Physician-scientists: a dying breed?☆

The winners of the Damon Runyon-Lilly Clinical Investigator Award are, by any measure, some of the leading lights in clinical research. Employed by leading hospitals and medical schools—the 2003 winners hail from the Dana-Farber Cancer Institute in Boston, from Stanford University School of Medicine in Stanford, California, from Vanderbilt Medical Center in Nashville, Tennessee, and from Memorial Sloan-Kettering Cancer Center in New York, among other top institutions—the young physician–scientists have all been guaranteed appointment to a tenure-track position, as well as research and office space, plus a promise that they will be able to devote 80% of their time to research. The award brings each winner an annual stipend of $100,000 for 5 years plus a research allowance of $75,000 annually for 4 years plus $100,000 in the final year. The scientists' mentors also receive a stipend, totaling $95,000 spread over 5 years.

All of which is to say that these rising stars are by no means whiners or sore losers.

Yet they are no more sanguine about the state of clinical research—also called patient research or translational research—than other observers. As a recent paper in the Journal of the American Medical Association by members of the Institute of Medicine's Clinical Research Roundtable found, “clinical research is increasingly encumbered by high costs, slow results, lack of funding, regulatory burdens, fragmented infrastructure, incompatible databases, and a shortage of qualified investigators and willing participants” (1). These factors, the authors continued, are “impeding the translation of basic science discoveries into clinical studies.”

That sense is already bringing a political fallout. After 5 years of steep funding increases that doubled the budget of the National Institutes of Health (NIH) from 1998 to this year, the 2004 budget request was $27.9 billion, which would barely cover inflation; apparently, the White House and Congress are both waiting to see what the nation's biomedical researchers can do with the billions already allotted. And in May of this year, Sen. Joseph Lieberman of Connecticut announced his proposal for an American Center for Cures. This project, said the contender for the Democratic presidential nomination in 2004, would have the explicit purpose of turning the discoveries of the Human Genome Project and other basic research into cures and treatments for chronic and life-threatening disease. The implicit message: NIH and the rest of the existing biomedical infrastructure is not doing the job.

View Large Image Download to PowerPoint

Fig 1.

The reasons for the dearth of clinical research, which moves discoveries from bench to bedside, are legion. In talking to some of the brightest lights in the nation's medical schools and teaching hospitals, one gets the sense of a system beset by practical, financial, and ideological barriers to clinical research. A constant refrain is that, to do clinical research, one needs “protected time,” free from the pressures to see patients and bring in clinical revenues. Yet protected time is exactly what this generation of physician–scientists doesn't get, as even the Runyon-Lilly awardees explained. “To do patient research, which is difficult and time-consuming, you can't be in the clinic generating billables,” said one.

Twenty years ago, says Ernest Beutler, chairman of Molecular and Experimental Medicine at the Scripps Research Institute in La Jolla, California, department chairs and hospital administrators didn't care if “the examining room was empty and no revenue was being generated” while a physician–scientist conducted research. But today, says Dr. Buetler, “physicians are no longer able to spend time performing clinical research unless [it's] paid for.”

The lack of protected time is driving countless physician–scientists out of research. Ellen Deibert, 40, found out the hard way. A former chief resident in neurology at the Johns Hopkins Medical School in Baltimore and then at Washington University School of Medicine, the neurologist had high hopes for combining research, teaching, and seeing patients in the brain-injury program she directed at Washington. But by this spring, she had decided that she “can't get out of academic medicine fast enough. Physician–scientists who spend 90% of their time in the lab are a dying breed: It's not financially viable for a department.”

Dr. Deibert was encouraged to sign onto the “clinician” track at Washington, where promotion is based on becoming nationally known (difficult without doing research). “But in the academic world there is little respect for the clinician, and academic advancement is still based on research. I tried many times to get research going, but with most of my time seeing patients with minimal office help, virtually no mentoring and no protected time, it was frustrating if not impossible. At this point I cannot get out of academics fast enough.”

One young researcher at a Harvard Medical School teaching hospital in Boston hasn't quite reached that point, but she's close. “Patient-oriented research isn't considered ‘real’ science, even though it is very hard to do and to do well,” she says. “As a clinician, they occupy all your time doing clinical work, leaving too little time to write grant proposals, let alone conduct the research.”

The problem is not a lack of will and desire on the part of physician–scientists; in fact, we're not doing too badly recruiting young people to this career track. The front end of the pipeline isn't the problem—retention is. “The efforts to increase the ranks of physician–scientists focus on college students and medical students,” says the Harvard clinician. “But the problem is not recruitment but retention. There just isn't a lot of support later on.”

Some of the blame for the leaky pipeline—young clinical researchers bailing out in favor of just treating patients, typically—goes to academic politics. Many clinical researchers speak of a bias against M.D.'s who would like to conduct patient research. They describe a two-tier system at teaching hospitals and other medical centers that encourages such research by M.D.–Ph.D.'s but discourages it among M.D.'s. “M.D.'s were given the shoulder and not welcomed into the lab,” says another young researcher at a Harvard teaching hospital. As a clinician, they occupy all your time doing clinical work, leaving too little time to write grant proposals, let alone conduct the research.”

