ML20245E818
| ML20245E818 | |
| Person / Time | |
|---|---|
| Issue date: | 06/30/1989 |
| From: | NRC |
| To: | |
| Shared Package | |
| ML20245E824 | List: |
| References | |
| FRN-54FR24468 AC81-2-043, AC81-2-43, NUDOCS 8908110392 | |
| Download: ML20245E818 (13) | |
Text
.- >s
)
l A DISCUSSION OF APPROPRIATE CUT-OFF LEVELS FOR AMPHETAMINES AND MARIJUANA IN WORKPLACE DRUG TESTING PROGRAMS U.S. NUCLEAR REGULATORY COMMISSION JUNE 1989 Developed with the assistance of the Battelle Human Affairs Research Centers B90B110392 890725 T*
PDR COMMS NRcc CORRESPONDENCE PDR Enclosure 3
_-_--_--___=a
i %
i
1.0 INTRODUCTION
In response to the NRC's proposed fitness-for-duty rule (53 FR 36795,1988), a .!
number of public commentors offered compelling arguments for using cutoff levels for mari proposed rule marijuana (juana and amphetaminesng/ml metabolites--100 lowerforthan those initial stated screen and in the NRC 15 ng/ml for confirmatory test; and amphetamines--1,000 ng/ml for initial ,
screen and 500 ng/ml for confirmatory test). Evidence indicates that lowering l these cutoff levels will enhance detection and deterrence of drug use.
- rther, current testing technology is capable of supporting these lower .
levels. !
Where technically possible, it would appear preferable to set the lowest possible cutoff levels to detect any of the drugs in question in order to !
achieve the objectives of a drug-free workplace. Because the amount of drug l or drug metabolites in the urine tends to decrease over time, lower' cutoff levels increase the probability of detecting the drug or drug metabolites over a greater period of time.
Commentors on the NRC proposed fitness-for-duty rule offered evidence that the lower cutoff levels can be expected to promote detection of drug use. Many nuclear power plant licensees have established programs that use lower cutoff levels, and they believe that industry experience shows that lower cutoff levels are effective in achieving the goal of a drug-free workplace (about 70 percent of licensees use cutoff levels for marijuana of 20 ng/ml or 50 ng/ml).
Further, the 20 ng/ml cutoff level for marijuana is used extensively in other safety-sensitive industries. Several commentors felt that the lower cutoff levels were especially important in detecting and deterring the casual drug user, such as the " weekend pot smoker." Two utilities that have well-established testing programs for marijuana reported that in 80 percent of the cases where specimens test positive for marijuana, the marijuana metabolite concentration is between 20 ng/ml snd 100 ng/ml. One of these utilities has been conducting tests for six years and has conducted 1,300 (pre-employment) tests for marijuana with 650 positive test results. This finding indicates that, at a 100 ng/ml initial test cutoff level, a large portion of marijuana !
users would not be detected. I 1.1 Tech d al Basis for Lowering Marijuana Cutoff Levels j
The scientifw literature supports industry experience. Recent use of marijuana (i.e. . within six hours after ingestion) may frequently result in urinary metabolita c.cacentrations that are above 20 ng/m1 but below 100 ng/ml (McBurney, Bobbie & Sepp, 1986; DeLaurentis, McNeil, Mann, Clark & Greenwood, !
1982; Baselt, 1984). l l
Because marijuana metabolites are lipid-soluble, they are generally excreted ,
into the urine over a greater period of time than are water-soluble drugs. l Thus, relatively low levels of marijuana metabolites may be excreted over a fairly long period of time. Consequently, although a 100 ng/ml cutoff level may detect very recent use of marijuana, a lower cutoff level would greatly increase the probability of detecting use of marijuana that occurred some time 1 1
f , ,
1 l
before the test. While marijuana metabolites in the urine may be present at levels between 20 ng/mi and 100 ng/ml for relatively long intervals, the levels rarely exceed 100 ng/ml (Schwartz & Hawks, 1985).
