Schaffer Online Library of Drug Policy Sign the Resolution for a Federal Commission on Drug Policy


Contents | Feedback | Search | DRCNet Home Page | Join DRCNet

DRCNet Library | Schaffer Library | Miscellaneous Statements on Drug Policy


Dr. John Morgan

KANSAS LAW REVIEW, VOL. 36, 1988, PP. 683-697 (PART II)


Among the fears fueling the drive to test is the specter of industrial accidents and the likelihood of intoxicated workers harming other workers and the public. The lack of data that drug use in industry is harmful or an important cause of accidents does not deter the frequent attribution of accidents to drug use. Nowhere is this discussion more focused than on transportation. It is impossible to raise arguments about testing without provoking commentary on intoxicated airplane pilots and locomotive engineers. Perhaps the strongest impetus to testing occurred on January 4, 1987, when a Conrail engineer, who later tested positive for marijuana, ignored a number of warning devices and pulled his train into the path of an Amtrak Metroliner. (30) Sixteen people were killed. (31)

From February 10, 1986 to January 15, 1987, under the auspices of the Federal Railroad Administration ("FRA"), most American rail carriers conducted an incident-based testing program. Any train accident or incident associated with a fatality, injury, or nontrivial property damage resulted in the testing of all operating personnel--- including the train crews, dispatcher, and policemen. One hundred seventy-five events in the year led to the testing of 759 employees. (32) Of the 759 employees sampled, 43 tested positive for controlled substances including alcohol. (33) Of the forty-three positives, the drugs identified were as follows: Alcohol = 9 (1.2%) Illicit drugs = 29 (3.8%) Other controlled substances = 14 (1.8%). (34)

The FRA made public the specific results only for the alcohol and illicit drug positives, so it was not possible to know which licit drugs (one assumes obtained by prescription) were identified in the fourteen workers. Of the twenty-nine (3.8%)positives for illicit drugs, eighteen had only THC carboxylic acid (marijuana metabolite). five had only benzoylecgonine (cocaine metabolite) and six had both. (35) One individual who tested positive for methamphetamine also had tested positive for marijuana and is included in the eighteen. Despite its listing as illicit, methamphetamine is still available in the United States as a licit appetite depressant.

The FRA document correctly noted that these "data are not conclusive of alcohol/drug role in industrial accidents." (36) Indeed, of the twenty-nine, at least seven had job classifications which would seem to place them at a distance from operating decisions (track patrolman, ticket taker, conductor, road master). (37) This costly program is not cited for what appears to be significant evidence that illicit drugs do not play an obvious role in industrial rail accidents. Furthermore, there is no reason to believe that an investigation of other industrial accidents would yield a different picture. VI. MARIJUANA ISSUES

Because most workplace positives are due to marijuana metabolite and most arguments revolve around the putative dangers of off-the-job marijuana use, the details of cannabinoid testing are important. A review of these details is, however, beyond the scope of this article. (38) However, three points involving marijuana that are commonly used to provoke testingt merit some discussion. First, the fat storage and carry-over effect of marijuana. Second, the airplane simuator testing study. Finally, the growing potency of domestically available marijuana.

A. Fat Storage and Effect

The persistence and storage of THC and other cannainoids after use is subject to much discussion and misunderstanding. Figure 1 depicts the length of time THC remains in the plasma after an individual smokes marijuana containing delta-9-THC (the intoxicating ingredient in marijuana). (39) The pattern of distribution follows, with some simplification, what pharmacologists refer to as a two-compartment model. The drug initially enters the blood stream (compartment I) and then is rapidly distributed to the deep compartment II, which is most of the body. The initial rapidly declining concentration curve represents the disappearance from the blood stream through distribution.

(Unable to show Figure 1)

Three to four hours after the initial ingestion, the concentration of delta-9-THC has fallen below 2-5 ng. ml in the blood. (40) This concentration represents the minimal effective concentration. (41) Below this concentration in the blood, the drug exerts no demonstrable effect. We assume that like other drugs, the concentration has fallen so that the complementary concentration in the brain and other receptor tissue is below the threshold which will produce an effect. However, delta-9-THC remains in the body.

