Translation frustration: When research doesn't reach
■ The pace of basic science is accelerating, but applying that knowledge to clinical trials is difficult. And findings often remain unavailable to in-the-trenches physicians.
- WITH THIS STORY:
- » External links
When John P.A. Ioannidis, MD, researched what happened to more than 100 major scientific discoveries published 20 to 25 years ago in the most significant scientific journals, he didn't find much.
All supposedly had the potential to become major medical developments. But according to his findings, which appeared in the January Journal of Translational Medicine, only five led to licenses for clinical use and only one made significant inroads into medical practice. Three-quarters had yet to be tested in a randomized trial.
"Lots of very good ideas get abandoned, and some are not given a fair shot," said Dr. Ioannidis, an adjunct professor of medicine at Tufts University School of Medicine in Boston. He is also chair of the Dept. of Hygiene and Epidemiology at the University of Ioannina School of Medicine in Greece.
While he notes that some of these innovations should never reach patients anyway, he is one of many who voice concern about the slow, bumpy and sometimes impassable road that scientific discoveries travel from the test tube to the clinical trial setting and finally to real-world medical practice.
The issue has repeatedly been the subject of medical society concern. Last June, for instance, the American Medical Association released a Board of Trustees report diagnosing a "bottleneck" in the translation of science into public health benefits.
Government agencies have also been involved in trying to improve the pathway. Last year, the National Institutes of Health released a "roadmap" -- a plan to change the structure of research in order to address some of the difficulties in transforming good science into good medicine.
"We know that today's scientific landscape demands new ways of thinking, and we know we need to introduce a new paradigm for the conduct of medical research," said NIH Director Elias A. Zerhouni, MD.
While science has accelerated, medicine has not always kept up. Thus, the investment in basic science may not be paying off its full dividend. Experts say that improving translational research, sometimes referred to as "bench to bedside to trench," is a crucial part of this process if people are to benefit from scientific innovation.
"The information just piles up waiting for the person who can sift through and understand how to use [it] for human purposes," said Philip R. Johnson, MD, president of the Columbus Children's Research Institute in Ohio.
But while translational research has become a scientific buzzword, those involved in the process admit the task will not be easy and the road is winding, hazardous and full of potholes. "The pathway to translate basic science and new knowledge in the laboratory to actually do something in a person is unbelievably long and frustrating," said Tim Cripe, MD, PhD, director of the translational research trials office at Cincinnati Children's Hospital Medical Center.
From test tube to trial
The most notable challenges on this road are two points sometimes described as valleys or gaps. The first is between bench science and clinical research. Bridging it involves transitioning from basic findings to attempts to do the research with human subjects. It also involves physician-scientists, an endangered species, who make their home at this point in the continuum linking the test tube and the clinical trials.
Twenty years ago, physician-scientists would easily move between these two points.
"It used to be that one could wander from the bench to bedside and back to the bench," said Ralph Snyderman, MD, chancellor emeritus at Duke University and a visiting professor at the University of California, San Francisco. "That was the most romantic vision for a bench scientist. Now, the ability to jump back and forth is vanishingly small."
One reason: The pressure of today's health care system has made this role less financially viable.
Clinical research is tougher to fit into busy academic medical practices, making the dual role of physician-scientist a less appealing career choice for medical students. Other specialties usually pay more and have better hours.
"Clinical research promises long hours, low pay and hard work," said William F. Crowley Jr., MD, director of clinical research at Massachusetts General Hospital in Boston. "And it delivers on all three."
But physician-scientists are crucial, say most experts, because it often seems that bench scientists and clinical researchers barely speak the same language.
"What's really missing are the individuals who can walk both sides of the street," said Dr. Johnson, who is also a professor of pediatrics at the Ohio State University's College of Medicine. "We've gotten away from training that kind of individual. And what you've got now is a large group of excellent basic scientists and a large group of excellent clinicians who really don't understand how to talk to each other."
Some academic centers have recognized the need, and are now creating translational research departments, but many haven't.
"Bench scientists would like to see translation happen, but they can't suddenly take their discovery and march over across the street to the hospital and start giving it to people," said Dr. Cripe. "And that infrastructure to do that is often lacking in academic centers."
Those who are interested in research are also more likely to stay at the benchside because of the regulatory requirements of human subject research. Experts say that human subject protection is important but may now be actually impeding rather than encouraging research.
For example, when Dr. Crowley, also a professor of medicine at Harvard Medical School, started running trials, the consent forms were about a page and a half and led to extensive discussions. These days, they're more than a dozen pages, most unread and undiscussed and, sometimes patients don't participate.
"Everyone who is doing clinical research wants it to be safe, but efficiency is not the enemy of safety," said Dr. Crowley.
Crossing the next divide
The next valley is between clinical research and patients beyond the trial setting.
This point is where most primary care physicians enter into the process. It is also a place that is the most complex while also the least funded or understood. It involves sending new information into the real world in the hope that physicians and patients will use it correctly.
This area is also the particular interest of the Agency for Healthcare Research and Quality, which has only a fraction of the budget of the NIH, the agency most focused on the first translational block.
"Many have wrung their hands over the inability to get advances in basic science applied meaningfully in medical practice," said Myron Genel, MD, former chair of the AMA's Council on Scientific Affairs and a member of the IOM's Clinical Research Roundtable. "We have guidelines that are produced by some of the most expert societies. We have ample literature and evidence. But the various parts just do not connect well with each other."
For example, in August the New England Journal of Medicine published a study finding that the drug spironolactone was not achieving its promise of reducing heart failure deaths by 30% because in actual practice physicians were prescribing it at higher doses and not monitoring for side effects as frequently as in the clinical trial setting. More recently, Dr. Lisa Simpson, professor of child health policy at the University of South Florida in St. Petersburg and former deputy director of the AHRQ, published in the September/October Health Affairs her own struggles to get evidence-based care for her child's ear infections.
"Just because we have evidence about what should be done does not mean that we know how to change practice or policy," wrote Dr. Simpson.
This area is getting particular attention now because of the belief that the significant improvement in information technology has transformative possibilities, although most recognize it is not a quick solution. "We're moving into an era where we're able to provide information in real time at the bedside, and with that comes a realization that this is a very complex problem and much more complex than we had realized," said Jean Slutsky, MSPH, director of the Center for Outcomes and Evidence at the Agency for Healthcare Research and Quality.
But this translation also does not purely fall to physicians, and many experts believe patients should be involved, too. For example, a study in the December 2002 American Journal of Medicine found that statins did not lower cholesterol as well in a real-world practice setting, primarily because patients weren't necessarily complying with recommendations to take them.
"Ultimately, personal responsibility for health care is something that has to be stressed more," said Dr. Crowley.
Experts say, though, that while potholes and gaps impede progress, the true problem is that everyone is focused on their own kingdoms.
Bench scientists publish and expect clinical researchers to pick up their findings. They then move on to their next discovery.
Clinical researchers complete trials with human subjects, and then expect physicians and patients to be familiar with the evidence.
Meanwhile, many physicians and patients take the process for granted, and there are not enough people focused on whether the beginning of the road actually links directly to the end or if somewhere along the way discoveries are getting lost.
"The problem is that it's not a system," said Enriqueta C. Bond, PhD, chair of the IOM's Clinical Research Roundtable and Burroughs Wellcome Fund president. "We need to knit this group of stakeholders together in a more seamless fashion to facilitate the translation of basic knowledge into things that work in people and then get it applied."