usage." He stated that this technique "would also tend to correct, to a certain extent, for the possible accumulation of these substances over a time from repeated usage."

Dr. Baselt stated that, without relying on the Huestis models, but only on the THC/ cTHC levels, and assuming that one averagesized marijuana cigarette was smoked, he "probably would have come up with a figure closer to two hours, I would think, because 5.5 is very high," but he observed that if the marijuana usage was at least three hours before the blood was drawn, the 5.5 ng/ml level "would have to represent substantially more than just a single average marijuana cigarette." He testified that gender, state of health, and body fat make no difference for estimation of THC levels or time of usage, but that gender does make a difference in alcohol calculations. He stated that his opinion as to recency of use was not affected by Sercey's alleged admissions regarding her use of alcohol and marijuana.

Dr. Bateh testified that the Huestis models are useful for targeting some time frame of possible exposure, but that they are not definitive, and are not scientifically reliable in this case. He stated that Dr. Baselt's estimate of 1-2 ng/ml baseline levels of THC and 10-20 ng/ml baseline levels of cTHC "may be a general statement, [but] that certainly doesn't encompass a whole population of chronic marijuana users," and opined that it is quite possible that a repetitive user could have a baseline THC level higher than 1-2 ng/ml. He testified that the Huestis models cannot be used without knowing an individual's baseline levels, past exposure history, and body mass. He stated that the Model I result in this case (one hour before the draw) "basically speaks for itself," that Model II should not be used with excessive cTHC levels, and that "there is no real documented evidence as to the cTHC partition ratio between plasma and whole blood" (to use Model II, the whole blood levels must be converted to plasma levels, using the partition coefficients of THC and cTHC). He felt that an estimate of the recency of use of marijuana cannot be made in this particular case.

Dr. Willey testified that it is not possible to estimate, to a reasonable scientific certainty, the recency of marijuana use based on a one-time blood draw. Based on his "working knowledge" of the Huestis models, he opined that they "have very limited applicability to chronic users," and found the result using Model I "manifestly absurd" in light of the facts of the case. He stated that there is a great deal of uncertainty as to baseline levels for chronic users.

Burr testified that the Huestis models are not reliable indicators of the recency of marijuana use in this case, because they presuppose the method of administration is smoking, they presuppose there are no residual levels of THC and cTHC, they are based on serum tests, and "I don't think there's sufficient evidence to show that the serum ratio and whole blood ratio are interchangeable."

[12] Dr. Goldberger testified that the differences in analyzing whole blood and blood plasma/serum depend on the drug:

In the case of THC, the active drug in marijuana, we know that serum contains about twice the amount that the blood does. So if you have a blood level of 5.5, it would correspond to a serum [level] of about 11.... In the case of alcohol—blood alcohol determination, the serum alcohol is roughly 15 percent greater than the blood alcohol.

He did not think the conversion from blood to serum affected the ability to quantitatively evaluate the level of THC in blood. He stated that the conversion factors are "known and published widely," are "generally accepted in the scientific community," and are "used all the time." He testified that he was not aware of a generally accepted conversion factor for cTHC, but that Dr. Baselt's study indicated "that it partitions similar to THC" and there is no contrary indication from the "chemistry of the molecules." He admitted that if the partitioning for THC and cTHC were different, that would affect the interpretations of the results under the Huestis Model II.

Dr. Baselt testified that it is generally accepted within the forensic community that for both THC and carboxy THC, the ratio of blood to plasma is 0.55, and the cTHC conversion factor "has been published in the literature, but, in fact, I did not need to employ it in this case because there is a ratio between the THC and carboxy THC that would cancel out any blood to plasma difference." His 1983 comparative study, which involved three separate laboratories and was peer reviewed by two anonymous toxicologists before publication in the Journal of Analytical Toxicology (a recognized journal in the field of toxicology), supported his opinion that the conversion factors were the same for THC and cTHC. He testified that significant variations in hematocrit can make a difference in the ratio, and that he had assumed Sercey's hematocrit was 0.50.

