From Clinical Trials to Clinical Practice

In the age of information, patients expect doctors to make treatment decisions based on the most current evidence available, but experience shows that doctors don't adopt lessons from clinical trials nearly as quickly as they could.

When patients walk into the doctor's office, they expect to receive treatments backed by solid evidence. The strongest evidence comes from clinical trials in which hundreds or thousands of people receive a particular treatment and are monitored to find out whether they fare better than people who receive a placebo or treatment with a different drug. The results of such trials could potentially save lives, but studies show that many doctors remain behind the times when it comes to learning from the literature. "In general the translation of evidence into practice is very slow," says Sumit Majumdar, an internist at the University of Alberta in Edmonton, Canada. "It usually takes on the order of 5 to 10 years from when it is published to when it is put into practice." Such delay can hurt patients, who might be denied access to superior drugs or given harmful treatments.

Ideally, doctors should make decisions based on the most current information possible, but they must learn about and digest the results of the latest studies first. For clues about how to help speed this medical translation, researchers are examining how information about past clinical trials flowed from investigators to clinicians. Two studies published in January in the Journal of the American Medical Association (JAMA) suggest that scientists should learn to exploit the media if they wish to convey their results to physicians and patients.

Researchers are beginning to realize that publishing their results in a journal is not enough to get doctors to adopt new practices. "Those of us in academia have thought that getting the evidence published or put in a guideline is the last step in translation, when in reality it's the first step," says Majumdar. In a perfect world, physicians would read journal articles and attend meetings to follow developments in their specialty. But "a general physician needs to read about 17 articles per day to truly keep up," says Majumdar. In the real world, doctors crunched for time end up receiving much of their information about new treatments by word of mouth. When the big-name physicians in a given community adopt a new practice, their peers usually fall in line. This herd mentality, says Majumdar, can be used to advantage: One study showed that when the leaders in the field were persuaded to adopt a particular treatment protocol, others quickly followed. In this case, getting the "opinion leaders" on board sped adoption of the practice by 20% to 30% compared to communities where opinion leaders weren't targeted.

But doctors don't just listen to their peers; they also pay attention to the news media. That's one conclusion that came from the JAMA papers. In these studies, Ronald Stafford, an epidemiologist at the Stanford Prevention Research Center in Palo Alto, California, and his colleagues examined how quickly physicians absorbed the lessons from two large-scale clinical trials. One of these trials, the Women's Health Initiative (WHI), showed that hormone replacement therapy (HRT) using a combination of estrogen and progestin failed to provide the anticipated protection against cardiovascular disease, instead sparking a small increase in the risk for breast cancer, heart disease, pulmonary embolism, and stroke. The other showed that patients taking alpha blockers, a class of drugs that reduces blood pressure, had a 25% greater risk of cardiovascular problems than did those taking another type of antihypertension drug.

In both cases, clinicians wrote fewer prescriptions for the questionable drug after the trials were published, according to Stafford's group. Prescriptions for the estrogen-progestin combination used in the HRT study plummeted 56% in the year following the trial's publication, and prescriptions for alpha blockers fell off about 30% during the 2 years following the heart disease trial, called ALLHAT.

These drop-offs occurred more quickly than the changes in prescription habits that researchers had noted in previous studies. Although the relatively rapid response to the WHI and ALLHAT trials cannot be pinned on a single factor, Stafford says that the resulting news coverage was surely at least partially responsible. "The media coverage of the HRT story was really unprecedented, and there was also considerable coverage of the alpha blocker trial when it came out," he says. The extensive attention the popular press gave these trials was unique. "These are two trials that terminated early because the harms from treatment exceeded the benefits, and that makes them much more newsworthy than the normal conclusion of a trial with positive effects," says Stafford. Researchers can take advantage of the media to help disseminate the results of their clinical trial to the public by talking to reporters and helping them put study results in context, Stafford says: "The researchers at WHI were aware of this and did a very good job of highlighting their finding."

Drug companies also use the popular press to spread their message. Before the WHI results came out, the pharmaceutical industry spent billions each year to market HRT, even though the drugs had never been tested in large-scale clinical trials. And although doctors might deny it, drug company marketing influences which drugs they prescribe. "When you survey physicians, they always say they get their information from textbooks and journals, but when you study their actual practice, it's much more in line with what the industry promotes," Majumdar says.

"The good news is that most of the compounds that are advertised and promoted are reasonably effective," says David Naylor, an internist and dean of the faculty of medicine at the University of Toronto, Canada. "The bad news is that sometimes there are safer and cheaper alternatives." When a trial produces positive results for a lucrative drug, industry marketing quickly disseminates the new evidence, but older drugs or generic ones that are shown to be superior or equivalent don't receive such marketing, says Stafford. Insurance companies might also play a role in promoting the latest evidence, but they too have a conflict of interest and could hesitate to publicize results that promote expensive treatments.

In the end, the patients themselves exert a strong influence over which drugs doctors prescribe. "There's no doubt that consumer demand played a role in the decline of HRT," says Stafford. "Generalist clinicians sometimes report that patients are more up-to-date about the latest pharmaceutical options than they are," says Naylor. Although no one believes that patients should bear the responsibility of keeping their doctors informed, those who want to ensure that their doctors prescribe treatments backed by clinical trials should not be afraid to arrive at the exam room with the relevant evidence in hand.

Christie Aschwanden is a freelance writer in Zuoz, Switzerland, who always arrives at the doctor armed with MEDLINE printouts.