Alvan R. Feinstein, M.D., HS ’54, joked in his 400th scholarly article that he hoped to continue writing until he faced “the ultimate rejection, by adverse pathophysiology.” When Feinstein did die in October—at age 75, while participating in the Gairdner Foundation conference in Toronto—his former student, Peter A. Singer, M.D., M.P.H., found himself taking Feinstein’s place on the podium. Instead of listening to Feinstein speak, Singer delivered a eulogy.

The first part of Singer’s tribute was what one might expect: He described Feinstein’s leadership in establishing the field of clinical epidemiology, in which researchers use scientific principles to study decisions about patient care—what one colleague called “conducting research at the bedside.” And he spoke of Feinstein’s “pattern of challenging orthodoxy.” But then Singer did something unusual: he quoted a comment Feinstein had written on a manuscript that Singer sent him.

“I don’t think your writing is actually bad,” Feinstein wrote to Singer, the Sun Life chair in bioethics at the University of Toronto. “After you fixed the tone and removed most of the sociobabble in this manuscript, the paper could be understood despite its prolixity, verbosity, and subadequate inter- and intra-sentence structure.”

That Singer would affectionately recall the man who offered such criticism reflects Feinstein’s mystique: he had almost impossibly high standards and never minced words, yet his students deeply valued his guidance. Many of those students were young physicians Feinstein supervised as founding director of the Robert Wood Johnson Clinical Scholars Program at Yale, which since 1974 has provided two years of post-residency training in research.

Although Singer now studies bioethics rather than the clinical epidemiology in which he immersed himself as a clinical scholar from 1988 to 1990, he says he uses what Feinstein taught him every day. “He didn’t teach us only clinical epidemiology. He taught us how to think,” said Singer in an interview. He said Feinstein also imparted to students “a fundamental irreverence for scientific authority and established beliefs. … He was one of the giants of 20th-century American science. It was really an honor to be able to learn from him.”

The core values that Feinstein established for the clinical scholars program will endure, said Ralph I. Horwitz, M.D., a program scholar from 1975 to 1977 and now chair of internal medicine at the School of Medicine. The scholars—four to five are admitted each year—will continue to study quantitative methods in depth, to think critically not only about methodology but also about the merit of research questions themselves, and to work closely with faculty mentors, said Horwitz, co-director of the program with Harlan M. Krumholz, M.D.

Second-year scholar Thomas M. Morgan, M.D., said that he and fellow scholars learn to recognize hidden assumptions underlying research and to find limitations to methodology. “They want us to be able to think critically about the techniques used rather than just to be able to plug numbers into a computer and get results,” said Morgan. “We deconstruct research and try to build it up from the rubble.” He said the scholars are taught to evaluate whether a question is even worth studying. “The difference between good research and great research, as defined by the mentors of this program, is determined by the quality of the question.”

The clinical scholars program is at a crossroads now: Yale faces unprecedented competition in seeking renewed funding for the 28-year-old program, because The Robert Wood Johnson Foundation will reduce training sites from seven to four beginning in 2005. To win a grant, Yale must successfully argue that its program not only ranks among the best four programs but also outflanks new competitors from top-ranked schools. Lewis Sandy, M.D., the foundation’s executive vice president, said it decided to fund fewer sites for two reasons. First, although the clinical scholars program was unique in 1974, other fellowships now provide similar training. Second, the foundation hopes that through consolidation it can provide more in-depth training in clinical epidemiology and health services research, with an option to extend the fellowship for as long as six years.

Whatever the future of Yale’s program, Singer believes that Feinstein’s model for rigorous research and his “unswerving devotion” to training his students will have a lasting effect. More than 100 clinical scholars studied with Feinstein. Those men and women, said Singer, compose Alvan Feinstein’s “living legacy.”