U.S. flag

An official website of the United States government

NCBI Bookshelf. A service of the National Library of Medicine, National Institutes of Health.

Center for Substance Abuse Treatment. Substance Abuse: Clinical Issues in Intensive Outpatient Treatment. Rockville (MD): Substance Abuse and Mental Health Services Administration (US); 2006. (Treatment Improvement Protocol (TIP) Series, No. 47.)

Cover of Substance Abuse: Clinical Issues in Intensive Outpatient Treatment

Substance Abuse: Clinical Issues in Intensive Outpatient Treatment.

Show details

Appendix B. Urine Collection and Testing Procedures and Alternative Methods for Monitoring Drug Use

Urine testing is the best developed and most commonly used monitoring technique in substance abuse treatment programs. This appendix describes procedures for implementing this service and other methods for detecting clients' substance use. The Substance Abuse and Mental Health Services Administration (SAMHSA) has a number of documents about drug testing available in the Workplace Resources section of its Web site, www.samhsa.gov.

Testing Schedule

Urine specimens are collected

  • As part of the intake process to confirm a newly admitted client's substance use history
  • As a routine part of therapy
  • To identify an intoxicated client or confirm abstinence

Each intensive outpatient treatment (IOT) program should consider establishing a schedule for urine testing that takes into account Federal and State requirements (e.g., for methadone programs) and balances the therapeutic needs of the population being served with costs to the program or payer. Clients generally need more frequent monitoring during the initial stages of treatment when they are trying to achieve abstinence but still may be using substances. Routine specimen collection after admission should take place in conjunction with regular clinic visits.

Under ideal conditions, the consensus panel believes that collection should occur not less than once a week or more frequently than every 3 days in the first weeks of treatment. It is important that the scheduled frequency of urine collection match the usual detection window for the primary drug. Too long an interval between urine tests can lead to unreliable results because most of the target drug and its metabolites will have been excreted. On the other hand, if the interval between tests is too short, a single incidence of drug use may be detected twice in separate urine samples. Multiple positive urine test results produced by a single ingestion (carryover positives) can be discouraging for the client and misleading for the clinician (Preston et al. 1999).

Once clients are stabilized in treatment, they require less intensive monitoring of abstinence. At this point, most programs reduce the frequency of scheduled tests and randomize the collection times. Even with a decreased and randomized testing schedule, specimen collection should be scheduled on clinic days following weekends, holidays, or paychecks—the times when clients are most tempted to use.

During IOT, monthly testing is standard in most programs. Random testing can be achieved by

  • Asking clients to produce specimens only on random days
  • Requiring that all clients provide a specimen on every visit but analyzing only a randomly selected sample

Collection Procedures and Policies

Urine sample collection procedures need to strike a balance between trusting clients and ensuring that specimens are not contaminated or falsified. Some programs insist that a staff member of the same sex accompany a client into the bathroom to observe urine collection. Others find that monitoring through an open door and having clients leave packages and coats outside are sufficient. A sink that is separate from the toilet area also discourages attempts to dilute samples (Bureau of Justice Assistance 1999). Many programs use temperature strips to make certain that urine specimens are produced on site and are body temperature. Tests of creatinine or specific gravity can determine whether a sample has been diluted with water or the client is consuming excessive fluids to lower the concentration of drugs below detectable levels (Preston et al. 1999).

Information about how to beat the drug testing system is widely available. Web sites advertise inexpensive products that can be added to urine specimens to absorb toxins as well as herbal remedies for consumption for a few hours before testing to cleanse the urine. Concentrated, “clean” specimens can be purchased for mixing with warm water at the test site. A variety of low-cost, self-testing kits also are available to preview likely results from more formal testing procedures.

As part of their orientation to the IOT program, clients need to be informed about the urine collection and testing procedures. Clients also should be advised that informed consent is necessary for release of toxicology results to anyone other than staff. Most IOT programs do not comply with workplace standards for testing or maintain an adequate chain-of-custody for specimens that would meet court challenges. If employers, representatives of the criminal justice system, or children's protection agencies feel that such reporting is necessary, they can be advised to conduct their own testing or to accept other clinical evidence of client progress in treatment.

