Article - Kidwell — USA v. Jittaphol (Dkt. 70-6, D. Mass.)

Record facts

U.S. District Court for the District of Massachusetts · No. 1:21-cr-10270-MLW · Doc. 70-6 · 2022-04-06 · Docket on CourtListener

Full text

Comparison of daily urine, sweat, and skin swabs
among cocaine users
D.A. Kidwella,*, J.D. Kidwella, F. Shinoharaa, C. Harperb, K. Roartyb,
K. Bernadtb, R.A. McCaulleyb, F.P. Smithb
aChemistry Division, Naval Research Laboratory, Washington, DC 20375, USA
bDepartment of Justice Sciences, The University of Alabama at Birmingham, Birmingham, AL 35294, USA
Received 16 October 2002; received in revised form 15 January 2003; accepted 17 January 2003
Abstract
This study (1) compares urine, skin swabs, and PharmChekTM sweat patches for monitoring drug use; (2) measures possible
environmental contamination in recent cocaine (COC) users; and (3) evaluates various immunoassays (IA) for screening COC in
diverse matrices. Unique aspects include daily urine monitoring of 10 participants for 4 weeks, multiple monitoring methods,
analysis for all specimens by IA and gas chromatography (GC)/mass spectrometry (MS), and the potential for continued illicit
drug use by participants. Urine served as the ‘‘gold standard’’ specimen for determining drug use. Only cocaine and related
substances were detected.
Trace amounts of drugs were found on the skin (<50 ng per swab) of urine-negative participants’ hands or forehead. In contrast,
larger quantities of COC were found on the skin of individuals with BE-positive urines or individuals living with drug users (up to
20 mg per swab). Patch COC amounts among the three regular users (250–9000, 0–240, 160–22,000 ng per patch) exceeded BE
(50–950, none, 30–2200 ng per patch). Pre-swabs, valuable for interpreting the source or time frame of positive patch results,
contained substantial COC (38–1160, 0–152, 34–762 ng per swab) prior to patch application; therefore, patch results may
represent current use, prior use, contamination, or a combination. In three individuals with no indication of cocaine use, false
positives (defined as sweat patch positive when urine specimens were <300 ng BE/ml) occurred at a 7% rate. Proposed cut-off
concentrations of 75 ng cocaine per patch and 300 ng BE/ml urine curtail the incidence of false positives in this limited population.
Three immunoassays were compared to screen specimens for cocaine: a modified, manual Microgenics CEDIA; a Cozart
ELISA; and an OraSure ELISA. CEDIA’s limit of detection (LOD) was 81 ng/ml, compared with LODs of 4 ng/ml for the
Cozart ELISA and 1.5 ng/ml for the OraSure ELISA. Cozart correlated with OraSure results for COC concentrations <2000 ng
per swab (n ¼ 117), r2 ¼ 0:79.
# 2003 Elsevier Science Ireland Ltd. All rights reserved.
Keywords: Environmental contamination; Sweat; Urine; Cocaine
1. Introduction
Historically, drug testing programs have relied on urina-
lysis as the gold standard for establishing illicit drug use.
Due to the pharmacokinetics of cocaine (COC) and the
normally applied immunoassay (IA) cut-off concentration
(300 ng BE/ml urine), drug testing programs generally
require urine collection every 2–3 days to determine an
individual’s use pattern [1,2]. Lower cut-off concentrations
extend the window of detection at the expense of possible
false positives from inadvertent environmental exposure,
including such sources as gross external contamination
(for example, crime lab workers handling cocaine) and
trace ingestion. Trace ingestion, considered to be 1–5%
(approximately 1–5 mg) of street use dose, may not result
in a noticeable euphoric effect, but would be capable of
producing a substantial urine positive. For example, in
consuming Inca Tea, an individual ingests 2–3 mg of
Forensic Science International 133 (2003) 63–78
* Corresponding author. Tel.: þ1-202-767-3575;
fax: þ1-202-767-3321.
E-mail address: kidwell@ccf.nrl.navy.mil (D.A. Kidwell).
0379-0738/03/$ – see front matter # 2003 Elsevier Science Ireland Ltd. All rights reserved.
doi:10.1016/S0379-0738(03)00051-3
Case 1:21-cr-10270-MLW     Document 70-6     Filed 04/06/22     Page 1 of 16

