Unlike most Institute of Medicine (IOM) studies, which deal purely with policy choice, this project's task was to develop a method by which the Department of Veterans Affairs (VA) could determine its physician manpower requirements. The implications of this difference are significant. To construct a new state-of-the-art tool for calculating physician staffing requirements, by specialty and at the facility level, calls for an element of creativity in the development and use of quantitative analytic methods, data bases, and professional judgment. While these efforts taxed both committee and staff, they responded admirably and have broken new ground.

The VA manages this country's largest and, arguably, one of the world's most important health care systems. It is critical for the VA's future that it have a sound plan for determining the number of physicians required for its three mission-connected responsibilities of patient care, education, and research. We believe we have created a tool for determining physician requirements that will be of great utility to VA decision makers in their policy roles.

The committee's background varied from "quantniks" to bedside physicians of many specialties. It took a significant effort to marshall this expertise to produce an approach that is methodologically innovative, capable of being applied systemwide in a relatively efficient fashion, and sufficiently detailed and concrete to be relevant to the realities of the clinical environment. Here also the members of the committee enriched each others' experience and understanding. No one could have asked more of a committee and its panels.

My thanks go to Sam Thief who was supportive over a longer-than-usual IOM project and who understood the uniqueness and complexities of our task. Division director Karl Yordy personally added his considerable experience to our effort, and the committee is grateful. However, to Joe Lipscomb, the staff director, go the committee's and my own personal thanks, admiration, and even

awe. His prodigious efforts made this complex project run smoothly. More important, he was a nidus of creativity around which the committee's efforts crystallized. He was vigorously seconded by Bobbie Alexander and the rest of the staff. In all my years of involvement in IOM endeavors, I have never seen such a hard-working group.

Now we pass the baton back to the VA.

David R. Challoner, Chair

Committee to Develop Methods Useful to the Department of Verterans Affairs in Estimating Its Physician Requirements

This study could not have been accomplished without the assistance of numerous individuals in VA Central Office and VA medical centers around the country.

The committee is especially indebted to the leadership and staff of the VA's Boston Development Center (Braintree, Massachusetts) for their unfailing diligence in providing the data and interpretive expertise that enabled the committee to produce the empirically based physician staffing models. Deserving of special praise are the contributions of Frank Holden, the center's director, and of Stephen Kendall and Michael Doyle.

The committee is grateful to the administrative and clinical officials at the four VA medical centers (labeled in the report as VAMCs I, II, III, and IV) whose locally generated data were crucial to the development of accurate depictions of these facilities in the expert judgment staffing exercises.

Over the course of the study, committee and panel members and staff conducted site visits at 16 different VA medical centers across the country. To the VA professionals who participated in the organization and conduct of these visits, the committee owes special thanks. The committee also expresses its appreciation to the administrative and clinical professionals in the 60 VAMCs that participated in surveys conducted by the affiliations and nonphysician practitioners panels.

The VA Liaison Committee, chaired by Elizabeth Rogers, offered the committee thoughtful commentary, and encouragement, throughout the study. This group of VA clinicians, administrators, and health services researchers improved the committee's understanding of the rapidly changing VA health care system and the role that a physician staffing methodology might play in it.

From the project's beginning to its conclusion, the committee has greatly benefited from the advice and support of its VA project officer, Gabriel Manasse. No one has had a better understanding of the subtle complexities—both administrative and clinical—that have characterized this complicated and lengthy endeavor. He, of course, does not shoulder the responsibility for what

the committee produced, but he has most certainly enhanced the quality of the product.

Finally, the committee expresses its deep appreciation to the members of the 11 panels appointed to advise this study. For their diligence and intellectual leadership, the chairs of these panels deserve special recognition: Robert M. Donati (other physician specialties); Daniel W. Foster (medicine); Ernest W. Johnson (rehabilitation medicine); Robert J. Joynt (neurology); Kerry E. Kilpatrick (data and methodology); W. Eugene Mayberry (affiliations); Harold M. Visotsky (nonphysician practitioners); Harold M. Visotsky and co-chair Robert L. Leon (psychiatry); David C. Sabiston, Jr. (surgery); Sankey V. Williams (ambulatory care); and Cheryl E. Woodson (long-term care).

For excellent editorial assistance, the committee thanks Julie Phillips, consulting editor; Leah Mazade, IOM staff editor; and Wallace K. Waterfall, director of the IOM Office of Communications.