Forensic

ForensicScienceInternational
Science
ELSEVIER 84 (1997) 87-111 Intern3iond

Distinguishing passive contamination from

active cocaine consumption: assessing the
occupational exposure of narcotics officers to
cocaine

Tom Mieczkowski
Department of Criminology, Uniuersi~ of South Florida, Florida, USA

Abstract

Hair analysis has been used in probationary and parole populations to monitor for
cocaine use, but only in very limited settings or circumstances. Its wider adoption has been
limited by questions regarding the ability to distinguish environmental contamination of hair
via casual contact from actual ingestion. To evaluate this capability we sought to identify
persons routinely exposed to cocaine, who were not cocaine users. Undercover narcotics
officers engaged in cocaine-centered enforcement activities and evidence room clerks who
have no history of cocaine use were identified as an appropriate example population.
Thirty-six active undercover officers and four evidence technicians were asked to voluntarily
submit hair samples for analysis. Additionally two cocaine contaminated (aqueous soaked),
three negative control samples, and hair from a self-reported crack smoker were also blindly
submitted to the testing laboratory. The hair samples were washed and after washing,
enzyme digested. The wash solutions and hair digest were each analyzed for the presence of
cocaine. The results indicate that nearly every person had trace amounts of cocaine
contamination in the wash fraction, and one person had cocaine present in their hair digest.
That person, when retested, was a negative. The laboratory correctly identified and charac-
terized the contaminated, negative, and positive controls. The study concludes that the
findings support the capability of hair analysis to distinguish cocaine use from exposure
under normal field conditions. The study results indicate that cocaine-abstinent persons who
are in chronic, casual environmental contact with cocaine are not likely to test hair positive
for cocaine using the analysis protocols followed in this project. The study also indicates that
passive microingestion of cocaine needs to be considered when examining persons who are
in cocaine intensive environments. 0 1997 Elsevier Science Ireland Ltd. All rights reserved

Keywords: Passive contamination; Active cocaine consumption; Occupational exposure;

Narcotics

0379-0738/97/$17.00 0 1997 Elsevier Science Ireland Ltd. All rights resewed

PZZ SO379-0738(96)02052-X
88 T Meczkowski /Forensic Science International 84 (1997) 87-111

1. Introduction

Cocaine is a drug which appears to be used at very high rates in populations

undergoing treatment for drug abuse as well as criminal offender populations.
However, it is relatively difficult to detect cocaine by urinalysis except in the
immediate day or two after it is consumed. As a consequence the true prevalence
rate for cocaine use remains unknown, and estimates for this rate which rely solely
on reporting of use have consistently proven to be underestimates [ll. Further-
more, cocaine’s rapid excretion rate also makes defeating urine testing relatively
easy. For example, in many criminological settings (such as probation or parole
management) the combination of high caseload, the client’s ability to delay an
appearance for a testing appointment, or deliberate evasive maneuvers such as the
use of diuretics, have all contributed to a relatively low credibility of the true
detection efficacy of urine testing in routine monitoring circumstances [2,3]. A
nascent industry has emerged, devoted to helping persons ‘beat’ their urinalysis
tests, and operates openly and legally in most major urban areas around the United
States.
Hair analysis has been proposed as a desirable alternative or supplement to
urine testing. It has already been employed as a monitoring technology in several
criminal justice contexts, including intensive probationary supervision programs,
work release programs, experimental programs designed to monitor routine proba-
tioners, and pretrial diversion programs. A literature has also accumulated on the
effectiveness of hair analysis in these roles, and generally these reviews have been
positive, especially in regards to the increased ability to detect cocaine exposure
[4-lo]. Hair analysis has also withstood court examination of its suitability as
evidence. To date hair assaysresults have been accepted in most criminal and civil
cases in which they have been introduced as evidence, up to the level of the
Federal District Court (see for example, Nevada Employment Security Department
v Cynthia Holmes, Supreme Court, State of Nevada 26157 [HI or United States v.
Anthony Medina. 1990. 749 F. Supp. 59; US District Court, New York [12]).

2. The fundamentals of hair assay technology

Hair analysis is based upon the premise that many drugs become entrapped and
stabilized in the keratin matrix of hair. This trapping appears to be especially
effective for cocaine [13] and is well demonstrated even at low dosages,as shown by
the work of Henderson et al. [14]. Entrapped drugs appear to enter hair by several
routes; from the plasma, transcellular diffusion during keratinization, sweat and
sebum bathing, etc. These materials, acquired in the development of the hair,
appear to be firmly held in microstructural elements of the hair. Once bonded to
these elements the materials are not able to be removed by extensive washing. In
hair analysis these analytes are accessedby hair digestion or extraction procedures
whose aim is to liberate and identify the particular chemical entities under
scrutiny. It appears that with cocaine and several of its various metabolic products
this sequestering is extremely stable [151. As well, the amount of cocaine use in
 T. Mieczkowski / Forensic ScienceInternational 84 (1997)87-111 89

terms of gross quantities of substances ingested, appears to bear a discernible

relationship to the concentration attained in hair, provided it is measured over a
sufficiently wide dosage range. Cocaine and other entities are also capable of
attachment to hair via environmental exposure, but they do not appear to attain
the same bonding or attachment strength, under normal circumstances, as do drugs
which are ingested. These environmentally acquired drugs, in most circumstances,
can be removed by appropriate washes.
The primary advantage of hair analysis is the relatively long retrospective
identification of classes of drugs which normally quickly disappear from blood
orplasma. Cocaine and several other popularly abused psychoactives are stable
when embedded in the hair and can be detected for months, and in some cases
even years after exposure. A second advantage is quantifying the drug recovered in
the hair and estimating from that value the amount of drug ingested (i.e. establish-
ing a dose/assay relationship). The dose/assay relationship appears to be limited
in its utility to a rank-order assignment of values. There are relatively high degrees
of inter-subject variability regarding the regression relationship between dose
consumed/drug recovered. However, hair appears useful for tracking drug expo-
sure within individuals, where a baseline value is established for any given individ-
ual and they can subsequently act as their own control [16-M]. Some research has
shown that good correlation exists between self admitted cocaine use and hair
assays values [19], that t>he probability of having a cocaine positive urinalysis
outcomes and the quantitative value of a cocaine positive hair specimen are very
strongly positively related [4], that cocaine concentration values attained in the hair
are positively related to the efficiency of the consumption method [20], and that the
correlation between maternal and neonatal hair assaysfor cocaine is quite strong
[21-241.
The use of hair as a specimen for toxicological identifications is not novel. Hair
was first used in criminal proceedings in the United States in the nineteenth
century, with testimony about hair analysis first admitted in 1882 in Knoll v. State,
55 Wis. 249, 12 N.W. 369 [25]. The first reported recovery of a psychoactive drug
from the hair of guinea pigs was published in the United States more than 40 years
ago [26]. In recent years concern with drug monitoring has created sufficient
demand to make development of a low-cost, immunoassay based screening tech-
nology economically attractive. This has resulted in several commercial laborato-
ries in both the US and Europe developing and offering hair analysis services to
detect psychoactive drugs. In a recent laboratory evaluation exercise, for example,
11 different laboratories participated in a round-robin review to detect cocaine and
morphine in hair samples [27].

3. Hair analysis of cocaine exposure and its controversies

The identification of cocaine and its metabolites in hair has, itself, not been
particularly controversial. Virtually all published research has shown that cocaine,
as well as its major metabolites, can be readily identified in hair by a wide variety of
analytic techniques, including the use of radioimmunoassay (RIA), high perfor-
90 T. Mieczkowski /Forensic ScienceInternational 84 (1997)87-111

mance liquid chromatography (HPLC), gas chromatography/mass spectrometry

(GC/MS), and gas chromatography/tandem mass spectrometry (GC/MS/MS)
[28]. As with all diagnostic procedures controversy has arisen, however, about how
to interpret the detection of drugs in hair [291.Clearly, the detection of a substance
in hair is an indication of exposure to that substance. Indeed, as Kidwell [30] has
noted:
‘Certainly, many positive hair analysis results are due to ingestion of drugs but passive exposure must
be considered when evaluating any particular case. What does hair analysis for drugs of abuse
measure? It measures exposure. Methods to distinguish use from exposure are still undiscovered.’