Clients should report any substance use to their counselor before a urine sample is submitted so that the substance use can be addressed therapeutically. It may be helpful to remind clients that the clinic conducts drug monitoring to support their recovery. Because there may be some likelihood of cross-reactivity and false positive results on screening tests, clients need to keep counselors informed about any prescribed medications or over-the-counter (OTC) drugs they have used.

Appropriate attention needs to be given to handling and storing collected specimens. Collection bottles that are sent to an offsite laboratory should be clean and tamper proof. Waterproof labels attached to the bottles should note either the client's name or identification number and be checked for accuracy by the client and the counselor or technician. Collected specimens need to be kept cool—or refrigerated—until transmitted to the laboratory and should be stored in a protected or locked room for security. Clients and staff members who touch the urine collection bottles need to be reminded to wash their hands thoroughly. Rubber gloves should be worn by technicians who perform onsite analyses.

Selection of Drug Batteries and Testing Techniques

Programs need to test for a standard battery of drugs, which may include such drug groups as amphetamines, barbiturates, benzodiazepines, cannabinoids, cocaine, methadone, methaqualone, opioids, phencyclidine (PCP), propoxphene, or euphorics (Ecstasy). In programs where the majority of clients use only a few types of substances, the standard battery can be small, and only selected individual clients need be tested for other specified substances. Programs should add substances to the routine battery, temporarily or permanently, if patterns of substance use change in the target population or in the community. It is helpful to stay up to date about local drug use patterns identified by the nearest Community Epidemiology Work Group (www.nida.nih.gov/CEWG/CEWGHome.html) or the Single State Authority. For example, oxycodone (OxyContin®) has become a serious drug of abuse in particular locales. Fads come and go for abuse of a wide variety of substances (e.g., Ecstasy, PCP, pentazocine [Talwin®], propoxyphene [Darvon®]).

Detection Limits for the Substances Being Tested

The length of time during which different licit and illicit substances or their metabolites can be detected in urine samples depends on many interacting factors, including

  • Chemical properties (e.g., half-life) of the selected drugs
  • Metabolism rates and excretion routes
  • Amount, administration route, frequency, and chronicity of the dose consumed
  • Sensitivity and specificity of the assay
  • Individual variations in clients' physical health, exercise, diet, weight, gender, and fluid intake that affect excretion rates

Most substances of abuse can be detected for approximately 2 to 4 days (see exhibit B-1). However, the higher the dose taken and the more frequently the substance has been used over an extended time, the more likely that it will be detected. Although substances are excreted at various rates, they accumulate in the body with continued use. Whereas a single use of cocaine may be detectable in urine for only a day or less, continued daily use is likely to be detectable for 2 to 3 days following its discontinuation (Preston et al. 1999). Chronic use of such drugs as marijuana, PCP, and benzodiazepines may be detectable for up to 30 days, whereas alcohol remains in the system for 24 hours or less. Realistically, it may be difficult to detect illicit substances in most clients who stop all use for several days before a drug screen. An accurate profile of a client's substance use over more than a few days requires both urine test results and a good retrospective history.

Exhibit B-1. Urine Toxicology Detection Periods for Different Substances

SubstanceTypical Urine Detection Period
Amphetamine or methamphetamine2–4 days
  Short-acting—Secobarbital1–2 days
  Long-acting—Pentobarbital2–4 days
Phenobarbital10–20 days
  Therapeutic dose3–7 days
  Chronic dosingUp to 30 days
Cocaine1–3 days
  Casual use1–3 days
  Daily use5–10 days
  Chronic useUp to 30 days
Ethanol (alcohol)12–24 hours
Opioids (e.g., codeine, morphine)1–3 days
Methadone2–4 days
Propoxyphene6–48 hours
Ecstasy/euphorics1–5 days
  Acute use2–7 days
  Chronic useUp to 30 days

Copyright © 1999 by the American Psychological Association. Adapted with permission. No further reproduction or distribution is permitted without the written permission of the American Psychological Association (Preston et al. 1999, p. 286).