cocaine. This has been shown to cause positive urine results
for 21–26 h [3]. Consuming larger amounts (25 mg) of
cocaine produces positive urine results for up to 36 h [4].
Even at this dosage level, only a slight numbing of the mouth
was observed during the consumption.
External continuous monitors, such as the PharmChemTM
sweat patch, offer substantial advantages over frequent urine
testing, such as convenience and cost [5]. Continual sam-
pling of an individual’s sweat for up to 7 days using the sweat
patch creates a wider time window for detecting drug use,
with the potential to trap and accumulate drugs and their
metabolites excreted in sweat. The polyurethane covering of
the patch is claimed to protect the collection pad from
external contamination and to allow water vapor to escape,
which permits long-term wear. However, this claim has
come under increased scrutiny.
Based on previous laboratory experiments, two sources of
contamination can occur [6,7]. First, drug contamination on
the external patch membrane cause what we have termed
‘‘contamination from without’’ (CFWO). Under these cir-
cumstances, rapid diffusion of drugs through the membrane
and into the moistened interior of the patch, within 1 min,
resulted in patch drug amounts above the suggested manu-
facturer’s cut-off concentration for determining a positive.
Proof of patch penetration by other molecules further sub-
stantiates our findings. For example, certain dyes, with
molecular weights (molecular weight influencing diffusion)
that exceed those of most illicit drugs, penetrate the patch
membrane. In the uncharged state, they readily diffuse
through the patch membrane and deposit in the patch test
pad when the interior pad is moist [6,7]. When the interior
pad is dry, penetration was not observed [8,9]. In another
example, Uemura et al. [10] reported finding d-5 cocaine in
50% of the sweat patches applied on top of sweat patches
spiked with d-5 cocaine and worn by cocaine-naive volun-
teers. Uemura’s results suggest that cocaine penetrates that
patch from the inside outward, also confirming our findings
[6,7] that the patch membrane is permeable to cocaine.
The second source of contamination, ‘‘contamination from
within’’ (CFWI), results from the presence of drugs on the
skin before application of the patch. CFWI can occur from at
least two sources: (1) an individual’s own previous drug use
or (2) an individual’s being ‘‘around drugs’’ unrelated to
intentional use by the individual in question [11]. Drugs
persist on skin, even though the skin is ‘‘cleaned’’ before
the patch is applied, and are difficult to remove [6,7,12–14].
We have shown that, after an initial application of only 10 mg
of drugs to the skin (an amount equivalent to 0.1–0.01% of a
dose), 6 days of regular hygiene followed by ‘‘cleaning’’ with
isopropanol wipes (as recommended by the manufacturer)
does not prevent positive patch results [6,7]. Unfortunately,
courts and drug treatment programs rely on the patch to
distinguish between current and prior drug use. No known
studies have been performed that address the interval during
which patches will remain positive after binge drug use
cessation, a critical question for future research.
Patch positives accompanied by urine negatives present a
legally challenging question (some examples of legal cases
are given in Appendix A) [15,16]. Often where urine and
patch results are obtained, the drug testing programs fail to
collect sufficiently frequent urine samples to demonstrate
cocaine abstinence. Thus, the trier of fact must decide
whether to punish an individual based on conflicting and
potentially erroneous interpretations of drug test results. The
positive patch and negative urine results often are explained
as due to the longer window of detection of the patch.
A unique aspect of the present research relates to our goal
to expand previous studies to ‘‘real life’’ conditions where
illicit drug use occurs. In this regard, participants were
current or recent cocaine users. For these outpatient parti-
cipants, their self-prescribed use of street drugs was not
limited by approved research protocols (such as hospital
ward settings) where the dosage of dangerous drugs is
appropriately restricted by ethical research oversight. Sepa-
rately, no effort was made to remove external contamination
in their living environments, including contamination from
previous or ongoing cocaine use. Furthermore, outpatient
participants were not prohibited from intimate contact with
possible cocaine-using spouses. Casual contacts with those
who may be cocaine-contaminated included others in their
environment (such as domestic cohabitants) as well as fellow
patients in the drug treatment program.
Broadly, our research seeks to improve drug use monitor-
ing by: (1) making it more convenient to both those who
administer drug testing programs and those who are subject
to drug testing; (2) making drug testing less invasive/more
dignified; and (3) extending the window of detection and
lowering costs, while maintaining the high reliability cur-
rently associated with forensic urine drug testing as per-
formed under the federal workplace drug testing guidelines.
Immediate research goals include: (1) examining the issue
of environmental contamination as it applies to urine, skin
wipe, and sweat patch testing, (2) revisiting the issue of cut-
off concentrations and their effect on drug testing reliability,
and (3) evaluating several immunoassays for their use as
pre-screens in diverse matrices.
2. Materials and methods
2.1. Volunteers
Research participants were recruited from adult patients
in cocaine dependence treatment and gave informed consent
in accordance with an IRB-approved protocol. Ages of the
10 participants ranged from 21 to 50, with 3 male and 7
female participants. The ethnic distribution roughly approx-
imates the local (Alabama) population: seven Caucasian,
two African-American, and one Hispanic. To qualify, parti-
cipants must have tested positive by urinalysis for cocaine
within 90 days, indicating the potential for recent use. We
predicted that this criterion of recent use might provide some
64
D.A. Kidwell et al. / Forensic Science International 133 (2003) 63–78
Case 1:21-cr-10270-MLW     Document 70-6     Filed 04/06/22     Page 2 of 16

‘‘real life’’ participants who were either: (1) currently
cocaine abstinent yet living in an environment contaminated
by their previous cocaine use, (2) currently cocaine abstinent
yet living in an environment contaminated by their spouse’s
(or other cohabitant’s) cocaine use, (3) currently using
cocaine occasionally, (4) chronically using cocaine, or (5)
currently living in relatively uncontaminated environments.
2.2. Specimens
Urine specimens and skin swabs of fingertips and forehead
were collected daily (except Sundays and holidays). Partici-
pants opened individually packaged (ordinary, commercially
available) isopropyl alcohol (70%) swabs, wiped their nor-
mally favored hand, including fingertips (right-handed people
swabbed their right hand), replaced the swab in the torn
package, and then placed the package in a plastic, zip-lock
bag. Participants swabbed their foreheads in a similar fashion.
Urine was obtained in standard 60 ml screw-cap cups,
supplied with temperature strips as an added safeguard.
Sweat patches were applied and removed according to the
PharmChekTM sweat patch package insert and training
video. For example, isopropyl alcohol (70%) swabs were
used by study personnel who wore new (clean), latex gloves
to prepare the patch application sites on the upper arms. In
addition to the manufacturer’s instructions, swabs were
saved and analyzed in the same manner as the sweat patch
pads. Two sweat patches were worn at all times, removed
and replaced on alternating arms every 3–4 days on a weekly
basis. Due to this alternating pattern, a patch was worn less
than 7 days during the first and last week of the 4-week study.
Patches were checked frequently; those that were not adher-
ing well were removed for testing and replaced. Any com-
promised patches were not included in the data analysis.
Participants were discouraged from tampering with patches
by loss of their weekly financial incentive (US$ 50 mer-
chandise gift card) when samples were not collected as
scheduled, including premature removal or compromised
patches. Participants who successfully completed the 4-
week study received one US$ 50 gift card at the end of
each week, plus a bonus US$ 50 gift card at the end of the
study, for a total of US$ 250 in gift card incentives. Speci-
mens were coded in a manner designed to prevent linking
specimen results to individual participants.
Even with the monetary incentive and anonymity, some
patches were compromised. Of the 97 patch results reported
in this study, two patches had an interior anomaly, such as
movement or tearing of the absorbent pad, but because
the outer membrane was intact and firmly attached to the
skin, these patches were considered uncompromised and their
results were considered valid. Five patches were removed and
replaced early because the patch was peeling back. Of these
five exposed patches, four were considered compromised. We
also observed tearing of the inner pad in our previous labora-
tory evaluations [6] involving heavy sweating and exercise.
In some drug use monitoring programs, authorities accuse the
wearer of violating a condition of their supervised release
when a patch is compromised. Based on the results of this
research, previous research, and the random nature of the
patch failure, it appears that such a position is untenable.
2.3. Extraction and analysis
Dried skin swabs or patches were placed in a 15 ml plastic
test tube held in place mechanically by a permeable divider
at the upper third of the test tube. (Patches were dried
because those still moist with isopropanol showed higher
background.) The swabs and patches were washed with two
1 ml portions of 0.1 M HCl separated by brief centrifugation
after each addition. Aliquots of either the swab or patch
extracts or the urine samples, varying from 10 to 1000 ml
(depending on the immunoassay results) were taken, diluted
with 2 ml of 0.1 M HCl, and then spiked with a deuterated
internal standard in ethanol. The aqueous extracts were
applied to DAU solid phase extraction (SPE) columns
(Ansys, Inc.) using a Zymark Rapid Trace automated SPE
extractor. The columns were conditioned with methanol,
0.1 M hydrochloric acid, and 20% aqueous acetone. The
columns were dried under positive pressure for 1 min, and
the drugs were then eluted with 40:10:1 methylene chlor-
ide:isopropanol:ammonium hydroxide. The eluate was then
concentrated to dryness under a stream of nitrogen and
derivatized using 70 ml 1% triethylamine in methylene
chloride, 50 ml acetic anhydride, and 20 ml pentafluoropro-
panol at 70 8C for 30 min. The excess derivatization reagents
were evaporated under a stream of nitrogen. The drugs were
reconstituted in 20 ml of ethyl acetate. Aliquots (2 ml) were
injected into a Varian 4 GC/MS with the following para-
meters: 30 m DB-5MS column (J&W Scientific), initial
temperature100 8C (20 s) ramped at 18 8C/min to 280 8C
then 5 8C/min to 300 8C for 2.9 min for a total run time of
17.1 min. Samples were ionized using isobutane chemical
ionization. Quantitation was performed by ratioing the peak
areas of the protonated molecular ions to their respective
deuterated internal standards. The limit of detections (LODs)
for the CI-GC/MS assay (calculated from the blanks, which
comprised 10% of all samples analyzed) were ca. 4 ng/ml for
cocaine and 2 ng/ml for BE (n ¼ 41 blanks). Alternatively,
the LODs from ‘‘blank’’ urines were 6 ng/ml for cocaine and
9 ng/ml for BE (n ¼ 73 from three individuals). LODs
determined from a series of different specimens/individuals
(rather than pooled specimens) are considered to be more
representative of the true LOD [17].
2.4. Immunoassays
Three immunoassays were investigated. Because the
manufacturers’ procedures were substantially modified to
accommodate the diverse matrices, the complete procedure
for each immunoassay is given. All were calibrated with
cocaine solutions in 0.1 M HCl at 5, 10, 50, 100, 500, and
1000 ng/ml. All assays were performed manually.
D.A. Kidwell et al. / Forensic Science International 133 (2003) 63–78
65
Case 1:21-cr-10270-MLW     Document 70-6     Filed 04/06/22     Page 3 of 16