The crux of the controversy is contained in the last statement, namely the
determination of the nature or cause of exposure. If the individual denies using the
drug, what is its source? This interpretative problem has intensely focused on the
issue of passive contamination and the ability to distinguish passive exposure from
active (i.e. willful, knowing) ingestion [31].
The term ‘passive contamination’ is generally used to describe cocaine which has
been environmentally deposited on the surface of the hair. A second, related issue
is passive ingestion. Passive ingestion generally is taken to mean the secondary
consumption of cocaine via inhalation of smoke, oral contact, or other similar acts
by which small traces of cocaine are actually consumed, but not consumed willfully
or knowingly by the person. This can result from known contact with persons who
use cocaine (e.g. kissing a person who has just smoked crack cocaine), or could
result from unknown contact with contaminated persons or objects. Passive and
active ingestion are not biochemically distinct, of course. However, as a practical
matter, under most clinical circumstances passive ingestion entails microscopic
(nanogram) quantities of a drug, while active ingestion usually involves taking
thousands of milligrams of the material.
Some confusion exists regarding the concept of contamination and the deposi-
tion of cocaine into hair from sweat or sebum. The sweat of a cocaine user contains
cocaine, and this source of cocaine is often reflexively treated as ‘environmental
contamination.’ Cocaine deposited into hair via the sweat of the individual cannot
be considered as an environmental contaminant, since the cocaine arises from an
endogenous source, namely the various somatic pools of drug in plasma, cellular
fluids, etc. Thus while sweat may make some contribution to the total quantity of
drug found in the hair of the drug user, its role does not represent a problematic
one from the point of view of identification of drug users. Contamination via sweat
between persons, which would be true environmental contamination, may occur
with chronic and prolonged skin-to-skin contact, or the sharing of wet clothing,
applied to hair, and heavily laden with sweat. However, no study has shown that
such contamination would be confused with cocaine arising from endogenous
sources, although specific studies examining this issue should continue to be
pursued. The circumstances under which this type of inter-person sweat transfer
might occur would be rather specialized, and do not appear to be frequently
encountered in clinical populations studied. Problems of this sort are normally
resolvable by extensive washing of samples with water or methanol.
 T. Mieczkoiwki /Forensic Science International 81(1997) 87-111 91

4. Clinical determinations of the source of drugs found in hair

One erroneous characterization of clinical application of drug analysis is that

practitioners do not utilize assays in a diagnostic fashion, i.e. as merely one of a
series of informational items upon which a decision is based, but rather in a sort of
‘all or none’ fashion. For example Kidwell [30] states that:

‘Those using hair analysis seek a definitive yes/no answer to the question: Did this individual ingest
drugs? An individual’s mere contact with drugs is seldom at issue.’

This statement is not accurate. While it may be true in some circumstances that
critical decisions are made on the basis of a single assay outcome, this is not true
for most other uses of drug analysis. In criminal justice and treatment monitoring,
as well as employee assistance programs, actions taken in response to apparent
drug use arise out of long-term relationships and assessmentperiods, and consider-
ation of multiple measures of drug involvement.
Furthermore, the term ‘mere contact,’ as Kidwell labels the phenomenon,
obscures the clinical significance of the difference between trivial, microscopic
levels of contamination and massive, overwhelming levels of contamination. The
determination of massive exposure to an illicit drug is indeed relevant to criminal
justice monitoring, treatment program monitoring, and a variety of security-based
drug monitoring contexts. For example, what is called ‘mere contact’ may be a
critical issue in a court-supervised diversion program where a condition of partici-
pation is to avoid contact with drugs, places where drugs are bought and sold, and
social associations with drug users. A person showing high degrees of cocaine
contamination on their clothing, etc. would be a situation of clinical and legal
concern. Likewise the massive contamination of an infant by cocaine would be of
importance. Kidwell’s statement implies that those who utilize hair analysis are (or
perhaps ought to be) unconcerned about cocaine exposure, but are (or can
legitimately be) concerned with cocaine use. It would be more accurate to state
that those using hair analysis for cocaine detection seek an answer to the question
‘did the individual experience trivial, incidental, or meaningless exposure that can
be plausibly explained by background effects, or is this exposure of a magnitude
that calls into question the person’s denial of cocaine use or involvement?’
In clinical use, the outcome of a single hair assay is rarely treated as a definitive
informational item. Rather, it is treated as one of a series of informational items,
all gleaned from various sources available to the clinician. The gathering of
information regarding a final decision on the interpretation of an assaydepends on
both available concomitant data (e.g. urine testing, saliva testing, interviewing, etc.)
as well as historic data (e.g. how does this assayvalue rank relative to other assays
taken in the past?). A little-mentioned but important aspect of clinic experiences
with both hair and urine testing in criminal justice and clinical treatment contexts
is that challenges are very rare and self-admissions are the norm [7,17]. A
simplified flow diagram illustrating this process is shown in Fig. 1.
92 T. Mieczkowski /Forensic Science International 84 (1997) 87-111

ContaminationArises From
Exposure

I Passive,
Innocent

c:-sI

1. ingested Drug Unknowingly

2. Had UnknowingEnvironmentalContact
3. ExperiencedBoth of These Events
4. MisrepresentedTheir Useor Exposureto Drugs

Fig. 1. A decision tree for assessing contamination.

 T. Mieczhmwski/Forensic ScienceInternational 84 (1997)87-111 93

4.1. Passive contamination and wash processes

Hair, being on the exterior of the body, is subject to environmental contact and
hence contamination. Drugs can be readily deposited on hair from environmental
sources, such as dust, smoke, or aqueous sprays. Critics of hair assayscontend that
hair analysis cannot adequately distinguish between active use and inadvertent
environmental contamination because environmentally deposited cocaine tightly
bonds to hair, and cannot be removed [31]. This effectively obliterates the differ-
ence between surface-contamination and cocaine which may enter the hair from
the consumption of cocaine. Alternately, others have argued that wash procedures,
properly done, can remove all or nearly all of any normally acquired environmental
contaminants [32,33].
The argument that any environmental contamination, no matter how small,
renders hair assaysunreliable is not accurate. Hair assay interpretation does not
need to assume that the hair is absolutely free of any trace of contamination. The
simple value of the drug extracted from hair is not used by itself to determine the
outcome of the assay.The procedure compares the outcome from specific sequen-
tial washes applied to the hair with drug recovered from the hair. Interpretation of
a hair assay, relative to passive environmental contamination, relies on several
pieces of information; how much drug, if any, is discovered in the various wash
series (which are done to cleanse the hair and identity any contaminants), and
second, how much residual drug remains in the hair after the wash procedure is
completed. Thus hair assay values/wash assay values must attain a series of
specific ratios to be properly interpreted. Baumgartner and Hill [34] have devel-
oped the most specific of these criteria. They are presented below in Table 1.
Positive specimens must pass all three criteria to be considered a cocaine
positive. Additionally, one can also utilize assaysto detect the presence of cocaine
metabolites, typically benzoylecgonine (BE), ecgonine methyl ester (EME), norco-
Caine (NE) and cocaethylene (CE). The presence or absence of these metabolic
products adds further information in interpreting a particular assay outcome. The
use of metabolites is premised on the observation that these arise largely or
exclusively from internal metabolic processeswhich occur in the body, or they are

Table 1
Wash kinetic criteria of Baumgartner and Hill

Criteria Calculation Required ratio

Extended (Amt. of drug per 10 mg hair in digest)/ 2 10

wash ratio (amt. of drug per 10 mg hair in last PO, wash)
Safety zone (Amt. of drug per 10 mg hair in digest/ 2 0.33
ratio (amt. of drug per 10 mg hair in all 4 PO, wash)
Curvature ratio (Amt. of drug per 10 mg hair in 3 PO,, wash)/ 2 1.3
(3 times the amt. of drug per 10 mg hair in last PO, wash)
94 T Mieczkowski / Forensic Science International 84 (1997) 87-111

not typically found in cocaine hydrochloride or ‘crack’ cocaine as it is vended.

