Impact of Lowering Confirmatory Test Cutoff Value in Pre-Enlistment Urine Cannabinoids
Screening: About Five Years’ Experience in the French Gendarmerie
YannickLecompte*, MartinePerrin, SophieSalle and OlivierRoussel
Toxicology Department, Institut de Recherche Criminelle de la Gendarmerie Nationale, Rosny sous bois, France
*Author to whom correspondence should be addressed. Email: firstname.lastname@example.org
The guidelines for screening of urinary cannabinoids require that all
specimens testing positive should be confirmed by gas chromatog-
raphy–mass spectrometry at a confirmatory test cutoff value of
15 ng/mL of 11-nor-9-carboxy-D9-tetrahydrocannabinol (THCCOOH).
To assess the impact of lowering the confirmatory test cutoff value
on the diagnostic sensitivity and efficiency of a cannabinoid testing
program, the results of 986 confirmation analyses of positive
screening tests, conducted in the framework of medical fitness
examinations prior to enlistment in the French Gendarmerie
between January 1, 2005, and December 31, 2009, were retro-
spectively studied. If the confirmatory test cutoff value of
THCCOOH is set at 5 ng/mL instead of 15 ng/mL as recommended
by guidelines, the number of confirmed results increases by 25.2%.
The positive predictive value of the initial screening test rises from
63.9 to 80.0%. Only one true-positive applicant has appealed. His
THCCOOH urinary concentration, which was incompatible with
passive cannabis smoke exposure, was confirmed by another la-
boratory. The use of a confirmatory test cutoff value lower than
that recommended significantly increases the diagnostic sensitivity
of the screening program for urinary cannabinoids without altering
Like the whole of society, the armed forces are directly con-
cerned by illicit drug use (1). This is why a systematic
pre-enlistment urine testing program was implemented in
1990 by the French Gendarmerie (police force with military
Based on the same principles as workplace testing, a typical
pre-enlistment drug testing program consists of an initial
screening immunoassay test in urine. If the immunoassay is
positive, confirmatory testing using gas chromatography–mass
spectrometry (GC–MS) is required to definitively establish the
identity of the drug or drug metabolite and report the speci-
men as positive. Cutoff concentrations are used to determine
whether the screening and the confirmatory results are posi-
tive or negative. Screening and confirmation cutoff levels are
set for each tested substance according to several parameters,
including pharmacokinetics (particularly metabolism and ex-
cretion, which are responsible for the detection window in
biological matrices), passive exposure (e.g., passive cannabis
smoke exposure) and food or licit drug interactions (e.g., for
opiates). Cutoff values usually comply with guideline values
set by the U.S. Substance Abuse and Mental Health Services
Administration (SAMHSA) or by an expert group like the
European Workplace Drug Testing Society (EWDTS) (3, 4).
With 12.4 million people admitting having used it more than
once in their lifetime, cannabis is the most popular illicit sub-
stance in France (5). Its use predominantly involves young
adults. In 2005, 47.5% of people aged 18 to 25 used cannabis at
least once in their lifetime and 8.7% are regular users (5). As
confirmed by recent epidemiological studies conducted within
the French Army and the French Navy, the use of cannabis is a
major issue for the armed forces, because the 18-to-25
age range is the target population for recruitment and the
major part of the military strength (1). The urinary analyte that
D9-tetrahydrocannabinol (THCCOOH). THCCOOH is the main
metabolite of the psychoactive substance D9-tetrahydrocannabinol
(THC). In 1994, due to the improvement in the specificity of
the antibodies directed against THCCOOH used for the im-
munoassay test, the initial screening test cutoff for cannabis
was lowered from 100 to 50 ng/mL. This resulted in a rise in
the diagnostic sensitivity of the tests (number of true positives
increasing from 23 to 53%) without reducing specificity signifi-
cantly (the loss of specificity is below 3%) (6, 7).