2.5. Cozart cocaine metabolite microplate enzyme
immunoassay (EIA)
In an antibody-coated plate provided by the manufacturer
was added 75 ml of a 0.1 M phosphate buffer, pH 8.8,
followed by 25 ml of the acid specimen extract or urine
sample. The plate was incubated at room temperature for
15 min shaking, 100 ml of enzyme conjugate added, and the
plate shaken for 30 min. The plate was then washed at least
six times with a wash buffer, provided by the company. After
washing, 100 ml of substrate solution was added and the
plate shaken at room temperature for 30 min. During this
time, the plate may be read at 620 nm to confirm the
absorbances are within range (0–2 o.d.) by noting that the
reading at 620 nm is ca. 1/3 of the final reading at 450 nm.
After the last incubation period, 100 ml of an acid stop
solution is added and the plate read.
2.6. Microgenics cocaine CEDIA DAU immunoassay
This assay was greatly modified from the manufacturer’s
procedure to increase the sensitivity. The solutions come in
three packages, two of which are to be mixed together in the
normal assay. In this modified assay, they were kept separate
and mixed in the well at the appropriate times. A microtiter
plate was used as the optical wells. To each well, 10 ml of
sample was added, followed by 50 ml of antibody solution.
The plate was incubated at 37 8C for 20 min while shaking.
After 20 min, 50 ml of enzyme acceptor, dissolved in 0.1 M
PIPES buffer with calcium chloride, was added. The plate
was incubated at 37 8C for an additional 20 min. Enzyme
donor (50 ml) was added, the plate incubated at 37 8C while
shaking, and the plate read at 492 nm every 5 min. Normally,
a 20–30 min incubation provided an acceptable calibration
curve.
2.7. OraSure technologies intercept cocaine metabolite
microplate enzyme immunoassay (EIA)
All manufacturer’s solutions were prepared as directed. In
an antibody-coated plate provided by the manufacturer was
added 100 ml of a 0.1 M phosphate buffer, pH 8.8, followed
by 50 ml of sample. The plate was incubated at room
temperate for 15 min shaking. Then, 50 ml of enzyme con-
jugate was added, the plate shaken for 30 min, and the plate
then washed at least six times with a wash buffer (phosphate
buffered
saline þ 0:05%
Tween1-20).
After
washing,
100 ml of substrate solution was added and the plate treated
as for the Cozart cocaine ELISA.
2.8. Data analysis
Plotting the log[cocaine] vs. log[(absorbance of standard/
absorbance of blanks)  100/((100  absorbance of standard/
absorbance of blanks)  100)] produced linear logit plots.
For the Cozart assay, the correlation coefficients ranged from
0.8 to 1 (average ¼ 0:96, n ¼ 24). For the OraSure assay,
the correlation coefficients ranged from 0.91 to 1.0
(average ¼ 0:96, n ¼ 11). Several blanks (3–6) were run
for each plate with the LOD determined by the scatter in
the blank results as transformed by the logit plot. For the
modified, manual Microgenics CEDIA assay the best LOD
was 81 ng/ml, compared with LODs of 4 ng/ml for the Cozart
ELISA and 1.5 ng/ml for the OraSure ELISA.
3. Results and discussion
The drug use pattern can be estimated for cocaine by
monitoring daily urine levels of BE [18,19]. Briefly, a
negative urine specimen, followed by a positive specimen
of a significant amount, can be interpreted to indicate
cocaine use sometime after the last negative specimen
and before the positive specimen. For example, Participant
#10 (Fig. 1b) whose urine tested negative Friday morning but
positive on Saturday morning then having decreasing con-
centrations the following Monday, Tuesday, probably
ingested cocaine on Friday afternoon/night. Participant
#10 would be positive through Tuesday morning (4 days),
at occasionally used 100 ng BE/ml cut-off. In another exam-
ple, Participant #4 (discussed below and shown in Fig. 3),
whose urine tested negative Saturday morning, then decreas-
ingly positive the following Monday, Tuesday, and (depend-
ing on which cut-off concentration is employed) Wednesday
mornings, probably ingested cocaine Saturday or Sunday.
Without extensive knowledge of the participant’s metabo-
lism of cocaine and an estimate of the physiological dilution
of the urine, the analysis of urine BE levels cannot tell how
much cocaine was ingested during the use period, which may
comprise a single use or a binge use over a short period of
time. Although not a perfect measure of dilution of the urine,
the creatinine levels were tested in all specimens. None
showed the abnormally low creatinine concentrations that
would be consistent with large fluid intake (‘‘flushing’’).
Flushing (and urine adulteration) shortens the window of
detection for single dosages, affecting the reliability of
estimating the post-drug use interval.
Our results show that the time window for detecting
cocaine use from urinalysis, sweat, and skin swabs depends,
in part, on use pattern. Fig. 1 contrasts two participants—a
chronic cocaine user and an individual with an apparent
single use. Among active, chronic cocaine users, urine
results seldom if ever decrease to negative (below cut-off
concentration). For example, urine BE concentrations in
one participant (Fig. 1a), ranged from approximately
900–160,000 ng/ml urine, remaining positive throughout
the study. Although the cocaine use by Participant #1 would
be evident from any urine sample during this 28-day mon-
itoring period, the single use by Participant #10 would not. A
random urinalysis program of one sample/month would have
only a 10% (3 days per month) chance of detecting the use of
cocaine by Participant #1. Even a once-per-week monitoring
66
D.A. Kidwell et al. / Forensic Science International 133 (2003) 63–78
Case 1:21-cr-10270-MLW     Document 70-6     Filed 04/06/22     Page 4 of 16