Dr. Bateh testified that there is "no real documented evidence as to the cTHC partition ratio between plasma and whole blood," and that Dr. Baselt presumed that the partition coefficients of THC and cTHC were the same, but "because THC is fat liking and cTHC is water liking, the partition coefficient is expected to be different-how different, we don't know."

Dr. Willey testified that he is "by no means certain" it is appropriate to assume the same partition coefficient for THC and cTHC, that he has found no reported coefficient for cTHC, and that the conversion factor for THC would depend on the hematocrit at the time of the blood draw. He stated that Sercey's hematocrit was 39-40 before the legal blood specimen was drawn, and 32 after the blood specimen was drawn. He opined that if Dr. Baselt had used her actual hematocrit, his 5.5 ng/ml whole blood THC result converted to plasma concentration would have been somewhere in the range of 7-8 ng/ml, instead of the 10 ng/ml he calculated.

Burr testified that the conversion factor for THC in plasma and THC in blood is generally accepted as .55, but that factor would not apply to cTHC, and he was not aware of any literature on the issue. He was not familiar with Dr. Baselt's 1983 peer reviewed article finding the conversion factors for THC and cTHC to be the same, and he knows of nothing that contradicts that finding.

[13] Dr. Goldberger testified that it is not possible to "back extrapolate" with marijuana, but that if Sercey smoked the marijuana between 9:30 and noon, "the level must have been very high at the time of the crash, which was about 12:30, because the blood levels three hours later were still very high." He stated that if she ate the marijuana, "she ate a lot of marijuana," but he would not be able to say what her levels were while she was driving.

Dr. Baselt stated: "In this case where we have a blood THC level of 5.5 and the time in question is the time of the blood draw, which happened several hours after the accident, there's no question in my mind at the time of the blood draw this person was impaired and at the time of the accident the level would have been much higher ... potentially as much as twice as high."

Burr testified that there was no evidence that would allow anyone to give an opinion as to what Sercey's THC/cTHC level was at the time she was driving, and that "when you don't know past use and you don't know [the method of] administration, you can't make even an educated guess as to earlier levels."

[14] Dr. Goldberger testified that there is no written consensus per se limit for marijuana or cocaine in blood, as there is for alcohol, but that Sercey's 5.5 ng/ml THC and 91 ng/ml cTHC "corresponds to measurable impairment." He stated that the current consensus statement on drug concentrations and drug impairment is from a 1985 article and is out of date in light of more recent studies which found that blood levels of THC as low as 1.0 ng/ml would correspond to impairment of driving, and that "ordinary amounts" of alcohol and marijuana taken together result in "severe intoxication" and "severely impaired" driving. He stated that there is a movement to establish a per se impairment limit for marijuana, and that "there are a few people out there that feel like they have enough information now that they could establish realistic limits." He testified that he could not give an opinion on impairment based solely on a single sample showing the presence of THC, without any other information, but opined that "when you look at all the other factors that compose this accident," including Sercey's observed erratic behavior, the THC/cTHC analysis results reflect that she was intoxicated and "severely impaired." He stated that Sercey's case parallels a federal civil case in which he is retained, in which the driver had "a moderately high level of cocaine in the blood" and was driving erratically. He opined that in both cases, looking at all the facts, including the toxicological data, "we come up with a clear picture that there was impairment at the time of the crash." He testified that he does not need the Huestis models to determine if someone is impaired.

Dr. Baselt testified that a person with a blood THC level of 1.0 ng/ml or higher "would be under the influence of marijuana at that time," and that Sercey's "very significant concentrations" of THC and cTHC would be "consistent with the person being under the influence of marijuana at the time of the blood collection."

In this case where we have a blood THC level of 5.5 and the time in question is the time of the blood draw, which happened several hours after the accident, there's no question in my mind at the time of the blood draw this person was impaired and at the time of the accident the level would have been much higher than 5.5. And, absolutely, the person would have been under the influence of marijuana.