The screening test cutoff of 50 ng/mL seems to be adequate
to detect the use of cannabis on the basis of both regular and
recent occasional use. Nevertheless, some authors have demon-
strated that a screening test cutoff of 25 or even 20 ng/mL
THCCOOH would improve the diagnostic sensitivity of the
test while preserving its specificity (8, 9). Established in 1988,
the SAMHSA urinary confirmatory test cutoff of 15 ng/mL
THCCOOH was not modified when the screening test cutoff
was lowered from 100 to 50 ng/mL. This confirmatory test
cutoff value continued to correlate with results provided by
immunoassays in use during that period (10). Today, this con-
firmatory cutoff appears to be very high compared to the de-
tection limits of GC–MS techniques currently used for the
The possibility of setting a lower confirmatory test cutoff is
rarely mentioned in the scientific literature. Studies on drug
testing programs primarily concern the performance character-
istics of the different screening immunoassay tests that have
been developed and marketed. They essentially determine the
sensitivity and specificity of the screening tests compared to
the results of confirmatory testing using the recommended
confirmatory cutoff. In these controlled studies, screening tests
are performed by operators with strong laboratory experience
and the samples are from drug-added specimens or donor
urine specimens at specified concentrations. In general, these
studies only include a few cases. Theses study designs do not
represent what can be observed in pre-enlistment or work-
place medical fitness examinations.
Applicants for enlistment in
undergo a medical examination, including a point-of-collection
screening immunoassay for urinary cannabinoids. The Forensic
Science Institute of the French Gendarmerie (IRCGN) per-
forms all confirmatory testing of screening tests conducted
the French Gendarmerie
# The Author . Published by Oxford University Press. All rights reserved. For Permissions, please email: email@example.com
Journal of Analytical Toxicology 2012;36:569–574
at the national level by medical centers of the French
The confirmatory test cutoff value used by this laboratory is
set at 5 ng/mL instead of the recommended 15 ng/mL. To
assess the impact of lowering the confirmatory test cutoff value
on the diagnostic sensitivity and efficiency of a cannabis testing
program routinely involving many cases, the results of confirm-
ation analyses, conducted in the framework of medical fitness
examinations prior to enlistment in the French Gendarmerie
between January 1, 2005, and December 31, 2009, were retro-
Materials and Methods
Urine collection and on-site analysis
The device used during the study period in the medical
centers of the French Gendarmerie is the monoparametric
Syva RapidTest Cartridge (THC) (Dade Behring, Paris, France).
The cutoff defined by the supplier of this test is 50 ng/mL. All
positive urine specimens are sent to IRCGN for confirmation
testing by GC–MS (2). Urine specimens that tested positive
were aliquoted by paramedical personnel into two tamper-
proof 30 mL plastic bottles. They were both identified by
means of an adhesive label with the following data: anonymity
number, year of collection, identification of the medical center
in which the screening test was performed, signature of both
the patient and the paramedical personnel who supervised the
collection. The samples are usually transported by post at
ambient temperature. On arrival at the authors’ laboratory, the
samples are stored at –208C. Only one sample is analyzed; the
second is kept in the freezer in the event of disputed results
and subsequent need for a second assessment.
The samples and results considered in this study were
obtained exclusively through the routine medical examinations
conducted in accordance with medical fitness standards in the
French Gendarmerie. In this context, applicants are previously
and regulatory informed that “drug use, detected by the neces-
sary diagnostic tests, is a general cause of unfitness for service”
(2). The applicants sign a statement on the chain of custody
form, certifying that the specimen identified on the form was
in fact the specimen provided by the donor. They also provide
informed consent for the work to be undertaken.
Standards and reagents
All reagents are of analytical grade. Methanol, hexane, and ethyl
acetate are purchased from Sigma Aldrich (Saint Quentin
Fallavier, France). Concentrated ammonium hydroxide and
glacial acetic acid are from Fisher Scientific (Illkirch, France).
The enzymatic hydrolysis is performed by beta-glucuronidase
from Escherichia coli 400 UI/L (APOH Technologies, Villeneuve
Saint Georges, France). The derivatizing agent is bis(trimethylsi-
lyl)trifluoroacetamide with 1% trimethylchlorosilane (BSTFA)
(Alltech, Templemars, France). Ultrapure type I water is pro-
duced through a Millipore gradient A10 system with Q-Gard
I and Quantum EX cartridges and filtered on a Millipore
0.22 mm Millipack filter (Millipore, Molsheim, France).