1 In a study conducted by Mul6, Lomax, and Gross (1988), the urine of subjects was tested subsequent to the subjects smoking one and two standard NIDA marijuana cigarettes (27 mg delta-9-tetrahydrocannabinol / cigarette). Three of the subjects were occasional marijuana smokers (smoked cr e per week) and five of the subjects were moderate smokers (smoked one to three times per ;
week). The marijuana metabolite levels of the subjects' urine specimens were measured quantitatively with the Amersham Cannabis RIA and non-quantitatively with the Abuscreenr using a 100 ng/mi cutoff level. Tests were performed on samples collected from two hours to five days after smoking. In all cases, the Amersham Cannabis RIA indicated that the subjects' urine had metabolite concentrations below 100 ng/ml. Of the 14 samples collected within two to four hours after smoking, 4 tested positive with tha Abuscreenr and 10 tested negative (with a 100 ng/ml cutoff level). In all tests of specimens l collected from one to five days after smoking, the samples tested negative l with the Abuscreenr . The authors concluded that the " occasional and moderate !
marijuana users, the vast majority of the marijuana user population, have apparently less than 100 ng/ml cannabinoids in urine after smoking and, I therefore, most would be undetected as users at a 100-ng/ml threshold" (p. 115).
1 The testing technology is capable of supporting cutoff levels for marijuana much lower than those currently used. Immunoassay techniques are available that can provide sensitive, reliable, and practical results at 50 ng/ml when J confirmed with GC/MS (Irving, et al., 1984; Black, Goldberger, Isenschmid, !
White & Caplan, 1984; Vereby, Mule, Alrazi & Lehrer, 1986; Cook, 1986; Sutheimer, Yarborough, Hepler & Sunshine, 1985).
Some studies indicate that immunoassay testing technology currently in use can support a 20 ng/ml cutoff levei, while other studies indicate that reliability of the test results may be compromised if this cutoff level is used.
Sutheimer, et al. (1985) report a study where the EMIT assay was used with a thin-layer chromatography confirmation with each test using identical cutoff levels. The false positive rate of the EMIT d.a.u., with a 20 ng/ml cutoff I level, was found to be only 4.6 percent (11 negative confirmation test results )
out of 242 presumptive positives). Frederick, Green and Fowler (1985) fourd I the EMIT d.a.u. to have a higher false positive rate. Three (7.1 percent) of i the 42 samples identified positive with the EMIT assay were negative when j tested with GC/MS with a 4 ng/ml cutoff level. Provided safeguards such as ;
those required in the NRC fitness-for-duty rule are taken, these levels of 1 false positive results are acceptable. These safeguards include using the ,
highly reliable GC/MS procedure for confirming pre:,umptive positive test i results, taking no actions against workers on the basis of screening test '
results, and ensuring that screening test results remain confidential.
However, false negative results may be unacceptably high when a 20 ng/ml cutoff level is used with the EMIT d.a.u. system. Schwartz and Hawks (1983) report that the system is capable of detecting concentrations of marijuana metabolites of 50 ng/ml with a 95 percent probability, but that metabolite !
concentrations of 20 ng/ml will only be detected 50 percent of the time. I 1
2 ii 1
l t
, t 1
l l !
Frederick, et al. (1985) found that the false negative rate of the EMIT l d.a.u., using the 20 ng/ml cutoff level, was 11 percent (a total of 46 )
presumptive negative test results).
Concern has been expressed regarding a lower initial screening cutoff level for marijuana at 20 ng/ml and the possibility of confirmed positives due to passive inhalation. Passive or involuntary inhalation of marijuana smoke produces levels in excess of 20 ng/ml only '. infrequently, and these levels occur only within an average time span of four to five hours after the person has been subjected to heavy exposure to marijuana smoke (Cone & Johnson, ,
1986; Cone, et al., 1987). l Further, in real-world situations, metabolite levels are apt to be much lower ,
than the 20 ng/ml observed when subjects " passively inhale" marijuana smoke t under the relatively extreme conditions often used in controlled experiments.
