Government Affairs Home > Research

AAMC Position Statement on NIH Centers of Excellence

The Association of American Medical Colleges (AAMC) is grateful for this opportunity to assist the Institute of Medicine's assessment of NIH "Centers of Excellence." Our views submitted here are based in part on the experiences of our member institutions and organizations, which include all 125 U.S. allopathic medical schools, more than 400 teaching hospitals, and 94 medical and professional societies representing more than 100,000 faculty. These institutions perform approximately half of all NIH funded extramural research, and receive nearly 70 percent of all NIH research and training awards to academic institutions. Our informal examination of the NIH CRISP system indicates that AAMC-member institutions are similarly well represented among the performers of various NIH-sponsored research centers.

Our observations in respect to extramural research centers, and particularly NCI's Comprehensive Cancer Centers, are based in part on the findings and recommendations of the AAMC's Task Force on Clinical Research. They are also based on the experiences of AAMC's leadership and staff, including David Korn, MD, AAMC's Senior Vice President for Biomedical and Health Sciences Research and representative presenting to the IOM committee. Dr. Korn is Dean Emeritus of Stanford University Medical School and served as chairman of the National Cancer Advisory Board from 1984-91.

Our comments here generally follow in outline the challenging and well-formulated questions detailed in the IOM's project scope for this study. The AAMC hopes that the outcome of this inquiry can guide the NIH in determining how best to employ extramural research centers to serve the NIH's public health mission, and to ensure that the basic and clinical research supported by the centers maintains a standard of excellence.

The Scope and research aims of NIH centers: The term "centers of excellence" encompasses many NIH activities and funding mechanisms and a spectrum of research emphases possibly as broad as the NIH itself. A comprehensive definition of centers will remain elusive. We assume that the IOM study will focus primarily on extramural research centers created under NIH's research program award mechanisms (designated as "P"-grants), such as center core grants (P30s) or specialized centers (P50s). We also recognize that some centers may be created under cooperative agreements or other mechanisms.

NIH extramural research centers may be viewed along a continuum, from those that provide shared technology or other special research resources supporting many disparate lines of investigation, to centers that support the integrated activities of multiple investigators or teams of investigators sharing a common research focus, methodology, or objective.

The AAMC's comments here focus on the latter type, and particularly on those extramural centers that support translational and patient-oriented research, and argue that such centers play an indispensable role integrating multi-specialty teams, including physician and non-physician scientists for concerted progress in biomedical research. Such centers can further provide an environment to foster the training and career development of new investigators, particularly physician scientists. These capabilities are exemplified perhaps most fully in an elite category of NIH center, the National Cancer Institute's Comprehensive Cancer Centers, about which we comment further below.

Centers for Shared Resources: We would like to remark briefly on the importance of the other type of centers, those providing shared research resources to support multiple investigators and lines of research. For example, General Clinical Research Centers-funded through an entirely separate administrative mechanism (M01) by NIH's National Center for Research Resources-are an irreplaceable resource for patient oriented research. Given the never-ceasing financial turbulence of the nation's health care delivery system, the continued support of dedicated, peer-reviewed research facilities provided by GCRCs within academically affiliated hospitals, clinics, and other venues of health care delivery will become exponentially more important over time. Other significant centers provide shared facilities and sophisticated, capital-intensive research resources, including NCRR-supported primate and other animal research centers, the P41 biotechnology centers, and similar programs. Twenty-first Century biomedical research increasingly relies on sophisticated facilities, instrumentation, informatics, and other capital-intensive infrastructure, as well as dedicated technical staff, which are often made available economically through centers programs. In turn, these centers encourage the growth of cross-disciplinary research (by bringing investigators from disparate fields together) and may even promote the growth of new disciplines.

We believe that the rationale for supporting these shared resource centers is based on the comparative advantages and efficiency of scale that they provide, the unique research that they enable, and their capacity-similar to other types of extramural research centers-to foster interdisciplinary and multidisciplinary research. The AAMC has argued strongly for increased NIH support of such centers, along with other types of capital investment-for shared instrumentation, computing and other technologies, and research facilities themselves-which have been substantially under-invested in the NIH budget. We remain deeply concerned that the past years of increased funding for NIH failed to take advantage of opportunities to upgrade and expand these resources, and that prospects for future funding of shared facilities, instrumentation, and infrastructure will now be much dimmer in a period of constrained NIH appropriations.

The relationship of centers to other NIH funding mechanisms: Given the broad focus of the NIH mission-to generate new knowledge useful in promoting health and combating disease and disability-and the complexity of its operations, the agency must have at its disposal a variety of mechanisms to implement its programs flexibly, effectively, and accountably. Accountability is especially important given the extraordinary degree of support for NIH from patient advocates, elected representatives, and non-scientists, specifically to see NIH research applied to fight human disease and disability. The AAMC views research centers as essential components of the NIH's portfolio and as complementary to other administrative mechanisms funding other types of research (and training). Generally, NIH requires an institution to demonstrate a strong base of research accomplishment prior to receiving a center grant. Established research centers support and facilitate a multiplicity of separate investigator-initiated or other research projects.