Calibrators and controls
The Ceriliant reference solutions, namely THCCOOH and
(THCCOOH-d3), each at the concentration of 0.1 g/L in
methanol, are purchased from LGC Standards (Molsheim,
France). These stock solutions are diluted with methanol to
obtain a THCCOOH solution at the concentration of 5 mg/L
and THCCOOH-d3 at the concentration of 1 mg/L. The
quality control material is Liquicheck urine toxicology
control from Bio-Rad (Marne la Coquette, France). The
target value for the free THCCOOH is 18.5 ng/mL.
Calibrators and specimen preparation
The calibration samples are prepared in duplicate from 2 mL
drug-free urine by spiking THCCOOH at 40 and 100 ng/mL
with the nondeuterated THCCOOH solution. A run comprises
the calibration samples, a negative urine, a quality control and
applicant samples. The method described in the following is
able to quantify total THCCOOH (i.e., both conjugated and free
THCCOOH). Therefore, a preliminary enzymatic hydrolysis is
needed. The urine sample of 2 mL is adjusted to a pH between
6.5 and 7.0 by adding diluted ammonium hydroxyde or acetic
acid. Then, 50 mL of beta-glucuronidase are added. The tube is
left for at least 14 h at 378C. After the hydrolysis, 20 mL of the
deuterated internal standard and 200 mL of 10% acetic acid
(in water) are added. The liquid–liquid extraction is performed
by adding 5 mL hexane–ethyl acetate (90/10, v/v). After gentle
shaking, the tubes are centrifuged 5 min at 1,780 ? g. The
upper organic layer is transferred to a second borosilicated
5 mL tube and subsequently evaporated at 408C under a gentle
stream of filtered air. Forty microliters of the derivatization
reagent (BSTFA þ 1% TMCS) are then added to the tube and
left for 20 min at 808C. One microliter of this preparation is
analyzed by GC–MS.
Analysis and quantitation are performed by GC–MS Shimadzu
GC 17A and QP5000 (Shimadzu, Champs sur Marne, France)
using a chromatographic column from Agilent Technologies:
J&W HP-1MS Ultra Inert capillary 12 m ? 0.20 mm ? 0.33 mm
(Interchim, Montlucon, France). The software is GCMSsolution
(Shimadzu). The injection is made in splitless mode at 2808C,
with a constant flow of helium at 0.9 mL/min. The temperature
program as follows: 908C for 4.5 min; from 908C to 127.58C at
37.58C/min; from 127.58C to 2108C at 13.58C/min; from 2108C
to 2508C at 58C/min; from 2508C to 3208C at 258C/min; 3208C
maintained for 4 min. The electron impact mass spectra are
acquired in the selective ion monitoring mode (SIM). Three
ions for the analyte and for deuterated internal standard are
monitored (quantification ions are in bold): THCCOOH m/z
473, 371, 488 and THCCOOH-d3 m/z 476, 374, 491.
The response function is a linear unweighted function (y ¼
ax). The method validation was based on the total error
concept (11, 12) and performed using e-noval software
(Chemcad, Obernay, France). Validation standards at four con-
centrations (3, 10, 50 and 100 ng/mL) were analyzed with
independent calibration curves on each of three series (inter-
mediate precision) of four replicates (repeatability). The
Lecompte et al.
accuracy profile of the analytical method is represented by
Figure 1. According to the acceptance criteria (acceptance
limits:+40% and b-expectation tolerance limits: 90%), the ana-
lytical method is accurate throughout the measuring interval of
3 to 100 ng/mL THCCOOH. The limit of quantification (LOQ)
is 3 ng/mL. The lower limit of detection (LOD) is estimated at
a third of the LOQ or 1 ng/mL. The maximal measurement un-
certainty is 30% (coverage factor k ¼ 2). This method is also
accredited according to the ISO 17025 standard by the French
Accreditation Committee (COFRAC). The confirmatory test
cutoff value used by the authors’ laboratory is set at 5 ng/mL
instead of the recommended 15 ng/mL.
In a retrospective study like the one presented in this paper,
not all performance parameters of the urine screening test can
be calculated, particularly diagnostic sensitivity and specificity.
