DRUG TESTING TAKES A HIT

New Studies Question Value Of Screening For Illegal Substances

Drug testing on the job, once a controversial practice at a few
companies, has become so pervasive that it now seems as common as
filling out a W-4 form or punching a time clock.

Want that high-profile new job at a Fortune 200 company? Here's your
cup, there's the bathroom. Give us a urine sample, then we'll talk
stock options, pal.

Want to stay employed in that construction job? Better watch what you
ingest over the weekend because you may be randomly selected to give a
sample before firing up the bulldozer Monday morning.

In 1986, only 21.5 percent of companies tested employees, according to
a survey by the American Management Association. By 1996, 81 percent
did.

The number of Fortune 200 companies that require pre-employment or
random drug testing grew from 6 in 1983 to 196 in 1996, the AMA found.

Eighty-three percent of employers surveyed believe that testing slows
employee drug use, according to the AMA study. But 80 percent of
companies in the same survey had never done a cost-effectiveness analysis.

Now, for the first time, several studies question the worth of
workplace drug testing.

In September, the American Civil Liberties Union issued a report based
on studies by the National Science Foundation and the AMA showing that
testing has been ineffective in reducing drug use and has no
noticeable impact on reducing either absenteeism or
productivity.

The National Academy of Sciences recently found that illegal drugs
contribute little to workplace accidents and that off-duty drug use
has about the same small effect on worker accidents as off-duty drinking.

And, in January's Working USA magazine, two researchers with the Le
Moyne College Institute of Industrial Relations surveyed 63 Silicon
Valley companies and found that productivity was 29 percent lower in
firms with pre-employment and random testing.

Still, few businesses have abandoned drug testing, even though the AMA
found it costs a company $77,000 to find one drug user by testing all
employees.

Trying To Look Good

``If drug testing didn't work, why would we see so many companies
instituting policies in an era where every department in a corporation
has to prove its worth?'' asked Mark de Bernardo, executive director
of the Institute for a Drug-Free Workplace and managing partner at
Littler Mendelson's Washington, D.C., office.

``I don't doubt for a second it works. It's ludicrous to think a
program that results in deterrence and detection of substance abuse
doesn't work.''

But Lewis Maltby, director of the ACLU's National Taskforce on Civil
Liberties in the Workplace, called for employers to rethink what he
calls ``an invasive and humiliating procedure for employees.'' Maltby
charges that most large and medium-size corporations know there is
little cost benefit or effectiveness in testing employees. They do it,
he says, for public relations.

``I have a friend who formerly was the CEO of a Fortune 500 company,
and I asked him why he spent all that money on a drug-testing program
even though he'd told me privately he knew it wasn't doing that good
for the bottom line,'' Maltby said.

``And he said, `I don't care if it improves bottom line. I'm a
publicly held company. I've got stockholder meetings and every year I
get asked what we're doing about the drug crisis in America. The
policy is my answer.' It's there for image.''

Education More Effective

Eric Greenberg, director of management studies for the AMA, said drug
education and awareness programs have proven more effective than
testing, according to his group's research.

``Ten years of survey data have not allowed us to make a statistical
case that drug testing makes a difference,'' Greenberg said.

SmithKline Beecham Clinical Laboratories, the nation's leading
drug-testing corporation, recently reported that positive test results
fell from 18.7 percent of pre-employment and random testing in 1987 to
5 percent in 1998.

But Greenberg argued that test-positive rates went down not because
fewer workers were doing drugs, but because more companies moved from
testing only ``for cause'' to more widespread random testing.

``In all of our years of researching, the only hard case we can make
in our data is that drug education and awareness programs in companies
deter use, whether they test or not,'' Greenberg said.

Bay Area corporations are believed to have a lower pre-employment
testing rate than companies in other regions, though no statistics
exist, because private-sector companies are not required to make
policies public.

Levi-Strauss, for instance, performs pre-employment drug testing on
employees in manufacturing and logistics, according to a
spokeswoman.

The Bay Area's other major apparel company, the Gap, does not require
pre-employment testing, according to a spokeswoman.

Intel instituted pre-employment drug testing shortly after Congress
passed the 1988 Drug Free Workplace Act, which requires federal
contractors and grant recipients to provide drug-free workplaces.
Though Intel does not do federal contract work, it instituted testing
to enhance quality control.

