Unconfirmed UAs aren't accurate but used anyway

After all the allegations that the Bexar County probation department made supervision decisions based on faulty, unconfirmed urinalysis tests, I was interested to see a USA Today story ("False results put drug tests under the microscope," Nov. 3) reported by scribe Mimi Hall describing the same problem with a national scope, the:

use of unreliable field drug-test kits as the basis to arrest innocent people on illegal drug charges.

The inexpensive test kits are used by virtually every police department in the country and by federal agents, including Customs officers at the nation's borders. The kits test suspicious materials, and a positive result generally leads to an arrest and court date, pending more sophisticated tests done after the sample is sent to a lab.

The kits use powerful acids that react with the substance in a plastic pouch. If the liquid turns a certain color, it is a considered a positive result. But a number of legal products and plants test positive: chocolate for hashish; rosemary for marijuana; and natural soaps for the "date-rape drug" GHB.

"The tests have no validity," says former FBI narcotics investigator Frederick Whitehurst. And as more organic products come on the market, "the potential for civil rights violations when these presumptive tests are out there is phenomenal." ...

Government officials say there are no records on the number of people who have been wrongly arrested because of the tests. Garrison Courtney of the Drug Enforcement Agency says the test kits are "not perfect but they give you a pretty good idea" whether a suspicious substance is an illegal drug.

Allen Miller of Forensic Source, which makes kits, says they find "families of chemical compounds" and are not meant to be definitive. Any arrest should be the result of good investigative police work, Miller says.

At a minimum, confirmation tests (or admission of use by the testee) should be mandatory before taking either judicial or administrative action, and people should only be required to pay for the confirmation test if it comes back positive - if it exonerates them, the probationer (or whoever's being tested) shouldn't have to pay.

Sloppy forensic science isn't just a problem in high profile rape and murder cases, which is the impression you might get looking only at the long string of Texas-based DNA exonerations - it's just as big a source of false accusations (if not much more common) in penny ante cases where the stakes may not be high enough to justify the risk and expense of fighting them.

CORRECTION: To my chagrin, a commenter corrects me to point out the USA Today article "is talking about field testing on the actual substances found... not urinalysis drug tests." Looking at the story again, that's totally right and I apologize for the error. That said, the principle is the same. When we know cheaper forensics have a higher error rate, access to confirmation tests shouldn't be optional or based on budget concerns.