STATE of Tennessee, Appellant,
v.
Dorris W. JOHNSON, Appellee.
Court of Criminal Appeals of Tennessee, at Nashville.
April 30, 1986. Permission to Appeal Denied September 8, 1986.*299 W.J. Michael Cody, Atty. Gen., Kymberly Lynn Anne Hattaway, Asst. Atty. Gen., Thomas H. Shriver, Dist. Atty. Gen., Paul DeWitt, Asst. Dist. Atty. Gen., Nashville, for appellant.
Edward M. Yarbrough, Hollins, Wagster & Yarbrough, Nashville, for appellee.
Permission to Appeal Denied by Supreme Court September 8, 1986.
OPINION
SCOTT, Judge.
The defendant was charged with driving under the influence of an intoxicant. He filed a motion to suppress the results of a breath test which was administered to determine his blood alcohol content. The trial judge granted the motion. The state sought and was granted an interlocutory appeal pursuant to Rule 9, T.R.A.P. The sole issue is whether the trial judge erred in granting the motion to suppress the evidence of the breath test.
The trial judge concluded that the Auto-Intoximeter model AI-1000 breath testing device "does not employ methods or techniques generally accepted by the relevant scientific community." According to the trial court, a blood test is the most accurate way to measure the alcohol content in an individual's blood. He concluded that the state must prove blood alcohol concentration only by means of a blood test.
Due to the carnage caused by drivers who operate vehicles under the influence of an intoxicant, the issue presented in this case is an extremely important one. It is important because it directly affects the methods by which law enforcement officers go about their work. It is equally important because it directly concerns the rights of those charged with driving under the influence of an intoxicant, an offense which has now, belatedly, caught the attention of the three branches of government and the public generally.
An extensive hearing was held and testimony was introduced from officers of the Metropolitan Police Department who operate the breath testing program in the city of Nashville. Dr. John Hopewell Hebb, Jr., the Director of Technical Operations for International Clinical Laboratories, and Dr. Hugho Pribor, the Medical Director for International Clinical Laboratories and a clinical professor of pathology at the Vanderbilt University School of Medicine, also testified for the state. The trial judge admitted as exhibits the expert testimony from the cases of State of Tennessee v. Frank T. Baxter and State of Tennessee v. Mark E. Jerikovec, at which the trial judge and another judge jointly presided. In those cases, the judges rendered divergent opinions concerning the admissibility of breath tests. From the record in those cases the appellant presented the testimony of Dr. Charles William Stratton, a clinical pathologist at the Vanderbilt University Medical Center, and the state presented the testimony of William L. Spitler, the Director of Training and Scientific Development for Intoximeters, Inc. of St. Louis, Missouri, the manufacturer of the Auto-Intoximeter.
*300 Tennessee law, like that of every American jurisdiction, prohibits the driving of motor driven vehicles upon the public roads and highways while "under the influence" of any intoxicant. TCA § 55-10-401(a). The degree to which an individual is "under the influence" of an intoxicant is determined by the amount of the intoxicant in the brain. Direct measurement of the amount of alcohol in the brain is not possible while the individual is alive. That measurement can only be made at an autopsy. Therefore, samples of other biological materials must be used to make that determination. Blood, urine, saliva, and the breath have been utilized for that purpose. McCormick on Evidence, § 205 A., p. 615.
Using these methods to determine whether an individual is intoxicated raises two problems the accuracy of the measurement itself and the extent of the correlation between the concentration of alcohol in the sample and the degree of intoxication. Id. Both of these problems were addressed in this case.
The proof revealed, and the authorities agree, that the most accurate and reliable method of testing the amount of alcohol reaching the brain is to test the blood, for it is the blood which carries the alcohol from the absorption surfaces in the intestines to the brain. Id. However, even that method is not entirely accurate, since there is no known method to entirely eliminate error from the examination of biological substances. The integrity and skill of the examiner, the methodology of the examination, and the accuracy of the reporting system, taken together, determine the accuracy and reliability of the testing.
Even assuming accuracy in the examination, the fundamental problem remains. The problem is moving from an estimated value for the blood alcohol concentration to a correct statement of the degree of intoxication during the crucial period, in this case, at the time of the driving. The substantial variability in tolerance to alcohol, absorption rates, and clearance rates, both among individuals and within the same individual from time to time, complicate the efforts to deduce the true extent of intoxication at the time of the driving. For these reasons, the extrapolations are more perilous than is generally recognized. Id.