``It's just the nature of this business,'' Intel spokeswoman Tracy
Koon said. ``It's a matter of safety and productivity. We want to make
sure our product is high quality.''

Koon said that in 1997, Intel had 0.39 percent positives out of 13,165
tests. In 1998, positives fell to 0.18 percent out of 3,696 tested.
This year, through October 19, Intel had 0.15 percent positives out of
6,294 pre-employment tests.

``We are convinced that our testing has helped these numbers fall and
raises our productivity,'' Koon said.

Testing Might Hurt Work

But many Silicon Valley firms, such as Cisco Systems, do not
test.

Eric Shepard, co-author of the Le Moyne study of drug testing in
Silicon Valley, said his researchers combined each company's
drug-testing data with its public financial information.

``We found that productivity was 16 percent lower in companies with
pre-employment testing than those that didn't test, and it was 29
percent (lower) in companies with both pre-employment and random
testing,'' he said. ``It's hard to determine exactly why that is,
because it's not easy to get companies to talk about drug testing at
all.''

Shepard said his survey didn't delve into the reasons productivity
declined, but he has a theory.

``If drug tests contribute a negative view toward the company, as
other surveys have found, then workers may not contribute as much in
return, or they may seek employment elsewhere,'' Shepard said. ``You
may lose your best workers to companies that don't test.''

Dan Abrahamson, a San Francisco attorney for the Lindesmith Center, a
national drug policy institute that opposes drug testing, said he
receives at least one e-mail a week from high-tech workers who smoke
marijuana away from the job and are concerned about drug testing at
work.

``There are a lot of smart, creative people who work in Silicon Valley
in programming and they feel it helps them intellectually to use
marijuana,'' Abrahamson said. ``So testing might actually hurt their
work.''

Ed, a 27-year-old financial analyst at Charles Schwab in San Francisco
who declined to give his last name, said he would have thought twice
about accepting an offer from the company six months ago if that
company required pre-employment drug tests.

``I don't use drugs,'' he said, ``but I would look at that company as
not as trusting (and) more rules oriented, as opposed to a place that
values its employees and entrusts them to do a good job.''

San Francisco was the first city in the country to pass legislation
limiting drug testing in the workplace. In 1985, the Board of
Supervisors passed a statute drafted by civil rights lawyer Cliff
Palefsky banning random drug testing except for workers in
safety-sensitive jobs.

Safety-Sensitive Jobs

Several years later, California also passed a statute allowing random
testing only for employees in safety-sensitive jobs. What constitutes
``safety sensitive'' is open to legal interpretation, however.

In 1996, a worker at the Fresno Irrigation District was fired for
failing a random drug test. Ron Smith, a ditch digger who had a
spotless record and won five safety awards in six years on the job,
won a suit in Fresno Superior Court saying the test violated his
constitutional rights because he wasn't employed in a
``safety-sensitive job.''

However, in May, the Fifth District Court of Appeals in Fresno
overturned the ruling, saying that Smith's job was safety sensitive.

``Anyone operating a big piece of heavy machinery, sure, that might
harm other people, but Smith barely even used a shovel,'' said Joseph
A. Davis, Smith's attorney. ``Fresno argued that Mr. Smith could
possibly fall into the canal and someone else may be imperiled going
in to save him. Well, if that's the case, we're all in
safety-sensitive jobs.''

For those in non-safety-sensitive jobs, some companies will be more
lenient in punishing drug-test offenders, notes Greenberg of the
American Management Association. Not that they will admit it publicly.

``I know of companies that test for drugs that are illegal but really
don't affect job performance,'' Greenberg said. ``They'll pick and
choose which positive tests to act on. Some companies just don't want
to know. Any advertising firm that gave its copywriters pre-employment
tests would have a real hard time filling positions, if you know what
I mean.''

CHART:

HOW LONG CAN DRUGS BE DETECTED?

Drug Usage Level Period Of Detection

Marijuana:

moderate (4 times a week) 4 days

heavy use 7 days

chronic heavy use 21-30 days

Cocaine:

any level 2-3 days

Opiates:

any level 2 days

PCP:

any level 3-8 days

Amphetamines:

any level 2-4 days