With changes in the financing of medical care, the funding flow that sustains much of innovative research is becoming uncertain
A revolution at risk

Annetine C. Gelijns

Conventional wisdom holds that the role of academic medical centers is to generate fundamental knowledge, while the role of industry is to conduct applied research and develop medical technology. In this view, medical centers have a public role--because some are state-owned, because the National Institutes of Health (NIH) underwrites a large portion of their research, and because much of their output is "public goods." Industry's role, by contrast, is said to be largely private. This view may be widely held, but a closer look at the ways in which medical innovation arises and spreads suggests that reality is much more complex and subtle than conventional wisdom maintains.

Academic medical centers play a variety of roles in university/industry interactions. But major changes in the health care environment--notably the rise of managed care--are putting these roles in disequilibrium. It is widely understood that university medical centers and managed care are on a collision course over two of the major functions of medical centers--patient care, which managed care would make less specialized and less technology-intensive, and medical education, in which managed care would emphasize training of more primary care physicians and fewer specialists. But, common wisdom holds that research--the third major function of university medical centers--should remain above the fray, at least as long as NIH and industrial funding sources remain strong. In turn, this protected status should preserve the contributions these centers make to medical innovation.

This view fails to fully capture the connections between functions within the centers--such as the complex and indirect flow of funds among care, training, and research activities--and the nature of the challenge that managed care's quest for cost-containment poses to medical innovation. At the crux of this challenge is the crucial, but sometimes undervalued, contributions of clinical research to the overall innovation process and the changing nature of the resources that have traditionally sustained this activity.

To speak of a "crisis" of research in the world of academic medicine may seem strange. Since World War II, university medical centers in the United States have expanded dramatically. This is largely the result of increased federal support for biomedical research (especially NIH support, which has been steady and strong, and which President Clinton has proposed to expand very dramatically) and the growth of insurance coverage of the American people, including the Medicare program, which has paid these centers handsomely for their patient care and educational activities. NIH gives signs of attaching higher priority to clinical research, but today, as the financing of medical care changes, the economic imperatives of managed care do not easily accommodate risk-taking research, especially the vast and diverse set of activities that come under the rubric of "clinical research."

Research activities at academic medical centers can be divided into five somewhat overlapping categories: At one end of the spectrum is the conduct of basic biomedical research in the test tube or in animal models. At the other end is clinical trials, mostly on new products that are being brought to market. In between is research into the basic mechanisms of physiology and disease in humans, the development of new technologies, and the discovery of new indications for existing technologies. Medical centers are well recognized for their spectacular achievements in basic biomedical research, as well as their work in discovering the basic mechanisms of disease in humans. More recently, the spotlight has illuminated their accomplishments in applied research and technology development, evidenced by the major increase in the number of inventions and patents generated at research-based universities.

University medical centers also are involved in finding new indications for existing products. In medicine, even after new technologies have been introduced into practice, uncertainty over their eventual uses remains high. This uncertainty may be the result of both the complexity of the human body and the heterogeneity of the human population.

For example, alpha blockers were first tested for their ability to treat hypertension. At the time of their introduction, it wasn't known that alpha receptors existed in the urological tract as well as in the arterial system and that, therefore, their blockage could constitute a treatment for benign prostatic hyperplasia. Initial trials, which focused on hemodynamics, did not pick up on the urological value. In fact, it took another 20 years to establish this because of the inherent complexity of the human body and our incomplete knowledge of it.

Another antihypertensive, reserpine, demonstrates the role of heterogeneity. Only after its widespread use in clinical practice did we learn that reserpine can induce clinical depression in predisposed individuals. This side effect, a result of serotonin depletion in the brain, subsequently led to the development of drugs to block the reuptake of serotonin, such as fluoxetine (Prozac).

Thus, carefully controlled clinical trials, which inevitably reduce the level of heterogeneity in the study population, are likely to conceal important dimensions of the usefulness or limitations of a new technology. The identification of new uses may only be possible after the extensive use of a medical innovation. This post-R&D phase is not unique to medicine, but is far more extensive and of far greater importance in the medical realm. The history of pharmaceutical innovation is replete with the discovery of new uses after drugs have been introduced into widespread use. For instance, of the top 20 best-selling drugs in 1993, more than 50 percent had significant post-introductory indications of use. Much of this "learning by doing" occurred in medical centers, where, as Pasteur once said, "chance favors the prepared mind."

The assessment of new indications for existing drugs and devices does not exhaust the contributions of clinical research at academic medical centers. Research also analyzes the efficacy, safety, and costs of newly emerging clinical interventions. This is, in fact, a major growth area for research at Columbia's College of Physicians and Surgeons, which has created an Office of Clinical Trials to promote the interface between researchers and private and public funding agencies.

These different types of research display quite distinct patterns of funding. Basic laboratory research is primarily funded by the NIH and, with the biotechnology revolution, by various arrangements with industry. Given its strong political support, NIH funding appears to have a rosy future. Clinical trials are heavily supported by industrial moneys, which show no sign of decreasing. However, other types of clinical research fail to attract sufficient public and private support and instead rely heavily on internal funding.

For example, the majority of research at Columbia's surgery department is supported internally by approximately 25 percent of the department's patient care revenues. This corresponds to a recent report by the Commonwealth Fund suggesting that 30 percent of faculty practice plan revenues are used to underwrite research, mostly clinical research.

As managed care exerts downward pressure on clinical revenues, creative cultivation of coping strategies may keep research on a firm ground within medical centers. For instance, centers under fiscal stress could create efficiencies of scale for doing research--such as establishing a central data coordinating center for the conduct of clinical trials. But what medical centers do with their internal economic options largely follows the balance of political power within them and the ability of leadership to direct that balance toward organizational enhancement. This means, for instance, that centers must renegotiate funding flows from physician practice plans to support research. Moreover, medical centers can form partnerships with industrial firms and managed care organizations to help increase support for and conduct of research. Finally, medical centers can help build strong political support for research as an important public good. They should document and clarify the public benefits of their research activities and help sharpen the debate about the relative roles of the public and private sectors in supporting medical research.

Related links:

ANNETINE C. GELIJNS, Ph.D., is associate professor of surgical sciences and public health and director of the International Center for Health Outcomes and Innovation Research at Columbia.

This article is adapted from a talk Dr. Gelijns gave at the President's Circle of the National Academy of Sciences, Washington, D.C., June 1997.