For example, Muls, et al. (1988) conducted a passive inhalation experiment designed to duplicate realistic, but still relatively extreme, exposure to marijuana smoke. Three subjects, who did not smoke marijuana, were confined for one hour in a 10'by 10'by 8' sealed room where four marijuana cigarettes 1 (27 mg delta-9-tetrahydrocannabinol / cigarette) were burned simultaneously. j The resulting concentration of delta-9-tetrahydrocannabinol in the air was !
quite high (5 micro mg/1). The subjects' urine was tested 24 hours2.777778e-4 days <br />0.00667 hours <br />3.968254e-5 weeks <br />9.132e-6 months <br /> after they !
had been exposed to the smoke, and all subjects had less than 6 ng j cannabinoids per ml of urine. ]
4 The cutoff level used in the confirmatory test for marijuana should be lower I than the cutoff level used in the screening test. Hawks (1986) explains that i this is a common concern when performing assays for many drugs, and he cites '
marijuana essays as a case where this is especially impnrtant:
. . . the confirmation cu tof f is generally set at a much lower level than the screening cutoff, because the immunoassay reacts additively !
with several metabolites from THC and the more specific confirmation methods are directed at only one. (p.37)
Vereby, et al. (1985) expla'.n that this phenomenon can account for data where l the marijuana metabolite concentrations measured in urine specimens by GC/MS are only half the concentration measured in the same specimens with a less specific immunoassey test.
These findings suggest that, in order to minimize unconfirmed positive marijuana test results, the confirmation cutoff level should be substantially lower than the screening cutoff level. The data reviewed by Vereby, et al.
(1985) indicate that, at a minimum, the confirmation cutoff level should be one-half the screening cutoff level. Many experts recommend an even wider margin between the cutoff levels. For example, the HHS final guidelines for federal workplace drug testing (53 FR 11970, 1938) stipulate a 100 ng/ml cutoff level for marijuana screening tests and a 15 ng/ml cutoff level for confirmation tests. This confirmation cutoff level is less than one-sixth the screening cutoff level. It is noteworthy that HHS lowered the confirmation cutoff level for marijuana to 15 ng/mi from the 20 ng/ml level stipulated in the notice of proposed guidelines published in 1987 (52 FR 3068, 1987). When 3
v s explaining the reason for this change, HHS stated that the change was made in accordance with Department of Defense experience. It appears that the .
confirmation test cutoff level for marijuana should be substantially below the !
screening level, and, at a minimum, it should be half the screening test cutoff level.
Current GC/MS testing technology can test wine for delta tetrahydrocannabinol at levels sufficiently low to support a 20 ng/ml j screening test cutoff level. I Theoretically,(GC/MS metabolites at extremely low levels. Joern can detect marijuana 1987) describ l
for detection of marijuana metabolites in urine that Las a sensitivity of l 1.8 ng/ml. Baker, Harry, Russell, and Myers (1984) describe a GC/MS urine test that can detect delta-9-tetrahydrocannabinol at levels as low as.1 ng/ml.
It should be noted that the procedure described by these researchers is relatively simple so that it is practical for conducting large-scale confirmation testing.
Although GC/MS tests are sensitive at these levels, higher cutoff levels may En an experiment described by Baker, be necessary)to et al. ensure (1984 , blank urine reliability.
specimens were tested as controls. The average marijuana raetabolite concentration measured in these clean specimens was 1.1 ng/m1, and the highest observed concent mtion was 3.7 ng/ml. Presumably, the speciraens became contaminated. If such contamination occurs in an experiment, it seems possible that similar errors could occur in a large-scale testing operation. Thus, the reliability td GC/MS confirmation at very low levels may be questionable. Baker, et al. (1984) stata that the detection limit of 10 ng/ml, however, can be reported with confidence when using their procedure. l The positive results obtained from blank specimens in the experiments of Baker, et al. (1984) should not be attributed to sloppy procedures or considered a problem that can be eliminated simply by executing reasonable care. Kogan, Razi, Pierson, and Willsor (1985) cite the erroneous results obtained by Baker, et al. (1984) and note that GC/MS has been known to detect 1 low levels of cannabinoids in blank control blood specimens as well. For thesereasons,Kogan,etal.(1985)concurwiththe10ng/mlcutofflevel recommended by Baker, et al. (1984) when performing GC/MS analysis of marijuana metabolites.
