Hair-testing for illicit drugs

Hair-testing for illicit drugs

Parkes, John


Human hair may contain deposits of illicit drugs. Testing of hair will provide an indicator of drug use at the time the hair was grown. Hair samples have several advantages over urine samples, particularly length of surveillance period (months rather than days) and resistance to tampering. Any form of drug-testing must be seen as a component of a clinical plan for the management of the patient’s drug misuse, mental disorder and offending.


Drugs and/or their metabolites which are present in the blood stream will be incorporated into hair being generated in the follicle. Once incorporated into the cortex of the hair, these traces of drug/ metabolite are permanently retained and substantially immune to any attempts to remove them. Analysis of the hair will constitute a proxy analysis for drugs present in the bloodstream at the time that part of the hair was grown. Since hair grows, analysis of the end of hair near to the scalp will indicate drugs consumed recently, while analysis of the end of hair distant from the scalp will indicate drugs consumed some time ago. The time period covered is limited only by the length of the hair sample.

A typical testing procedure involves clipping a small sample of hair from the head. At the laboratory this sample is washed to remove surface contamination, then dissolved. Initial testing would be by immuno-assay technology. Any positive results are confirmed by gas chromatogram/mass spectrometer analysis. Careful ‘chain of evidence’ and confidentiality procedures are essential to protect the patient, for example barcoding the sample at all stages of testing (DuPont & Baumgartner, 1995; Baumgartner & Hill, 1996).

The literature suggests support for drug-testing of hair samples in multiple countries and for a variety of quasi-judicial purposes. For example, Ricossa et al (2000) reports use of hair-testing in Italy for people who have been disqualified from driving following heroin or cocaine use. Re-instatement of the driving licence is contingent on successful monitoring by regular hair-testing, and the authors present evidence that hair-testing acts as an incentive to abstinence. In Canada, hair-testing has been used in monitoring drug use in relation to child care cases (Klein et al, 2000; Lewis et al, 1997). It is suggested that hair-testing is particularly useful to allow parents to demonstrate consistent abstinence, while hair-testing of a young child provides a non-invasive measure of the child’s exposure to dangerous substances in the home. The potential value of hair-testing in a UK drug treatment setting is noted in Department of Health et al (1999).

Hair-testing may have important clinical applications in mental health settings. Swartz et al (2003) reports a comparison of self-report and hair-testing to identify use of illicit drugs in a group of 203 patients. Sixteen per cent of the patients reported that they had used illicit drugs, while only 12.4% showed positive results from a urine test. However, when a hair sample was tested, 31% of the patients showed positive for amphetamines, cocaine, marijuana, opiates or phencyclidine over the last three months. The authors note the clinical significance of this finding in relation to their existing work on the risk-increasing effects of illicit drug use in patients with severe mental health problems. Schwartz & Swanson (1998) and McPhillips et al (1997) report a similar, but smaller-scale, study in the UK.

Key clinical advantages of hair as a test sample

Increased surveillance window

Urine-testing will usually detect drugs consumed within days of the test; see Table 1, below.

Hair will grow at approximately one centimetre per month, so a three-centimetre sample of hair will provide a ninety-day surveillance window. Maximum period of detection by hair-testing is limited only by the length of the hair sample (DuPont & Baumgartner, 1995). The possibly very long surveillance window is the main clinical advantage of hair-testing – Box 1, opposite.

Resistance to evasion

Urine tests are subject to multiple widely known strategies for cheating, such as the following.

* Substitute a drug-free sample of urine which has been concealed and carried into a toilet.

* Drink large quantities of water shortly before the test, greatly diluting the small amounts of drug/metabolites in the urine to below the cut-off level for reporting a positive result.

* Contaminate the sample with certain domestic or commercially marketed substances.

* Simply abstain from drug use for several days prior to the test.

Hair samples collected by a trusted tester are immune to all the above strategies.

By way of balance, it is appropriate to question possible strategies which might allow evasion of hair-testing. The greatest interest appears to be raised by products which allegedly decontaminate hair before a test; evidence for their effectiveness is weak. As an example, Rohrich and colleagues (2000) evaluated a shampoo product alleged to be of value in cleaning illicit drugs from hair. Laboratory testing showed that the shampoo reduced the concentration of drug metabolites (by between 5% and 36%, depending on the metabolite) but not by enough to avoid detection. Sample substitution is the only strategy likely to provide a purposeful false negative, and procedures are needed to obviate this risk, for example by ensuring that samples are taken by a trusted person and that chain of evidence procedures exist to prevent later interference with the sample.

Greater probability of detection

A combination of long surveillance window and resistance to evasion may result in a higher probability of detecting covert drug use. For example Mieczkowski and colleagues (1998) assessed the use of cocaine in 418 American young offenders using three strategies: self-report, urine-testing and hair-testing. Hair-testing was found to identify significantly more of those who had used cocaine (Figure 1, below).

