If you are a typical psychiatrist practicing in a setting other than a substance abuse clinic, chances are good that you rarely order drug testing on most of your patients. But should you be doing this more? Possibly. Here are some reasons why you might want to order drug testing:

1. To monitor the sobriety of patients who are acknowledged drug/alcohol addicts or abusers, and who want to get clean.
2. To determine whether a patient who has acknowledged abuse of one substance (such as alcohol) is also secretly abusing other substances. In one study of 248 treatment-seeking alcoholics, two thirds (68%) reported using illicit drugs in the past 90 days: 33% powder cocaine; 29% crack cocaine; 15% heroin, and 24% cannabis (Staines GL et al., J Addict Dis2001;20(4):53–69).
3. Finally, to determine whether an apparent primary psychiatric disorder is actually caused, or worsened, by drug use. Many patients with mood, anxiety, and psychotic disorders are abusing drugs, whether you know it or not.

The estimated lifetime prevalence of drug or alcohol abuse in depression is 16% to 27%, in bipolar disorder 43% to 56%, and in schizophrenia 20% to 65%, depending on the study methodology used (see Bradizza C et al., Clinical Psychology Review 2006;26(2):162–178). The lifetime prevalence of drug abuse in the general population is around eight percent (Compton WM et al., Arch Gen Psychiatry 2007;64(5):566–576), and alcohol abuse is around 18 percent (Hasin DS et al., Arch Gen Psychiatry 2007;64(7):830–842).

Most substance abuse clinics have a standard procedure for doing random urine toxicology screens, often including semi-supervised urination to prevent patients from substituting someone else’s clean urine for their own. But assuming that you work in a private practice with a middle class clientele, your approach to drug testing will likely be different. When should you consider asking a patient to get tested? Examples would include: Patients who are not improving despite aggressive treatment; patients to whom you are prescribing frequently abused substances, such as stimulants or benzodiazepines; patients who are being treated with opioids by another physician; any patient with an acknowledged history of substance abuse, even if remote.

Broaching the topic of drug testing can be uncomfortable for both you and your patient. You can introduce it by saying something like: “You might have heard that there is an epidemic of drug abuse in the U.S., and doctors are being encouraged to test most of their patients, especially those who are not getting better on standard treatment. How would you feel about getting tested?” Generally speaking, most patients will agree to it. Those who are particularly resistant are likely to be those who know they won’t test “well.” Hopefully, the whole conversation will encourage patients to be forthcoming about any hidden drug use, which would be the best possible outcome.

Assuming your patient agrees, how do you go about getting the test done? Some psychiatrists have testing kits in their offices, but most might feel uncomfortable handing a patient a urine cup and asking him or her to go into the office bathroom and return with a sample. A more genteel approach is to ask the patient to go to the lab sometime within the next six hours. You can write out an order for a urine tox screen on your prescription pad and ask that the results be faxed to your office. The lab report will include the time that the sample was delivered, allowing you to verify that your patient complied with the six hour time limit.

What about the patient who tests positive but swears the test is wrong? There’s no question that false positive results, also known as cross reactivity, are common, especially with urine drug screens. False positives reported in the literature include amphetamine with diphenhydramine (Benadryl), hlorpromazine (Thorazine) (Adhami S, Psych BulletinJ Clin Toxicol 2001;39(2):181–182), bupropion (Wellbutrin) (Weintraub D et al., Depress Anxiety 2000;12(1):53–54) and many others. THC can come up as a false positive with the proton pump inhibitor pantoprazole (Protonix) (Srinivas B et al.,Curr PsychPsychiatric Bulletin 2003;27:17–19).

Any chemistry lab will be able to give you a cross-reactivity booklet with a dismayingly long list of very common drugs that can cause false positives. For example, both coffee and ibuprofen are often listed as cross-reacting with the amphetamine test. Generally, if there is a positive result, labs will confirm it by sending it out to a different lab for gas chromatography, which is more precise, though not always perfectly accurate.

The table below lists a variety of drug testing options. Most are capable of testing for the usual panel of drugs, including amphetamines, barbiturates, benzodiazepines, cocaine, methadone and other opiates, LSD, PCP, and THC. (Acknowledgments to Susan Hochstedler, RN, of Addison Gilbert Hospital in Gloucester, Mass., and Karen Toscano, Core Lab Supervisor at Anna Jaques Hospital in Newburyport, Mass., for providing some helpful information for this article)

Correction: Since the publication of this article, it has come to our attention that they are additional false positive, false negatives, and interactions that were not listed in this story. Please click here for the updated correction.

2005;29:276), trazodone (Desyrel) (Roberge RJ et al., 2006;5(8)), and LSD can cross-react with sertraline (Zoloft), chlorpromazine, and paroxetine (Paxil) (Acosta-Armas AJ,

Available Drug Tests
Test	Detection Time Frame	Notes
Urine	6 to 24 hours	Test can be faked; false positive due to cross-reactivity; false negative due to dilution
Blood	6 to 12 hours	It hurts!
Hair	7 days to several months	Expensive, but good for discovering use in more distant past
Saliva	24 hours	Up and coming
Breathalyzer (for alchohol only)	A few hours	Often used by patient’s family to assess driving safety