Because only the urine specimens for which the screening test
proved positive are sent to the laboratory for confirmation pur-
poses, no information is available about the negative screening
tests. The positive predictive value (PPV) is the only parameter
that can be calculated. In the hypothesis that all specimens
sent to the lab had a positive screening test, the PPV corre-
sponds to the percentage of confirmed samples.
During the study period, 62,135 medical fitness examinations
involving urine screening for cannabis use were conducted
prior to enlistment. Of these, 1,004 had positive screening
results, requiring subsequent confirmation testing at the
IRCGN toxicology department. Among the 1,004 samples
received, only 986 have been analyzed. The remaining 18 speci-
mens had urine volumes less than 2 mL. For the majority of
these samples, the bottle had not been sealed properly and
contents emptied during transport.
Number of confirmations and performance
of screening test
In the study period, 789 urine tests (80.0%) were confirmed at
a THCCOOH concentration equal to or over 5 ng/mL; 197
samples did not confirm (absence of THCCOOH or concentra-
tion under the cutoff of 5 ng/mL) (Figure 2). Among the 789
samples confirmed, only one applicant has appealed. His
THCCOOH urinary concentration (.15 ng/mL), which was in-
compatible with any passive cannabis smoke exposure (13, 14),
was confirmed by another laboratory (Institute of Aerospace
Medicine of the French Military Health Service).
Distribution of urinary THCCOOH concentrations
and influence of the confirmatory test cutoff
The distribution of THCCOOH concentrations in GC–MS con-
firmed urine samples for the entire study period is shown in
Figure 2. During the study period, 55.5% of urine samples sub-
jected to confirmation analysis showed a concentration lower
than the initial screening test cutoff (50 ng/mL) and 20.2% of
confirmed samples had a THCCOOH concentration value in
the concentration range 5 to 15 ng/mL.
Table I presents the distribution in number of samples con-
firmed when different confirmatory test cutoffs are applied a
posteriori. Using a confirmatory test cutoff of 10 ng/mL instead
of the recommended cutoff of 15 ng/mL raises the number of
true positives (samples confirmed) by 12.5%. The PPV comes
to 71.9 instead of 63.9%. When the confirmatory test cutoff
used in the authors’ laboratory (5 ng/mL) is applied, the
number of true positives rises by 25.2%, and the PPV is 80%.
Using a cutoff higher than 15 ng/mL induces a dramatic fall
in the number of confirmed samples and decreases the PPV
Syva RapidTest, like other point-of-collection urinary cannabin-
oid screening immunoassays, will yield a very small number of
false positives and presents a PPV higher than 90% for a con-
firmatory test cutoff of 15 ng/mL (15–18). Although a con-
firmatory test cutoff of 5 ng/mL was applied, the PPV obtained
in this study is lower than the values reported in the scientific
literature. This illustrates the differences in performance
between a study under standardized conditions and a study in
real conditions of use. These unexpectedly low PPV values
probably result from an excess of false positives combined with
a lack of true positives in the use of routine tests. False posi-
tives can result from a misreading of the screening test by un-
initiated personnel. Indeed, these tests require an inverse
reading (the presence of the colored band indicates a negative
result). Moreover, when confronted with a doubtful screening
result, many medical practitioners prescribe a confirmation
analysis, which contributes to an excess of false positives.
Furthermore, the specificity of screening immunoassays is
limited due to cross-reactions of the anti-THCCOOH antibodies
with molecules like nonsteroidal anti-inflammatory drugs
(NSAIDs) (19) and, to a lesser extent, proton pump inhibitors
(Efavirenz) (15). Concerning the deficit of true positives, this
could result from the failure to request confirmatory testing
Figure 1. Accuracy profile of the THCCOOH analytical method: measured standard
validation concentration, acceptance limits (...), b-expectation tolerance limits (---)
and average relative error (solid gray line).
Impact of Lowering Confirmatory Test Cutoff Value in Pre-Enlistment Urine Cannabinoids Screening 571
when the positive screening test is reinforced by the confes-
sion of the applicant during the medical examination.