We urge the IOM to avoid dichotomous thinking that sets centers in direct opposition to investigator-initiated grants or other types of funding mechanism. Traditional research project grants, such as R01s, extramural centers, and other project and program mechanisms are distinct facets of a multifaceted NIH mission. Too frequently, the investigator community is tempted to judge an NIH center or other programmatic entity in terms of its opportunity cost for R01-funded research. This seems no more sensible than judging R01s as a lost opportunity for integrated multi-disciplinary research, or judging research project grants generally as a lost opportunity for new facility construction. The AAMC envisions that all NIH mechanisms should function complementarily.

Distinctive aspects of centers: Centers provide an integrated environment for research, training, and career development, and for research that integrates multiple disciplines and medical specialties. The idea of a research center is to:

• bring focus and emphasis to high priority area or research
• bring together separate but interrelated components
• provide critical infrastructure, and
• provide seed funding for innovative ideas.

They are also, in the case of the Comprehensive Cancer Centers, which are sui generis and discussed in detail below, designed to provide comprehensiveness and outreach to the community. Many centers enable the integration of basic and translational research with extension into medical practice, health promotion and disease prevention in the surrounding communities. Centers can effectively integrate fundamental research with health care delivery and thus play a critical role in patient-oriented research.

The important role played by NIH extramural research centers in supporting clinical research is evidenced in the 1997 Nathan panel report, which estimated that approximately 60% of the NIH centers budget was allocated to clinical research (as defined by the panel). In contrast, 38% of overall NIH research funding was allocated to clinical research. Centers programs were the only NIH research mechanism predominantly dedicated to the full spectrum of clinical research. While the Nathan panel data are now more than six years old, the NIH estimates that presently, approximately the same level of overall funding is dedicated to clinical research, under the panel's definition, and we believe that centers continue to be leading performers of that research.

The Nathan panel also recognized and made recommendations to address formidable impediments to the career development of new clinical researchers, particularly physician-scientists. Physician-scientists with clinical care duties are probably among the most "multi-tasked" individuals in our society. The competing demands of patient care and other duties are not conducive to the intensive, highly focused career track that leads to support as a principal investigator on a traditional research project grant. Centers provide a structured research environment and, we believe, more opportunities than traditional research projects for physicians to participate and contribute to clinical research while contending with other clinical demands.

Example of the NCI Comprehensive Cancer Centers: Beginning in 1998, the AAMC's Task Force on Clinical Research undertook an introspective examination of clinical research within the Association's constituent organizations. As the task force searched for models of institutions that had succeeded in integrating clinical research programs with centers of health care delivery, medical school and hospital leadership consistently identified the Comprehensive Cancer Centers as exemplary and often unique venues for such integration. In almost all other instances aside from these centers, the vicissitudes of health care financing had increasingly restricted or preempted the joint sharing of clinical care and research resources and facilities and the integration of research with the delivery of care. The Task Force reported that,

Comprehensive Cancer Centers represent a model for NIH collaboration with medical schools and teaching hospitals in support of the integrated mission of clinical care, research, and community service. NCI supports the costs of research projects and infrastructure, and the NCI endorsement of the comprehensive centers brings additional repute and credibility to the clinical care programs for cancer patients and their families, as well as a cachet of general clinical excellence to the medical center. The model benefits the programs of the National Cancer Institute, the clinical, research and educational programs of the medical school or teaching hospital, and the care of patients with cancer. The model should be instructive to other areas of clinical practice and to other institutes at the NIH.

The Task Force went on to state that the NIH merit review process submits centers to the same critical evaluation as other NIH awards. Centers must continue to compete within the NIH system for funding, and should never be funded by inertia. Our experience has been that centers can in fact compete intensively, and that NIH can and does withdraw a centers-of-excellence designation on basis of stringent scientific review.

The Task Force noted the two-step process for approval of the Comprehensive Cancer Centers. First, determination by peer review that the center fulfills the research-related criteria, including the scope of the center's research and the presence of a high degree of interactivity among its programs; the center's overall priority score determines whether the application will be funded, in accordance with the NCI's funding plan for the Cancer Centers Program. As the second step, centers are asked to report their institution's efforts in providing outreach, education, and cancer information to the communities they serve (this information is updated by the centers annually). The Task Force noted in particular that some institutions with excellent NIH-funded cancer research have not been able to gain the "center" designation, or the "comprehensive center" designation, because of inability to resolve internal organizational and political issues or to devise structures that meet both the requirements of NIH and the needs of the medical school and teaching hospital.