Rather than rely on the simple presence or absence of these substances, one can
use ratios of the parent drug to the metabolite in order to help in interpreting the
assay outcome. For example, Cone [35] has suggested that the presence of CE and
NE are indicative of active drug use and that BE/cocaine ratios which exceed a
value of 0.05 are also indicative of active drug use. Koren et al. [33] have argued for
a similar interpretive approach, as have Baumgartner and Hill [34].
The problem of passive, inadvertent ingestion is one which presents a somewhat
different interpretive issue. Because passive ingestion can produce the same
qualitative biological outcome and same metabolic outcomes as ‘knowing use’ one
must take the same approach as has been done with urinalysis. This requires an
operational assumption that casual, inadvertent ‘use’ (i.e. ingestion) normally
differs substantially in quantitative dimensions from active use. This is essentially a
statistical distinction. Inadvertent users and deliberate users have the same experi-
ence, in a qualitative biological sense. However, they do not have the same
experience in a quantitative sense. Inadvertent use (except under the most bizarre
circumstances) is an event premised on minor exposure, while willful use is an
event which typically is characterized by large-scale consumption, especially for
cocaine abusers.

4.2. The use of threshold criteria

When it is necessary to control for passive ingestion of cocaine (e.g. by passive

inhalation of crack smoke, by contamination due to exchange of body fluids such as
semen or sweat), hair analysis uses the same methodology as urinalysis, statisti-
cally-based cut-off values. The value of the assay must exceed a specific threshold
in order to be labeled as a diagnostic positive outcome. This is quite distinct from a
technical positive. A technical positive is defined by the analytic technology’s limit
of detection CLOD). A diagnostic negative test may result from a specimen that is
technically positive. That is, the specimen has a technically detectable amount of
the particular analyte present, but an insufficient amount of the drug is recovered
to ‘cross’ the cutoff threshold.
Where should such a threshold be established? Unless one decides that the
LOD is appropriate, there is no technically imposed answer to this question. The
threshold represents a marker at which it is generally recognized that explanations
of passive or inadvertent exposure are implausible. This is intrinsically a statistical
phenomenon and is related to the scatter caused by biochemical clinical individual-
ity and the correlation between dosage and cocaine levels in hair. Unlike urine, the
statistics of hair analysis are not effected by excretion kinetics. Kintz and Mangin
[29] have addressed this issue and recommend a ‘stand alone’ value for cocaine of 1
ng/mg of hair (i.e. by ‘stand alone’ is meant that hair is used in the absence of any
other corroborating specimen), and suggest that this cut-off may be lowered to 0.5
ng/mg when ‘supported by other evidence of drug intake.’ Baumgartner and Hill
[36] have argued for the use of a 0.5 ng/mg cut-off value.
 T. Mieczkowski /Forensic ScienceInternational 84 (1997)87-l 11 95

In clinical practice there is no need to require a universal cutoff value for all
monitoring circumstances. Indeed, in current practice there are a wide variety of
thresholds employed in urinalysis testing for cocaine, dependent on the perceived
needs of the testing program and its goals. For example, over the last 2 decades the
thresholds for cannabinoids has been consistently edged downwards in response to
epidemiological reports that initial values were so high as to produce substantial
numbers of diagnostic false negative urinalyses. Quite unlike other illicit drugs,
marijuana has been shown to be often over-reported; that is, more people report
marijuana use than are identified by either hair or urine assays[37]. The lowering
of cannabinoid cutoff values was not based on any theoretical model, but solely on
clinical experience.
The selection of thresholds varies considerable in clinical practice. Commercial
test kits are readily available to test urine for cocaine at a threshold of 25 ng/mg,
even though in the US the federal government employs a threshold in its work-
place testing more than ten times this value. Some clinical programs have utilized a
‘zone’ approach to interpretation of assaysvalues [6]. In this setting - criminal
justice-based treatment program - persons testing positive for cocaine in hair at
values between 0.5 and 3.5 ng/mg are monitored with increased scrutiny but are
not presumed to be using the drug. Those who test positive at values greater than
3.5 ng/mg are treated as users, unless some compelling alternative explanation is
apparent. Over the history of this program these particular cutoffs, based on actual
clinical conditions and experiences of the treatment staff, have proven to be very
useful. The clients are also randomly urine tested, consistently interviewed and
counseled, and subject to surface contamination analysis. Program clients also have
long-term clinical histories and a baseline hair assayvalue determined at program
intake. Thus hair assay value changes for an individual are tracked over time. All
this data is weighed and evaluated as potentially corroborating or clarifying
information available to lend support to a particular case analysis. The effective-
ness and utility of hair analysis in this context serves as an example of the prudent
and careful use of hair assay technology. It contradicts the assertions that hair
assaysare inevitably used as simple binary indicators of cocaine ingestion.

5. Laboratory-based contamination studies

Laboratory studies have shown that hair can be contaminated by cocaine. These
studies have been criticized, however, because they have generally used ‘extreme
contamination scenarios’ in order to simulate contamination [38]. Laboratory
studies have typically depended on prolonged aqueous soaks of hair in concen-
trated cocaine solutions, or suspensions of hair samples over a pyrolized cocaine
base [13]. There are problems in linking these in vitro studies to the hypothesis that
casual contact, such as inadvertent touching of cocaine contaminated objects, could
contaminate the hair of a non-user by subsequent touch. While clearly a contami-
nation event is likely to occur, the issue is whether this contamination is so severe
as to obliterate the distinction between use and non-use. It has been suggested, for
example, that simple and transitory touching of contaminated objects could result
96 T. Mieczkowski /Forensic Science International 84 (1997) 87-111

in a positive hair assay and might lead to the labeling of an innocent person as a
cocaine user [14]. On a practical level, for example, could a barber who cut the hair
of a crack smoker then transfer sufficient cocaine to other customers by physical
contact to the degree that wash procedures and cutoff values could not distinguish
such persons from the crack smoker?
Such an event has never been demonstrated in any field setting and studies
which have reported on field-based samples do not support the argument that such
a phenomenon represents an impediment to the use of hair analysis. Contamina-
tion by physical touch appears to require rather specific or peculiar conditions to
be significant enough to confound a properly executed assay.Avolio et al. [39] have
shown, for example, that dry hair samples placed in physical contact with cocaine-
impregnated silica, removed periodically, and washed with methanol, did not begin
to acquire methanol-resistant cocaine contamination even at the picogram level
until after approximately 7 days of continuous contact.
In general, the literature reveals that studies have distinguished quite readily
between known cocaine users and known cocaine abstainers when those studies
have been done under controlled conditions (e.g. see the work of Cone et al. [40]).
Koren et al. [33] reported on both laboratory based hair sample manipulation as
well as the exposure of human volunteers to cocaine contamination. In one study
they had human volunteers exposed to 100 mg of vaporized cocaine in a confined
space, and reported an average concentration for cocaine of 27 ng/mg of hair, and
no detection of BE. Subsequent washing of the volunteers’ hair resulted in removal
of both cocaine and BE below the LOD. Wang and Cone [ 131exposed hair samples
in vitro and also exposed human volunteers to similar amounts of vaporized
cocaine under conditions similar to Koren et al. They reported comparable values
for initial contamination concentrations. They reported that hair samples - vapor
contaminated by 100 mg of vaporized cocaine, and soaked for 24 h in a small
volume of mild commercial shampoo - showed substantial loss of the contamina-
tion. The contaminated hair had initial values of 19.6 ng/mg in the wash fragment
and 7.3 ng/mg in the extract. After one shampoo wash cycle, no cocaine could be
detected in the wash fragment, and 0.7 ng/mg were detected in the hair extract.
After the third shampoo wash cycle the cocaine values in the extract reduced to 0.4
ng/mg. A very similar pattern was reported for the cocaine pyrolysis metabolite
anhydroecgonine methyl ester (AHE).
The two cocaine vapor-exposed human volunteers examined by Wang and Cone
attained a mean value for cocaine of 29.5 ng/mg in the wash fraction and 7.1
ng/mg in the extract. They had concentrations of 6.2 ng/mg and 1.8 ng/mg for
AME. After 8 days of routine hair hygiene the wash values for cocaine had reached
0.0 for one volunteer and 0.5 ng/mg for the second. The extract values were 0.6
and 0.5. The AME values were zero in both wash and extract fractions for both
volunteers after 8 days. Wang and Cone [13] found that aqueous contamination, as
opposed to vapor, resulted in elevated values for initial contamination readings in
hair samples. However, under mild and moderate soaking conditions repeated
shampooing (up to ten cycles) removed substantial amounts of the cocaine con-
taminants. In the mildest soaking scenario (0.01 mg/ml HCl solution), cocaine
 T. Meczkowski /Forensic ScienceIntemational84 (1997)87-111 91

recoverable from washing was zero and the concentration reduction in the extract
fraction was 94.3% after the second wash cycle. By the tenth wash cycle the
cocaine reduction was slightly greater than 99.5%. As the soaking solutions
concentration were increased, the washing was less effective in removing the
cocaine. Some hair samples were subject to a fivefold and a hundredfold increase
in soaking concentrations (0.05 and 0.1 mg/ml). In each case, after ten shampoo
washing cycles, the percentage of original contaminate removed exceeded 99% of
the original contaminate concentration. Also notable in Wang and Cone’s report is
that in the cocaine hydrochloride aqueous soaks BE was either absent or quickly
removed by a single shampoo cycle. And in a contamination scenario in which the
hair was contaminated with cocaine and added BE, the BE values dropped rapidly
after shampooing, and approached or attained zero values.

