ABSTRACT The objective of this review is to evaluate whether the results of blood drug concentrations could be used by expert witnesses as a basis for scientifically acceptable opinions on driving impairment in adversarial proceedings. Research findings on actual driving performance will be used whenever available.

The adverse effects on driving performance of one drug, alcohol has been well established. Experts can testify to its effects based upon blood and breath alcohol concentrations.

The effects of a few other drugs on actual driving performance have been compared to its likely effects at various blood alcohol concentrations (BACs).

In an actual driving study the impairing effects of the highest smoking dose of marijuana, 3.7% THC, never exceeded those of alcohol’s at BAC of 0.8 mg/mL.. Several studies came to the conclusion that it appears to be impossible to conclude anything about a driver’s impairment based on THC and THC-COOH blood concentrations. A study of chronic heavy marijuana users which included those who drove trucks, buses and taxis concluded that, no real consequence of prolonged use of the drug was uncovered. Amphetamines and cocaine can improve the performance of fatigued drivers.

The driving impairment BAC equivalencies following the therapeutic doses of drugs used in this review were:

Less than 0.5 mg/mL.. BAC; lorazepam, fluoxetine, flunitrazepam, nitrazepam, paroxetine, loratidine, pseudoephedrine, terefenadine, zoplicone, and doxepin (chronic).

0.5 - 1 mg/mL.. BAC: diphenhydramine, triprolidine and clemastine.

More than 1 mg/mL.. BAC: diazepam, barbiturates, flurazepam, loprazolam, mianserin and doxepin (single dose), all depending on dose and the time between administration and blood sampling.

The ranges of drug concentrations resulting from therapeutic doses, the lack of studies of therapeutic and higher doses for most drugs and combinations, and other factors make expert opinions of drug effects on driving performance questionable.

Keywords: forensic science, forensic toxicology, drug testing, driving performance, expert interpretation, drug concentrations.

The widespread availability of drug testing have led legislators and the public to believe that specimens obtained from drivers can be easily tested and that the results of such tests can be correlated with drug impairment. The objective of this review is to evaluate whether blood drug concentrations could be used by expert witnesses as a basis for scientifically acceptable opinions on driving impairment or improvement in adversarial proceedings.

The adverse effects on driving performance of one drug, alcohol has been well established. Experts can testify to its effects based upon blood and breath alcohol concentrations. Assuming that no alcohol is ingested between the time of an incident and the time the blood is collected, they can calculate within an acceptable range the blood concentration at the time of incident. They can estimate the probable amount of alcoholic beverage which should be ingested to produce certain blood alcohol concentrations. All states have "per se" concentrations. There are many different types of analyses for alcohol which are relatively easy to perform and are inexpensive.

Most of what is known about alcohol and driving performance is not available for most drugs. In 1983 a panel of experts reached a consensus concerning drug concentrations and driving impairment which was reaffirmed in 1989 [1,2]. The panel reported: "In order to establish that use of a drug results in impairment of driving skills and to justify a testing program to respond to this hazard, certain facts must be available. (1) The drug can be demonstrated in laboratory studies to produce a dose related impairment of skills associated either with driving or with related psychomotor functions. (2) Concentrations of the drug and/or its metabolites in body fluids can be accurately and quantitatively measured and related to the degree of impairment produced. (3) Such impairment is confirmed by actual highway experience. (4) Simple behavioral tests such as can be done at the roadside by police officers with modest training, can indicate the presence of such impairment to the satisfaction of courts. (5) A range of concentrations of the drug can be incorporated in laws relating to impaired driving as ipso facto evidence. These criteria have been met for ethanol. It is not certain they can be met for most drugs that are now of concern to highway safety."

The effects of marijuana on actual driving performance have been studied in the Netherlands in a project supported by the U.S. Department of Transportation [3]. Performance of subjects, after smoking standardized marijuana cigarettes, who drove in traffic and on highways for 64 km (40 miles) at speeds up to 100 km/h (62 mph) was evaluated. Plasma specimens were analyzed for tetrahydrocannabinol (THC) and its carboxy metabolite (THC-COOH).