Under the Tennessee statute, evidence that ten-hundreths of one percent (.10%) or more by weight of alcohol is in the defendant's blood creates a presumption that the defendant was "under the influence" of an intoxicant, and that "his or her ability to drive was impaired thereby" sufficiently to constitute the offense of driving under the influence of an intoxicant. TCA § 55-10-408(b). By driving a motor vehicle in this state, the driver gives "implied consent" to "a test" to determine the alcoholic or drug content of his blood. TCA § 55-10-406(a)(1).
In order to facilitate testing procedures, breath tests have been devised for use by law enforcement agencies. Breath testing has the advantages of mobility of the testing devices and the less intrusive nature of the test. However, it has been recognized that measurements of alcohol content of the blood by sampling the breath is less reliable than direct testing of the blood. Testing of urine is even more unreliable, and saliva testing is "still more infrequent and problematical." McCormick on Evidence, supra, at n. 10, p. 616.
The testing of breath for alcohol content of the blood is derived from the fact that blood and breath "meet" at the alveoli in the lungs. Oxygen is imparted from the breath to the blood in exchange for wastes which are imparted from the blood to the breath. Since it is impossible to actually go down into the lungs for testing, scientific principles must be applied in order to accomplish the testing. The principle applied to this process is Henry's Law.
This law of physics, observed and first described by William Henry in 1803, recognizes that the concentration of a volatile substance dissolved in a liquid is directly proportional to the vapor pressure of the volatile substance above the liquid.
As applied to blood alcohol testing, it is believed that the concentration of alcohol in the blood is directly proportional to its concentration *301 in the deep lung air of the person being tested. The trick is how to formulate the proper ratio of alcohol found in the breath to the alcohol found in the blood. This has been the focus of disagreement in the scientific community, but the generally accepted ratio is 2100:1. Whatever the concentration of alcohol may be in the blood, the concentration of alcohol in the person's deep lung air is 1/2100th as much. To compute the blood alcohol content, it is necessary to obtain a sample of deep lung air of known volume, determine the amount of alcohol in the breath sample and then mathematically convert that quantity to a percentage of blood alcohol, using the 2100:1 ratio.
The Auto-Intoximeter accomplishes this in the following manner: The person to be tested blows into a mouthpiece attached to the machine to insure that only deep lung air is tested. A solenoid is connected to a diaphragm valve located on a fuel cell assembly. The solenoid activates after the person has expelled a certain amount of air. The diaphragm is triggered and it draws one-half cubic centimeter of air from the last portion of the person's breath into the fuel cell assembly. The fuel cell is a mylar chip covered with a grid of platinum. The breath passes over the active surface of the fuel cell and a chemical reaction occurs between the alcohol and the platinum which converts the alcohol in the breath into acetic acid and frees one electron from each molecule of alcohol. The freed electrons then generate an electrical signal that is proportional to the amount of alcohol present. The signal is amplified and translated by the computer in the machine into a blood alcohol level, using the ratio derived from Henry's Law.
Dr. Stratton, the clinical pathologist, was called by the defense to outline the deficiencies that he saw in the machine. First, he saw that there was no guarantee that the machine was actually testing deep lung air. A failure to sample deep lung air would result in an inaccurate reading. He noted that spitting, burping and regurgitation, conditions which frequently occur among those who have been drinking alcoholic beverages, can produce unreliable results. He challenged the validity of the 2100:1 conversion ratio and noted that the ratio is not consistent from individual to individual, nor even consistent in the same person at different times. He contended that body temperatures and atmospheric pressure were factors affecting the accuracy of the machine and that these were not taken into account by the Auto-Intoximeter. Dr. Stratton concluded that all of the current scientific literature doubts the accuracy of breath analyzers in determining the blood alcohol content of a person.
The state countered this position with the results of a double blind scientific study designed by International Clinical Laboratories, Inc. and conducted by the laboratory and the Metropolitan Police Department. The study compared the breath alcohol tests and blood alcohol tests of 134 persons arrested for driving under the influence of an intoxicant. In 79% of the cases the blood alcohol test revealed a higher alcohol content than was shown by the breath test. In 21% of the cases the breath test was higher than the blood test, and for 2 people the tests revealed identical results. For the 27 individuals for whom the breath test was higher, the mean difference between the breath and blood tests was .012%. As in all biological research, there were "outriders," cases that were far from the median.