He opined that even a small amount of alcohol (as low as .02 or .03 percent) in a person's system at the same time there is THC present in the blood stream "greatly exacerbates the effect of the marijuana." He stated that in forming an opinion as to impairment, he looks first to the blood test results and then to "other investigative facts," and that his opinion is based upon "numerous studies, probably more than three dozen" done over the past 20 years which administered known amounts of marijuana to individuals and then measured "psycho-motor parameters that would be indicative of significant alteration in the person's ability to perform mental or motor functions," as well as studies (in several of which he has been involved) "that have looked at the relationship between administering marijuana and the determination of blood THC and carboxy THC concentrations," and "hundreds of actual forensic cases over the years, typically involving vehicle accidents and blood samples having been drawn from individuals in correlation with the impairment that was observable by police officers as well as statements made by the various individuals." He stated that he has been testifying to this type of thing for at least 15 years, that he could provide the names of numerous scientists who have the same opinion as he about the impairment level of THC, and that his opinion is generally accepted within the scientific community of toxicologists ("As far as I can tell, this information is generally held by my other colleagues").

Dr. Baselt testified that he was not aware of a consensus in the scientific community on per se levels of blood THC that equal impairment, but "a number of investigators have concluded that people who are chronic users of marijuana, who have [residual] blood levels of THC in that range [1-2 ng/ml], are always under the influence and do show impairment of their psycho-motor ability." He stated that these are "fairly recent" studies (within the last year or two) which do not establish a per se level, but "just a general correlation" between THC level and psycho-motor impairment, but that it appears to be generally accepted in the scientific community that impairment is predicted at certain levels of blood THC. He disagreed with Dr. Bateh's opinion that no one can say whether a person was impaired by marijuana three hours earlier, based upon a single blood draw, noting: "I have spent many, many hours in the last ten years doing exactly that, in various courts around this country."

Dr. Bateh testified that it is generally accepted in the scientific community that THC causes impairment at some levels, but that based on a one-time blood draw three hours after the accident, with numerous compounding variables, "there is no way to go back and say, beyond a reasonable doubt, that at 12:30 that person was impaired," and that an opinion on any past impairment based upon back calibrations is "scientifically indefensible." He stated that there have been studies in which baseline blood THC levels for chronic users have been above one (up to 4 ng/ml), and he would disagree that anyone above one "would always be impaired."

Dr. Willey testified that an extensive literature research did not establish that there is a consensus as to any specific level of THC in the blood which correlates to impairment for the purpose of driving a vehicle, nor that there is a consensus that impairment by marijuana can be determined from a one-time blood draw. He stated that a study sponsored by the National Highway Traffic Safety Administration determined that it was not possible to predict driving impairment by plasma concentrations of THC and/or cTHC. He opined that it is not possible for anyone to form an opinion, within a reasonable degree of scientific certainty, as to whether Sercey was impaired by marijuana at the time of driving, and in his opinion, she was not impaired by alcohol. He stated: "I cannot dismiss some sort of combination producing a level of impairment, but that is, on my part, simply speculation."

Burr testified that there are no per se levels of THC associated with impairment, that "you cannot make a level of impairment [opinion] based solely on a single toxicology test" without other evidence, and that speeding would not be evidence of impairment, nor the fact that there was an accident. He admitted that some studies indicate that at low levels (1-2 ng/ml in blood) "a large percentage [90%] of the people exhibited some signs of impairment" and that "a couple of those studies have even suggested that we can say, based on these studies, that people are under the influence," but he stated that he did not think "that those things have been generally accepted in the scientific community, the basis of these particular studies" and "I don't think they can apply them to the population as a whole." He stated that not very many of these studies have been done on driving impairment, and that "some studies of driving show people do just as well, or better, when they have levels of marijuana." He testified that he "would not be able to give an opinion on whether [Sercey] was or wasn't impaired at the time of driving, by marijuana," but he admitted that he "would expect somebody with that level to have, for the marijuana, to be having some effects on them, sure," and "at that level, yeah, I would expect to find some impairing effects of that marijuana, yes." He stated that "you can demonstrate impairment in individuals at that particular level, even in more individuals than not," but that "you can't tell in any one particular individual whether they're impaired because there's not enough data."