Plasma THC concentrations after smoking 100,200,300 mcg/kg were: 3.3-45.9 ng/mL at 40 min.(minutes), 0.3-15.2 ng/mL at 100 min., 0.5-6.8 ng/mL at 160 min. and 0-5.1 ng/mL at 220 min. The THC-COOH concentrations were from 0-96.4 ng/mL. It was concluded that "THC’s effects of SDLP (standard deviation of lateral position) were equivalent to those associated with BACs in the range of 0.3-0.7 mg/mL. Other driving performance measures were not significantly affected by THC." "THC’s effects after smoking doses up to 300 mcg/kg never exceed those of alcohol’s at BAC’s (blood alcohol concentrations) of 0.8 mg/mL.. It appears not possible to conclude anything about a driver’s impairment on the basis of his/her plasma concentration of THC and THC-COOH determined in a single sample.." A common standardized test was used that measures driving impairment from vehicular weaving , SDLP (standard deviation of lateral position).

The effects of chronic marijuana use have been reported in a study funded by the National Institute on Drug Abuse [4]. The subjects were 86 chronic marijuana users and 156 non-users. The users smoked an average of 10 (2.5-40) marijuana cigarettes a day for a minimum of 10 years and an average of 17 years. The cigarettes contained 1.3 to 3.7% THC. The report states: "No hard data were obtained regarding the effect of marijuana use on driving ability. However, some of the user subjects did earn their living by driving trucks, buses or taxis, and some preferred to drive while under the influence of the drug." It was concluded that, no real consequences of prolonged use of the drug was uncovered. This was found to be in keeping with the controlled studies carried out in Jamaica and Greece. Until actual driving studies are performed which report blood concentrations in heavy chronic users, one can only speculate as to what its effects might be.

Another NIDA study reported that: "The performance effects of several drug classes were examined using repeated measures design. Eight volunteers were administered two doses of ethanol (0.3and 1.0 g/kg.), marijuana (1.3% and 3.9% THC), amphetamine (10 and 30 mg), hydromorphone (1 and 3 mg), pentobarbital (150 and 450 mg), or placebo on separate days [5]. The larger dose of each increased subjective drug strength; however only, ethanol and pentobarbital impaired performance on circular lights, digit symbol substitution, and serial math tasks. Both ethanol and pentobarbital impaired performance on card-sorting tasks; impairment was evident at lower doses as the cognitive load increased. Results illustrate differences among drugs in producing performance impairment at doses that cause subjective effects. Increasing cognitive requirements uncovered performance impairment at lower doses." "Marijuana had a significant effect on only one of the fourteen performance measures in the present study. On the Serial Addition/Subtraction task, response time was significantly slowed (46%) by the 3.9% THC marijuana cigarette at the 30-minute time point. Our results differ from those of several studies that have shown performance impairment after smoked marijuana. "

Hypnotic drugs were taken by subjects nights before they were to be tested by the aforementioned actual driving method [6]. "All the mean performance changes which occurred after two nights of drug treatment were significant in morning or afternoon tests, or both, except those following nitrazepam 5 mg. and temazepam 20 mg.. The magnitudes of some changes were relatively small for: lorazepam 1 mg., nitrazepam 10 mg., zoplicone 7.5 mg., and flunitrazepam 2 mg. These were equivalent to the amount of impairment caused by BACs (blood alcohol concentrations) in a range from just under 0.5 mg/mL. to about 0.6 mg/mL.. Slightly greater impairment was produced by flurazepam 15 mg. in the morning test. However a very serious degree of impairment, greater than the equivalent of a BAC of 1 mg/mL., was caused by the residual effects of secobarbital 200 mg., flurazepam 30 mg., and loprazolam 2 mg.."

Another study in which temazepam (15 mg.) and temazepam plus ethanol (breath ethanol concentrations, at three testing times were: 30 min. 8 mmol/L, 90 min. 7 mmol/L, 150 min. 4 mmol/L) were administered concluded [7]: "Previous studies have shown that appropriate use (e.g. a therapeutic dose taken before bed with testing the following morning) does not result in residual impairment. This study showed that temazepam, especially coupled with ethanol, does result in impairment by tracking tasks over three hours, the divided attention test providing a more sensitive measure of these effects. The subjects' perception that their performance was unimpaired after taking both drugs in combination is especially important in the light of the measureable reduction in performance seen with the psychomotor tasks." "The mean plasma temazepam concentrations for the six subjects at 150 minutes was 372 (SD 144) ng/mL."