6. Field-based contamination studies: background contamination by cocaine

Another source of information on the plausibility of the contamination problem

for hair assays is to take field measurements in circumstances where cocaine
contamination is likely to occur. Cocaine is known to be present as an enviromnen-
tal contaminant. For example, it is widely recognized that a sizable percentage of
United States currency is cocaine-contaminated. Also, it is quite plausible that any
number of public objects are touched by cocaine users, potentially contaminated by
them, and then these objects may, in turn, contaminate innocent persons touching
these objects at a later time. Public drinking fountains, pay telephones, door
handles and a host of other objects might be sources of cocaine transfer from users
to non-users.
Field studies attempting to assessthe degree to which this type of contamination
occurs have shown that measurable amounts of cocaine are not easily transferred
to hands by simply touching contaminated objects. Maloney et al. [41], for example,
have shown that after handling cocaine-contaminated objects such as crack pipes,
non-users failed to transfer measurable amounts of cocaine to their hands. Maloney
and his colleagues also assayed the hands of bank tellers in a pre/post design to
measure contamination of the hands based on handling cocaine-contaminated
currency. Fifteen tellers from three different banks handled contaminated currency
for the entirety of a normal 4-h shift (and refrained from washing their hands at
any time during the work period). Tests on the currency showed it to be contami-
nated with cocaine, but no cocaine was detectable on the hands of the tellers. They
likewise were able to demonstrate that cocaine did not transfer to individuals who
drove cocaine-contaminated vehicles which were seized from drug dealers by the
police, even though the steering wheels of the cars tested cocaine positive. Nor
could the researchers detect cocaine on public objects likely to be used by cocaine
users or sellers such as pay telephones in high drug trafficking areas. The only
scenario under which cocaine was readily transferred from contaminated person to
negative control was under condition of direct skin-to-skin contact when volunteers
handled ‘rocks’ of crack cocaine, and then rubbed hands with negative control
volunteers.
98 T. Mieczkowski /Forensic ScienceInternational 84 (1997)87-111

Ulvick et al. [42] and Demirgian et al. [43] have also evaluated cocaine contami-
nation, including the transference of cocaine from currency to persons. They
concluded that while the total percentage of American currency in circulation
contaminated with cocaine is ‘very high’, currency with high levels of cocaine near
the surface are ‘much rarer.’ By examining contaminated currency with scanning
electron microscopy, they determined that currency can be contaminated by direct
contact with cocaine, but ‘soon after initial contact most of the cocaine falls off.’
Some remnant cocaine penetrates the subsurface, but this cocaine, which is
trapped in the currency’s fiber matrix, does not contaminate the hands unless the
surface is abraded by ‘hard rubbing’. These findings explain why Maloney et al. [41]
were unable to detect any contamination on the, hands of bank tellers, even after
many hours of handling currency which was cocaine contaminated. Demirgian et
al. [43] also found that ‘contamination did not occur by normal handling of highly
contaminated bills’ and Ulvick et al. [42] also concluded that their data ‘indicate
that transmission of cocaine from currency to a person is unlikely.’ As well,
Demirgian et al. also examined the potential for cocaine contamination by placing
highly contaminated experimental subjects into a motor vehicle with negative
control subjects. These people spent several hours driving around together, includ-
ing periodically stopping and switching seating positions. They found that ‘con-
tamination did not occur from riding in vehicles with contaminated people.’

6.1. Contamination of children exposed to cocaine

Smith et al. [44,45] have reported on a field study of the cocaine contamination
of the hair of children of cocaine smokers. Several of these children, presumably
residing primarily in the parental household, were characterized as having hair
assayvalues indistinguishable from their cocaine-using parents. The study reported
on 20 adult, cocaine-using parents and 29 associated children. Sixteen of the 20
adults and 29 of the children were reported as cocaine positive by hair analysis.
The authors concluded that:
‘These results show that cocaine-related compounds were deposited in the hair of children when
cocaine was present in the environment. ChildEn living with a cocaine-dependent adult exhibited
both cocaine and cocaine metabolite in their hair - if one assumes that young children are not
intentional cocaine users and are not intentionally given cocaine by adults, these results show that
their hair can become cocaine positive through unknowing exposure when they live with a cocaine
user. Saliva and shin swabs suggest that external contamination, not ingestion (emphasis original), was
the source of cocaine-related substances in the children’s hair. Neither wash-out kinetics, metabolites,
cut-off concentrations, nor re-test would exclude many of the children, presumed to be innocent
non-users of cocaine, from being identified as cocaine users.’

Fig. 2 reproduces the outcomes for the 29 children and infants. Smith et al.
reported these 29 cases as cocaine positive, including 5 cases at levels below
quantification (which they refer to as ‘trace’ positives). These trace positive cases
are indicated by the five bars which have no vertical dimension in Fig. 2.
Smith et al. conclude that ‘cut-off concentrations’ would not exclude ‘many of
the children’ but this statement is not consistent with the data. In Fig. 2 there are
 T. Mieczkowski /Forensic Science International 84 (1997) 87-111 99

ng cocalne/mg hair

8
7.5
7
6.5 a
6
5.5
5
4.5
4
3.5
3
2.5
2
1.5
1
0.5
I I
I , Y
30 9 P 1213 Ad15

Grouping of Children (Years)

Fig. 2. Children reported as cocaine positive by Smith et al. [44].

two cutoff criteria superimposed on the graph, indicated by the lines marked
‘Mangin and Kintz (1.0 ng)’ and ‘Baumgartner (0.5 ng).’ These represent two
cut-off values for cocaine suggested in the literature. As each criterion line
indicates, few of the children would be positive by either threshold criteria.
Applying the 0.5-ng cutoff suggested by Baumgartner and Hill, only 7 of these 29
children would be classified as cocaine positive. Using the Kintz and Mangin
‘stand-alone’ criteria of 1.0 ng, only five children would be positive.
The authors further state that the wash-out kinetics as developed by Baumgart-
ner and Hill would not identify these children as ‘contaminated’. Unfortunately,
they do not provide the complete wash data. However, based on their description
the Baumgartner and Hill criteria are not applied appropriately in this study. There
were major modifications to the wash method as described by Baumgartner and
Hill and there is only a single criterion used. The wash kinetic procedure alluded to
in the study requires all three criteria to be met simultaneously. Furthermore, they
do not utilize a digestion method on the post-wash hair. The Baumgartner and Hill
protocol specifically states that:
‘It is also important to realize that the presently defined kinetic criteria (particularly the numerical
values) are only valid if the residual drug in the hair fiber is measured by a method that guarantees
the complete release of the entrapped drug, e.g. a digestion procedure.’ [34]