[15] See State v. Weitz, 500 So. 2d 657 (Fla. 1st DCA 1986), disapproved on other grounds, State v. McClain, 525 So. 2d 420 (Fla.1988). See also State v. Katiba, 502 So. 2d 1274 (Fla. 5th DCA 1987). And see State v. Brigham, 694 So. 2d 793 (Fla. 2d DCA 1997); State v. Tagner, 673 So. 2d 57 (Fla. 4th DCA), rev. denied, 677 So. 2d 841 (Fla.1996); State v. Pagach, 442 So. 2d 331 (Fla. 2d DCA 1983); State v. Snell, 391 So. 2d 299 (Fla. 5th DCA 1980). But see State v. Barber, 783 So. 2d 293 (Fla. 5th DCA 2001), rev. granted, 800 So. 2d 612 (Fla.2001); State v. Sandt, 751 So. 2d 136 (Fla. 2d DCA 2000), approved in part, quashed in part on other grounds, 774 So. 2d 692 (Fla. 2000); State v. Sybers, 743 So. 2d 619 (Fla. 1st DCA 1999).

[16] See Williams v. State, 710 So. 2d 24, 33-34 (Fla. 3d DCA), rev. denied, 725 So. 2d 1111 (Fla.1998).

[17] Federal courts have rejected the assertion that even when the underlying scientific theories or methods are generally accepted in the scientific community, the expert opinions based upon those theories or methods must also be generally accepted in the scientific community. See Cella v. U.S., 998 F.2d 418 (7th Cir.1993); Christophersen v. Allied-Signal Corp., 939 F.2d 1106 (5th Cir.1991), cert. denied, 503 U.S. 912, 112 S. Ct. 1280, 117 L. Ed. 2d 506 (1992); Peteet v. Dow Chemical Co., 868 F.2d 1428 (5th Cir.), cert. denied, Dow Chemical Co. v. Greenhill, 493 U.S. 935, 110 S. Ct. 328, 107 L. Ed. 2d 318 (1989); Osburn v. Anchor Laboratories, Inc., 825 F.2d 908 (5th Cir.1987), cert. denied, Rachelle Laboratories, Inc. v. Osburn, 485 U.S. 1009, 108 S. Ct. 1476, 99 L. Ed. 2d 705 (1988); Ferebee v. Chevron Chemical Co., 736 F.2d 1529 (D.C.Cir.), cert. denied, 469 U.S. 1062, 105 S. Ct. 545, 83 L. Ed. 2d 432 (1984).

[18] See Troedel v. State, 462 So. 2d 392, 396 (Fla.1984), in which the defendant challenged admission of gunshot residue test results. The supreme court found that he had not objected to the evidence at trial, then observed:

Moreover, appellant has not presented any sufficient reason why the chemical analysis should not have been admitted into evidence. He does not challenge the scientific basis for the neutron activation test. The validity and reliability of the tests are widely recognized among scientists and the evidence from them are therefore generally accepted as evidence in the courts. Appellant's challenge is to the method in which the test was administered.... We see the appellant's objection as pertaining to the weight and credibility of the evidence rather than to its admissibility. Any perceived weakness in the testimony of the expert and the detective could have been brought out through cross examination.

See also Moore v. U.S., 374 A.2d 299, 303 (D.C.1977), abrogated on other grounds, Thomas v. U.S., 650 A.2d 183 (D.C.1994), in which the court found that

appellant's attack focused upon the manner in which the government's expert conducted the tests and not upon the inherent validity of those types of tests. Essentially, then, appellant's argument deals with the weight to be accorded to the testimony of the government's expert and is properly addressed to the finder of fact.

And see State v. St. Pierre, 693 So. 2d 102 (Fla. 5th DCA 1997), in which the court reversed an order suppressing blood alcohol results on the ground that the .15 whole blood control had expired eight months prior to the test date. Based on the implied consent rule, the court found that the state had the burden of showing "that the test of the defendant's blood was performed in substantial compliance with the applicable administrative rules and statutes." It found that "there is no evidence in the instant record to support a finding that substantial compliance was not met."