program might not detect drug use in Participant #10 if the
specimens were taken only during weekdays and moderate,
physiological dilution of the urine had occurred. Because
high BE levels may not be reached for the occasional user
(Participant #10, plotted in Fig. 1a on the same scale as
Participant #1) physiological dilution of the urine would
more easily evade detection of drug use.
A device that can continuously monitor drug use, such as
the patch, would not have the window of detection problem
apparent in Fig. 1a. Additionally with detection of drugs in
sweat, physiological dilution that occurs with urine would
not be a concern. In the case of chronic cocaine users, the
patch worked quite well. The patch results for the three
chronic cocaine users, followed in this study, are shown in
Fig. 2. The patches of two participants were positives at quite
high levels (340–10,000 ng cocaine per patch for Participant
#1 and 160–21,000 ng cocaine per patch for Participant #9).
The patches of Participant #2 were lower (40–131 ng
cocaine per patch) as was the frequency of cocaine use,
according to spikes in the urine BE concentrations. Although
all these urine samples were positive at the LOD of GC/MS,
two specimens contained less than 300 ng but more than
100 ng BE/ml urine for Participant #9. Two specimens
contained 77 ng BE/ml and 271 ng BE/ml for Participant #2.
Of great concern to the forensic analysis is a false
positive. Two such instances of false patch positives are
Fig. 1. Comparison of chronic (A) vs. occasional use (B). Consistently high-positive urine results are characteristic of chronic cocaine use.
Urine positives spanning 3–4 days and returning to negative are characteristic of occasional use. Participant #10, in Fig. 1B, is also plotted in
Fig. 1A to emphasize the different BE levels. Because of the high BE values, the window of detection of cocaine use in Participant #1, a
chronic user, is much longer than that of Participant #10, an occasional user.
D.A. Kidwell et al. / Forensic Science International 133 (2003) 63–78
67
Case 1:21-cr-10270-MLW     Document 70-6     Filed 04/06/22     Page 5 of 16

Fig. 2. Urine BE levels and patch cocaine levels of chronic users. Chronic cocaine use patterns may include short periods of urine specimens
testing below cut-off concentrations. For Participant #1, no urine samples had a BE concentration <900 ng/ml. For Participant #9, 22/24 urine
samples were >300 ng BE/ml and 24/24 were >100 ng BE/ml. For Participant #2, 20/22 urine samples were >300 ng BE/ml and 21/22 were
>100 ng BE/ml. Sweat patches tested positive throughout the study at high levels (compared to other participants) showing a range of 164–
22,000 ng per patch for Participant #1, 390–10,000 ng per patch for Participant #9, and 40–240 ng per patch for Participant #9.
68
D.A. Kidwell et al. / Forensic Science International 133 (2003) 63–78
Case 1:21-cr-10270-MLW     Document 70-6     Filed 04/06/22     Page 6 of 16

shown in Fig. 3; Participant #7’s urine specimens tested
consistently negative, yet two patches were above the
recommended 25 ng per patch commercial cut-off. The
positive patches followed several negative patches, indicat-
ing that carry-over of drugs in the skin from prior use is
probably not the explanation for these two positive sweat
patch results. Additionally, since the skin swabs prior to
patch application were consistently blank, there is no evi-
dence that CFWI (cocaine on the skin prior to patch applica-
tion) caused these positive patch results. Since prior research
shows that CFWO (cocaine penetrating the protective
membrane while patches are worn) can occur [6,7] and does
occur with patches applied to research volunteers [10], it is
reasonable to conclude that these positive patch results are
consistent with cocaine contamination of the external patch
membrane. Furthermore, the pattern in this example of
decreasing cocaine in patches over time suggests decreasing
exposure to environmental contamination over time. Traces
of BE also were detected in the cocaine positive patches
from this participant illustrating that BE can result from
sources other than ingestion. In vitro degradation of cocaine
to BE has been demonstrated for cocaine applied to hair
(another protein matrix) and human skin [6,7,20].
Other than Participant #7, only two other participants had
completely negative urines during the study period. All 18
patches from these other two participants tested negative
for cocaine at a 25 ng per patch cut-off level. Taken as a
group, the results from the three participants, where cocaine
abstinence is evident throughout the study, show that the
patch correctly reported cocaine use 25/27 times (93%) and
incorrectly reported cocaine use 2/27 times (7%). Both
incorrect/false positive patches from Participant #7 appear
to indicate that this volunteer was environmentally exposed
to cocaine.
An analysis of the commercial cut-off level for the patch
as applied to the participants in this study is given in Table 1.
Raising the cut-off level from 25 to 75 ng per patch increased
the specificity of the assay to 100% without substantially
decreasing the efficiency of the patch. Nevertheless, it is
unlikely that a simple cut-off level for the patch would be
definitive of use because of the difficulty in cleaning the skin
of prior drug residues and the potential to contaminate an
individual with arbitrarily high concentrations of drugs.
Besides false positives, false negatives also occur with the
patch. Fig. 4 shows the urine and patch results from Parti-
cipant #4; based on urine results, this participant likely used
cocaine twice during the study period. The second use
occurred after the urine sample was taken on Saturday
and was detected by urinalysis Monday. The patch correctly
detected cocaine use during the early part of the study but
both patches missed the second use. False negatives occur
frequently with the patch due to poor transfer of drugs from
the skin to the patch [10,13,21]. We have suggested the use
of glycerol to increase transfer [13], and a full report will be
published later. Increasing the commercial cut-off to 75 ng
per patch to reduce false positives would not have changed
Fig. 3. Participant #7 showing false positive sweat patches with negative urine BE levels. Consistently blank urine specimens accompanied by
above cut-off positive sweat patches are consistent with environmental contamination and inconsistent with cocaine ingestion. The four
patches, with no cocaine present, are indicated by double arrows and offset from the X-axis for clarity.
D.A. Kidwell et al. / Forensic Science International 133 (2003) 63–78
69
Case 1:21-cr-10270-MLW     Document 70-6     Filed 04/06/22     Page 7 of 16