Smith et al. also make direct comparisons between the assayvalues of the adults
and children, noting that ‘in some cases the child’s hair contained quantities
100 T. Mieczkowski /Forensic Science International 84 (1997) 87-111

greater than the adult user’s hair.’ Without providing concentrations relative to
body weight, these comparisons are difficult to assess,since they could result from
the children chronically consuming very small dosages, given their much lower
body mass. The pattern in Fig. 2, indicating that the smallest children have the
highest cocaine values, is a finding consistent with a hypothesized passive ingestion
of small quantities of drugs, likely to be most pronounced in infants who are very
orally active and who have very low body weights compared to other youths and
adults in the sample. The data in Fig. 2 is also consistent with the risk and quantity
of ingestion lessening as a child ages and matures. Aging would result in a decrease
of the constant oral exploration and crawling activity of infancy. As well it would
increase time spent outside the home, and increase the likelihood of eating food
prepared out of the home environment (e.g. school cafeteria lunches). Fig. 2 may
also be demonstrating another aging effect, that the impact of passive ingestion
would also be lessened, given a constant rate of exposure, by increasing body
weight due to growth.
Smith et al., however, reject passive ingestion as a potential basis for explaining
their findings, stating that ‘there is no evidence to support the hypothesis that
ingestion was the primary route of cocaine entering hair.’ They base this statement
on the results of saliva analysis for cocaine, which was negative for most of the
children and many of the adults, and skin swabs, which were positive for every
subject in the study. This, they argue, indicates that all subjects had been passively
exposed to cocaine.
It does not appear that the ruling out of passive ingestion can be made on the
basis of this data. Skin swab data, while it may indicate passive contamination, is
also compatible with the sweat excretion of ingested cocaine. The method by which
the authors differentiate cocaine found on the skin (from passive deposition) versus
cocaine found on the skin as a result of sweat (or sebum excretion) is not
elaborated. It is a plausible explanation that these children continuously ingested
small amounts of cocaine over a long period of time via actions such as the placing
of their hands or contaminated objects in their mouths, and continuously living in
an environment in which their foodstuffs, eating utensils, play items, clothing, etc.
are grossly contaminated by cocaine vapor. Certainly if the subjects themselves are
contaminated, as the authors emphasize, it must also be true that their physical
environment is contaminated as well. Two children, for example, are reported as
having positive saliva assays, indicating possible recent ingestion of cocaine, or
placement cocaine-contaminated objects into their mouths. Adult crack users, who
constitute the parent group, typically ‘cook up’ or ‘rock up’ cocaine in microwave
ovens, often using ordinary household utensils [46]. The assessmentof the degree
of potential passive ingestion versus contamination would be strengthened by an
evaluation of the degree of environmental contamination of the households and
the clothing, eating utensils, foodstuffs, and play items of the children.
The interpretation of Smith et al.3 hair assay data is further complicated
because hair samples collected for this study were taken in the course of a cosmetic
styling of the hair. Such a collection process would presumably involve only the
most distal ends of the hair shaft. During the course of the trim, the subject’s hair
 T. Mieczkowski / Forensic Science In temational84 (1997) 87-1 I I 101

was allowed to fall into a collection bag, producing a random mix of hair lengths,
loci, and hair shaft orientations. Such a procedure would randomly mix hairs of
varying lengths, and would fail to preserve root-to-distal end shaft orientation,
make any longitudinal analysis or comparison of hair assay data to other drug use
indicators impossible. Thus long-term interpretations of the outcomes of the hair
assaysrelative to either saliva or urine assays- short-term measures - cannot be
made. Also, this sample collection method is, unfortunately, not comparable to any
other reported field studies, which have collected hair samples specifically cut from
the scalp, and preserved them in proper orientation for the purposes of sectioning
and comparison to urine tests and self-reports of drug use (see, for example,
Marques et al. [211X

6.2. Narcotics oficers: field exposureto cocaine

Field studies of naturally-occurring cocaine exposure are not easy to conduct.

Because cocaine is a controlled substance with extremely limited medical use, it is
difficult to identify occupational groups which have meaningful, known, and chronic
environmental exposure to cocaine. However, one such group consists primarily of
undercover narcotics officers and evidence custodians who manage the inventory of
seized drugs. These individuals, in the course of their duties, have continuing
contact with cocaine, cocaine-rich environments, cocaine users, and cocaine deal-
ers. The officers function in environments where cocaine is consumed, they handle
cocaine in the process of buying and selling it, they intimately handle cocaine
during covert penetrations of drug selling organizations, when they make arrests
and seize the associated contraband, and they transport and process the seized
drug as part of the securing of the chain of evidence. Some of these officers also
routinely handle cocaine as part of training exercises. Considering these factors,
narcotics officers would appear to be a good study group for evaluating the degree
of contamination acquired via incidental environmental exposure and, as well, the
resistance of contamination to wash-based cleaning procedures. Certainly their
exposure to cocaine far exceeds that which is likely to be incurred by the general
public.
It is important to recognize that these officers often play covert roles as drug
users and drug dealers. The narcotics control strategies they pursue includes posing
as drug users, and convincing illegal drug sellers that they are ‘customers’, i.e. drug
users themselves. This precludes their ability to take precautions against contami-
nation as that would have meaning in any ordinary laboratory setting. They cannot
wear gloves or masks, etc. as this would betray their attempts to pass as drug users.
These officers, in other circumstances (such as handling evidence once it is in
police custody) may employ the conventional prophylactic measure of using gloves.
Based on the exposure these persons have to cocaine, and assuming none of
these persons consciously abuse cocaine, a series of simple hypotheses are sug-
gested:

(1) It is hypothesized that the individuals in this sample are exposed to

102 i? Mieczkowski /Forensic Science International 84 (1997) 87-l I1

detectable levels of cocaine via environmental contamination. This contami-

nation can be detected in the hair.
(2) If their environmental contact and contamination results in micro-ingestion
which emulates consumption characteristically seen in abusers of cocaine,
these persons should have significant amounts of cocaine in their hair, after
the hair is cleansed and tested. These amounts should be comparable to the
values associated with self-admitted users of cocaine.
(3) If they do have cocaine on or in their hair, but they are not similar in profile
to willful consumers of cocaine, the wash procedures and ratio criteria
designed to detect contamination as opposed to ingestion should identify
these officers as contaminated non-ingesters of cocaine.

This study presents data based on the analysis of hair samples and responses to
survey questions of 40 persons, 36 narcotics officers drawn from a number of police
departments situated in a major metropolitan area of the southeastern United
States, and 4 evidence room clerks who handle cocaine on a routine basis. Six
challenge samples were also sent to the testing laboratory, which was unaware of
their use in the study. Four officers were sampled twice, separated by approxi-
mately a 4-month interval.’ One officer (case 19) did not complete a questionnaire.
Thus there was a total of 50 samples analyzed for this project including six
challenge samples. Thirty-nine questionnaires out of a possible 40 were completed.
The participating officers and evidence clerks are employed as part of a county-wide,
multi-departmental narcotics enforcement task force. These officers volunteered to
provide a scalp hair specimen and answer a 24-item questionnaire on their
undercover experience, their perceived exposure to cocaine, and their hair hygiene
habits. The hair samples were gathered by two narcotics officers, who also adminis-
tered the survey. All specimens and survey instruments were anonymous. Samples
and surveys were common-coded to allow comparison of responses to values
determined by assay of the hair specimen.
The hair was analyzed for cocaine by the Psychemedics Corporation, of Culver
City, CA, using radioimmunoassay and the preparatory method described by
Baumgartner and Hill [34]. All positive cases were confirmed with GC/MS. The
hair samples ranged from 1 to 4 cm in length, and consisted of 20-40 strands of
hair, cut at the scalp by surgical scissors. The hair was preserved with the root to
distal orientation maintained. The hair was subject to an initial anhydrous isopro-
panol wash, and three subsequent phosphate buffer washes. After the third
washing in buffer, the hair was digested by a proteinase enzyme at a neutral pH.
Each wash and final hair digest was assayed by RIA. Complete technical descrip-
tion of the Psychemedics sample preparation and assessment, and confirmation
procedure has been published elsewhere [47]. The criteria required to consider a
sample cocaine positive are discussed in detail in Baumgartner and Hill [34,361.
The laboratory values are reported in units of ng/mg. The data reported here
‘The duplicate samples were as follows: cases 4 and 10, 1 and 14, 6 and 15, 2 and 18.
 T. Mieczkowski /Forensic Science International 84 (1997) 87-111 103

Table 2
Narcotics officer, descriptive data (N = 39)

Age (mean) 36 years

Gender
Male 33
Female 6

Ethnicity
White 31
Black 4
Hispanic 4

Experience (mean)
Years in narcotics 4.64
Years in undercover 4.56

included values for three phosphate buffer (PO,) washes as well as the RIA values
for the final hair digest.
As part of a laboratory challenge component, the field samples sent to the
laboratory also contained two positive contamination samples, which were fortified
by aqueous soaks in cocaine (24 h in 0.01 mg/ml cocaine HCI) per the method of
Wang and Cone [13], three negative blanks, and one hair sample from a self-ad-
mitted, chronic crack cocaine smoker. The laboratory identified all challenge
samples correctly.
There were 39 completed survey questionnaires. Table 2 presents basic descrip-
tive information on the officers and their level of experience.