Non-intrusive sample

In order to counter the risk of sample substitution (above), it is often deemed necessary to subject the patient to intrusive supervision while collecting a urine sample. Hair is a much less intimate sample, and the procedure of cutting a small lock of hair is likely to be seen as substantially less invasive and embarrassing to the patient.

Health and safety of collector/handler

Urine samples may give rise to concerns or genuine risks about cross-infection, hair samples much less so.

Capable of quantitative analysis

Concentration of a drug/metabolite in hair is determined at the time of deposition and is not subject to significant change over time. It is therefore possible to measure concentration of a substance in hair as an indicator of the amount consumed at the time the hair was laid down. It should be noted that some authorities urge caution in the interpretation of quantitative results from hair, because of potential variations in take-up of metabolites into hair, particularly between different individuals (Wennig, 2000). For example, differences in hair colour between individuals may alter the concentration of codeine laid down in hair, black hair absorbing higher concentrations than brown or red hair (Rollins et al, 2003).

Capable of historical analysis

The part of a length of hair nearest to the scalp/follicle has been deposited most recently. Tests of sections nearer to the distal end of the hair will provide a test of hair grown longer ago. By cutting successive one-centimetre sections from a long hair sample and analysing them separately, it is possible to provide a degree of analysis over time (DuPont & Baumgartner, 1995).

Repeat fesfs possible

Given the short surveillance window and the time taken for a laboratory to receive, process and report the sample, it is usually not possible to repeat the test if a patient contests a positive urine test. A person who had abstained since the first test would now be beyond the detection window of urine-testing and could test negative, even if the original sample was genuinely positive. Because of the long surveillance window of hair, a second, independent sample can be taken which can cover the same period of alleged consumption (Box 2, overleaf).

Compliance with prescribed medication

Substances other than illicit drugs can be detected in hair, for example prescribed medication where compliance is contested or for research purposes.

Williams and colleagues (2002) studied women prescribed anti-epileptic medication. Using hair assays to measure compliance, pregnant women were compared with a control group of nonpregnant women. Fifteen per cent of the former showed a pattern of anti-epileptic (carbamazepine or lamotrigine) in distal portions of the hair sample, but little or none in the hair proximal to the scalp, indicating that pregnant women may reduce or discontinue anti-epileptics during pregnancy because of fears for the safety of the foetus. Only one of these women disclosed to the doctor that she had discontinued. Cirimele et al (2000) detected prescribed clozapine in hair, suggesting that the presence or absence of the drug could be determined. However, the degree of correlation between prescribed dose and hair concentration did not appear sufficient to support quantitative analysis to determine whether the patient had followed precisely the dose prescribed.

Criticisms of hair-testing

The primary criticism of hair-testing is related to concerns that hair might be contaminated externally by drugs in the environment (Kidwell & Blank, 1996). It is suggested, for example, that a person sitting in a room where cocaine is being used or handled might have traces of the drug deposited on the surface of the hair, which might then be dissolved (for example in sweat) and absorbed by the hair. The person might later test positive for cocaine in hair, even though they had not ingested it. Blank and Kidwell (1995) report the use of radioactive labelling to demonstrate that cocaine can be absorbed into hair from aqueous solutions and that subsequent attempts at decontamination will remove some, but not all, of the drug. External contamination may be greater in hair which is bleached, ‘permed’ or greased than in untreated hair (Thorspecken et al, 2004).1

In response to these criticisms, testing procedures have developed over time in order to reduce the risk of false positive results. These procedures may include ‘kinetic washing’, where hair samples are washed in various solvents. It is not assumed that all external contamination could be washed out. The concentration of contaminants in the wash-off from multiple washes are measured and a specific pattern of declining concentrations in successive washes is considered to be indicative of external contamination. Most important, it is now considered good practice to test not for the drug itself, but for metabolites of the drug which are produced by its metabolism in a living mammal. It is argued that the base drug might be deposited externally by innocent exposure, but it is improbable that metabolites of the drug could be laid down in the matrix of the hair in this way. The issues and responses regarding contamination are discussed in detail in Baumgartner & Hill (1996).

Hair-testing also has some disadvantages in comparison with urine-testing. First, hair will take approximately one week to grow from the follicle to the point where it can be clipped above the scalp, so urine is a more appropriate sample for detecting short-term drug use (blood or saliva samples are appropriate for detection of the immediate presence of drugs). Urine samples are able to detect alcohol use, while hair samples do not. Hair samples could not give a full picture of all possible causes of intoxication, which may be significant in the light of findings that mentally disordered offenders are three times as likely to report alcohol use as illicit drug use (D’Silva & Ferriter, 2003). Finally, hair-testing is likely to be many times as costly as urine-testing, although this difference may be less apparent if long-term monitoring is required, since the higher cost of a single test may be offset by the longer surveillance period and reduced number of tests required.


