IMPAIRED DRIVING, BY PRESCRIPTION

WASHINGTON -- Drivers on drugs are involved in thousands of fatal accidents
on American roads each year, according to Federal statistics. Most attention
is focused on recreational drugs like marijuana and cocaine, with which any
use is considered abuse. But new research in Britain shows that in many
accidents, the drugs involved are perfectly legal.

The study, by researchers at the University of Dundee, in Scotland, examined
nearly 20,000 crashes over a three-year period, from August 1992 to June
1995, cross-checked the drivers against prescription records and found a
very strong link between crashes and use of tranquilizers. Most of the
compounds they looked at are used in various prescription and
nonprescription drugs, including benzodiazepines, which are widely
prescribed to reduce anxiety. Benzodiazepines (pronounced
ben-zoe-dye-AZ-e-peens) are found in Ativan, Halcion, Librium, Valium and
Xanax. Zopiclone, sold in the United States as Inovane, a short-acting
tranquilizer used as a sleeping pill, was also cited. The problem of
impairment by prescription is overshadowed by drunken driving. A
Transportation Department report in 1992, of 1,882 drivers killed in traffic
accidents, found that 51.6 percent had alcohol in their blood streams,
including 42.6 percent who had blood-alcohol levels of 0.10 percent or
more -- fitting the definition of drunken driving in most states. The
runners-up were cannabis and cocaine, at 6.7 percent and 5.3 percent; the
top-ranking legal drug was benzodiazepine, at 2.9 percent. But people who
drink and drive face a drumbeat of warnings and, sometimes, invervention
from friends and family; no such stigma attaches to prescription drugs.

In addition, researchers have always wondered whether people using marijuana
or cocaine tend to be involved in crashes because of the drugs themselves,
or because they have risk-taking personalities; in contrast, people who use
tranquilizers are not believed, as a group, to have the same predilection
for risky behavior.

At Hoffman-La Roche, which uses benzodiazapines in Valium and Librium, as
well as several drugs inused only in hospitals, Martin D. Hirsch, a
spokesman, said that the company did not have much information on the study.
But he said that the company was already warning people not to drive while
using these drugs, which act on the central nervous system. He pulled out a
package insert and quoted from it: "It says, 'As is true of most
preparations containing CNS-acting acting drugs, patients receiving Valium
should be cautioned against engaging in hazardous occupations requiring
complete mental alertness such as operating machinery or driving a motor
vehicle.' "

It is hard to tell how many American drivers are involved in fatal accidents
with alcohol in their systems. The National Highway Traffic Safety
Administration keeps a data base of traffic accidents, but it reported that
of all the drivers involved in fatal crashes in 1997, 60 percent were not
tested for drugs. It was unclear whether another 18 percent were tested. Of
the remainder, about 15 percent were tested but no drugs were found; for
nearly 2 percent, drugs were found but the type was not determined.
"Everybody openly admits we don't test enough drivers in fatal crashes,"
said Terrance D. Schiavone, executive director of the National Commission
Against Drunk Driving, an advocacy group based here. And if investigators
test and find alcohol, then "they're done with testing," he said, although
alcohol may not be the only substance in the driver's bloodstream. Mr.
Schiavone said he did not favor any approach that would take the emphasis
off alcohol.

A focus on alcohol would help even drug cases; many of those found to be
using drugs had also been drinking.

In the United States, drivers in fatal crashes who were known to be tested,
and for whom investigators recorded the drug type found, accounted for fewer
than 5 percent of the drivers involved in fatal accidents. But they may not
represent all drug users involved in accidents. Among those tested, the most
popular drugs were stimulants, followed by cannabinoid (marijuana-like)
drugs, depressants and narcotics.

The British researchers found, though, that prescription drugs were a big
problem for the people taking them. Of 19,386 drivers who had their first
accident in the study period, 1,731 were users of the drugs being studied.
People younger than 45 had triple the risk of being involved in an accident,
compared with nonusers; for people over 45, the risk was about double that
for nonusers.

The research, by the Medicines Monitoring Unit of the University of Dundee,
published in October in The Lancet, the prestigious British medical journal,
contrasts with previous research in this country that has focused mostly on
drunken driving. An October 1992 report by the Departent of Transportation
found only limited data that drivers on drugs, legal or illicit, had an
increased chance of causing an accident, and said that when drugs were
combined with alcohol, "no drug or drug group exhibited a responsiblity rate
significantly different from alcohol by itself."

The British report said that women were more likely to be using the
prescription drugs, a finding echoed in this country. A report by the
National Highway Traffic Safety Administration in December found that men
were overrepresented among those testing positive for alcohol and for every
class of drugs except benzodiazepines.

The report also found that nine million people, or 5 percent of drivers,
said that in the previous year they had driven within two hours after drug
use; 45 million people, or 27 percent of drivers, had driven after drinking
alcohol. Some people said they had done both -- not all of them
concurrently.