l l The 10 ng/mi confirmation cutoff is sufficiently high that GC/MS positive i results can be reported with confidence. As this level is only one-half the l
proposed screening level of 20 ng/ml, it is anticipated that some unconfirmed positives will occur if the 20 ng/ml screening cutoff level and 10 ng/ml confirmation cutoff level are used. Occasional unconfirmed positives may also
, occur with the 50 ng/ml screening test cutoff level; however, as the ratio of these cutoff levels is 5:1, it is expected that this will occur only infrequently.
It should be noted that the recommendation of a 10 ng/ml confirmation test cutoff level for marijuana is based upon the reliability of GC/MS procedures, not the sensitivity of the procedures. As previously stated, GC/MS procedures are sensitive to marijuana metabolites at levels as low as 1 ng/ml and 1.8 4
1
, s j
ng/ml. The concern is that the GC/MS confirmation tests may not reliably I yield negative test results when analyzing metabolite-free specimens at these levels. Should testing facilities be able to demonstrate that they can l perform GC/MS testing for marijuana metabolites at relatively low levels l (e.g., 3 ng/ml) without sacrificing testing reliability, then such levels could be used. If the reliability of GC/MS testing at such levels can be i firmly established, then a confirmation cutoff level in the 3 ng/ml or 4 ng/ml range is recommended along with a 20 ng/ml screening test cutoff level.
1.2 Technical Basis for Lowering Amphetamine Cutoff Levels :
The current drug screening technology will allow for amphetamine (amphetamines and methamphetamine) testing with a 300 ng/ml initial test cutoff level and a GC/MS confirmatory test of 300 ng/ml. Reducing cutoff limits to these levels may increase the likelihood that those who abuse amphetamines are detected.
The concentration of an amphetamine that may occur in urine subsequent to amphetamine use varies considerably. In addition to the total amphetamine dose ingested, the rate of ingestion and urinary pH influence the concentration of amphetamine that will be excreted in the urine. Kaistha (1977) reports that about 30 to 40 percent of the amphetamine an individual l ingests is usually excreted unchanged in urine within 48 hours5.555556e-4 days <br />0.0133 hours <br />7.936508e-5 weeks <br />1.8264e-5 months <br />. However, I after a single larce dose, the amphetamine is slowly excreted over a five- to seven-day period. Thus, following large doses of amphetamine, the drug may be present in the urine for a relatively long period of time at relatively low levels. In urine with a high pH, amphetamine concentration may be relatively low. According to Foltz, et al. (1980), the amount of unchanged amphetamine that is excreted in urine varies from 10 to 60 percent of the ingested amphetamine, depending upon the pH of the urine. Kaistha (1977) reports that the amount of unchanged drug excreted can be as low as 2.9 percent in urine with a pH of 8. These factors suggest that the urinary concentration of 1 amphetamine may frequently be well below the 1,000 ng/ml cutoff level in the HHS Guidelines, especially following a single large dose of the drug and when drug users have acidic urine. Thus, a lower cutoff level should be used to promote detection.