Costs and Weaknesses of Centers: The costs associated with establishing and maintaining centers of excellence are varied, and the NIH has the most authoritative information on budgetary costs. The Comprehensive Cancer Centers, in particular, represent a substantial investment by the NCI and this fact probably limits the NIH's ability to replicate this model widely.

There are several weaknesses of centers, some of which have been sharply exhibited, at least in the past, in the case of Comprehensive Cancer Centers. One is in part related to the substantially high levels of investment that are required from both NCI and the awardee institution. The centers represent such a level of commitment, and therefore, of sunk costs, and they have such a high profile in their community, that it has proved politically difficult (but not impossible) to retire a center or even, as noted above, to remove its "comprehensive" status. Thus, there is the legitimate concern that once a center is initiated, it can be very hard to discontinue it if its research quality flags, or the areas of research that it serves are no longer productive. A second concern is that the center's "umbrella" can be used to shelter research or researchers that would not be able to win competitive research support independently. The perception that centers shelter uncompetitive or un-productive research is a major problem for the investigator community.

To counter these concerns and maintain credibility, NIH must demonstrate systematic, rigorous, and dispassionate scrutiny of centers, and must exercise its authority to phase out or discontinue centers when they cease to be competitive. In our experience, centers can and have been discontinued under such circumstances, given strong and dedicated leadership within the NIH institute and strong support from the institutes' advisory councils. Thus, during Dr. Korn's tenure as NCAB chairman, the NCI reviewed, revised, and strengthened its review criteria for Cancer Centers (and Program Project grants) several times, in all instances emerging with a stronger program.

Conclusions and consideration of complementary research models: Any policy directed toward the diminution of the role of extramural research centers in the NIH portfolio would most likely further erode the environment for patient oriented and other clinical research and would have negative repercussions for the next generation of clinical researchers. The Comprehensive Cancer Centers have achieved in microcosm a remarkable level of integration of research with patient care and community engagement. The costs of these centers represent a substantial investment in resources and credibility by the NIH, and the NIH must preserve authority to determine when this mechanism is appropriate and assure that these or other kinds of centers maintain their cutting-edge science.

However, to impose extramural centers where there is no solid base in competitive research, or to act politically to shelter centers from the judgment of competitive merit review would greatly diminish the capacity and credibility of the NIH overall. The essential truth, which correctly lies at the heart of the academic community's concerns for investigator-initiated R01s noted earlier, is that the NIH program must continue to be based on peer review, merit-based allocation, and pursuit of the most promising research opportunities. Centers or other programs established to facilitate NIH-sponsored research to address specific public health objectives must take special care that they remain scientifically and programmatically competitive within the NIH portfolio.

For these reasons, we believe centers awards should be established only where institutions demonstrate a base of competitive research and capacity for continuing this research. Notably, NIH P-30s and other center mechanisms are linked to continued successful applications and awards for investigator-initiated research project grants that provide the programmatic substance of the centers.

There is often demand from patients and communities to implement various types of centers, both within the NIH (to serve as a champion and source of funding for disease-focused research) and extramurally (to "jump start" research in areas where progress is perceived to be slow or unsatisfactory). While AAMC supports the role of centers for disease-oriented and patient-oriented research, we believe that centers must be established on a critical mass of demonstrated research expertise and opportunity, and not as a contemporary form of the "political spoils system."

The NIH has already begun to improvise novel types of organizations and networks specifically to address areas of limitation in current scientific knowledge and methods. For example, the AAMC has strongly supported NIH's establishment of the Osteoarthritis Initiative (OAI), a collaboration with pharmaceutical companies that

[Pools] funds and expertise for a public repository of osteoarthritis patient data, radiological information and biological specimens. Scientists will be able to use this public resource to test much-needed biochemical and imaging markers of disease progression, to further the development of OA drugs, and to improve public health. Neither the federal nor private sector alone would be able to develop such a resource.

A second model, the NCRR's Biomedical Informatics Research Network (BIRN), combines groups of shared resource centers, including GCRCs, to collaborate closely towards a unified scientific goal. The initial goal is to address the needs of biomedical investigators across the country effectively to share and mine data in a site-independent manner for both basic and clinical research. Initially, BIRN is concentrating on research involving neuroimaging to take advantage of the relatively advanced level of sophistication of this community in the use of information technology. An objective of the program is to create infrastructure that can be deployed rapidly at other research centers throughout the country that may have different research emphases.

The OAI and BIRN provide examples of innovative organizations (networks or consortium-like arrangements) that attempt to address specific areas of paucity of scientific knowledge and medical capability. In the case of the OAI, the paucity is of systematic data on the pathogenesis of the disease. In the case of BIRN, the impediments are computational capacity and models for analyzing data. The IOM should encourage the NIH and advise its champions to consider developing such innovative organizational arrangements where impediments to research progress are perceived, rather than to diminish the quality and credibility of the research center mechanisms.