As part of the agreement that would have guaranteed them higher wages, members of the Hawaii teachers union agreed to a random drug testing regime. In 2007, during the negotiations over the next employment contract, the state's Republican governor Linda Lingle proposed the compromise, saying that a job like teaching, though it wasn't dangerous per se, was still unpredictable enough to make some kind of a drug-testing program acceptable.
Although no drug tests have yet been administered, now some teachers are regretting the bargain they've made and want the provisions requiring random drug testing revoked. This resistance comes on the heels of recent court cases that put similar programs to the legal test in North Carolina and Virginia. But getting lost the in the rhetoric and the legal maneuvering is one important question: what impact does a random drug-testing program have on rates of drug use?
No studies I found have looked at the specific issue of whether random drug tests affect substance use among teachers. But several studies have examined the impact of random testing in another school population — students. In the most comprehensive study on the subject to date, a 2003 University of Michigan study involving 894 middle and high schools found that random student drug-testing tends to reduce marijuana use slightly (about 5%) but actually increase the use of other drugs (about 3%). The authors theorize that drug-using kids may think that prescription and other drugs are harder to detect by urinalysis, so they switch from pot to something else.
But how much about the behavior of teachers can be extrapolated from that of their students? Answering that would only give rise to more questions: like if the society is serious about stamping out drug use, shouldn't everyone who comes into contact with children be tested? Why aren't they?
One answer is cost. In the West Virginia drug-testing case, which is currently working its way through the federal court system, Judge Joseph Goodwin of the U.S. District Court noted that it costs about $44 a pop to do urine tests, which would cost the West Virginia school district in question about $37,000 a year. (Here's a PDF of Goodwin's preliminary injunction against drug-testing.) That same $37,000 could easily pay for a full-time teacher, meaning that drug-testing would have to be sufficiently valuable to displace an entire teaching position.
Another issue is that the scale of the problem posed by drug use by teachers could be overstated. A major study published by the Department of Health and Human Services in 2007 found that, of the 19 occupations ranked by prevalence of illicit drug abuse, teachers were ranked 18th. Compared to 18% of construction workers who reported that they had used drugs in the past past month, only 4% of teachers reported likewise. Even the National Education Association, the national umbrella group for teachers unions, agrees that drug testing of teachers who, by their behavior, indicate that there might be a problem, is sound policy. It's the random testing of those teachers who aren't exhibiting any signs of drug use that union members find demeaning.
The matter won't be resolved without further studies on whether random drug-testing actually reduces drug use. But the data so far suggest that random drug-testing is a costly, ineffective solution to a non-problem.