the analysis for this individual (a substantially lower cut-off
level would be necessary to detect the second, apparently
single use).
Interpretations of the second possible use (or inadvertent
ingestion) of cocaine by Participant #4, depicted in Figs. 4
and 6, are ambiguous. In support of use during the weekend
is that the concentration of BE in urine on Monday (2200 ng/
ml) is consistent with 1–2 days after cocaine use. Addition-
ally, both the skin swabs and patches recorded some cocaine
use or exposure during this time. Alternatively, inadvertent
ingestion may have occurred (as discussed below). The low
BE levels present in the urine are consistent with oral
ingestion of 25 mg of cocaine 1 day prior to testing or as
little as 1–2 mg a few hours before testing [3,4]. Monitoring
of low to moderate level (above 1–2 mg but below the 50–
100 mg required for a substantial physiological effect) of
cocaine ingestion through the use of skin swabs or patches
has not been investigated in the published literature and
needs of further research.
Previous use is a concern if an individual abruptly stops
cocaine use and the patch may be positive for an unknown
period after cocaine cessation. In a legal setting, the issue of
when drug use occurs is crucial. For example, judges require
drug abstinence as a standard condition of probationary
release. Prior drug use may be irrelevant to meeting this
condition. The probationer is required to remain drug-free
from the time of sentencing or release onward. If the sweat
patch is used as a stand-alone test for determining drug use
Table 1
Statistical parameters for the patch from all 10 participants
Patch/urine cut-off (ng per patch or ml)
25/100
50/100
75/100
25/300
50/300
75/300
True positive
33
30
26
30
28
26
False positive
4
1
0
7
3
0
False negative
9
12
16
8
10
12
True negative
51
54
55
52
56
59
Sensitivity (%)
78.6
71.4
61.9
78.9
73.7
68.4
Specificity (%)
92.7
98.2
100.0
88.1
94.9
100.0
Efficiency (%)
86.6
86.6
83.5
84.5
86.6
87.6
Sensitivity ¼ ðTPÞ=ðTP þ FNÞ; specificity ¼ ðTNÞ=ðTN þ FPÞ; efficiency ¼ ðTN þ TPÞ/total number of patches.
Fig. 4. Occasional use was easily detected by frequent urinalysis in Participant #4. Sweat patches were positive when worn during the first
documented use period, but not during the second use period. Meanwhile, one positive sweat patch was worn during a period of consistently
negative urine results. The two patches, with no cocaine present, are indicated by double arrows and offset from the X-axis for clarity.
70
D.A. Kidwell et al. / Forensic Science International 133 (2003) 63–78
Case 1:21-cr-10270-MLW     Document 70-6     Filed 04/06/22     Page 8 of 16

status during the period that the sweat patch is worn, then it
is essential that the patch does not falsely report previous
drug use as ‘‘current drug use’’. In Fig. 4, it appears that
previous cocaine use by Participant #4, including that docu-
mented by urinalysis during the first week of this study,
may have caused or contributed to some cocaine in the
sweat patch (barely negative at the 25 ng per patch cut-off)
result during the second week, when urine results were
consistently blank.
For chronic users, as shown in Fig. 2, it is not clear
whether cocaine appearing in the patches came from current
drug ingestion, previous drug ingestion, previous drug con-
tamination, current drug contamination, or a combination of
the above. From previous research, drugs were shown to bind
to skin and persist for many days [7,12]. We proposed to save
and analyze the ‘‘cleaning’’ swabs to help distinguish the
source of the cocaine [6,7]. Fig. 5 shows the results for
selected ‘‘cleaning’’ swabs from these three chronic cocaine
users. The coloration on the clean swab is a combination of
dirt and dead skin cells. Because melanin containing skin
cells are easier to observe, some individuals may have more
cleaning applied than others. In this study, the skin was
cleaned until the cleaning pads appeared visually clean (note
variable number of swabs used for cleaning in Fig. 5),
according to the sweat patch instructions. Even with four
swabbings (Participant #9), cocaine is still present. Based on
previous research [6,7] this contamination on the cleaning
pads likely will be accompanied by cocaine persistence on
skin sufficient to cause positive patch results even if this
individual had ceased cocaine use. Although the sequential
patches were placed in similar locations to those removed,
the skin contamination (removed by the swabs) may origi-
nate from use rather than from exposure during use or from
the environment. Alternatively, drug may move on the skin
during the swabbing process, allowing drug in skin areas not
covered by one patch to contaminate subsequent patches.
Given the high levels of cocaine found in the patches of
these two users (up to 21,000 ng cocaine per patch), prior
skin contamination likely contributed some of the cocaine in
the patches. For Participant #1, the cocaine present in the
cleaning swabs is a substantial fraction of that found in the
patch (Fig. 5). In prior publications, we proposed an arbitrary
10:1 ratio for patch:swab drug concentration to distinguish
CFWI from use [7,13]. If this criterion were applied to
Participant #1, four of the seven patches would have been
positive and three negative (indicated by asterisks in Fig. 5).
If this criterion were applied to Participant #9, three of the
four patches would have been positive and the fourth would
have been negative. However, this patch was on the volun-
teer for only 3 days compared to the 5–7 days for most other
patches. Also, based on the urine data, Participant #9 used
cocaine only once (or during 1 day as the frequency and
quantity within the time frame cannot be estimated) during
the time represented by the wearing of this patch. In addition,
based on the excretion curve and the presence of cocaine in
the urine, the use was likely to have occurred over a holiday
and a mere 1 day before the patch was removed. Either
reason could account for the lower levels of cocaine (237 ng)
in the patch. Likewise for Participant #1, the patches that
would fail the 10:1 criterion are those worn during lower
cocaine use periods. For Participant #2 a similar number of
patches fail this criterion. Of the three chronic cocaine users,
Participant #2 used cocaine the least often and had lower
amounts of cocaine in the patches.
Skin swabs measure both exposure and use of cocaine.
Besides swabbing the area of patch placement, daily skin
swabs of the hands and foreheads of the participants were
also obtained. Fig. 6 shows the skin swab cocaine results
from two participants as compared to their urinalysis results.
For Participant #1, a chronic cocaine user, the skin swabs
were also consistently positive and at high levels (micro-
grams of cocaine per swab). Likewise, skin swabs for
Participant #4 are consistent with cocaine use on the day
before the holiday. The skin swabs on the following day also
were positive at fairly high levels. Additionally, the skin
swab for the second likely use by Participant #4, indicated by
an arrow in Fig. 6, appears positive but far below that of the
earlier use. Although the timing of cocaine use may be
speculative, if the use occurred on the Saturday, 2 days
would have intervened between use and skin swabs. How-
ever, as expected for a matrix that measures environmental
exposure, there are several cases where the skin swabs are
positive but the urine samples are negative. No correlation
could be found for the quantities of cocaine on the skin and
the BE levels in urine. This is reasonable because the skin
swabs measure environmental contamination AND removal
rate by hygiene. While environmental contamination that
affects sweat patch results should be correlated with the
incident of use, it should not be correlated with the amount of
cocaine used.
For most of the non-using participants in this study, skin
contamination was low, while continual users had substan-
tial skin contamination. Frequent cocaine users contaminate
themselves with high levels of drug (Figs. 5 and 6). It is
conceivable that individuals that live with drug users but who
are not users themselves could also become contaminated
[11]. Therefore, skin swabs should only be used as an
indicator of the potential for cocaine use and as basis to
obtain other biological specimens. For example, an indivi-
dual with no cocaine on their skin is an unlikely user of
cocaine and need not be subjected to additional testing with
an alternative testing procedure such as urinalysis. Alter-
natively, an individual who has cocaine present may be a
cocaine user and urinalysis should be the deciding factor.
One such individual may be Participant #7. This volunteer
claims to have a spouse who is a cocaine user. Some of
Participant #7’s skin swabs (Fig. 7a) reach substantial values
(2117 ng cocaine per swab for fingertips and 344 ng per
swab for forehead) whereas his/her urine BE levels are low
or negative.
Table 2 gives the statistical parameters for the fingertip
and forehead swabs for all the participants in this study.
D.A. Kidwell et al. / Forensic Science International 133 (2003) 63–78
71
Case 1:21-cr-10270-MLW     Document 70-6     Filed 04/06/22     Page 9 of 16