After a week of taking diazepam 5 mg. 3xd and lorazepam 2 mg. 2xd, driving performance was impaired more than that produced at a BAC of 1 mg/mL..[8].

The driving of subjects using first and second generation antihistamines was evaluated [9]. Single doses of diphenhydramine 50 mg., clemastine 2 mg. and multiple doses of triprolidine 5 and 10 mg. produced changes equivalent to those produced by BAC’s of 0.5 - 1 mg/mL..

Terfenadine [9] a second generation "non-sedating" antihistamine, was taken in single doses up to 180 mg. and multiple doses over 4 days up to 120 mg. 2xd. Single doses and multiple doses of 60 mg. 2xd and 120 mg. 4xd never produced a significant rise in SDLP. On the contrary, there was a tendency for 60 and 120 mg. to produce a slight fall in SDLP, suggesting a mild stimulating activity of the drug. When subjects took doses of 120 mg. 2xd for 4 days, impairment was equivalent of up to that of 0.05% BAC.

Loratidine [9] in single doses of 10 and 20 mg. produced no significant rise in SDLP. When given in 20 mg. doses 4xd for 4 days, the impairment was similar to that of terfenadine.

"Cetirizine’s [9] effects on SDLP is a matter of contention between different groups of investigators. One showed a significant impairing single-dose effect of cetirizine 10 mg., while the other found no effect of that dose on either the first or fourth day of that dose.

"In Europe acrivistine is available alone in 8-mg. doses and combined with pseudoephedrine in two formulations: acrivistine 8 mg., pseudoephedrine 60 mg. (instant release) and acrivistine 12 mg. (slow release)[9]. Only the combination, Semprex D™, acrivistine 8 mg. and pseudoephedrine 60 mg. (instant release) is available in the U.S.A." "Acrivistine 8 mg. had no effect on mean SDLP but the 8 as well as the 16 and 24 mg. combination preparations had a salutary effect on driving performance. The women who were impaired after 8 mg. of acrivistine alone were not affected by the same dose in combination with pseudoephedrine 60 mg.. Moreover, the men who were treated with that combination and drove, after 4 days of treatment, had a significantly lower SDLP than after placebo. It would appear that the pseudoephedrine’s mild stimulating activity physiologically antagonizes acrivistine’s correspondingly mild sedating activity, when present; and the former predominates when the latter is low or absent."

The study included two other drugs, mizolastine and ebastine [9]. Mizolastine taken in single doses of 5, 10, and 20 mg. produced effects less than those of 0.5 mg/mL. BAC and at 40 mg. less than those of 0.8 mg/mL. BAC. Ebastine was taken 4xd. "The effects of the 10 mg. doses were stimulating both days. The 20 mg. doses lowered SDLP on day 1, though not significantly." By day 5, the 30 mg. doses produced impairment less than that of 0.5 mg/mL. BAC.

Fluoxetine 20 mg. was administered to 18 healthy volunteers for 22 evenings [10]. "Mean plasma concentrations and (s.d.) for fluoxetine and norfluoxetine were respectively 34.47 (14.41) and 42.47 (17.47) ng/mL on day 8 and 57.83 (24.88) and 75.78 (28.29) ng/mL on day 22 of treatment." "No significant effects were found on any parameters in either the highway driving or the car-following test."

"Amitriptyline 75 mg./day produced severe drowsiness and strikingly impaired performance on nearly every test on the first day but its effects were practically gone after 1 week of treatment [11]. Paroxetine 20 mg., the usual anti-depressant dose, had no effect on performance. Paroxetine 40 mg. did not affect road tracking but slightly impaired performance in some psychomotor tests in a persistent manner."

"Depression itself and the chronic use of one antidepressant, amitriptyline, are associated with greater than normal risk of traffic accidents [12]. Otherwise, impairments associated with depression generally resolve in those patients showing a favorable response to antidepressant therapy, regardless of the drug."

Mianserin 10 mg. 3xd and doxepin 25 mg. 3xd were administered for 8 days. "On day 1, mianserin and doxepin impaired driving [13]. Impairment dissipated after 8 days of treatment with doxepin but not during treatment with mianserin."