Current advances in the state of urinalysis technology now permit lower ;
cutoff levels to be established witt.out sacrificing the reliability of test ~
results. Roche Diagnostic Systems advertises that the Abuscreenr system uses a 1,000 ng/ml cutoff level as the standard for amphetamine testing, but i states that their assays are capable of detecting amphetamines at lower leveis (Roche Diagnostic Systems, 1985). Cutoff levels lower than the 1,000 ng/ml standard provided can be prepared with the reagents provided in the Roche ;
assay kits. The EMIT assay uses a 300 ng/ml cutoff level (Hanson,1986). -
Highly reliable confirmation tests using GC/MS at this cutoff level are j feasible. Foltz, et al. (1980) describe a rapid GC procedure for testing j urine for amphetamine that incorporates diethylamine as the internal standard and has a detection limit of 100 ng/ml. Another chromatographic technique described by these authors could detect amphetamines at levels as low as l
5 i
0.1 ng/m1, although reliability of this method suffered when concentrations I were lower than 2 ng/m.l. j Concern has been expressed regarding the cross-reactivity of over-the-counter medications and licit amphetamine use at lower initial ard confirmatory cutoff levels than those proposed. Confirmatory testing with GC/MS and the !
review of all test results by the Medical Review Officer should eliminate !
such problems. l 1.3 Lower cut-off levels and deterrence I
As is discussed above, decreasing the cutoff levels for marijuana and '
amphetarnines can be expected to significantly increase the probability of detecting marijuana and amphetamine use. Further, lowering the cutoff levels l is likely to significantly increase the deterrent effect of drug testing ;
programs. In fact, it appears that increasing the probability that drug users i will be detected, by lowering the cutoff levels, may have a stronger l deterrent effect than would increasing the severity of the sanctions imposed '
upon those who are detected through the drug testing process.
Ross, Campbell, and Glass (1970) and Beshai (1984) found that the enactment of legislation increasing the probability of detection and punishment for driving while intoxicated appeared to have a considerable deterrent effect.
These findings suggest that workers' unacceptable use of drugs is a behavior that can be deterred, and that the probability of detecting workers' unacceptable drug use is directly related to the degree to which the drug use is deterred. Because lowering cutoff levels would enhance detection, these findings suggest that lowering cutoff levels can be expected to enhance ,
detarrence as well.
Research on deterrence in a context broader than just the deterrence of drug use provides further support for the deterrent value of lower cutoff levels.
Literature on criminal benavior irdicates that both increases in the i probability (or certainty) of punishment and increases in the severity of l punishment are associated with lower crime rates (Gibbs, 1966; Tittle, 1969). l The relative importance of each of these factors varies, depending upon j whether the behavior in question is deterred by formal sanctions (such as fines, imprisonment, or discharge from one's employment) or by informal ;
sanctions (such as disapproval by society or one's peers). Tittle (1980) found that increasing tie probability of punishment had a greater deterrent i effect than did increasing the severity of punishment when formal sanctions ,
were involved, whereas the reverse was true when infomal sanctions were !
involved. Tittle (1969) and Logan (1971) found the probability (certainty) j of punishment is associated with lower crime rates for most offenses. On the other hand, when probability of punishment is held constant, the severity of punishment was found to deter only one type of criminal behavior--homicide (Tittle,1969). These findings suggest that increasing the probability of :
detecting workers' use of drugs, by lowering cutoff levels, may be expected !
to have a greater deterrent value than would increasing the severity of j l sanctions taken in the event of a confirmed positive test result. l l
6 i
I
c ,
It has been argued that individuals prone to commit crimes are also individuals prone to take risks, and that this fact explains, at least in part, why certainty of detection is more strongly correlated with crime rates than is severity of punishmer,t (Becker,1974; Smigel,1965; Ehrlich,1967).
Criminal behavior certainly involves risks, and it seems reasonable to assume .
that people who commit crimes are, therefore, prone to taking risks. A study {
by Claster (1967) corroborates the assumption that persons who are prone to 1 criminal behavic: tend to be risk takers. Claster compared the beliefs of a sample of incarcerated delinquents with a sample of non-delinquents and found that, in spite of the fact that there was no difference in the perceptions of each group about the general probability of arrest and conviction for various offenses, delinquents had lower estimates of their own personal 3 probabilities of arrest and conviction. Delinquents appear to think they can 3 eat the odds. 3 l'
It also appears reasonable that risk takers would be more strongly deterred by increases in the probability of detection than they would by increases in the consequences that follow detection. For individuals adverse to risk taking, any risk is too great. Risk takers, however, believe they can beat the odds and will only be deterred by a relatively high probability of detection.