The efficiency of swabs is high for detecting cocaine users
(86–88% at a cut-off of 100 ng cocaine per swab). By
excluding Participant #8 from the statistics, the false positive
rate drops by about 50% and the efficiency increases 1–2%.
Clearly, skin swabs are not definitive of drug use for
individuals living in a drug-using environment yet abstaining
from drug use. Also, Participant #8 is responsible for three of
the seven false positive patches for the total cohort shown in
Fig. 5. Selected, matched swab and patch levels of chronic users. Each bar series represents swabs used to clean the skin prior to applying the
patches and the subsequent patch results. The patch results are divided by 10 to keep the data on scale and make visual comparison to the
swabs easier. Asterisks indicate series that failed the 10:1 criterion and n/a (not available) indicate missing data because swabs were not taken.
Positive arm swabs from the cleaning prior to patch placement often accompany positive sweat patches.
72
D.A. Kidwell et al. / Forensic Science International 133 (2003) 63–78
Case 1:21-cr-10270-MLW     Document 70-6     Filed 04/06/22     Page 10 of 16

Table 1, indicating that sweat patches should be used with
caution in such an environment. Interestingly, all these
patches would have been considered contaminated using
the 10:1 patch:swab criteria, discussed above.
Incorrect interpretations of positive results can occur with
urine testing as well as patch testing, if too low a cut-off were
applied. Low urine positives have occurred from the casual
handling of drugs [22]. In this research, several participants
Fig. 6. Skin swabs of hands and forehead with comparison to urine levels for Participants #1 and #4. As expected, almost no linear correlation
was observed for Participant #1 of hands:urine (r2 ¼ 0:03) and forehead:urine (r2 ¼ 0:07).
Table 2
Statistical parameters for the for skin swabs using urine BE >100 ng/ml
Swab cut-off (ng per swab)
Fingertips 50
Fingertips 100
Fingertips 250
Forehead 50
Forehead 100
Forehead 250
True positive
69
61
48
57
48
38
False positive
28
11
4
10
6
3
False negative
9
17
30
14
23
33
True negative
125
142
149
124
128
131
Sensitivity (%)
88.5
78.2
61.5
80.3
67.6
53.5
Specificity (%)
81.7
92.8
97.4
92.5
95.5
97.8
Efficiency (%)
84.0
87.9
85.3
88.3
85.9
82.4
D.A. Kidwell et al. / Forensic Science International 133 (2003) 63–78
73
Case 1:21-cr-10270-MLW     Document 70-6     Filed 04/06/22     Page 11 of 16

volunteered information that their spouse or significant other
continued cocaine use during their abstinence. In one such
volunteer,
two
urine
specimens
contained
detectable
amounts of BE, above the 100 ng BE/ml cut-off and below
the 300 ng BE/ml cut-off (Fig. 7b). Based on the urine BE
levels and no sign of dilute urine (creatinine normal), this
participant is unlikely to have used physiologically active
amounts of cocaine. Three sweat patches during these
periods were positive (above 25 ng cocaine per patch) but
below 75 ng cocaine per patch. In addition, the swabs used in
‘‘cleaning’’ for these three patches had trace amounts of
cocaine present, indicating prior exposure and casting suspi-
cion on the patch results. These data suggest that the higher
cut-offs (300 ng BE/ml urine and 75 ng cocaine per patch)
would provide greater protection to individuals from allega-
tions of intentional drug use.
Other derivatives of cocaine (BE, cocaethylene, and
methyl ecgonine) also appeared frequently in the skin swabs
of Participant #1 and have been attributed to human meta-
bolism and excretion in the sweat [23]. These substances
tended to be most prevalent when the cocaine concentration
was high. Although their source is unknown, they may
originate from placement in the environment during prior
cocaine use followed by replacement on the skin. For
Participant #4, the BE concentration was not useful in
distinguishing use from exposure as it paralleled that of
the cocaine concentration. The other two cocaine derivatives
were not measurable for this individual. Thus, the usefulness
of metabolites to determine the source of the drugs in the
patch is not clear in this population.
Immunoassays are often used to screen specimens and
any positive samples submitted to GC/MS confirmation.
Fig. 7. Swab, Urine and patch data for Participant #8 claiming inadvertent ingestion. Two patches were positive (at 25 ng per patch) during
this period. Forehead and fingertip swabs often showed the presence of cocaine.
74
D.A. Kidwell et al. / Forensic Science International 133 (2003) 63–78
Case 1:21-cr-10270-MLW     Document 70-6     Filed 04/06/22     Page 12 of 16

Thus, an immunoassay that has few false negatives (to avoid
discarding positive samples) and one that has few false
positives (to avoid a heavy workload on the more time-
consuming and expensive GC/MS) would be desirable.
This study allowed evaluation of immunoassays on matrices
not commonly tested (skin swabs and patches). For cost
considerations, a modification of the Microgenics CEDIA
cocaine assay for urine was attempted to adapt it to these
Table 3
Statistical parameters for the swab ELISA assays using GC/MS cocaine >10 ng/ml
ELISA Cut-off (ng/ml)
Cozart-25
Cozart-50
Cozart-75
OraSure-25
OraSure-50
OraSure-75
True positive
19
16
14
7
4
2
False positive
32
15
9
5
1
1
False negative
3
6
8
0
3
5
True negative
101
118
124
20
24
24
Sensitivity (%)
86.4
72.7
63.6
100.0
57.1
28.6
Specificity (%)
75.9
88.7
93.2
80.0
96.0
96.0
Efficiency (%)
77.4
86.5
89.0
84.4
87.5
81.3
Table 4
Statistical parameters for the patch ELISA assays using GC/MS cocaine >10 ng/ml
ELISA Cut-off (ng/ml)
Cozart-25
Cozart-50
Cozart-75
OraSure-25
OraSure-50
OraSure-75
True positive
15
15
11
8
8
7
False positive
5
1
0
1
1
0
False negative
1
1
5
0
0
1
True negative
45
49
50
9
9
10
Sensitivity (%)
93.8
93.8
68.8
100.0
100.0
87.5
Specificity (%)
90.0
98.0
100.0
90.0
90.0
100.0
Efficiency (%)
90.9
97.0
92.4
94.4
94.4
94.4
Fig. 8. Correlation of Cozart and OraSure ELISAs for skin wipes. Only those samples whose quantitations were in the linear range of the
assays were compared.
D.A. Kidwell et al. / Forensic Science International 133 (2003) 63–78
75
Case 1:21-cr-10270-MLW     Document 70-6     Filed 04/06/22     Page 13 of 16