"Cocaine effects on driving performance have been examined in a series of studies performed at SCRI (Southern California Research Institute). Twenty-four healthy male subjects, ages 21-40 years, who were self-reported moderate users of cocaine were used. An initial experiment with cocaine (96 mg., intranasally) and alcohol 0.58 g/Kg b.w., found no impairment of driving-related laboratory tasks by cocaine [14]." "In a second experiment with 96 mg. cocaine, subjects performed better with cocaine than with placebo with greatest difference observed during a test battery beginning three hours after dosing. Since that second test time coincided with the afternoon slump, the findings raised questions about the drugs effects with circadian rhythm [15]. Time-of-day differences associated with cocaine’s effects were further studied in a nighttime experiment."

In the nighttime experiment [16], "Subjects participated in three two-day treatment sessions. Day 1 began between 18.00 h and 19.30 h. Subjects slept overnight and were awakened at 08.00 h to begin day 2." Each treatment of 96 or 126 mg. of cocaine was divided into three equal amounts given intranasally at half hour intervals. Blood specimens obtained 10 min. after each dose had the following concentrations of cocaine/benzoylecgonine; 3/57, 64/214, and 189/363 ng/mL. "D-A (divided-attention) and VIG (vigilance) data agree with previously-reported data [15] in demonstrating that the effects of cocaine on performance persist past the period of acute stimulation. When subjects were tested near midnight, scores were better with cocaine than with placebo. It was only in the placebo condition that overall D-A performance was significantly worse at the late night hour. D-A RTs were faster with 96 mg. cocaine whereas 126 mg. cocaine prevented slowing of VIG RTs (response times). These data suggest that cocaine effects may be task dependent as well as dose dependent."

A study of methamphetamine and driving impairment concluded: "The net conclusion of the material reviewed in this study was that the circumstances under which any methamphetamine induced performance increment is possible are extremely narrow, and is not guaranteed because of typical side effects associated even with low dose use. Furthermore, there is ample evidence from the epidemiological, clinical, case report and toxicological data to conclude that the behavior displayed in the (28) cases we reviewed is consistent with impairment as a result of methamphetamine use, withdrawal, or combined use of methamphetamine and some other drugs [17]. The author's hysteresis plot showed improved reaction time, relief from fatigue, and euphoria with blood methamphetamine concentrations of 0.0l to 0.09 mg/L becoming fatigue and exhaustion on withdrawal with the same blood methamphetamine concentrations.

A study of the effects of methadone, as used in methadone maintenance programs, on performance related to driving has been reported [18]. The mean dose of methadone was 70 mg. (range 15 to 150 mg.) . The test battery was sensitive to the effects of alcohol (mean BAC 0.64 mg/mL.) and diazepam 15 mg. orally. "Both alcohol and diazepam produced a significant decrement in performance on the test battery by the control groups and the stabilized methadone clients. However, there was no evidence for an interaction between methadone and either alcohol or diazepam in the group of methadone clients stabilized on the program.." "The insensitivity of these tests of skill performance to the acute effect of methadone on the clients within the methadone maintenance program indicate that these clients should not be considered as impaired in their ability to perform complex tasks such as driving a motor vehicle."

Patients whose pain was controlled with slow-release morphine sulfate tablets in a daily dose range of 60 to 1100 mg. had the following plasma concentrations: morphine 4.5-337ng/mL., morphine-6-glucuronide 20-1014 ng/mL. and morphine-3-glucuronide 139-4857 ng/mL. [19]. The authors stated that, "In conclusion, long-term analgesic medication with stable doses of morphine does not have psychomotor effects of a kind that would be clearly hazardous in traffic."

In another study, subjects who took oral single 10 and 15 mg. doses of morphine sulfate, "had minimal impairment of cognitive and psychomotor function with possible improvement in one test [20]..

Diabetics and epileptics require such drugs as insulin, diphenylhydantoin and phenobarbital in order to live and to control physical and mental conditions that would make driving hazardous if drugs were not used to normalize their driving performance. "Diabetic hypoglycemia produces cognitive motor slowing [21]. Driving performance was not disrupted at mild hypoglycemia nor after recovery from moderate hypoglycemia. Moderate hypoglycemia disrupted steering, causing more swerving, spinning, time over midline and time off road. It also resulted in an apparent compensatory slowing, with more very slow driving." Mean blood glucose levels: Control, euglycemia 6.3 nM/L., mild hypoglycemia 3.6 nM/L., moderate hypoglycemia 2.6 nM/L.