[19] In Durham v. State, 956 S.W.2d 62, 66 (Tex.App.-Tyler 1997), the appellate court rejected the claim that the trial court had erred in allowing a toxicologist "not only to testify regarding the presence of cannabinoids in the blood sample but also to offer his opinion concerning whether and to what extent Appellant was intoxicated by marijuana at the time of the accident, based on the levels of the chemical components found in the blood sample taken at least two hours after the accident occurred." The appellant had not challenged the expert's qualifications, nor had he objected to the use of GC/MS to determine the presence of marijuana. His objection was directed only to whether the expert could reach a scientifically acceptable conclusion regarding the time and degree of impairment from marijuana ingestion based upon the levels of cannabinoids found in the blood sample. The expert had explained his findings and had testified that it is fairly well documented in scientific circles that reasonable judgments and estimations may be made based on the half-life of the cannabinoids concerning the level of intoxication and effects of intoxication at a particular time.

In Munroe Auto Equipment Co. v. International Union, United Automotive, Aerospace and Agr. Implement Workers of America (UAW), Monroe Auto Equipment Co., Unit of Local 878, 981 F.2d 261 (6th Cir.1992), cert. denied, 508 U.S. 931, 113 S. Ct. 2396, 124 L. Ed. 2d 298 (1993), involving dismissal of an employee for violating a drug-free workplace policy, the expert toxicology witness described the effects of marijuana and conceded that there was no legal definition or standard for determining when a person is under the influence of marijuana as there is for alcohol, but noting that the employee's test level was far above suggested levels both for being under the influence and for being impaired, he opined that the employee was under the influence of marijuana when he was tested (no Frye issue was raised in that appeal).

[20] See State v. Weitz, in which this court reversed a trial court order which had suppressed evidence of drugs found by urinalysis because the effect of the drugs could not be linked quantitatively to impairment. This court held that the evidence was admissible because the existence of drugs went to the element of "under the influence" necessary for a DUI prosecution, and noted: "The extent of that influence is a separate element, provable by other evidence, including the observations of the police officers and the fact that appellee caused the accident." 500 So.2d at 659.

In State v. McClain, the supreme court resolved the apparent conflict between the 4th District's opinion in that case and this court's decision in Weitz. It observed that evidence of even a trace amount of cocaine in the blood of a defendant charged with vehicular manslaughter while intoxicated "would have some relevance." 525 So.2d at 421. It distinguished the "unquantified" amount of drugs in Weitz from the "unquantifiable" trace amount in McClain, and explained that this court was "correct when it rejected the trial court's conclusion that it was necessary for the toxicologist to estimate the degree of impairment caused by the existence of the drugs." Id. at 423. It explained the application of section 90.403, Florida Statutes, in the two cases:

It may be that McClain and Weitz can be reconciled when the challenged evidence is viewed in light of its relationship to the other evidence. In both cases, it could be said that the prejudicial impact of permitting the jury to hear that the defendant had taken illegal drugs was equal but that it was the difference in probative value which tipped the scales. In Weitz, the defendant's low blood alcohol test belied the other evidence of his intoxication. Thus, the presence of even a small amount of drugs in the defendant's urine was significant because it provided an explanation for his impaired conduct. In the instant case, McClain's blood alcohol level substantially exceeded the figure necessary to raise a presumption of impairment. Therefore, evidence of a trace amount of cocaine in McClain's blood added little to the state's proof of intoxication.

Id.

In State v. Tagner, the trial court's exclusion of evidence of cocaine in the blood of a defendant charged with DUI manslaughter based upon its finding of "no measurable effects" was reversed. The appellate court held that the state's "inability here to show a specific measurable effect of the cocaine on defendant did not per se render the evidence inadmissible," and that the testimony of the toxicologist "established the probative value of the cocaine on the issue of `under the influence.'" 673 So. 2d 57 at 60. The court observed that

the experts here concede that they could not provide a quantitative range of impairment correlating to the cocaine in defendant's blood because this body of information does not exist in a fashion similar to the research on alcohol. If the inability to quantify the degree of impairment was a basis for disallowing evidence of cocaine impairment, such evidence would necessarily be suppressed in every case.

