Drug testing works, but maybe not as well as bosses think

Employers screen their workers and job applicants with the expectation that testing will deter worker drug use. It’s a cause-and-effect relationship that many worksites rely on, and a belief that fuels a multibillion-dollar drug-testing industry.

The study from the University of California Irvine examines alternative explanations to test the link between drug testing and lower rates of substance abuse.

“My results don't definitively prove that drug testing directly reduces drug use, but they are the strongest evidence to date,” said Christopher Carpenter, a health economist at UC-Irvine’s Paul Merage School of Business.

Other workplace drug policies — like a written “zero tolerance” standard or employee assistance programs — don’t explain away the association between testing and less worker drug use, found the study in the online edition of Health Services Research.

Carpenter also considered the health profile of employees at worksites with lower drug-use rates to determine if healthier workers self-select workplaces that are more likely to screen their employees.

Because other policies and workforce characteristics likely dampen drug use to some degree, and because previous research did not account for those effects, Carpenter says, past studies may have overstated the testing-drug use link.

Failing to account for other workplace characteristics and drug policies may bloat the testing-drug use association by as much as 25 percent, Carpenter said.

That’s valuable information for budget-conscious personnel managers who are on the fence as they weigh the costs and benefits of establishing a drug-testing program.

One national drug-testing firm charges from $25 to as much as $65 per drug test, according to its Web site.

“If you tell an employer that workplace drug testing will reduce worker drug use by 25 percent less than they expected, this may affect an employer's decision to implement drug testing in the first place,” Carpenter said.

Mirroring previous study designs, the Irvine research compiles data on marijuana screening at private, for-profit firms across the country. Marijuana is the drug that turns up most often in failed worker drug tests.

John Hoffmann, a sociology researcher at Brigham Young University, says despite new evidence that drug testing works, its true value and best uses are still not clear.

“It tells us nothing about the degree of drug use or impairment of the employee, whether workers are problematic drug users or maybe weekend pot smokers,” Hoffmann said. “We are simply saying to workers, ‘If you use drugs, you cannot work here,’ rather than trying to find those people who might need help.”

Hoffmann was not involved in the UC-Irvine research, but has led national studies on workplace drug testing. Questions also linger about the economic value of drug testing, said Hoffmann, a professor in Brigham Young’s College of Family, Home and Social Sciences.

“If there aren’t good cost-effectiveness studies out there, there need to be — before employers embrace this completely,” Hoffmann said.

Health Services Research is the official journal of the AcademyHealth and is published by Blackwell Publishing on behalf of the Health Research and Educational Trust.

Carpenter CS. “Workplace drug testing and worker drug use.” Health Services Research (online), 2006.