This discussion suggests that increasing the probability of detection would have a very strong value in increasing deterrence in a workplace drug testing program. Drug abuse and misuse entails many risks: risks to health and )
i personal safety; in the case of illegal drug use, risk of arrest; and, j especially when drug use continues when a workplace drug testing program is 1 in place, risk of losing one's job or suffering other employment sanctions.
Thus, most individuals who continue forbidden drug use after a workplace drug testing program has been implemented belong to one the two groups:
(1) individuals who have a drug habit that is so pronounced that they essentially cannot be deterred, or (2) individuals who are willing to take substantial risks. A high probability of detection is desirable for both types of individuals. 1 It should be noted that aeople are not directly motivated by empirical evaluations of reality, aut by their personal perceptions of reality. That is, potential drug users will not be deterred by the actual probability of detection and consequences of detection; they will be deterred by what they perceive to be the probability of detection and by what they perceive to be ,
the severity of the consequences of being detected. A number of studies have !
demonstrated that increasing individuals' awareness of the undesirable i consequences of unacceptable actions is likely to result in increased i deterrence of the undesired behavior. For example, reminders of punishments ]
have been shown to result in increased honesty in income tax reporting l (Schwartz & Orleans,1967), decreased cheating in task performance (Sinha, j 1967), and decreases in classroom cheating (Tittle & Rowe, 1973). To be i effective, such reminders need not be received in the form of verbal or !
written reminders, but can be received in the form of knowledge of peers being !
punished. Bandura (1969) and Bandura and Walters (1963) found that knowledge !
of peers being punished increased compliance with rules. Because the lower i cutoff levels will enhance detection of drug users, more drug users will be 7
l
I o 1 detected. As the detection rates rise, the probability increases that other I drug users (or potential drug users) will observe the detection of their peers. The evidence presented hare indicates that this increase in the frequency at which drug users are observed being detected will result in increased deterrence.
Further, drug users who go undetected may encourage other workers to use drugs. Detecting drug users and rehabilitating them or removing them from the workplace can be expected to promote a workplace attitude consistent with the drug-free workp uce philosophy.
Finally, a high probability of detection is especially important in deterring those drug users who receive little or no deterrence through informal i sanctions. for example, some drug users may be involved in a drug-using subesiture, where their drug use is seen as acceptable or even desirable behavior. Waldo and Chiricos (1972), in a study of college students, found that the perceived certainty of formal sanctions had a stronger deterrent )
effect on marijuana use than on theft. They interpreted this difference to be '
an indication that certainty of formal sanctions may have a greater deterrent impact when informal sanctions condemning the behavior are absent or weak. A study by Tittle (1980) provides further support for this interpretation.
Tittle found that the certainty of punishment had the greatest deterrent I impact when undesired behaviors were regarded as highly desirable by the greatest number of persons. These findings indicate that a high probability of detection is important to deter those drug users who do not receive ,
societal or peer pressure to avoid drug use. l
1.4 CONCLUSION
The information presented here suggests that if employee drug testing programs are to ensure that the workplace is free of the effects of drugs that can endanger the workers' and the public's health and safety, then the testing programs should use cutoff levels for marijuana and amphetamines lower than those currently stipulated in the HHS guidelines. These lower cutoff levels are technically feasible and using these lower levels in workplace urine testing programs can be expected to enhanc.e the detection and deterrence of the use of these drugs. The screening and confirmation test cutoff levels recommended here are presented in Tables 1 and 2.
i l
8
- d. ;&
TABLE 1. Proposed Initial Test Cutoff Levels l
Specific drug or metabolite Cutoff level j Drug for which test is conducted in ng/ml l i
Marijuana Delta-9-i.ctrahydrocannabinol 50 Amphetamine Amphetamine 300 TABLE 2. Proposed Confirmatory Test Cutoff Levels Specific drug or metabolite Cutoff level- l Drug or drug class for which test is conducted in ng/mi Marijuana Delta-9-tetrahydrocannabinol 10 Amphetamine Amphetamine 300 Methamphetamine 300 9
o j
i
2.0 REFERENCES
52 FR.3068, 1987.
53 FR 11970, 1988. '
53 FR 36795, 1988.
Baker, T. S. , Harry, J. V. , Russell, J. W. & Myers, R. L. (1984, November / December) . Rapid method for the GC/MS confirmation of 11-nor 4 carboxy-delta-9-tetrahydrocannabinol in urine. Journal of Analytical i Toxicology,g, 255 259.