other matrices and increase its sensitivity. Although it was
capable of detecting 1 ng/ml of cocaine under certain cir-
cumstances, precise timing requirements made the precision
poor when it was run in the manual mode. Two ELISA
cocaine assays were evaluated. Both ELISAs gave similar
results, with the OraSure ELISA having slightly better
efficiency than the Cozart ELISA (Tables 3 and 4). The
intraday %CVs for the Cozart ELISAwas 22% (at 10 ng/ml,
n ¼ 29) and 13% (at 50 ng/ml, n ¼ 25). The intraday %CVs
for the OraSure ELISA was 28% (at 10 ng/ml, n ¼ 12) and
24% (at 50 ng/ml, n ¼ 12). The linear correlation for the
Cozart ELISA versus the OraSure ELISA showed an r2 of
0.79 (Fig. 8). This is consistent with similar cross-reactiv-
ities with cocaine-related substances and immunologically
similar materials. The slope of 1.3 also indicates similar
cross-reactivities.
4. Conclusions
Three patterns of drug use (chronic, occasional, and no-
use) were identified readily by daily urinalysis; however,
patch results were less clear. The patch identified some of the
occasional cocaine use episodes and virtually all of the
frequent, chronic uses. Some positive patches from partici-
pants could not be attributed to cocaine use while the patch
was worn. This confirms, in a real life environment, previous
experiments regarding external contamination and false patch
positives. In some of these positive patch/negative urines
instances, skin swabs identify cocaine and BE on the skin
prior to the application of patches, which later tested positive.
Prior skin contamination (CFWI) is the only identifiable
source of these positive patch results. In other positive
patch/negative urine results, skin swabs did not reveal
cocaine-related substances, suggesting that external contam-
ination of the patch (CFWO) may also play a role. Further-
more, in cases of concurrent cocaine use, it is not known how
much of the cocaine and BE in patches originated from
CFWI, CFWO, or within the body (from ingestion), as many
of the ‘‘cleaning’’ swabs from frequent cocaine users had
substantial amounts of cocaine and BE present.
In prior publications [6,7] we had suggested saving the
‘‘cleaning’’ swabs and testing them for the presence of
cocaine if exposure was suspected. A ratio of >10:1, patch:
swab,was proposed to screenforactive use. In the casesfound
here, that ratio appears reasonable and warrants further study
and corroboration. Additionally, it is proposed to increase the
sweat patch cut-off concentration. Based on this research with
a limited number ofparticipants,a cut-off of75 ngcocaine per
patch would reduce the false positive rate. Both criteria must
be considered because of it is difficult to clean the skin of prior
drug residues and arbitrarily high levels of cocaine may be
reached in contaminated environments.
Both the Cozart and OraSure cocaine immunoassays
performed similarly and showed a reasonably strong corre-
lation with each other. In contrast, although the modified
Microgenics assay showed the requisite sensitivity for the
matrices examined, it had poor precision when run in a
manual mode. Unfortunately, automation was unavailable to
allow more complete evaluation.
The most reliable method for detecting drug use appears
to be daily urinalysis, followed by frequent skin wipes,
which are approximately equivalent to sweat patch results.
Intermittently, both skin wipes and sweat patches will miss
drug use as well as wrongly indicate drug use. Where
external contamination is an issue, urinalysis will provide
more reliable proof of drug use.
Numerous legal challenges have asserted that positive
patch results were not the result of current drug ingestion.
Most of the individuals in these cases had some number of
negative urine results to buttress their legal positions of not
using drugs. In cases lacking frequent urine results, an expert
could argue against external contamination as the cause of
positive patches by claiming the defendant’s cocaine use was
undetected by urinalysis. This study clearly demonstrates
that patch positives can arise under real life conditions from
sources other than drug use; therefore, interpretation of
sweat patch results must proceed with caution.
Acknowledgements
This research was supported by US Naval Surface War-
fare Center grant number N00178-01-1-9002. Opinions
expressed are strictly those of the authors and do not
necessarily reflect those of the US Department of Defense
or the US Government. The authors thank Rachel Gill for
technical assistance and helpful comments with manuscript
preparation and Lissette Drinkwater, Celina Rodrigues, and
Inas Mahdi for assistance with participants.
Appendix A. Some legal cases
November 1999, US v. McLemore, Las Vegas, NV. US
District Judge Lloyd George, Franny Forsman, Federal
Public Defender, Las Vegas, NV. The US Attorney withdrew
their prosecution after PharmChem testified about their in-
house research. McLemore’s frequent (including every-
other-day) urine specimens tested negative while her sweat
patches tested positive. She lived in a contaminated apart-
ment with her young child, playing with child on the carpet.
The apartment manager testified that previous tenants left
behind cocaine paraphernalia sufficient to fill three large
trash bags.
January 2000, US v. Self, Kansas City, MO. Federal
District Magistrate Judge Rob Larson, Laine Cardarella,
Assistant Federal Public Defender. Self’s positive sweat
patch results conflicted with occasional negative urinalysis
results. Self had extensive contamination, sweated on the
job, and used lotion for scabies. Judge Larson ruled not to
revoke.
76
D.A. Kidwell et al. / Forensic Science International 133 (2003) 63–78
Case 1:21-cr-10270-MLW     Document 70-6     Filed 04/06/22     Page 14 of 16