Epileptics should not drive unless they are seizure-free or their seizures are controlled by anticonvulsants and sedatives. "EEG and driving behaviour were monitored in six patients with subclinical focal and generalized eleptiform EEG discharges during 420 km (260 miles) of actual motorway driving in a suitable instrumented vehicle [22]." "Evidence of impaired driving performance during subclinical discharges was significant in three subjects and was suggestive in one. The two patients with greatest impairment had active epilepsy, whereas the others had been seizure-free for upwards of 4 years. No other features appeared to be predictive of altered driving behaviour during discharges."

Adequate methods are available for the identification and determining the amount of drugs in blood, urine, hair, sweat, saliva and other specimens. The major problem is in relating the drug concentrations in the specimens to driving impairment. Studies have reported the numbers of drivers, injured or dead in crashes, who had drugs in their bodies. It is not known whether the drugs were factors in causing the crashes.

The U.S. Department of Transportation issued regulations that require testing of safety-sensitive employees in transportation industries "for use, in violation of law or Federal Regulation, of alcohol and drugs listed in the Controlled Substances Act." [23] Drivers may use controlled substances, "when the use is pursuant to the instructions of a physician who has advised the driver that the substance does not adversely affect a driver’s ability to safely operate a commercial motor vehicle." Could such a statement be supported scientifically? The stated intent of the Federal workplace drug testing program is to identify individuals who use illegal substances [24]. Urine specimens are tested for amphetamines, phencyclidine and the metabolites of marijuana, opiates, and cocaine.

Some "legal drugs" which are controlled substances have adverse effects on actual or simulated driving and must be obtained by prescription. Some of these are: diazepam, flurazepam, loprazolam, barbiturates, mianserin, and clemastine. Diphenhydramine and triprolidine are available without prescriptions. Tests for the above drugs and many others are rarely performed on impaired drivers. If two or more drugs are found, it is essential that the combined effect be evaluated. Combining an antihistamine with pseudoephedrine can overcome the impairing effect of the antihistamine.

Problems such as fatigue, lack of attention, vigilance deficits, suicidal and aggressive tendencies can cause crashes. Many drugs can create such problems. They can influence vision, vigilance and impulsiveness. Concentrations of drugs and metabolites in body fluids can be determined but the concentrations of most drugs cannot be correlated with impairment or improvement of driving.

Specimens other than blood are useful in determining drug use but none is helpful in determining whether there is an active drug in the body which is affecting driving performance. Interpretation of the effects produced at various concentrations of drugs in blood specimens depends on many factors not generally available to an expert witness for use as a basis for formulating acceptable scientific opinions. Some of the factors are: the impossibility of reliably back calculating concentration to a prior time, individual differences in metabolism, single or chronic dosing, tolerance, withdrawal, inter and intralaboratory methods and variances, multiple drug use, method of use, and the ranges of drug concentrations produced in different individuals ingesting the same size dose.

About twenty-five drugs are reviewed in this paper. There are thousands of drugs available and millions of combinations of these drugs. It is improbable that by any of the methods now available, that the problems of relating drug concentrations to impairment or improvement of driving performance will be solved.

In criminal court, it must be proven beyond a reasonable doubt that a person drove while his or her physical or mental faculties, or both, were appreciably impaired by an impairing substance. In civil courts the standard of proof is by the preponderance of the evidence or that impairment is more likely than not.

Based on the reports in this review, forensic scientists appearing as experts in adversarial proceedings should be able to offer some opinions. Based on the blood concentrations of THC and/or THC-COOH in this review, an expert could not say with scientific certainty that a driver’s impairment would be greater than that of a driver with a 0.5 mg/mL. BAC.

Based on the blood drug concentrations in this paper, knowledgeable experts should be able to rebut opinions of significant impairment by marijuana, cocaine, pseudoephedrine, amphetamines, lorazepam, fluoxetine, terfenadine, paroxetine, loratidine, nitrazepam, zoplicone, flunitrazepam and chronic doxepin. They appear to have little or no impairing effect on driving performance in the concentrations cited in this review. Opinions that higher concentrations results in impairment must be backed by scientifically acceptable evidence.