Selecting an Appropriate Testing Technique

A program should consider a variety of factors in selecting a method and source for drug testing. None of the methods are inexpensive, with costs ranging from less than $5 to more than $100 per assay for a particular drug. Turnaround time in receiving results is another important determinant. Whereas onsite methods can provide results in a matter of minutes, more accurate and expensive commercial laboratory analyses may take several days or longer. Reliability is a major consideration. However, substance abuse treatment programs that are using results for clinical purposes do not require the same accuracy (i.e., workplace standards) as agencies that make important, one-time decisions about such issues as employment, safety, eligibility for sports competitions, or probation or parole violations. Some cities and States have assumed responsibility for selecting a single vendor for providers under their jurisdiction to use and choosing a standard battery of drugs to be tested. Providers may wish to create a buying collective to negotiate the best discounts from a local drug-testing laboratory.

Two categories of urine tests are available:

  • Screening tests. These detect only the presumptive presence or absence of a class of drugs in the urine specimen, return results rapidly, are relatively inexpensive ($1 to $5 per assay), can be set to detect low concentrations of drugs (have high sensitivity), and are relatively simple to perform. But these screening tests—the ones most frequently used by substance abuse treatment programs—do not distinguish specific drug metabolites (only groups), provide only qualitative results (yes or no), and may mistake other chemically similar medications, OTC preparations, or substances for the target drug class (Preston et al. 1999). This potential for cross-reactivity is of more concern in detecting amphetamines, benzodiazepines, and opioids than cocaine or marijuana. More specifically, the following cross-reactive results may occur:
    • Some cough suppressors in OTC preparations may be reported as a positive result for opioids.
    • Phenylpropanolamine or ephedrine in cold remedies can cause false positives for amphetamines.
    • Ibuprofen and other anti-inflammatories may be interpreted as positives for marijuana on the enzyme-multiplied immunoassay technique (EMIT) test.
    • Amitriptyline (an antidepressant) can be mistaken for opioids.
    • Some antibiotics may cause false positives for cocaine.
    • Diazepam has been mistaken for PCP.
  • Confirmatory tests. These provide more definitive information about the quantitative concentrations (nanograms/milliliter) of specific drugs or their metabolites in urine specimens and are more accurate than drug screens (have higher specificity and sensitivity). They are much more expensive (up to $100 per assay), technically complex, labor intensive, and time consuming—often taking days to complete. If the results of a drug test will be used as a basis for actions taken against an individual (e.g., in a justice system context), positive findings should be followed by a confirmatory test of equal or greater sensitivity and better specificity (Bureau of Justice Assistance 1999). Although results from these quantitative tests can be more useful than a simple positive or negative for monitoring intermediate changes in drug consumption patterns, the concentration in urine might be the same for a small amount of a drug administered recently as for a large amount of the drug consumed several days ago. In addition, concentrations can be affected by fluid consumption levels and may be misleading (Preston et al. 1999).

The Meaning of Test Results

Urine test results can be inaccurate. Counselors should keep this fact in mind when discussing findings with a client. Asking the client whether results are accurate and, if so, when and how much of a particular substance was used can be the beginning of a therapeutic discussion that includes the circumstances surrounding substance use and the client's triggers.

In interpreting test results, clinicians should know the following:

  • Positive results show a presumptive or confirmed presence of targeted substances at a detectable level. Positive results also mean that the amount of the substance detected is above the cutoff point for labeling a specimen positive. (SAMHSA has established Federal guidelines for cutoff levels; see workplace.samhsa.gov/DrugTesting/RegGuidance/UrineConcen.htm.) Findings cannot determine when, how much, or how a drug was administered or the degree of impairment the drug produced (Bureau of Justice Assistance 1999).
  • Negative results do not guarantee that the individual did not consume the substances tested. Despite a client's use of the targeted substance, results could be negative because (1) most evidence may have been excreted or metabolized before testing took place, (2) the specimen may have been diluted or switched, (3) the client may have consumed an excessive amount of fluids to dilute the urine, or (4) the test may not have been sufficiently sensitive (Bureau of Justice Assistance 1999).
  • False-positive results that mistakenly find the presence of a substance can result from laboratory errors (e.g., outdated reagents and labeling mistakes), specimen tampering, or cross-reactivity of an immunoassay test with a substance of similar chemical structure.

Urine-Testing Techniques

Most screening tests are immunoassays that take advantage of antigen-antibody interactions—using enzymes, radioisotopes, or fluorescent compounds—and compare the specimen with a calibrated quantity of the substance being tested (Bureau of Justice Assistance 1999).