All of the experts agreed that in order to be acceptable in the scientific community, the correlation coefficient between the tests must be at least .85 to .90. (A perfect correlation is 1.0.) Dr. Hebb, who directed the study by ICL and the Metropolitan Police Department, found that the correlation coefficient between breath and blood was .904. Dr. Pribor confirmed this correlation coefficient, and both agreed that the Auto-Intoximeter is an excellent, reliable method of measuring the level of alcohol in the blood. They noted that the breath testing machine was actually biased on the low side in the majority of the cases, thus giving the accused the benefit of any doubt.
*302 The experts, while admitting that there is an error factor of ± 5% in breath testing, noted that this same error factor is present in blood testing. For example, a finding of .10% on either test could be as low as .093% or as high as .107%.
Charles Campbell, a Metropolitan Police Officer, who has been administering the breath testing program for eleven years, testified that the Auto-Intoximeter is as good as or better than any of the devices he had previously used. Another officer detailed how the Metropolitan Police Department requires a twenty minute waiting period before the test is administered. The purpose of the waiting is to allow any alcohol in the mouth to evaporate and eliminate the possibility of the reading being too high because of the presence of alcohol in the mouth. Officers are also required to observe the individual during the waiting period to insure that no hiccupping, belching or vomiting occurs to distort the breath test. The officer who administered the test to the defendant testified that there was no foreign matter in his mouth at the time of the test and that he registered .25%.
Mr. Spitler testified at great length concerning the design and functioning of the Auto-Intoximeter. He recounted that it is widely used in a number of states, including Georgia, Tennessee, Florida, Wisconsin, New Mexico, Indiana, New York and Nevada. It has been approved by the U.S. Department of Transportation as a mobile unit and was given a qualification testing by the U.S. Government in 1978. The results were published in the Federal Register. In addition to the states using this device, the U.S. Navy Bureau of Medicine and Surgery has placed this machine in its trauma centers, the system of Veterans Administration Hospitals has over 300 of these units in service and the U.S. Air Force is using the Auto-Intoximeter at some of its bases located in Florida.
Dr. Kurt Dubowski, a recognized authority in the field of the analysis of alcohol content of breath and blood, was quoted extensively during the trial by all of the experts. Dr. Dubowski wrote in one exhibited article:
Breath and blood are acceptable body fluids for alcohol determination. Whenever the patient can deliver a suitable breath sample, breath-alcohol analysis is the preferred procedure for clinical purposes. It is inherently simple, rapid, and noninvasive, and it reflects the alcohol content of the arterial circulation, which is physiologically and clinically more significant than the venous BAC, especially during alcohol absorption, since the blood in alcohol-equilibrium with the breath has essentially the same alcohol concentration as that reaching the brain. Alcohol Analysis: Clinical Laboratory Aspects. Part I, Laboratory Management, March 1982, 43 at 45.
Dr. Stratton testified that the 2100:1 conversion ratio is dubious at best and is no longer considered an appropriate ratio. Yet, in another article filed as an exhibit, Dr. Dubowski and Dr. M.F. Mason recognized that the Committee on Tests for Intoxication of the National Safety Council has determined that the ratio has been determined by chemical tests and has been the value almost invariably used for many years. They further noted:
These various studies had raised questions regarding the validity of the 2100:1 ratio, but the ratio was reaffirmed after a reconsideration (but not experimental verification) by a group with international representation in 1972. Mason and Dubowski, Breath-Alcohol Analysis: Uses, Methods, and Some Forensic Problems-Review and Opinion, 21 J. Forensic Sci. 9, 23 (1976).
TCA § 55-10-407 provides as follows:
Upon the trial of any person charged with a violation of this chapter, the results of any test made of the person so charged shall be admissible in evidence in criminal proceeding. Failure of a law enforcement officer to request the administering of a test shall likewise be admissible in evidence in a criminal proceeding.
*303 The term "test" as used in this section of our code is defined in TCA § 55-10-405(5) as follows:
Any chemical test designed to determine the alcoholic or drug content of the blood. The specimen to be used for such test shall include blood, urine or breath;
Thus, the legislature has seemingly expressed its desire that the results of any test shall be admissible in evidence in prosecutions for driving under the influence of an intoxicant. However, our Supreme Court has held that test results are admissible only when the tests are shown to have been administered by qualified personnel and when it is shown that the testing device is scientifically acceptable and accurate for the purpose that it is being employed. Crawley v. State, 219 Tenn. 707, 413 S.W.2d 370, 373 (Tenn. 1967), citing Fortune v. State, 197 Tenn. 691, 277 S.W.2d 381, 384 (1955), and Pruitt v. State, 216 Tenn. 686, 393 S.W.2d 747 (1965).