Assays performed in the 789 urine samples for which the
presence of THCCOOH was confirmed show a large dispersion
of concentrations. More than half (55.5%) of urine samples in
which the presence of THCCOOH was confirmed had a lower
confirmatory concentration than the initial screening test
cutoff of 50 ng/mL, and samples with THCCOOH concentra-
tions less than 15 ng/mL represented 20.2% of confirmed ana-
lyses. THCCOOH concentrations far below the screening test
cutoff are commonly observed (8). This phenomenon is due to
cross-reactions between the many metabolites of different can-
nabinoids present in marijuana smoke, which contribute to the
screening positivity rates (20). Moreover, cold chain failures
when transporting specimens to the laboratory generate a de-
crease in THCCOOH sample concentration between screening
and confirmatory testing and thus contribute to the measure-
ment of lower confirmatory concentrations. According to
Skopp et al., there is a 13.4% decrease in THCCOOH when the
urine sample is stored at room temperature (208C) for 48 h.
There is a 21.2% THCCOOH loss when the sample remains at
room temperature for 120 h. Mechanisms of decarboxylation
and bacterial or fungal degradation of THCCOOH predominant-
ly account for this decrease in THCCOOH concentration,
which also contributes to the increase in the number of false
recommended confirmatory test cutoff value of 15 ng/mL in
urine samples, which tested positive with an initial screening
cutoff of 50 ng/mL, is an argument that supports the use of
lower confirmatory test cutoff values. In this study, the cutoff
of 5 ng/mL increases the number of true positives by 25.2%.
Few publications report the criteria for setting the confirma-
tory test cutoff of THCCOOH in cannabis testing programs. In
a study using the 20 ng/mL initial screening test cutoff,
Wingert et al. recorded a 7.3% increase in the number of true
positives when reducing the confirmatory test cutoff from 10
to 5 ng/mL (9). THCCOOH can be detected in urine from the
first micturition after cannabis use in the case of an acute ex-
posure. Nevertheless, when analyses are performed on the first
urine voided after cannabis smoking, Huestis et al. demon-
strated that only half of the subjects who have smoked a cigar-
ette containing 15.8 mg of THC, and only 83% of subjects who
have smoked a cigarette containing 33.8 mg of THC, present a
urinary THCCOOH concentration above the cutoff of 15 ng/
mL (22). The use of a lower confirmatory test cutoff can also
effectively reduce the number of falsely unconfirmed screening
in urine specimens adulterated by dilution (whether with the
intention to suborn the test or as a result of ingesting large
amounts of liquid in the hours before collection) as well as the
decrease in urinary THCCOOH during the transfer of samples
(23, 24). Some authors recommend using no confirmatory test
cutoff at all, instead identifying THCCOOH by its mass spec-
trum only (based on the mass pattern identification criteria)
(8). This approach makes it necessary to define the analytical
performance of the method, notably the LOD. This prerequisite
is critical when several laboratories perform confirmation ana-
lyses within the same drug testing program.
The possibility of passive inhalation of cannabis smoke is an
argument often suggested against lowering the confirmatory
test cutoff value. Subjects have indeed sometimes used the
concept of passive inhalation of cannabis smoke to justify posi-
tive tests. Available studies on THCCOOH detection in urine
after passive exposure are related to acute exposure and ex-
perimental conditions largely more drastic than in real-life
situations (confined space, no ventilation, large number of
of confirmatory concentrationsbelowthe
Figure 2. Results of the 986 GC–MS confirmatory tests, and THCCOOH concentration distribution in confirmed urine
Influence of Confirmatory Test Cutoff Values on the Number of Confirmed Screening Tests during
the 2005–2009 Period
Variation in the number of
confirmed screening tests
(over the confirmatory test
cutoff of 15 ng/mL)
*Percentage expressed in relation to number of analyzed samples.
Lecompte et al.
smokers and very high THC level of the cigarette) (25, 26). If
THCCOOH is detected by GC–MS in urine after passive expos-
ure, these experiments establish that despite drastic exposure
conditions to sidestream cannabis smoke, the immunoassay
screening test results are still negative at a cutoff value of
50 ng/mL (13, 14).
Consequently, it seems very unlikely that passive exposure to
cannabis smoke should solely account for a positive urine
screening test. Even if the urine sample of an applicant who
has really been passively exposed to cannabis smoke would in-
appropriately be subjected to confirmatory testing (due to a
doubtful screening test, a misreading or a cross-reaction, for
example), these studies also show that THCCOOH concentra-
tions higher than the 5 ng/mL confirmatory test cutoff value
can only be measured within six hours after exposure (13, 14).