The second phase of the disapearance curve is nearly level because its decline is so gradual. This phase represents to some degree the drug's redistribution from compartment II to compartment I. This slow decline is used to calculate the terminal half-life (T 1/2) of THC. A low concentration, say 1.0 ng ml, will take more than twenty-four hours to fall to .05 ng / ml. (42) Actually, it probably takes even more time, but an accurate calculation is difficult because of the inability to measure extremely low concentrations.

This slow decline is the source of the popularly-held concept that THC persists in the body's fat. From this another speculation usually follows. The speculation is that the THC still is exerting some effect. Current evidence simply says this is not so. There are many speculations about the long-lasting effects of THC, but in the nearly thirty years since delta-9-THC was discovered, nothing has been verified. The brain actually contains little fat and there is no evidence to support the idea that the drug particularly persists there. Frequent users will carry the drug and its metabolites with them--even to the workplace--but they also will carry some amount of organochlorine pesticide, lead, and the hydrocarbon solvent used at work. We might better and more profitably examine the long term effects of these substances.

B. Airplane Simulator Testing In Private Pilots

To justify the intrusion of testing, studies purporting to show a lasting effect of marijuana have been widely cited. Marijuana users and their defenders frequently state that marijuana use-off-the-job is no different than alcohol use off-the -job. The effects of marijuana and the selective impairment generated by its use dissipate within three to four hours and generally correlate with the decline in serum tetrahydrocannabinol (THC) to levels of less than 5 ng / ml. (43) The studies of Herbert Moskowitz and Robert Petersen that indicate impairment of some psychomotor function for up to eight hours (44) are puzzling and not entirely scientifically acceptable. However, assuming they are correct, they still do not indicate that off-the-job drug use causes on-the-job impairment. A study by Dr. Jerome Yesavage is another matter. (45) This study using flight simulators has been widely discussed and previously used as a strategic justification for urine testing. (46) The study identifies significant impairment twenty-four hours after smoking marijuana in a group of volunteer pilots who performed on a flight simulator. (47) These conclusions are striking and have not yet been replicated or confirmed.

After careful analysis of the article, I think that the experimental methodology was so flawed that the study does not prove impairment at twenty-four hours. (48) Briefly, the study was uncontrolled and, therefore, scientifically unreliable. The subjects were studied with no placebos or attempts to conceal the active treatment. Subject efforts at a control run were compared to their efforts after they, and the investigators, knew that they had consumed an active drug. This violates elementary rules of study design and constitutes a serious flaw in the study. The study also fails to deal with the characteristics and prior drug experience of the volunteers, the issues of repeated testing and training upon the simulator, and the generation of a large number of simulator-generated measures some of which may have varied by chance. (49)

C. THC Content (Potency) of Domestically Available Marijuana

A commonly stated and widely accepted belief is that marijuana's past record of causing minimal harm to users was based on the use of low-grade material. Many plants, particularly those growing wild, have a low content of delta-9-THC. (50) It is often stated that the THC content has increased greatly. In fact, critics and proponents of testing sound like dealers extolling the power of Hawaiian and California grown material. "A glance at the percentages of THC in confiscated marijuana would....[indicate] that marijuana is now as strong as hashish in THC content," (51) This particular claim also has been difficult to document. However, NIDA has funded a THC analysis program at the University of Mississippi since 1980. Under the program, the University analyzes plants received chiefly by law enforcement agents, both federal (Drug Enforcement Agency) and a variety of city and state criminal agencies. The accompanying table lists the findings since 1980. (52)

Year # of samples Mean THC% 1980 35 4.64 1981 129 2.92 1982 148 2.57 1983 387 1.98 1984 330 2.55 1985 842 2.21 1986 800 1.86 1987 (Sept 100 2.5

The sampling is irregular but chiefly reflects material seized by these agencies from the back yards, frms, and even indoor hydroponic systems of growers. The most suspect year, 1980, of course, is the only one in which the mean percentage exceeded three percent. There has been no real change and certainly no real increase in marijuana potency in the 1980's.