March 2000, US v. Gillette, Kansas City, MO. Federal
District Magistrate Judge Rob Larson, Laine Cardarella,
Assistant Federal Public Defender, Kansas City, MO. After
testimony by the expert called by the prosecution, Judge
Larson stopped the proceedings before the scheduled testi-
mony of the expert to be called by the defense. Judge Larson
ruled not to rely on positive sweat patch results to revoke
Gillette’s probation.
June 2001, US v. Snyder, Syracuse, NY. Federal District
Judge Howard Munson, James Greenwald, Assistant Federal
Public Defender, Syracuse, NY. A cohabitant of Snyder’s
home used crack cocaine during positive patch periods.
Snyder sweated on the job and working out; his patch
was exposed during both. Judge Munson ruled not to rely
on positive sweat patch results to revoke probationary
release. Judge Munson’s published decision is located at
the following web address: http://www.nysd.uscourts.gov/
courtweb/pdf/D02NYNC/02-01702.pdf.
December 2001, US v. Redd, Kansas City, MO. US
District Judge Gary A. Fenner, Defense Attorney Bruce
Houdek, AUSA Marietta Parker. Judge Fenner ruled in favor
of patch test results as evidence of intentional drug use
during the patch wear period. While a prosecution expert
testified, no defense expert testified.
May 2002, State of California v. Ian C., Sonora, CA.
Presiding Judge Eric DuTemple, Defense Attorneys Dennis
Dunn (Office of the Public Defender, County of Tuolumne)
and Julian Gross (Drug Policy Alliance, Oakland, CA).
The defendant’s apartment was inhabited previously by a
methamphetamine user. The defendant sweated on the job.
The defendant’s patch tested positive for methamphetamine
while limited urinalysis was negative. Judge Eric DuTemple
ruled that the proceedings would be expanded into a Kelly-
Frye hearing to determine the admissibility of the sweat
patch. Subsequently, the prosecution withdrew their petition
to terminate the defendant’s parental child custody rights.
October 2002, US v. Zubeck, Kansas City, MO. US
District Judge Dean Whipple, Defense Attorney Elizabeth
Carlyle. Zubeck admitted to recent binge methamphetamine
use, but not during patch wear period. Zubeck reported heavy
sweating. Questions concerning Zubeck’s credibility were
raised through other testimony. Zubeck’s patch tested positive
for methamphetamine while limited urinalysis was negative.
Issues raised included longer-than-recommended sweat patch
wear period and other collection site concerns. Judge Whip-
ple ruled in favor of patch test results as evidence of inten-
tional methamphetamine use during the patch wear period.
References
[1] K.L. Preston, M.A. Huestis, C.J. Wong, A. Umbricht, B.A.
Goldberger, E.J. Cone, Monitoring cocaine use in substance-
abuse-treatment patients by sweat and urine testing, J. Anal.
Toxicol. 23 (1999) 313–322.
[2] J. Ambre, T.T. Ruo, J. Nelson, S. Belknap, Urinary excretion
of cocaine, benzoylecgonine, and ecgonine methyl ester in
humans, J. Anal. Toxicol. 12 (6) (1988) 301–306.
[3] G.F. Jackson, J.J. Saady, A. Poklis, Urinary excretion of
benzoylecgonine following ingestion of Health Inca Tea,
Forensic Sci. Int. 49 (1) (1991) 57–64.
[4] R.C. Baselt, R. Chang, Urinary excretion of cocaine and
benzoylecgonine following oral ingestion in a single subject,
J. Anal. Toxicol. 11 (2) (1987) 81–82.
[5] M. Burns, R.C. Baselt, Monitoring drug use with a sweat
patch: an experiment with cocaine, J. Anal. Toxicol. 19 (1)
(1994) 41–48.
[6] D.A. Kidwell, F.P. Smith, Susceptibility of PharmChek(tm)
drugs of abuse patch to environment contamination, Naval
Research Laboratory Memorandum Report #6170-99-8414, 3
November 1999.
[7] F.P. Smith, D.A. Kidwell, Susceptibility of PharmChekTM
drugs of
abuse
patch to environmental
contamination,
Forensic Sci. Int. 116 (2–3) (2001) 89–116.
[8] G. Skopp, L. Potsch, H.P. Eser, M.R. Moller, Preliminary
practical findings on drug monitoring by a transcutaneous
collection device, J. Forensic Sci. 41 (6) (1996) 933–937.
[9] D.J. Crouch, C.L. Metcalf, M.H. Slawson, J. Baudys, An
assessment of the potential for vapor phase contamination of
the PharmChekTM sweat patch, Bull. Int. Assoc. Forensic
Toxicol. 32 (3) (2002) 7–10.
[10] N. Uemura, L. Lester, R.P. Nath, J. Mendelson, M. Harkey,
G.L. Henderson, Cocaine levels in sweat patches decline over
time, in: Abstracts of Papers Presented at the 2001 Annual
Meeting of the American Society for Clinical Pharmacology
and Therapeutics, Orlando, FL, March 2001, p. 62 (Abstract
PII-117).
[11] F.P. Smith, D.A. Kidwell, Cocaine in hair, saliva, skin swabs,
and urine of cocaine users’ children, Forensic Sci. Int. 83
(1996) 179–189.
[12] D.A. Kidwell, M.A. Blanco, F.P. Smith, Cocaine detection in
a university population by hair analysis and skin swab testing,
Forensic Sci. Int. 84 (1997) 75–86.
[13] M. Long, D.A. Kidwell, Improving the PharmChekTM sweat
patch: reducing false positive from environmental contamina-
tion and increasing drug detection, NRL Memorandum
Report #6170-01-8597, 19 December 2001.
[14] D.J. Crouch, C.L. Metcalf, M.H. Slawson, An assessment of
the effectiveness of the PharmChekTM sweat patch skin
cleansing procedures, Bull. Int. Assoc. Forensic Toxicol. 32
(2) (2002) 5–8.
[15] J.A. Levisky, D.L. Bowerman, W.W. Jenkins, D.G. Johnson,
J.S. Levisky, S.B. Karch, Comparison of urine to sweat patch
test results in court ordered testing, Forensic Sci. Int. 122 (1)
(2001) 65–68.
[16] H. Weingarten, Sweating the patch: a case report, in:
Proceedings
of
the
Paper
Presented
at
the
California
Association of Toxicologists Annual Meeting, 3 August
2002, San Jose, CA, vol. 30, no. 3, 2002.
[17] F.P. Smith, S.A. Reuschel, K.C.Jenkins, Online opiate immuno-
assay evaluation: precision, Sci. Justice 35 (1) (1995) 65–71.
[18] M.A. Huestis, E.J. Cone, Differentiating new marijuana use
from residual drug excretion in occasional marijuana users, J.
Anal. Toxicol. 22 (6) (1998) 445–454.
[19] B.A. Goldberger, B. Loewenthal, W.D. Darwin, E.J. Cone,
Intrasubject
variation
of
creatinine
and
specific-gravity
D.A. Kidwell et al. / Forensic Science International 133 (2003) 63–78
77
Case 1:21-cr-10270-MLW     Document 70-6     Filed 04/06/22     Page 15 of 16

measurements in consecutive urine specimens of heroin users,
Clin. Chem. 41 (1) (1995) 116–117.
[20] G. Romano, N. Barbera, I. Lombardo, Hair testing for drugs of
abuse: evaluation of external cocaine contamination and risk of
false positives, Forensic Sci. Int. 123 (2-3) (2001) 119–129.
[21] K.L. Preston, M.A. Huestis, C.J. Wong, A. Umbricht,
B.A. Goldberger, E.J. Cone, Monitoring cocaine use in
substance-abuse-treatment
patients
by
sweat
and
urine
testing, J. Anal. Toxicol. 23 (1999) 313–322.
[22] M.A. ElSohly, Urinalysis and casual handling of marijuana
and cocaine, J. Anal. Toxicol. 15 (1) (1991) 46.
[23] E.J. Cone, M.J. Hillsgrove, A.J. Jenkins, R.M. Keenan, W.D.
Darwin, Sweat testing for heroin, cocaine, and metabolites, J.
Anal. Toxicol. 18 (1994) 298–305.
78
D.A. Kidwell et al. / Forensic Science International 133 (2003) 63–78
Case 1:21-cr-10270-MLW     Document 70-6     Filed 04/06/22     Page 16 of 16

File and source

File

gov.uscourts.mad.237964.70.6.pdf

Size

450,353 bytes

SHA-256

8f5a7729df96cc9c3902879d3adcbca9044259406fd4b6b8abc21369ae253dd0

Our copy

gov.uscourts.mad.237964.70.6.pdf

Original

PACER (login required)