As heretofore set forth, this appeal relates only to the acceptability and accuracy of the testing device, not the qualifications of the personnel.
For many years, the first prerequisite for admission of scientific evidence has been acceptability within the scientific community, the so-called Frye test. The thing from which scientific deduction is made must be "sufficiently established to have gained general acceptance in the particular field in which it belongs." Frye v. United States, 293 F. 1013, 1014 (D.C.App. 1923).
However, the modern trend, expressed in the Federal Rules of Evidence, is away from the Frye test. Rule 702, Fed.R.Evid. specifies that scientific, technical, or other specialized knowledge is admissible if it will "assist the trier of fact to understand the evidence or to determine a fact in issue" and requires that an expert witness be qualified by knowledge, skill, experience, training or education in order to give opinion testimony. There is no requirement in the rule that the scientific principles be generally accepted in the scientific community. The only requirement is that the data or facts be "of a type reasonably relied upon by experts in the particular field." Rule 703, Fed.R.Evid.
One federal appellate court has held that the Frye test is at odds with the spirit, if not the exact language, of Rule 702, Fed.R. Evid. That court suggests an inquiry into three factors prior to admission of "novel scientific evidence." They are (1) the soundness and reliability of the process or technique used in generating the evidence; (2) the possibility that admitting the evidence would overwhelm, confuse or mislead the jury; and (3) the connection between the proffered scientific research or test result and the particular factual issues in the case. U.S. v. Downing, 753 F.2d 1224, 1237 (3d Cir.1985).
One state court, applying its rules which track the federal rules, has rejected the Frye test as a special condition for admission under the state rule which parallels Rule 702, Fed.R.Evid., since Frye is not mentioned in either the rule or the advisory commission notes. State v. Williams, 388 A.2d 500, 503-504 (Me. 1978).
Saltzburg and Redden suggest that courts are now going in three separate directions regarding the Frye test. Some adhere to it, others tailor Frye to the facts of a particular case, while still others are changing to a pure relevance analysis. Saltzburg and Redden, Federal Rules of Evidence Manual, Rule 702, p. 633 (4th ed. 1986). It has been suggested that Frye was accepted and cited in the early cases as an easy way to deal with a ticklish problem. Giannelli, The Admissibility of Novel Scientific Evidence: Frye v. U.S. a Half Century Later, 80 Colum.L.Rev. 1197 (1980).
Our Supreme Court has employed the language, "scientifically acceptable and accurate for the purpose for which it is used." Pruitt v. State, supra, followed in Crawley v. State, supra. Earlier, the Court simply stated that the testing device must be "a proper one and in good and accurate order at the time it was used." Fortune v. State, supra.
*304 It is obvious from the wide use being made of the Auto-Intoximeter by both state and federal governments for both law enforcement and medical purposes that it is accepted in the scientific community. Over twenty years ago, in Pruitt v. State, supra, our Supreme Court addressed the admissibility of breath alcohol tests derived from the use of the Breathalyzer, an earlier testing device, and found the device to be "a reliable device for measuring intoxication." 393 S.W.2d at 750. In that case our Supreme Court also noted the use of the 2100:1 conversion ratio. Id. See also: Gray, Attorney's Textbook of Medicine, § 133.73(1) (3d ed. 1985), earlier edition cited in Pruitt v. State, Id.
The other requirement for admissibility set forth in our Supreme Court's case is a showing that the device is "accurate for the purpose for which it is used."
While Dr. Stratton questioned the accuracy of Auto-Intoximeters, his opinion was based entirely on a survey of the scientific literature. His only familiarity with breath testing instruments was five hours of training he received in 1971 or 1972, when he and other residents drank alcohol and were tested on breath machines, undoubtedly a pleasant way to learn about their accuracy. That training was not on the Auto-Intoximeter. Dr. Stratton worried that test results may be skewed by the individual belching or vomiting. However, other proof revealed that it is physiologically impossible to belch or regurgitate while exhaling air.
On the other hand, the other witnesses had all had hands-on experience with this particular testing device. Each testified that the Auto-Intoximeter is a reliable and accurate machine.
Although the reliability of breath testing devices is not universally accepted, the majority of courts have admitted such evidence. McCormick on Evidence, supra, n. 17, at 617. Examples of courts that have accepted such evidence include our own Supreme Court in Pruitt v. State, supra, which is, of course, controlling on this Court. In addition, this method of testing has been approved by the courts of other states, e.g., State v. Helmer, 278 N.W.2d 808, 813 (S.D. 1979); State v. Buch, 595 S.W.2d 386, 389 (Mo. App. 1980); People v. Kozar, 54 Mich. App. 503, 221 N.W.2d 170, 173 (1974).