Hair-testing is now a mature technology which offers significant advantages over the use of urinetesting in certain circumstances. In particular, the long surveillance period and the difficulty of evasion, even in the absence of oppressive security precautions, offer substantial advantages in a forensic setting. Hair-testing is a useful tool, but it is essential that any testing regime be integrated into an appropriate clinical treatment and management plan for the individual patient.

1 It should be noted that urine tests can also provide false positive results. For example, eating poppy seeds, sometimes sprinkled on bread buns, may result in a positive urine test for opiates, but is shown not to result in a positive hair test, because of the higher cut-off levels used for reporting a positive result in hair-testing (Baumgartner & Hill, 1996).


Baumgartner WA & Hill VA (1996) Hair analysis for organic analytes: methodogy, reliability issues and field studies. In: P Kintz (Ed) Drug Testing in Hair. Boca-Raton, Florida: CRC Press.

Blank DL & Kidwell DA (1995) Decontamination procedures for drugs abuse in hair: are they sufficient? Forensic Science Internationally (13) 13-38.

Cirimele V, Kintz P, Gosselin O & Ludes B (2000) Clozapine dose-concentration relationships in plasma, hair and sweat specimens of schizophrenic patients. Forensic Science International 107 (1-3) 289-300.

Department of Health, Scottish Office Department of Health, Welsh Office, Department of Health and Social Services of Northern Ireland (1999) Drug Misuse and Dependence – Guidelines on Clinical Drug Misuse and Dependence- Guidelines on Clinical Management. London: The Stationery Office.

D’Silva K & Ferriter M (2003) Substance use by the mentally disordered offender committing serious offences – a high security hospital study. Journal of Forensic Psychiatry and Psychology14 (1) 178-93.

DuPont RL & Baumgartner WA (1995) Drug testing by urine and hair analysis: complementary features and scientific issues. Forensic Science International70 (1-3) 63-76.

Kidwell DA & Blank D (1996) Environmental exposure – the stumbling block of hair testing. In: P Kintz (Ed) Drug Testing in Hair. Boca-Raton, Florida: CRC Press.

Klein J, KaraskovT & Koren G (2000) Clinical applications of hair testing for drugs of abuse – the Canadian experience. Forensic Science International 107 (1-3)281-8.

Lewis D, Moore C, Morrissey P & Leikin J (1997) Determination of drug exposure using hair: application to child protective cases. Forensic Science International 84 (1-3) 123-8.

McPhillips MA, Kelly FJ & Barnes TR (1997) Detecting co-morbid substance misuse among people with schizophrenia in the community: a study comparing the results of questionnaires with analysis of hair and urine. Schizophrenia Research 25 141-8.

MieczkowskiT, Newel R & Wraight B (1998) Using hair analysis, urinalysis and self reports to estimate drug use in a sample of detained juveniles. Substance Use and Misuse 33 (7) 1,547-67.

Ricossa MC, Bernini M & Ferrari F (2000) Hair analysis for driving licence in cocaine and heroin users. An epidemiological study. Forensic Science International 107 (1-3) 301-8.

Rollins DE, Wilkins DG, Krueger GG et al (2003) The effect of hair color on the incorporation of codeine into human hair. Journal of Analytical Toxicology 27 (8) 545-51.

Rohrich J, Zorntlein S, Potsch L, Skopp G & Becker J (2000) Effect of the shampoo Ultra Clean on drug concentrations in human hair. International Journal of Legal Medicine 113 (2) 102-6.

Swartz MS & Swanson JW (1998) Violence and severe mental illness: the effects of substance abuse and nonadherence to medication. American Journal of Psychiatry 155 226-31.

Swartz S, Swanson JW & Hannon MJ (2003) Detection of illicit substance use among persons with schizophrenia by radioimmunoassay of hair. Psychiatric Services 54 (6) 891-5.

Thorspecken J, Skopp G & Potsch L (2004) In vitro contamination of hair by marij uana smoke. Clinical Chemistry 50 (3) 596-602.

Wennig R (2000) Potential problems with the interpretation of hair analysis results. Forensic Science International 107 (1-3) 5-12.

Williams J, Myson V, Steward S et al (2002) Self-discontinuation of anti-epileptic medication in pregnancy: detection by hair analysis. Epilepsia 45 (8) 824-31.

John Parkes


Address for correspondence

John Parkes, Senior Lecturer, School of Health and Social Sciences, Coventry University, Priory Street, Coventry, CVl 5FB.

Copyright Pavilion Publishing (Brighton) Ltd. Dec 2004

Provided by ProQuest Information and Learning Company. All rights Reserved