To Judge Sleep Aids, U.S. Looks at Drowsy Driving in the Morning

The Food and Drug Administration is taking heightened interest in the issue, as new evidence suggests what many people have long suspected: the effects of common prescription sleep aids like Ambien can persist well into the next day. Of particular concern is whether people who take the drugs before bed can drive safely the next morning.

Consumer advocates have warned for years about possible links between sleep drugs and car accidents. In one prominent example, Kerry Kennedy, the former wife of Gov. Andrew M. Cuomo, was arrested last year after tests showed she had taken a sleep aid before swerving her car into a tractor-trailer.

The F.D.A.’s actions are part of a robust national conversation about how to cope with the throngs of drivers who take to roads every day under the influence of prescription drugs. Law enforcement authorities have struggled with how to prosecute those who are impaired, especially when they have a prescription. A government survey in 2007 found that nearly 5 percent of daytime drivers tested positive for prescription or over-the-counter medications.

Doctors wrote close to 60 million prescriptions for sleep aids in the United States last year, according to the research firm IMS Health, but experts say testing how these drugs affect driving is not easy. Nonetheless, the F.D.A. has been unusually active.

Last month, it rejected an application by Merck to approve a new sleep drug, suvorexant, in part because tests showed that some people had trouble driving the next day. In May, the agency warned patients taking common allergy drugs like Benadryl against driving, noting that the sedating effects can sometimes last into the following day. In January, citing similar concerns, the F.D.A. took the unusual step of requiring that all manufacturers of zolpidem, the generic name of Ambien, cut in half the dosage for women.

The agency has since said that it is taking a closer look at all insomnia drugs on the market, and will ask manufacturers to conduct more extensive driving tests for all new sleep drugs. It will also more closely scrutinize any drug that causes drowsiness.

Part of the problem is that the standard warnings on many drugs — against driving or operating heavy machinery — are often brushed aside. “It would be so convenient and it would be so good if you could just tell people, don’t drive unless you feel O.K.,” Dr. Ronald Farkas, the clinical team leader for the F.D.A.’s division of neurology products, told a group of industry experts at a conference in February. “I think this has penetrated now that this is not adequate. It is still good advice that, if you feel impaired, don’t drive. But if you feel fine, you might be impaired.”

For years, traffic safety officials have called, without success, for a so-called “safe list” of drugs that do not affect driving. Some people react more strongly to drugs than others do, making it difficult to predict a person’s individual risk. And driving tests can vary significantly, like studies conducted on roads or those done in computer simulators.

“It seems like a really simple question, but it’s really difficult,” said Dr. Matthew Rizzo, a professor of neurology at the University of Iowa, who studies driving abilities in people with cognitive impairments caused by conditions like Alzheimer’s disease.

Pharmaceutical companies have long conducted driving tests for certain drugs, but the tests were motivated mainly by marketing goals and not seen as pivotal in getting a drug approved by the F.D.A., said Gary G. Kay, whose company, the Cognitive Research Corporation, leases a driving simulator to pharmaceutical companies. Mr. Kay, for example, helped conduct some of the research showing that the allergy drug Claritin did not cause drowsiness, unlike older antihistamines like Benadryl.

Until recently, the F.D.A. evaluated driving tests by looking at the reaction of the average patient. But that thinking has shifted as research has shown that some people react more strongly to drugs than others, Dr. Farkas said in a telephone interview. “Some people have higher blood levels than others, and some seem to be more sensitive than others,” he said.

In 2011, the agency recognized this variability when it approved the sleep drug Intermezzo, which treats patients who wake up in the middle of the night, in two doses: one for men and a smaller dose for women. Data had shown that the drug’s effects lingered in women for longer than they did in men. The agency’s experience with Intermezzo, along with other research, led it to order in January that the dosage for all drugs containing zolpidem — the active ingredient in Intermezzo, as well as Ambien and related generic sleep drugs — be halved for women.

In the case of suvorexant, the Merck drug, four women had to stop the driving tests because of extreme drowsiness, and the F.D.A. concluded that the drug was linked to an increase of impaired driving among patients who had taken it. Last month, Merck said that the F.D.A. indicated it would eventually approve the drug if the company lowers the proposed doses.

The attention has not been limited to insomnia drugs. In 2011, the F.D.A. approved the drug Horizant to treat restless leg syndrome after the drug makers, XenoPort and GlaxoSmithKline, conducted driving tests because the drug also causes drowsiness. Horizant carries a warning against driving until patients know how the drug affects them, and notes that the effects could last until the morning after it is taken.

Dr. Eric Bastings, the acting director of the F.D.A.’s neurology drug unit, said the agency was writing guidelines to help pharmaceutical companies understand what is expected of them.

Until then, however, there is a wide range in how pharmaceutical companies measure driving ability. The most well-established test takes place in the Netherlands, where participants drive in specially outfitted cars — and alongside a driving instructor — on open roads. It is the only place in the world where such tests are conducted on roadways.

Drug companies still conduct the tests in the Netherlands — Merck’s drug was tested there, as was Intermezzo — but driving simulators are also growing in popularity. Dr. Kay’s company leases the MiniSim, a scaled-down version of a state-of-the-art simulator at the University of Iowa. Users sit in a dummy driver’s seat, complete with steering wheel and pedals, while three computer screens stand in for the windshield.

Despite the increased scrutiny for sleep drugs, however, some sought to place the issue in perspective. “What we have to weigh here is that people who don’t sleep well are a hazard on the road because they do fall asleep while driving,” said Robert Voas, a senior research scientist at the Pacific Institute for Research and Evaluation and an expert on driving safety. “There’s risks on both sides.”