In fact, in People v. Gower, 42 N.Y.2d 117, 397 N.Y.S.2d 368, 366 N.E.2d 69, 71 (1977), the New York Court of Appeals even overstated the acceptance of breath testing as follows:
Breathalyzer equipment and procedures have become familiar and their use is now commonplace. Reliability has been demonstrated and the results of such testing where properly performed are universally accepted.
Their acceptance is not "universal" and there is, as shown from the proof in this case, a lack of unanimity in the medical profession as to whether intoxication can be determined by measuring the breath. However, this objection goes to the weight of the testimony and not its admissibility. People v. Bobczyk, 343 Ill. App. 504, 99 N.E.2d 567, 570 (1951).
In Stewart v. State, 165 Ga. App. 62, 299 S.E.2d 134, 136 (1983), the Georgia Court of Appeals considered the reliability of the Auto-Intoximeter, the machine here under consideration, and admitted the results of tests conducted on that machine.
The record further revealed that the Metropolitan Police Department carefully trains the personnel who operate the Auto-Intoximeter. They are taught the proper calibration of the machine, its maintenance and safeguard procedures, all of which are designed to insure accurate test results. The scientific study designed by the laboratory and conducted by the police department and laboratory clearly demonstrated the accuracy, dependability and reliability of this particular machine.
The findings of the trial judge at a hearing on a motion to suppress are binding upon this Court if there is any evidence to support that determination. State v. Chandler, 547 S.W.2d 918, 923 (Tenn. 1977). Stated differently, the findings are conclusive on appeal unless the *305 appellate court finds that the evidence preponderates against the trial judge's findings. Braziel v. State, 529 S.W.2d 501, 506 (Tenn.Cr.App. 1975).
It is clear that the finding that the breath test is totally unreliable and must be excluded in every situation is not supported by the evidence. The evidence clearly preponderates against that finding. Although Dr. Stratton expressed his concern about the reliability of the testing, there was also testimony that there is no infallible test of a biological substance. There is always a margin of error when any substance, be it blood, breath, urine, semen or whatever, is tested. Furthermore, even though Dr. Stratton's credentials are impeccable, his familiarity with breath testing was based entirely on his reading of the literature and an experiment in which he participated years ago. The evidence simply does not support the trial judge's conclusion that these machines are unreliable.
The connection between the test results and the factual issues in DUI cases is obvious. The soundness and reliability of the process and technique of generating the evidence by use of the Auto-Intoximeter has been demonstrated. Furthermore, it cannot be said that the evidence would overwhelm, confuse or mislead the jury. Thus, the test devised in U.S. v. Downing, supra, is also satisfied.
It is commendable that the trial judge was greatly concerned, as we are concerned, that an innocent citizen may be wrongfully convicted because the testing was inaccurate. However, it is a rare case in which the only evidence introduced to establish that a DUI offender was under the influence is either a breath or blood test. In most cases there is the testimony of the arresting officer and others at the scene concerning the arrestee's driving pattern and their observations of the individual, including his speech, breath and eyes. Often a field sobriety test has been administered. Scientific tests are corroborative evidence and may serve to exonerate an accused in a close case. Indeed, the purpose of all the testing is to provide objective scientific data to eliminate guesswork and speculation and to supplement the fallible observations of humans. Peterson v. State, 261 N.W.2d 405, 409 (S.D. 1977).
The breath test result merely creates a rebuttable presumption of intoxication. TCA § 55-10-408(b). The state must establish the competency of the operator, the proper operation of the machine and that the testing procedures were properly followed. The defense is then free to rebut the state's evidence by calling witnesses to challenge the accuracy of the particular machine, the qualifications of the operator, and the degree to which established testing procedures were followed. The defendant's greatest challenge to the accuracy of the breath test is the independent blood test to which each accused has a statutory right. TCA § 55-10-410(e). This provides sufficient protection for his due process rights. State v. Helmer, supra, at 812.
In short, any inaccuracies in particular test results go to the weight of the evidence, not its admissibility. The weight to be given any evidence is a question of fact for the trier of fact in each case.
The evidence does not support the trial judge's blanket exclusion of all evidence derived from the Auto-Intoximeter. The order suppressing the evidence is reversed and the cause is remanded for trial on the merits.
DUNCAN and DAUGHTREY, JJ., concur.