More MS news articles for August 2000

Clinical Trials Bring Out Your Inner Liar

Study says 30 percent said they took drug but didn't

By Adam Marcus

HealthSCOUT Reporter

WEDNESDAY, Aug. 16 (HealthSCOUT) -- Most people consider themselves honest. So what goes wrong when they enroll in a clinical trial?

As many as 30 percent of clinical subjects pretend to take drugs that they really just throw out, new research shows. The finding alarms researchers, who say that deception in drug studies can mislead scientists, companies and regulators -- and ultimately harm patients.

Researchers aren't naive, and have long known that some people who participate in clinical trials "dump" their medication instead of taking it as instructed. The trouble, however, is how to prove it.

In pill studies, for example, an empty vial is merely that -- it says nothing about whether the drugs ended up in the subject's bloodstream or in a trash can.

Hoping to skirt this information void, the researchers in the latest study took a different tack. Bringing the mountain to Mohammed, they used home monitoring devices to measure dumping rates among 101 patients enrolled in the Lung Health Study, a 5,000-subject look at the effects of inhalers and smoking cessation counseling on men with chronic obstructive pulmonary disease.

The devices -- "nebulizer chronologs," which, unlike the patients, don't lie -- recorded that 30 percent of subjects who didn't know they were being monitored emptied their inhalers within a three-hour period at least once in the first year of the study. Most of the "dumping" episodes happened shortly before a visit to the clinic.

On the other hand, only one patient out of another 135 who did know about the tracking dumped.

"Deception among noncompliers occurs frequently in clinical trials, is often not revealed by the usual methods of monitoring, and cannot be predicted by data readily available in clinical trials," write the researchers.

Noncompliance accounted for

Yet studies where dumping is common don't necessarily impugn the efficacy of the drugs they test. On the contrary.

Dr. Donald Tashkin, a lung specialist at the University of California at Los Angeles and a co-author of the study, says drugs that show marginal to modest benefits in clinical studies may in fact be better than those results imply.

"In clinical trials the efficacy of inhaled medications may be underestimated," says Tashkin. Although compliance with inhaled drugs, such as those for asthma, is notoriously poor, the same probably goes for pills, too, he says.

Dr. Jacob Lalezari, director of Quest Clinical Research, a private company that conducts studies of new antiviral drugs, calls patient compliance and deceit "a serious problem" for scientists.

"Even the most conscientious, dedicated-appearing patient can deceive you," says Lalezari. He recalls one man, a psychologist enrolled in an early trial of the AIDS drug AZT, who eventually confessed that he took not a single pill.

Most researchers account for non-compliance by enlarging their studies enough to preserve their statistical power, Lalezari says.

"We try to draw conclusions based on reproducible studies, but a certain amount of it is artwork," he says. On the other hand, he adds, "there's nothing we can do instead."

Unfortunately, says Tashkin, systems like the timed dose meter are too expensive to implement widely in clinical trials. Each meter runs $300, not including the cost of collecting and processing the data it collects.

Ultimately, he says, behavioral researchers may be able to identify the personality types most likely to cheat in a study. But Tashkin and his colleagues found no tip-offs for which subjects would be most likely to lie in their trial.

"What would motivate someone to do that rather than confess is something that I don't completely understand," he says. Some subjects may find they can't tolerate the drugs but want to please the scientists. Or others may simply have no intention of complying to begin with.

What To Do

If you're interested in experimental drugs and procedures, visit ClinicalTrials.gov, which is run by the National Library of Medicine.