  • EMIT test is the least expensive, most widely used, and simplest test to conduct. It often is used on site at a cost of about $5 per screen. It also has the poorest performance record, returning up to 30 percent false positives. Although EMIT can be used to test for a wide variety of drugs and alcohol, some sources report that as many as 300 OTC preparations cause false-positive readings.
  • Fluorescent polarization immunoassay TDx is highly sensitive and highly specific.
  • Radioimmunoassay (RIA) is a more sensitive test than the EMIT and is used extensively by the military.
  • Kinetic interaction of microparticles in solution is a screening test used with most substances.
  • Thin-layer chromatography (TLC) involves the addition of a solvent to the specimen that causes the target drugs and metabolites to move up a porous strip, leaving colored spots at different distances that can be compared with known standards. The results are reported as positive or negative, without any quantitative information, and require skill to interpret. Because TLC returns many false positives, it is no longer used widely.

Confirmatory urine testing methods include

  • Gas liquid chromatography
  • High performance liquid chromatography
  • Gas chromatography/mass spectrometry (GC/MS) (the gold standard for drug detection, but costly at $25 to $100 a test)

Alternative Testing Methods

Several other body products are gaining prominence in the search for simpler, less expensive, noninvasive, and more accurate techniques for detecting the recent and current use of substances. Exhibit B-2 compares the effectiveness of urine, breath, saliva, sweat, blood, and hair testing methodologies for detecting drugs.

Exhibit B-2. Effectiveness of Drug Detection Methods That Use Different Biological Products

Body ProductDrug Detection TimeMajor AdvantagesMajor LimitationsPrimary Use
Urine2–4 daysMature technique; established cutoffs for detecting many drugs of abuseDetects only recent use; needs costly confirmation to be accurateMonitors recent drug use in many populations
Breath (alcohol)12–24 hoursEasy to use; readily available and well-established methodShort detection timeConfirms observed intoxication or impairment
Saliva12–24 hoursEasy to obtain samples; good correlation with blood levels for some substancesVery short detection time; new method; oral cavity is contaminated easilyLinks positive drug test to behavioral impairment and intoxication
Sweat1–4 weeksCumulative measure; relatively tamper-proof collection methodHigh potential for contamination; new techniqueDetects recent and less recent drug use
Blood12–24 hoursAccurate results; established methodInvasive method; expensive; detects only current use or intoxicationDetects drug effects on crashes, medical emergencies
Hair4–6 monthsMeasures long-term drug use; readily available samples; accurate resultsNew technique; costly and time-consuming; no dose-response relation establishedConfirms drug use in past 4 to 6 months; prevalence studies

Copyright © 1999 by the American Psychological Association. Adapted with permission. No further reproduction or distribution is permitted without the written permission of the American Psychological Association (Preston et al. 1999, p. 299).

Breath-Testing Techniques

Because alcohol is metabolized rapidly at an average rate of 15 to 25 milligrams per hour—and the detection period is hours, not days—drinking usually is not monitored by urine or blood tests. Instead, clinicians frequently rely on other observations of current use (e.g., an odor of alcohol, slurred speech) or an easily administered Breathalyzer test to confirm alcohol intoxication or drinking within the past several hours. Blood alcohol concentrations—measured in milligrams (mg) of alcohol per deciliter (dl) of blood—usually are expressed as a percentage (i.e., 100 mg/dl equals 100 mg percent or 0.1 percent) and correspond closely with measures of alcohol on the breath. One drink increases the breath alcohol level (BAL) by approximately 0.025 percent.

For most men, some impairment is observable at 0.05 percent BAL, and driving ability is appreciably affected at 0.07 percent. A woman weighing 150 pounds would reach a BAL of 0.1 percent if she consumed approximately four drinks in an hour (compared with six drinks in an hour for a 200-pound man), although individuals' metabolism of alcohol varies considerably according to gender, age, simultaneous ingestion of food, and physical condition, as well as weight and consumption rate. BALs between 0.10 percent and 0.20 percent without obvious signs of intoxication usually indicate tolerance for alcohol and regular, heavy drinking characteristic of dependence (CSAT 1997a ).