This window of detection should be in reality relatively
short, because these data were derived under particularly
extreme experimental conditions of passive cannabis smoke
exposure. In addition, medical fitness standards provide
advanced information to applicants before the drug screening
test and medical examination take place during working hours
and away from an eventual passive exposure. Moreover, experi-
ence accumulated by the authors’ laboratory has demonstrated
that use of a confirmatory test cutoff of 5 ng/mL did not gener-
ate an excessive number of disputed results and appeals. Only
one appeal was recorded during the five-year study period, for
which the results were confirmed by another laboratory and
for which THCCOOH urinary concentration was incompatible
with any passive cannabis smoke exposure (.15 ng/mL). This
implicitly reflects the acceptance of confirmatory testing
results by the applicants and attests the reality of their cannabis
The retrospective study of 986 urine confirmation testing
results over five years has examined the influence of the
urinary cannabinoid confirmatory test cutoff in pre-enlistment
drug testing programs, such as the one set up within the
French Gendarmerie. The choice of a confirmatory test cutoff
of 5 ng/mL THCCOOH over the recommended cutoff of
15 ng/mL considerably increases the PPV of initial screening
tests and therefore increases the diagnostic sensitivity of the
cannabinoids screening program as a whole. Choosing a lower
confirmatory test cutoff may compensate for sample adulter-
ation by dilution or degradation of the analyte during transpor-
tation to the laboratory. Mass spectrometry easily ensures the
analytical specificity of confirmatory testing at this level of con-
centration. In addition, recent studies have confirmed that
passive exposure to cannabis smoke cannot solely generate
positive results in initial screening tests using the re-
commended screening test cutoff of 50 ng/mL THCCOOH.
Finally, if detection of THCCOOH in the urine of an individual
proves cannabis use, the determination of the confirmatory test
cutoff value cannot aim to differentiate between occasional
and chronic users, to diagnose a problematic consumption
(harmful use and dependence) and to determine the medical
fitness to hold some employment. Assessment of addictive
behaviors should primarily be based on interview and clinical
data collected during medical fitness examinations, and may
involve questionnaires and validated scales such as the cannabis
abuse screening test (CAST) (1).
1. Gheorghiev, C., Arvers, P., de Montleau, F., Fidelle, G., Queyriaux, B.,
Verret, C. (2009) Cannabis in the army: Between past and topical-
ity. Annals of Medical Psychology, 167, 429–436.
2. Ministry of Defence. (2008) Ministerial instruction No. 22000/DEF/
GEND/RH on the standards of medical fitness of military personnel
of the Gendarmerie.
3. European Workplace Drug Testing Society. European laboratory guide-
lines for legally defensible workplace drug testing: Urine drug testing.
4. United States Department of Health and Human Services. (2008)
Substance Abuse and Mental Health Services Administration.
Mandatory guidelines for federal workplace drug testing programs.
Federal Register, 73, 71858–71907.
5. Legleye, S., Le Ne ´ zet, O., Spilka, S., Beck, F. (2008) Drug use among
adolescents and young adults between 2000 and 2005, France.
Bulletin E´pide´miologique Hebdomadaire., 13, 89–92.
6. Huestis, M.A., Mitchell, J.M., Cone, E.J. (1994) Lowering the federal-
ly mandated cannabinoid immunoassay cutoff increases true-
positive results. Clinical Chemistry, 40, 729–733.
7. Rowland, B.J., Irving, J., Keith, E.S. (1994) Increased detection of
marijuana use with a 50 micrograms/L urine screening cutoff.
Clinical Chemistry, 40, 2114–2115.
8. Madhavaram, H., Couch, R.A. (2010) Utilization of a detection level
of 25ng/mL for cannabinoids in urine using a CEDIA THCPLUS im-
munoassay: Application of this cut-off to urines of school children.
Forensic Science International, 198, 28–30.
9. Wingert, W.E. (1997) Lowering cutoffs for initial and confirmation
testing for cocaine and marijuana: Large-scale study of effects on
the rates of drug-positive results. Clinical Chemistry, 43, 100–103.