7. Reports of exposure

The officers reported relatively frequent handling of cocaine. Almost all (97.3%)
reported handling cocaine during purchases and arrests, and every officer (100%)
reported handling contaminated objects when making purchases and arrests. Fifty-
six percent handled cocaine several times weekly or more frequently. Table 3
reports frequencies for handling.
The majority of the cases handled by these officers were cocaine cases (mean

Table 3
Frequency of handling cocaine

How frequently do you handle cocaine?

Daily or near daily 5 (12.8%)
Several times weekly 17 (43.6%)
Several times monthly 14 (35.9%)
Rarely 3 (7.7%)
104 T. Mieczkowski /Forensic Science International 84 (1997) 87-111

value 65.9%; range from 2 to 100%). And the majority of the cocaine caseswere
crack cases as opposed to powder cocaine cases (mean value 60.1%: range from 0
to 100%). Nearly all officers (97.3%) reported consistent and ongoing activities
relative to cocaine, which included handling, purchasing, seizing, field testing, and
transporting cocaine. The handling of crack cocaine by many of these officers
included intense, unprotected contact. Examples of this kind of contact included
‘tongue tasting’ (the touching of one finger to cocaine and the subsequent touching
of the tongue) and ‘bumping.’ Bumping is a form of tipping. When one purchases
cocaine via an introduction through an intermediary, it is customary to smoke some
of the crack with the intermediary as a gratuity. Since these officers cannot smoke
crack, they alternatively give a ‘bump’ as a gratuity. This is usually done by the
officer crumbling or breaking of a piece of the ‘rock’ with their fingernail, and
giving the fragment to the intermediary.
As Table 3 indicates, these officers have frequent contact with cocaine. Many
also spend significant time in social settings with persons who use cocaine, who use
cocaine in their presence, and periodically attempt to induce them to use cocaine.
Table 4 reports the counts of contact type as reported by the officers.
The officers were also requested to self-estimate their own level of exposure. In
Table 5 the officers’ views on their degree of exposure are presented. More than
half of these officers estimate their exposure to be from moderate to extreme.
Although 64% report that they use some form of precautions (typically wearing
rubber gloves) when handling cocaine, these measures are only employed in two
circumstances. One is during the execution of some search warrants, and second
during the transferring of cocaine within the department after it has been seized
and taken into custody. More than l/3 of the officers report that they do not use
precautions under any circumstances. None of the officers, of course, use gloves or
masks when making covert buys in the field.

7.1. Hair hygiene and cosmetic treatment

It is generally recognized that the washing and cosmetic manipulation of hair

affects both its capacity to absorb and shed drugs and other materials it acquires
from the environment. Examining the hair treatment practices of the sample
revealed no notable departure from what one might perceive as normal washing
patterns. Table 6 reports the frequency of hair washing by the officers.
As Table 6 indicates, daily washing of hair is clearly the modal practice. Other
aspects of hair hygiene and cosmetic treatment are presented in Table 7. The types
of cosmetic reported by the officers consisted primarily of ‘perms’ (7), followed by
the use of hair sprays, gels, and mousses(2), with a single report each of dyeing and
bleaching of the hair. Of the seven perms reported by the officers, four were
reported by males and three by females. Since the study was retrospective, their
was no concern with efforts on the part of officers to either avoid or engage in
special hair treatment or hygiene which would affect the assaysprocedure.
 T. Mieczkowski /Forensic ScienceInternational 84 (1997)87-111 105

Table 4
Number of officers reporting types of cocaine contacts

Frequency of Present around Present when Present in environment

contact powder cocaine crack is smoked where crack is smoked

Daily 4 1 1
Several times weekly 8 3 6
Weekly 2 2 2
Several times monthly 7 4 9
Monthly 4 4 5
Less than monthly 12 19 10
Never 2 6 6

Table 5
Officers’ self-estimates of exposure to cocaine

Degree of perceived exposure Number of officers (%)

Non-existent 4 (10.3)
Slight 13 (33.3)
Moderate 14 (35.9)
Heavy 6 (15.4)
Extreme 2 (5.1)

Table 6
Frequency of hair washing

Frequency of hair washing Number of officers (%)

More than daily 2 (5.3)

Daily 31 (81.6)
3-5 times weekly 4 (10.5)
Once weekly l(2.6)

Table 7
Shampooing and cosmetic treatment of hair

Do you use... Yes No

A regular commercial shampoo? 38 1
A cream rinse? 23 16
Do you...
Cosmetically treat your hair? 11 28
106 T. Mieczkowski /Forensic Science International 84 (1997) 87-111

8. Hair analysis data

Table 8 displays the data outcome for the hair assaysfor cocaine for all subjects
in the study. Samples for these casesrepresent approximately 90 days of retrospec-
tion - roughly 3.9 cm in length. Examination of the table reveals that every officer
had some measurable amount of detectable cocaine on their hair, with the
exception of two caseswhich had a zero value for every wash and the digest.
One sample, 27A, attained sufficient value for cocaine (0.52 ng/mgl to be
considered a positive assay by the 0.5 ng/mg cutoff. A repeat sample (designated
27B in Table 8) was obtained from the subject. This sample was, on subsequent
analysis, negative. The second sample, however, was not comparable in time
sequence to the first, because a 4-month interval had elapsed. Given the circum-
stances of this case, it is quite possible that the first sample outcome represents
passive ingestion of cocaine, since this sample met all wash and metabolite criteria.
Only one other sample, 26 (which was obtained from an evidence technician of
approximately 60 years of age), had a measurable amount of cocaine in the digest,
0.072 ng/mg, a value well below the cutoff. All other samples had zero values for
the digest. The alcohol wash values are dominated by zero outcomes (35 samples),
with nine caseshaving values above zero. The mean value for the alcohol wash for
the group as a whole is 0.006 ng/mg. The mean value for the series of phosphate
buffer washes are as follows: first PO, wash, 0.0531 ng/mg; second PO, wash,
0.0049 ng/mg; third PO, wash, 0.0010 ng/mg. In each case the series of PO,
washes shows that in each subsequent wash step the concentration value was
lowered, or reduced to zero. While 42 of the 44 samples has a value greater than
zero in the first PO, wash, only 18 caseshad cocaine in the second PO, wash, and
only four cases had any cocaine remaining in the third wash.

8.1. Control and fortified samples

Table 9, below, presents data for the three negative control samples, two ‘spiked’
or fortified samples, and a sample submitted collected from a self-reported crack
smoker.

9. Discussion

Based on the data presented here several observations may reasonably be made
in reference to the suggested hypotheses.
The first hypothesis is consistent with the findings reported here. These persons
are exposed to cocaine in the course of their work, and such exposure generally
results in the environmental contamination of their hair. This contamination is
detectable by RIA. The levels of the detected contamination are comparable to the
amount of contamination reported by other in vivo contamination experiments.
The values for the officers in this study are, for example, very close to the
background contamination values reported by Koren et al. [33] for persons never
reporting any cocaine use.
 T. Mieczkowski /Forensic Science Inremational84 (1997) 87-111 107

Table 8
Wash and hair digest assay values: cocaine (ng/mg)

Sample Alcohol PO4 Buffer Washes Hair

No. wash No. 1 No. 2 No. 3 digest

1 0.09 0.13 0.014 0 0

2 0 0.10 0.01 0 0
3 0 0.15 0.014 0 0
4 0 0.34 0.03 0.02 0
5 0 0.08 0.008 0 0
6 0.11 0.12 0 0 0
7 0 0.34 0.01 0.01 0
8a 0 0.11 0.007 0 0
9 0 0.15 0.018 0 0
10 0 0.05 0 0 0
11 0 0.03 0 0 0
12 0 0.01 0.01 0 0
13 0 0.06 0.01 0 0
14 0 0.07 0.01 0 0
15 0 0.04 0.01 0 0
16 0 0.01 0.01 0 0
17 0 0.10 0.01 0 0
18 0.01 0.02 0 0 0
19 0 0.04 0 0 0
20 0 0.05 0 0 0
21 0 0.06 0 0 0
22 0 0.05 0 0 0
23 0.004 0.30 0 0 0
24 0 0.004 0 0 0
25a 0 0.008 0 0 0
26a 0.002 0.018 0.009 0.003 0.072
27A 0.006 0.03 0.019 0.016 0.52
27Bb 0 0 0 0 0
28 0 0.004 0 0 0
29 0 0.005 0 0 0
30 0 0.008 0 0 0
31 0 0.007 0 0 0
32 0 0.013 0 0 0
33 0 0.013 0 0 0
34 0 0.008 0 0 0
35 0 0.011 0 0 0
36 0 0.02 0 0 0
37 0 0.007 0 0 0
38 0 0.008 0 0 0
39a 0 0.008 0 0 0
40 0.011 0.007 0 0 0
41 0 0 0 0 0
48 0.020 0.021 0.007 0 0
49 0 0.010 0.005 0 0
50 0.011 0.018 0 0 0

a These cases are evidence clerks.