Bandura, A. (1969). Principles of behavior modification. New York: Holt, !
Rinehart and Winston.
Bandura, A. & Walters, R. H. (1963). Social learning and personality development. New York: Holt, Rinehart and Winston.
Becker, G. S. (1974). Crime and punishment: An economic approach. In Becker, G. S. & Landes, W. M. (Eds.), Essays in the economics of crime and punishment. New York: Columbia University Press.
Baselt, R. C. (1984, September / October). Unusually high cannabinoid concentrations in urine [ Letter to the editor]. Journal of Analytical l Toxicoloov,@,16A.
Beshai, N. (1984, July). California DUI law: One year implementation.
Abstracts and Reviews in Alcohol & Driving, }(3), 11-15.
Black, D. L. , Goldberger, B. A. , Isenschmid. D. S., White, S. M. &
Caplan, Y. H. (1984, September / October). Urine cannabinoid analysis: An integrated multi-method approach. Journal of Analytical Toxicology, g, 224-227.
Claster, D. S. (1967). Comparisons of risk perception between delinquents and non-delinquents. Journal of Criminal Law, 18,80-86.
Cone, E. J. & Johnson, R. E. (1986, September). Contact highs and urinary cannabinoid excretion after passive exposure to marijuana smoke. Clinical
, Pharmacology and Therapeutics, 40(3),247-256.
Cone, E. J. , Johnson. P..E. , Darwin, W. D. , Yousefnej ad, D. , Mell, L. D. ,
Paul, B. D. & Mitchell, J. (1987,May/ June). Passive inhalation of marijuana smoke: Urinalysis and room air levels of delta tetrahydrocannabinol. Journal of Analytical Toxicolocy, M(3), 89-96.
l Cook, E. (1986, July-December). Analytical methodology for delta tetrah 2(3-4)ydrocannabinol
, 79-92. and its metabolites. Alcohol, Drugs, and Driving, 10
)
. . i
- DeLaurentis, M. J., McNeil, K., Mann, A. J., Clark, S. & Greer. wood, H. M.
(1982). An EMIT assay for cannabinoid metabolites in urine. In Hawks, R. 1 (Ed.), Analysis of cannabinoids (NIDA Research Monograph #42). Washington, I DC: National Institute on Drug Abuse, Department of Health and Human J Services.
Ehrlich, I. (1967). The supply of illegitimate activities. Unpublished manuscript. New York: Columbia University.
Foltz, R. L., Fentiman, A. F. , Jr. & Foltz, R. B. (1980). GC/MS Assays for Abused Drugs in Body Fluids: Research Monograph Series #32. Rockville, MD:
U.S. Department of Health and Human Services.
Frederick, D. L., Green, J. & Fowler, M. W. (1985,May/ June). Comparison of six cannabinoid metabolite assays. Journal of Analytical Toxicology, 9_,
116-120.
Gibbs, J. P. (1966). Conceptions of deviant behavior: The old and the new.
Pacific Sociological Review, 9, 9-14.
Hanson, D. J. (1986). Drug abuse testing programs gaining acceptance in workplace. C&EN, June 2, 7-14. l Hawks, R. L. (1986). Analytical methodology. In Hawks, R. L. & Chiang, C. N. (Eds.), Urine Testing for Drugs of Abuse, NIDA Research Monograph 73. 1 Rockville, MD: Department of Health and Human Services, National Institute on Drug Abuse.