10. Liu, R.H., Edwards, C., Baugh, L.D., Weng, J.L., Fyfe, M.J., Walia, A.S.
(1994) Selection of an appropriate initial test cutoff concentration
for workplace drug urinalysis-cannabis example. Journal of
Analytical Toxicology, 18, 65–70.
11. Hubert, P., Rozet, E., Boulanger, B., Dewe ´ , W., Laurentie, M., Dubois,
N. et al. (2006) Harmonisation of validation strategies and estima-
tion of the associated uncertainty in the framework of assay labora-
tories accreditation. Acta Clinica Belgica, 1, 54–56.
12. Feinberg, M., Boulanger, B., Dewe, W., Hubert, P. (2004) New
advances in method validation and measurement uncertainty aimed
at improving the quality of chemical data. Analytical and
Bioanalytical Chemistry, 380, 502–514.
13. Niedbala, R.S., Kardos, K.W., Fritch, D.F., Kunsman, K.P., Blum, K.A.,
Newland, G.A. et al. (2005) Passive cannabis smoke exposure and oral
fluid testing. II. Two studies of extreme cannabis smoke exposure in
a motor vehicle. Journal of Analytical Toxicology, 29, 607–615.
14. Rohrich, J., Schimmel, I., Zorntlein, S., Becker, J., Drobnik, S.,
Kaufmann,T. etal. (2010)
tetrahydrocannabinol and 11-nor-9-carboxytetrahydrocannabinol in
blood and urine after passive exposure to cannabis smoke in a
coffee shop. Journal of Analytical Toxicology, 34, 196–203.
15. Moeller, K.E., Lee, K.C., Kissack, J.C. (2008) Urine drug screening:
Practical guide for clinicians. Mayo Clinic Proceedings, 83, 66–76.
16. Peace, M.R., Tarnai, L.D., Poklis, A. (2000) Performance evaluation
of four on-site drug testing devices for detection of drugs of abuse
in urine. Journal of Analytical Toxicology, 24, 589–594.
17. Kadehjian, L.J. (2001) Performance of five non-instrumented urine
drug-testing devices with challenging near-cutoff specimens.
Journal of Analytical Toxicology, 25, 670–679.
18. Samyn, N., Areschka, V., Verstraete, A. (2000) Evaluation of roadside
drug testing equipment. Ann. Toxicol. Anal., 12, 105–115.
19. Boucher, A., Vilette, P., Crassard, N., Bernard, N., Descotes, J.
(2009) Urinary toxicological screening: Analytical interference
Impact of Lowering Confirmatory Test Cutoff Value in Pre-Enlistment Urine Cannabinoids Screening 573
between niflumic acid and cannabis. Archives of Pediatrics, 16,
20. Goulle, J.P., Saussereau, E., Lacroix, C. (2008) Delta-9-tetrahydro-
Franc ¸aises, 66, 232–244.
21. Skopp, G., Potsch, L. (2004) An investigation of the stability of free
and glucuronidated 11-nor-delta9-tetrahydrocannabinol-9-carboxylic
acid in authentic urine samples. Journal of Analytical Toxicology,
22. Huestis, M.A. (2007) Human cannabinoid pharmacokinetics. Chemistry
and Biodiversity, 4, 1770–1804.
23. Dasgupta, A. (2007) The effects of adulterants and selected
ingested compounds on drugs-of-abuse testing in urine. American
Journal of Clinical Pathology, 128, 491–503.
24. Fraser, A.D., Zamecnik, J. (2003) Impact of lowering the screening
and confirmation cutoff values for urine drug testing based on dilu-
tion indicators. Therapeutic Drug Monitoring, 25, 723–727.
25. Hayden, J.W. (1991) Passive inhalation of marijuana smoke: A critic-
al review. Journal of Substance Abuse, 3, 85–90.
26. Niedbala, S., Kardos, K., Salamone, S., Fritch, D., Bronsgeest, M.,
Cone, E.J. (2004) Passive cannabis smoke exposure and oral fluid
testing. Journal of Analytical Toxicology, 28, 546–552.
Lecompte et al.