bThis represents the values of a second sample on case 27 which is to a later time frame. The first
sample was. as indicated, slightly above the cut-off threshold.
108 T Mieczkowski /Forensic Science International 84 (1997) 87-111

Table 9
Assay outcomes for negative and positive control samples (ng/mg)

Sample Alcohol PO, buffer washes Hair

type wash 1 2 3 digest

42 Negative 0.018 0.008 0 0 0

43 Spiked 46.7 146.0 38.0 34.8 296.4
44 Negative 0.014 0 0 0 0
45 Spiked 26.6 53.3 17.8 8.9 103.6
46 Negative 0.008 0.017 0.006 0 0
47 S/R user 2.1 5.5 2.5 2.0 21.4

Second, it appears that although these officers are chronically exposed to cocaine
through their work, their exposure as measured by hair analysis is slight. If they are
also micro-ingesting cocaine, it is at a level so low as to clearly distinguish them
from cocaine users. However, as described in this paper, contamination must
always be considered as an aspect of assay interpretation. The findings related to
case 27 serve as a reminder that intense exposure to cocaine may lead to
contamination via microingestion. Bear in mind some officers in this study reported
field practices which would lead directly and indirectly to oral contamination. Such
techniques as ‘tongue tasting’ or ‘bumping’ must be considered when interpreting a
low level positive, which is near the cutoff.
Third, it appears as a consequence of these findings that the alcohol and
phosphate buffer wash procedures are an adequate method for removing external
contamination from hair, at least for the type of exposure experienced by these
individuals. Thus our data supports the findings of Koren et al. [33], and others who
have argued that environmental contamination is not an insurmountable problem
to the interpretation of hair assays under most normal circumstances where
contamination is likely to occur.
This study also indicates that passive contamination of hair specimens as
practiced in laboratory scenarios, at least based on the in vitro contamination
processes reported to date, are not likely to be accurate reflections of ‘real world’
contamination for many groups of interest in criminal justice practice. Laboratory
studies exposing hair to aqueous cocaine soaks or pyrolized cocaine vapors have
reported contamination at concentrations many orders of magnitude greater than
what appear in this field study. More intensive study of field populations and
determining background levels of exposure for the purposes of developing interpre-
tive guidelines is probably a far more useful approach to the issue of cutoff
determination than synthetic laboratory contamination approaches.
The results of this study, especially when considered in light of the findings of
Koren et al. [33], Maloney et al. [41], Ulvick et al. [421 and Demirgian et al. 1431
encourage an approach to studying contamination problems in real-world environ-
ments. It is apparent that to accurately gauge the likelihood of misinterpretation of
 T. Meczk~wski / Forensic ScienceInternational M (1997)87-111 109

assaysbased on contamination as opposed to ingestion, background values need to

be empirically determined in field settings. It is a reasonable conjecture that
establishment of these field-based parameters would likely show that for most
routine situations, wash procedures and sample preparation techniques similar to
those used in this study are adequate safeguards against confusing contamination
with meaningful cocaine ingestion.
Surely there are going to be extraordinary cases when a person is significantly
contaminated with cocaine. Perhaps, for example, by deliberate sabotage, through
the deliberate ‘spiking’ of food. It may also be possible that with chronic, intimate
skin-to-skin contact, augmented by exchange of body fluids through sexual activity
that an ‘innocent’ person becomes contaminated via contact and ingestion and
could attain sufficient concentration in the hair to cross the lowest threshold as an
evidentiary positive. Persons (such as undercover narcotics officers) who are
peripheral but chronically ‘dabble’ with cocaine may be detected as positive at low
values. This may also be true for persons who sell or package cocaine, but do not
regularly or recreationally use the drug in any active manner. However, there is
little in the way of data to support this as a commonplace event, and considerable
data to support the view that such events are rare. It seems implausible that such
persons could attain the values we consistently find in active cocaine users.
The criticism made that a distinction between contamination and use is a ‘fatal
flaw’ for hair analysis does not appear to be viable. In many criminal justice
situations the difference between exposure and use is a moot issue in any event.
Furthermore, there is always going to be some degree of environmental contamina-
tion when a person is using cocaine, and there is always some level of microinges-
tion when a person is contaminated with measurable quantities of cocaine. Indeed,
the same condition is true for urine-based testing, and is the raison d’etre that
cutoff values for assaysexist. The same logic is applicable to hair analysis. While
there may be some contention about precisely where those cutoffs ought to be
placed, there does not appear to be evidence which indicates that a reasonable
cutoff threshold is unattainable or would work any less efficiently and effectively
than it works for urine, or plasma, or any other matrix used to test for drugs.

Acknowledgements

Presented at the 1st European Meeting on Hair Analysis, Clinical, Occupational,

and Forensic Applications, Torre Cambiaso Center, Genoa (Pegli), Italy, June
17-19, 1996. The author would like to acknowledge the assistance of officers Joe
Rindosh and Deborah Schnitzler in conducting this research.

References

[l] D. Brookoff, E. Campbell and L. Shaw, The underreporting of cocaine-related trauma: drug abuse
warning network reports vs. hospital toxicology test. Am. J. Pd. Health., 83(3) (1993) 369-371.
[2] E. Wish and B. Gropper, Drug testing by the criminal justice system: methods, research, and
applications. In M. Tonry and J. Wilson (eds.), Drugs and Crime, vol. 13, University of Chicago
Press, 1990, pp. 321-392.
110 T. Mieczkowski /Forensic Science International 84 (1997) 87-111