Normally, with little or no tolerance for alcohol, the following impairment levels are observed:

  • 0.40 percent = lethal
  • 0.30 percent = unconscious
  • 0.20 percent = decreased consciousness
  • 0.10 percent = intoxication
  • 0.07 percent = impaired driving ability
  • 0.05 percent = detectible effect

In addition to Breathalyzer tests, several other simple-to-use but accurate techniques now exist for determining either a client's BAL or his or her approximate blood alcohol concentration. One is a relatively inexpensive, portable, and disposable unit the size of a cigarette containing crystals that turn a particular color—from yellow to blue—to signify a blood alcohol concentration of 0.02 percent, 0.08 percent, or 0.10 percent within 30 seconds after someone blows into the unit for 10 seconds.


For alcohol, saliva is correlated closely with blood concentrations 2 hours after consumption. However, routes of drug administration that contaminate the oral cavity can change the pH levels of saliva. These changes can distort correlations of other drugs found in saliva with blood plasma levels (Magerl and Schulz 1995; Preston et al. 1999). One advantage of saliva testing is the ready availability of saliva specimens and the packaging for onsite testing. However, the short time window for detecting substances limits the effectiveness of this method to ascertaining only recent drug use (e.g., for accident investigations and for pilots or other employees about to engage in safety-sensitive activities). Most substances disappear from both blood and saliva within 12 to 24 hours of use; cannabinoids may be detectable for only 4 to 10 hours after marijuana is smoked. The U.S. Food and Drug Administration (FDA) recently approved limited use of RIA-based saliva tests. Kits that detect tetrahydrocannabinol (the active component of marijuana), opioids, and cocaine are available for about $30


Although a number of licit and illicit substances can be detected in perspiration (probably diffused from blood), perspiration is difficult to collect for monitoring purposes. Manufacturers have introduced a “sweat patch” with a tamper-proof adhesive that is worn for about a week. It has been used successfully to detect amphetamines, cocaine, ethanol, methadone, methamphetamine, morphine, nicotine, and PCP. The drugs are absorbed gradually into the pad, which must be applied carefully on clean skin and removed carefully for analysis. Although no rapid methods for analysis are available, and the pads must be mailed to laboratories, the FDA has approved their use for detecting cocaine, amphetamines, and opioids. The pads are used primarily to monitor offenders on parole or probation.


Hair analysis can be used for detecting illicit substance use in the workplace and for drug treatment screening. The exact mechanism by which drug metabolites are absorbed into hair follicles remains unclear. Trace amounts of metabolites in the bloodstream enter hair follicles; these metabolites then are trapped in the core of each hair strand. It seems to take about a week after substance use for hair follicles to absorb drug residues. Because hair grows at a rate of about 1/2 inch per month, a 2-inch strand retains the record of a person's substance use over approximately the past 4 months—a much longer historical record than can be found through urine testing (Mieczkowski et al. 1998).

The advantages of this technique are

  • The presence of larger concentrations of the substance use than in urine samples
  • The ease of specimen collection; hair usually is taken from the scalp, but any body hair can be used
  • The difficulties in falsification or tampering and the simplicity of storage and shipping

Certain objections to this technique have not been resolved. Few laboratories conduct the analyses. Questions exist about potential environmental contamination of hair, the relationship of dose to the concentrations of the substance in hair, and whether biophysical attributes affect outcome. However, a large random study of hair analysis found little evidence of any bias in assay results associated with hair color, race, or ethnicity (Kelly et al. 2000). Because hair grows slowly and recent drug use cannot be detected reliably, the methodology has limited application for routine monitoring of treatment compliance. It could be useful for corroborating an intake drug history and conducting prevalence research (Preston et al. 1999).

Hair testing involves dissolving about 50 strands of hair in solvents and testing the liquefied sample with GC/MS. The technique appears to be highly reliable for detecting cocaine and crack, opioids (heroin), methamphetamines, PCP, and synthetic substances such as methylenedioxyamphetamine and 3–4 methylenedioxymethamphetamine or Ecstasy. It may be less reliable for detecting marijuana (Mieczkowski and Newel 1997).


  • PubReader
  • Print View
  • Cite this Page
  • PDF version of this title (853K)

Recent Activity

Your browsing activity is empty.

Activity recording is turned off.

Turn recording back on

See more...