Id. at 59.

[21] See State v. Andres, 552 So. 2d 1151, 1153 (Fla. 3d DCA 1989) ("Relevant evidence is inherently prejudicial; however it is only unfair prejudice, substantially outweighing probative value, which permits exclusion of relevant matters"). See also U.S. v. Baller, 519 F.2d 463 (4th Cir.), cert. denied, 423 U.S. 1019, 96 S. Ct. 456, 46 L. Ed. 2d 391 (1975) ("Unless an exaggerated popular opinion of the accuracy of a particular technique makes its use prejudicial or likely to mislead the jury, it is better to admit relevant scientific evidence in the same manner as other expert testimony and allow its weight to be attacked by cross-examination and refutation"). In Weitz, this court observed: "While it is true that knowledge that a defendant had ingested illegal drugs may prejudice some prospective jurors, it is quite another matter to say that because of such possible bias no juror in a trial for driving under the influence of alcohol or drugs may hear that the defendant had ingested drugs prior to the incident." It noted that the trial judge's beliefs about the relevancy of drugs in the body and the prejudicial effect of that fact on the jury "appear to virtually eliminate the possibility of a conviction for DUI/controlled substances in his court." 500 So.2d at 659, n. 9.

[22] It is our understanding that THC and cTHC are determined in a GC/MS analysis by comparing the readout from the unknown sample with a "standard curve," a graphic representation of the readouts of a series of known concentrations of THC and cTHC (from zero to a concentration somewhat greater than expected in the unknown). Because THC and cTHC have an affinity for glass and plastic, there are likely to be minute losses of the THC/cTHC during the analysis, particularly during the solvent extraction procedure, and these losses must be measured and compensated. As previously discussed, the accuracy of the GC/MS analysis may be insured by methods employing both "internal standards" and "external standards." There appear to be two basic methods of using external standards. In the first method, the GC/MS instrument is calibrated using a series of standard solutions of THC/cTHC in the same solvent in which the extracted sample residues are dissolved (ethanol in this case); included with the unknown sample (and treated in exactly the same manner as the unknown sample) is at least one sample of clean blood and at least one sample of clean blood spiked with a known amount of THC/cTHC that is similar to the amount expected to be found in the unknown sample. In the second method, in addition to the unknown sample, a series of standard solutions are prepared in the same matrix as the unknown sample (in this case blood), are treated in exactly the same manner as the unknown sample, and are used to calibrate the GC/MS instrument just prior to injection of the unknown sample.

Dr. Bateh is apparently familiar with the first method, which is somewhat simpler than the second because it requires extraction of only two spiked blood samples (the manual solvent extraction process is time consuming and requires specialized glassware, so there is a limit on how many samples can be extracted). Dr. Baselt's lab opted for the second method, which requires more analyst time and more glassware, but which has the advantage that every calibration standard (to which the unknown sample will be compared) has been put through the exact same procedure as the unknown sample. In addition to the calibration standards prepared in blood, Dr. Baselt's lab includes deuterated internal standards in all the calibration standard samples and in the unknown sample prior to the extraction process. The lab also monitors random and systematic errors in the GC/MS analysis by participating in proficiency studies (comparing the lab's GC/MS results with the GC/MS results from many other labs, all of them analyzing the same stable sample) and by routinely duplicating the analysis of samples by other labs (a suggested method when stable control materials are not available). Dr. Baselt testified that this system of checks "gives us a good appreciation" of the accuracy of the procedure, and "from what I have seen, we are very accurate." His lab's results tend to read "slightly low" (5-10% low) 90% of the time, and once in a while (1-2% of the time), they get a reading as high as 5% high. His impression is that this accuracy rate is accepted within the scientific community, but he stated that "there are no rules for that carved in stone."