[31 T. Mieczkowski, R. Newel, Cl. Allison and S. Coletti, Final Report: Hair Assays for Drugs of Abuse
in a Probation Popularion. Report to the National Institute of Justice, Grant 92-IJ-CX-KOlO, June,
1994.
141 T. Mienkowski and R. Newel, Comparing hair and urine assaysfor cocaine and marijuana. Fed.
Probat., 57(2) (1993) 59-67.
[51 J. Baer, W. Baumgartner, V. Hill and W. Blahd, Hair analysis for the detection of drug use in
pre-trial, probation, and parole populations. Fed. Probat., March, (1991) 3-10.
(61 T. Mieczkowski, R. Mumm and H. Connick, The use of hair analysis in a pre-trial diversion
program in New Orleans. Int. J. mnder Thu. Comp. Criminal., 3x3) (1993) 222-241.
[71 K Knight, G. Rowan-Szal, M. Hiller, L. Chatham and D. Simpson, Hair analysis: a tool to identify
probationers in need of drug treatment. Fed. Probat., 50(3) (1995) S8-62.
181 S. Magma, R. Freemen, Q. Siddiqi and D. Lipton, The validity of hair analysis for detecting
cocaine and heroin use among addicts. Inf. 1. Addicr., 27(l) (1992) 51-69.
191 S. Magma, S. Kang and J. Shapiro, Measuring cocaine use by hair analysis among criminally-
involved youth. Paper presented at the Annual Meetings of the American Criminology Association,
Phoenix, Arizona, October 27-30, 1993.
1101 S. Magma and S. Kang, The validity of self-reported cocaine use in two high risk populations.
Paper presented at the National Institute on Drug Abuse Technical Review on the Validity of
Self-Reported Drug Use: fmprouing the Accuracy of Survey Estimafes, September 7-8, 1994,
Rockville, Maryland.
1111Nevada Employment Security Department v Cynthia Holmes, In the Supreme Court of the State
of Nevada No. 26157. 1996.
[121 United States v. Anthony Medina. 1990. 749 F. Supp. 59; US District Court, New York.
1131 W. Wang and E. Cone, Testing human hair for drugs of abuse. IV. Environmental cocaine
contamination and washing effects. Forensic Sci. Inf., 70(1-3) (1995) 39-51.
[141 G. Henderson, M. Harkey and R. Jones, Final Report: Hair Analysis for Drugs of Abuse. Report to
the National Institute of Justice, Grant 90-NIJ-CX-0012, June 1993.
WI L. Cartmell, A. Aufderheide, A. Springfield, C. Weems and B. Arriaza, The frequency and
antiquity of prehistoric coca leaf chewing practices in northern Chile: a radioimmunoassay study
of a cocaine metabolite in human mummy hair, L&n American Antiquity, (1991) September.
I161 R. Martz, B. Donnelly, D. Fetterolf, L. Lasswell, G. Hime and W. Heam, The use of hair analysis
to document a cocaine overdose following a sustained survival period before death. J. Anal.
Tox-icol. 15 (1991) 279-281.
[171 C. Brewer, Hair analysis as a tool for monitoring and managing patients on methadone
maintenance. Forensic Sci. Int., 63(1-3) (1993) 277-283.
b31 J. Williams, Sectional hair analysis as a potential index of therapeutic compliance in an epileptic
population, In R. de Zeeuw, 1. Hosani and S. Al Munthiri teds.) Proceedings of the 1995
International Conference on Hair Analysis in Forensic Toxicology, Ministry of the Interior, Abu
Dhabi, United Arab Emirates, 1995, pp. 443-466.
[191 J. Hoffman, E. Wish, J. Koman, S. Schneider, P. Flynn and J. Luckey, Hair, urine, andselfreported
drug use concordance ar treatment admission. Paper presented at the American Society of
Criminology Annual Meetings, Phoenix, Arizona, October 27-30, 1993.
ml T. Mieakowski and R. Newel, Epidemiological data and hair analysis: current tindings on
self-reported drug use, urinalysis, and hair assays. Paper presented at the Society of Forensic
Toxicology Conference on Drug Testing in Hair, October 29 and 30, Tampa, Florida, 1994.
I211 P. Marques, A. Tippetts and D. Branch, Cocaine in the hair of mother-infant pairs: quantitative
analysis and correlations with urine measures and self-reports. Am. J. Drug Alcohol Abuse., 1%2)
(1993) 1599175.
La K. Graham, G. Koren, J. Klein, J. Schneiderman and M. Greenwald, Determination of gestational
cocaine exposure by hair analysis. /. Am. Med. Assoc., 262(23) (1989) 3328-3330.
[=I C. Callahan, T. Grant, G. Phipps, G. Clark. A. Novack, A. Streissguth and V. Raisys, Measurement
of gestational cocaine exposure: sensitivity of infant’s hair, meconium, and urine. J. Pediatr.,
12Of5)(1992) 763-768.
[241 R. Welch, S. Martier, J. Ager, E. Ostrea and R. Sokol, Radioimmunoassay of hair: a valid
technique for determining maternal cocaine ahuse. Substance Abuse, ll(4) (1990) 214-217.
 T. Mieczkowski /Forensic Science International 84 (1997) 87-111 111

WI E. Imwinkelreid, Forensic science: hair analysis. Grim. Law Bull., 27 (5) (1991) 447-456.
Dl R. Goldblum, L. Goldbaum and W. Piper, Barbiturate concentrations in the skin and hair of
guinea pigs. J. Znuest. Dermatol., 22 (1954) 121-128
W'l M. Welch, L. Sniegoski and C. Ahgood, Interlaboratory comparison studies on the analysis of hair
for drugs of abuse. Forensic Sci. ht., 70(1-3) (1993) 295-303.
1281 M. Moeller, Drug detection in hair by chromatographic procedures. J. Chromatogr., Biomedical
Applications, 580 (1992) 125-134.
[291 P. Kintz and P. Mangin, What constitutes a positive result in hair analysis: proposal for the
establishment of cut-off values. Forensic Sci. ht., 70(1-31(1995) 3-11.
I301 D. Kidwell, What does hair analysis for drugs of abuse measure? SAFS Newsletter, 24(l) (1996)
9-18.
I311 D. Kidwell and D. Blank, Decontamination procedures for drugs of abuse in hair: are they
sufficient? Forensic Sci. ht., 70(-3) (1995) 13-38.
[=I W. Baumgartner and V. Hill, Hair analysis for drugs of abuse: decontamination issues. Proceedings
of the 2nd International Congress of Therapeutic Drug Monitoring and Toxicology, Barcelona,
(October 9-12, 1990).
[331 G. Koren, J. Klein, R. Forman and K. Graham, Hair analysis of cocaine: differentiation between
systemic exposure and external contamination. J. Clin. Pharmacol., 32 (1992) 671-675.
[341 W. Baumgartner and V. Hill, Hair analysis for drugs of abuse: Forensic issues. Paper presented at
the International Symposium on Forensic Toxicology, FBI Academy, Quantico, Virginia (June
15-19, 1992).
[351 E. Cone, The pharmacology and pharmacokinetics of cocaine in hair. Paper presented at the
Society of Forensic Toxicology Conference on Drug Testing in Hair, Tampa, Florida (October 29
and 30, 1994).
[361 W. Baumgartner and V. Hill, Hair analysis for organic analytes: methodology, reliability issues,
and field studies. In P. Kintz ted.), Dmg Testing in Hair, CRC Press, Boca Raton, Florida, 1995,
pp. 223-266.
I371 T. Mieczkowski, The accuracy of self-reported drug use: An evaluation and analysis of new data.
In R. Weisheit (ed.), Drags, Crime, and the Criminal Justice System, Anderson Publishers,
Cincinnati, Ohio, 1990, pp. 275-302.
[381 W. Baumgartner and V. Hill, Comments on the paper by D.A. Kidwell and D.L. Blank. Forensic
Sci. ht., 63(1-3) (1993) 157-160.
[391 J. Avolio, L. Kim and P. Radwanski, Ion mobility spectrometry (IMS) detection of cocaine in hair.
Paper presented at the Annual Meetings of the International Association of Forensic Toxicology,
Tampa, Florida, (October 3 1, 1994).
I401 E. Cone, D. Yousefnejad, W. Darwin and T. Maguire, Testing human hair for drugs of abuse. II.
Identification of unique cocaine metabolites in hair of drug abusers and evaluation of
decontamination procedures. J. Anal. Tonicol., 15(5) (1991) 250-255.
[411 B. Maloney, L. Barbato, B. Ihm, H. Nipper and R. Cox, The qualitative determination of trace
amounts of cocaine obtained through casual contact. Microgram, 27(6) (1994) 185-187.
1421 S. Ulvick, J. Cui, T. Kunz, J. Demirgian, B. Hwang, C. Tin, S. Roche, S. Su and S. Rigdon, Cocaine
contamination on currency. Houston Advanced Research Center, Argonne National Laboratory,
US Coast Guard Research and Development Center (1996).
[431 J. Demirgian, J. Hwang, J. Ewing, C. Roche, S. Ulvick, T. Kunz, S. Su and S. Rigdon, Cocaine
contamination studies on Chicago currency. Paper presented at the Pittcon Meetings on Analytic
Chemistry and Spectroscopy (19961.
[441 F. Smith, D. Kidwell and F. Cook, Passive incorporation of cocaine into hair as reflected in the
children of cocaine users. Unpublished manuscript (1994).
[451 F. Smith, D. Kidwell and F. Cook, Cocaine in children’s hair when they live with drug dependent
adults. Paper presented at the Society of Forensic Toxicology Conference on Drug Testing in
Hair, Tampa, Florida (October 29 and 30, 1994).
[461 T. Mieczkowski, Some observations on the scope of crack use and distribution. In T. Mieczkowski
(ed.1, Drags, Crime, and Social Policy, Boston: Allyn and Bacon, 1992, pp. 32-54.
[471 W. Baumgartner, C. Cawing, T. Donahue, G. Hayes, V. Hill and H. Scholtz, Forensic testing by
mass spectrometric analysis of hair. In J. Yinon ted.) Forensic Application of Mass Spectrometty,
CRC Press, Boca Raton, Florida, 1995, pp. 61-94.