Expert opinions might be offered that driving impairment is probably greater than that of a driver with a 1 mg/mL. BAC when blood specimens have therapeutic blood concentrations of the following: barbiturates, diazepam, flurazepam, loprazolam, mianserin and doxepin (single dose). Impairment equal to that of a driver with a BAC of 0.5 - 1 mg/mL. was found in drivers with therapeutic blood concentrations of the following: clemastine, diphenhydramine, and triprolidine.

In many countries: only a BAC of < 0.2 or of < 0.5 mg/mL. is acceptable and 1 mg/mL. is considered as an unacceptable impairment level.

Law enforcement officers, including drug recognition evaluators, DRE’s, who try to evaluate the performance of drivers should be aware of the reports in this review and elsewhere, before they offer opinions of the source of various signs and symptoms which might be produced by drugs and other factors, and affect the physical and mental conditions of drivers.

The establishment of "per se" concentrations of drugs is not scientifically sound. A discussion of the problems of such an approach is presented in the Consensus Report [1].

Making it a crime to drive while possessing drugs or finding drugs in specimens obtained from drivers cannot be related scientifically to driving impairment for most drugs.

Whatever expert opinions are offered, they must be supported by scientific documentation, experience and by other evidence. Much more scientific research is needed on the effects of drugs and drug combinations on actual driving performance. Experts must be able to show that an impairing substance appreciably adversely affected the driver’s physical and/or mental faculties. In adversarial proceedings where performance is a factor, the mention of the possible use of drugs is prejudicial and should be excluded as irrelevant, unless it can proven that performance was adversely affected by the use of drugs.

The ranges of drug concentrations resulting from therapeutic doses and the lack of studies of therapeutic and higher doses for most drugs and combinations, make expert opinions of drug effects on driving performance questionable.

5. W. B. Pickworth, M. S. Rohrer, and R. V. Fant. (NIDA) Effects of abused drugs on

psychomotor performance. Exper. Cllin. Psychopharmacol. 1997;5:235-241.

6. J.F. O’Hanlon, Benzodiazepine influence on driving proficiency. Workshop on clinical

differentiation of benzodiazepines; 1985 June 14; pp 31-37; Amsterdam.

17. B.K. Logan, Methamphetamine and driving impairment. J. Forensic Sci. 1996;41:457-464.

Amitriptyline (Elavil) 50mg(S) 16-35*(2-4 hr), 150mg(P) 38-162.

Diazepam (Valium) 10mg(B) 148 (1hr), 37 (24hrs), 30mg/d(P) 700-1500.

Diphenhydramine (Benadryl) 50mg(P) 83(3hr), 49(6hr), 9(24hr)

.

Doxepin (Sinequan) 75mg(P) 24(2hr), 113mg/d(P) 5-115.

Flurazepam (Dalmane) 90mg(B) 13, none after 30mg/d.

Lorazepam (Ativan) 2mg(P) 18(2hr), 9(P)(24hr), 10mg/d(P) 140-240.

Methadone 15mg. 75(4hr) 30(24hr), 100-200mg/d 570-1060(4hr),

280-790(24hr)

Mianserin (Norval) 20mg(P) 26-30(1.7hr), 60mg/d(P) 12-81.

Morphine 20-30mg above 20(4-6hrs).

Nitrazepam (Mogadon) 5mg(S) 35(2hr), 5mg/d(P) 39.

Secobarbital 200mg(B) 2000(3hr), 1300(20hr)

. Temazepam (Restoril) 10mg(P) 205-430(15-90m), 20mg(P) 363-856(15-75m)

.

PDR. 1997 [26]

Acrivistine (Simplex D) 8mg(P)max.393 + pseudoephedrine 60mg(P) max.1308

Cetrizine (Zyrtec) 10mg/d(P) 311±40(1.0±0.5hr)

Paroxetine 30mg/d 61.7

Terfenedine (Seldane) 60mg(P) 263-423(2.5hr)

*ng/mL, Blood(B), Serum(S), Plasma(P), mg/d=mg/day, m=minutes.

DD 10rev w6