Interestingly, significant variation in THC content of marijuana does not necessarily result in greater delivery. In a study of the delivery of THC into smoke under standard conditions, 16-19% of the THC content of a cigarette actually appeared in the smoke. (53) Increasing the THC content under some circumstances did not influence delivery. For example, cigarettes carefully prepared to contain 1.6% THC and 3.1% THC delivered the same amount of THC in smoke. (54)


This article has examined a series of rationales for the imposition of urine testing onto unimpaired American workers. I have tried in good faith to examine the arguments from the perspective of empirical studies and appropriate analysis. I have accused, not subtly, the proponents of testing of zealotry and improper use of statistics and data to support a moral stance. To some degree my bias as a civil libertarian has affected my evaluation. In the long run, however, the truth about this conflict will depend upon ideas and data and not biography. Urine testing is a wrong-headed intervention into the lives of some by those with power over them. It is not a search for illness but a search for deviance conducted in an un-American manner. Drug abuse is a specter of the 1980's resembling the specter of domestic communism of the 1950's. Under the guise of helping, the proponents of testing are creating enormous harm. Urine testing is simply drug abuse abuse.


30. Se N.Y. Times, Jan. 5, 1987, at 1, col. 2.

31. Id.

32. Federal Railroad Administration, U.S. Dep't of Transp., Summary of Post-Accident Testing Events Feb. 10, 1986 through Jan. 15, 1987 (summarized in 53 Fed. Reg. 46,641 (summary includes data through Dec. 31 1987).

33. Id.

34. Id.

35. Id.

36. Id.

37. Id.

38. See generally Morgan, "Marijuana Metabolism in the Context of Urine Testing for Cannabinoid Metabolite, 20 J. PSYCHOACTIVE DRUGS 107 (1988). [Editor's Note: See Miike & Hewitt, "Accuracy and Reliability of Urine Drug Tests. 36 KAN. L. REV. 641 (1988).

39. Id. at 109 (reprinted with permission).

40. Augerell, Lindgren, Ohlsson, Gillespie, & Hollister, "Recent Studies on the Pharmacokinetics of Delta-1-Tetrahydrocannabinol in Man," in THE CANNABINOIDS: CHEMICAL, PHARMACOLOGIC, AND THERAPEUTIC ASPECTS 167 (1984).

41. See id. at 179.

42. Cf. E. Johansson, Prolonged Elimination Half-life in Plasma of delta-9-Tetrahydrocannabinol in Chronic Marihuana Users 41 (abstract of presentation given Sept 4, 1987 ( copy on file at the Kansas Law Review).

43. See supra notes 40-42 and accompanying test.

44. H. Moskowitz & R. Petersen, MARIJUANA AND DRIVING -- A REVIEW (1980).

45. Yesavage, Leirer, Denari & Hollister, "Carry-Over Effects of Marijuana Intoxication on Aircraft Pilot Performance: A Preliminary Report," 142 AM. J. PSYCHIATRY 1325 (1985) [hereinafter Yesavage].

46. See, e.g., Declaration of Dr. Marian W. Fischman Associate Professor, Department of Psychiatry and Behavioral Sciences, Johns Hopkins University School of Medicine, at 2, submitted with Department of Justice Memorandum of Law in Support of Motion for Summary Judgment, American Fed'n of Gov't Employees v. Dole, Civ. No. 87-1815 GAG (D.D.C. Sept. 30, 1987); Declaration of J. Michael Walsh, Ph.D., supra note 19, at 6.

47. Yesavage, supra note 45, at 1328.

48. See Morgan, "Carry-Over Effects of Marijuana," 144 AM. J. PSYCHIATRY 259, 259-60 (1987) (letter to the editor); see also Greenblatt, "Marijuana Test Fails in Scientific Design," N.Y. Times, Apr. 30, 1986 at A-30 col. 4 (letter to the editor).

49. Whenever one measures many variables using a five percent level of significance, one in twenty will vary by chance.

50. See Mikurya & Aldrich, "Cannabis 1988 Old Drug, New Dangers, The Potency Question," 20 J. PSYCHOACTIVE DRUGS 47, 52 (1988).


52. Telephone interview with Carol Abel, University of Mississippi Research Institute (Feb. 4, 1988).

53. Davis, McDaniel, Cadwell & Moody, "Some Smoking Characteristics of Marijuana Cigarettes," in THE CANNABINOIDS: CHEMICAL, PHARMACOLOGIC, AND THERAPEUTIC ASPECTS 102 (1984).

54. Id.

Contents | Feedback | Search | DRCNet Home Page | Join DRCNet

DRCNet Library | Schaffer Library | Miscellaneous Statements on Drug Policy