B., Foltz, R. L., Cook, C. E., Dursey, J. T. & Willette, Irving,(J.,
R. E. Leeb, 1984, July / August). Evaluation of immunoassays for cannabinoids in urine. Journal of Analytical Toxicology, 8, 192-195.
l Joern, W. A. (1987, March / April). Detection of past and recurrent marijuana use by a modified GC/MS procedure. Journal of Analytical Toxicology, 1_1, 49-52.
Kaistha, K. K. (1977). Guide to urine testing in drug abuse prevention and i multi-modality treatment programs. Journal of Chromatography, 141, 145-196.
Kogan, M. J., Razi, J. A., Pierson, D. J. & Willson, N. J. (1985, April).
Confirmation of syva enzyme multiple immunoassay technique (EMIT) d.a.u. and :
Roche abustreen radioimmunoassays (RIA) (1251) urine cannabinoid immunoassays !
by gas chromatographic / mass spectrometric (GC/MS) and bonded-phase !
adsorption / thin layer chromatographic (BPA/ TLC) methods. Journal of Forensic Science, 31(2),494-500.
Logan, C. H. (1971). On punishment and crime (Chiricos and Waldo), 1970:
Some methodological commentary. Social Problems, H , 280-284.
McBurney, L. J., Bobbie, B. A. & Sepp, L. A. (1986, March-April). GC/M5 and .
EMIT analyses of delta-9-tetrahydrocannabinol metabolites in plasma and urine !
of human subjects. Journal of Analytical Toxicology, M (2). 56-64.
I 11
_ -- __m- __ - - - - - - - - - -- -
Mul6, S. J., tomax, P. & Gress, S. J. (1988,May/ June). Active and realistic passive marijuana exposure tested by three immunoassays and GC/MS in urine. Journal of Analytical Topicology, 12, 113-116.
5:hwartz, R. H. & Hawks, R. L. (1985, August). Laboratory detection of sarijuana use. Journal of the American Medical Association, 254(6),788-792.
Schwartz, R. D. & Orleans, S. (1967). On legal sanctions. ,The University of Chicago law Review, 34, 274-300.
Sinha, J. B. P. (1967). Ethical risk and censure-avoiding behavior. Journal of Social Psychology, 71, 267-275.
Smigel, A. (1965). Does crime pay? An economic analysis. Unpublished Master's thesis. New York: Columbia University.
Roche Diagnostic Systems. (1985). Abuscreen: Radioimmunoassays for ,
amphetamine HIGH SPECIFICITY (product brochure). Nutley, NJ: Roche Diagnostic Systems, Division of Hoffmann-La Roche, Inc.
Ross, H. L., Campbell, D. T. 3 Glass, G. V. (1970). Determining the social effects of a legal reform: The British 'breathalyser' crackdown of 1967.
American Behavioral Scientist, 13, 493-509.
Sutheimer, C. A., Yarborough, R., Hepler, B. R. & Sunshine, I. (1985, July-August). Detection and confirmation of urinary cannabinoids. Journal of Analytical Toxicology, 9, 156-160. j Tittle, C. R. (1980). Sanctions and social deviance. New York: Praeger Publishers.
Tittle, C. R. (1969). Crime rates and legal sanctions. Social Problems, 16, -
409-423.
l!
Tittle, C. R. & Rowe, A. R. (1973). Moral appeal.. sanction threat, and j deviance: An experimental test. Social Problems (forthcoming).
Verebey, K., Mule, S. J., Alrezi, J. & Lehrer, M. [LettertotheEditor]. I (1985). One hundred EMIT positive cannabinoid urine samples confirmed by i BPA/ TLC, RIA, and GC/MS. Journal of Analytical Toxicology, 10, 79.
Waldo, G. P. & Chiricos, T. G. (1972). Perceived penal sanction and self- I i reported criminality: A neglected approach to deterrente research. Social l Problems, 19, 522-540.
l 12 I
1
_ _ _ - - _ - - - - _ _ _ - _ _ _ _ _ _ _ _ - - _ - _ - - _ - - _ - ._ - _ ._.