The day-to-day practical hurdles to clinical research are only the beginning. Many critics point to what one might call the sociology, or workplace culture, at leading biomedical institutions, and how that translates into a value system. “Because clinical medicine is getting more complex and basic science is getting hugely more complex,” says David Nathan of Harvard Medical School, patient research requires collaboration between basic biologists and physicians. “That means you'll need a new system of academic awards.” Currently, promotions are often based on how many papers a physician–scientist writes, but institutions typically count only those papers on which someone's name appears first (meaning he or she did most of the actual work) or last (usually the senior investigator or laboratory head who oversaw it) in the list of authors. Numerous clinical researchers are wondering why they should collaborate if their name will be third of seven.

And full-time clinicians who control access to the patients that researchers need are sometimes dubious of their would-be collaborators. “Some clinicians will stereotype you as someone without a real interest in clinical medicine, as someone who is only using their patients to further his own career,” says one young investigator. Adds another, “Many look at you as having a divided loyalty. You are not fully committed to treating patients, and you're not solely doing research. If, when you come out of medical school, you want to put clinical medicine and research together, bridging basic and patient research, you don't really fit in either world.”

Comp:Please drop the first 〈CAB〉 from the end of the article here. Follow style. Use red.

Then there's the plain fact that patient research is hard—harder than research on Drosophila or mice or any other animal model. “Biomedical science has become more technically and intellectually sophisticated,” one Runyon-Lilly awardee says. “There has been a tendency to encourage physician–scientists to veer toward basic rather than patient research, because you have a much greater chance of succeeding quickly there. It's simply a lot easier to excel in the basic arena.” Even newly minted M.D.–Ph.D.'s with all the intention in the world of entering the world of patient research can be dissuaded by the allure of a simple slime mold model.

The reason is obvious. “Human experiments are much more time-consuming and more difficult than animal studies,” says James Krueger of Rockefeller University in New York City. His human research includes trying to correlate gene activity and changes in immune system cells with the progression of psoriasis. “There are also funding issues,” he adds. “It's much easier to write a successful grant proposal for animal experiments. Animals are homogeneous, and let you say ‘aha!’ in a neat, clean experiment.” For a young investigator in need of publications and clear successes, the choice is all too clear. Humans, after all, are not inbred; you can't order them from Jackson Laboratory and receive delivery of genetically homogenous batches. Humans are not only genetically diverse but also behaviorally diverse. As a result, it is a real challenge to figure out whether some aspect of their disease or their reaction to a therapy reflects the disease alone, their DNA, how they live—or some messy permutation of all three. It's a recipe for an ambiguous experiment that does little to advance a clinical researcher's reputation.

“The entire tenure process at major research institutions demands that the physician–scientist compete head-to-head with the very best Ph.D. scientists doing cutting-edge basic research,” says biologist Fred Turek of Northwestern University in Evanston, Illinois, himself a Ph.D. whose sympathies nevertheless lie with his M.D.–Ph.D. colleagues. “Quite often, they can't compete at the basic science level since they have so many clinical obligations, so they fall short of the ‘tenure bar’ that has been set by basic scientists.” Clinical researchers say they have a tougher time progressing through the tenure track, and a more difficult time with promotions, than colleagues doing either basic research or purely clinical medicine. Says one young investigator, “It's much harder to evaluate excellence in clinical research than it is in basic research. Clinical research takes longer, and the results are often more ambiguous and less clearcut.”

There is another, subtler barrier to translational research that would-be patient researchers often mention: the very name is a turnoff. To “translate” is to take someone else's original creation and tweak it a bit, make it relevant to (in this case) Homo sapiens rather than Mus musculus. The “real” science lies in discovering and cloning the Drosophila gene for, say, circadian rhythms. Curing insomnia, in comparison, is viewed as one step up from butterfly collecting. By science's prevailing values, human research is less intellectually demanding than the kind you do in slime molds.

Many of the brightest scientists have, therefore, plunged into the minutiae of roundworm genes and fruit-fly receptors, instead of human diseases. “Most of our best people work in lab animals, not people,” argues immunologist Ralph Steinman of Rockefeller University. “But this has not resulted in cures or even significantly helped most patients,” he wrote in a recent issue of the journal Cerebrum (2).

Comp. Please drop the second 〈CAB〉 from the end of the article here. Follow style. Use red.

Physician–scientists have traditionally moved discoveries from bench to bedside. But “the physician–scientist is an endangered species,” says Judy Swain, chairwoman of medicine at Stanford University School of Medicine in Palo Alto, California. They number some 14,000 in the United States today (a 9% drop since 1980), fewer than half the number many believe we need. Despite the truckloads of promising ideas spilling out of molecular and cellular biology, the dearth of researchers qualified to do patient research means that few discoveries turn into treatments, cures, or preventions.

Reversing this trend will be a significant challenge, as its origins lie in financial, cultural, and even sociological factors that have become ingrained in the nation's biomedical establishment. In fact, in a recent paper Floyd Bloom, president of the American Association for the Advancement of Scientist and a neuropharmacologist at The Scripps Research Institute in La Jolla, California, asked whether these obstacles can be surmounted with anything less than a complete overhaul of the nation's health system. “Unless steps are taken soon to undertake a comprehensive restoration of our [health] system, the profound advances in biomedical research so rapidly accruing today may never be effectively transformed into meaningful advances in health care for society” (3). He adds, “When the only goal of a health care system is financial solvency or profit through cost control and increased patient throughput, one can only imagine what the future might hold.”