Emeritus Professor of Health Care Sciences, Computer Medicine, and Medicine, George Washington University
Tom Piemme, MD, one of the founding leaders of ACMI and AMIA, passed away on April 17, 2021, in Peoria, Ariz., near Phoenix. Tom grew up in Pennsylvania and attended college and medical school at the University of Pittsburgh, where he served as Chief Resident in Medicine under Jack Myers (of Internist-1 fame), who was chair of the department at that time. Tom later taught at Pitt for some years before moving to Washington, D.C., in 1970, to become director of the Division of General Medicine and subsequently Director and Associate Dean for Continuing Medical Education (CME) at the George Washington University School of Medicine. He became involved with the National Board of Medical Examiners and developed a strong interest in computer-based testing and in the role of computers in clinical education. Tom retired from GW in 1998 and moved, with his wife Judy, first to Ft. Lauderdale but soon thereafter to Sun City, Ariz.
As Director of CME at GW he developed a top-notch meeting management organization. After an informal group of D.C.-area medical computing practitioners began the Symposium on Computer Applications in Medical Care (SCAMC) in 1977, it soon became clear that the meeting’s rapid success required a more professional organization to manage and fund the annual symposium. They approached Tom about taking on the meeting and soon thereafter SCAMC was incorporated with Tom as its Executive Director. Starting in 1979, Tom and his organization managed the SCAMC meeting, which grew from its initial size in 1977 (~225 attendees) to a vibrant national meeting of about 2,000 attendees in the early 1980s. The meeting was held in Washington, D.C., annually and all members of the Board were initially from the D.C. area. As SCAMC grew in national scope and influence, however, its Board evolved to include national leaders in the informatics field.
In 1983, at the suggestion of Morris F. Collen, Tom convened a group of five informatics leaders, including Marsden Scott Blois (UCSF), Donald A.B. Lindberg (University of Missouri), and Edward H. Shortliffe (Stanford) in addition to Collen. The focus was on the possibility of creating a college of elected fellows for the field of medical informatics. The group created a ballot with 100 names of informatics leaders and Tom sent out the ballot to all 100 names, asking them to select 50 names to be elected as founding fellows. In this way the American College of Medical Informatics was founded and ultimately incorporated with Dr. Blois serving as the first ACMI President.
With the subsequent growth of ACMI and SCAMC, plus a large membership organization known as the Association for Advanced Medical Systems and Informatics (AAMSI), it gradually became clear that it was logical to bring the three entities together as a single membership organization with a major annual symposium and a college of elected fellows. This led, with Tom’s assistance, to agreements to begin the American Medical Informatics Association (AMIA) by the end of the 1980s. Although the new organization ultimately established its own staff and secretariat, Tom continued to be closely involved with both AMIA and ACMI, and a regular attendee at the annual meetings.
Tom was also an early champion of the physician assistant profession. He was the founding director of the PA program at George Washington University in the early 1970s and quickly became involved in the administration of the profession at the national level. In 1972, he helped establish the joint office of the then Association of Physician Assistant Programs and the American Academy of Physician Assistants. He was also a key figure in the development of the National Commission on Certification of Physician Assistants, serving as its first president and as the chair of its Standards Setting Committee for many years, as well as helping to develop the Clinical Skills portion that was a feature of the early certifying exam. He served as the Association’s second president, in 1973-1974, and was one of the first honorary PAs named by the AAPA. In 2017, he was honored for his service to the profession with the Physician Assistant Education Association’s Lifetime Achievement Award.
In his retirement years he honed his skills as an actor and offered several superb performances in lead roles in the Phoenix community-theater environment. Particularly noteworthy was his role as Clarence Darrow in a local production of Inherit the Wind. He also served as a judge for annual local awards for theatrical productions and acting performances. Although his own health began to fail after his wife passed away, his mind remained sharp as always and he could opine on everything from current politics (which he followed voraciously) to analyses of past and current informatics events and discussions.
It happens that he recorded a Zoom call with Ted Shortliffe only a few days before his death. He was
eager to offer an interview on the impact of Don Lindberg during the SCAMC and early AMIA years.
Access the video here.
Tom Piemme was a wonderful, smart leader and influencer, as well as a cherished friend to many colleagues in the informatics community. We hope that this page and his personal history of contributions to our field will help others to appreciate why many former colleagues remember him with great fondness and respect.
The following personal note by Dr. Piemme, sent to a colleague only weeks before his death, captures a lot about his personal style and substance:
Let me tell you about my relationship with Jack Myers. Jack arrived at the University of Pittsburgh from Duke as chair of medicine in 1955. I was a sophomore student. As luck would have it, Jack was my instructor in physical diagnosis. Moreover, he happened also to be the attending physician during my medical clerkships in both my junior and senior years. He asked me to remain at Pitt for my internal medicine “internship” with the promise that he would see to my appointment as a resident at a program of my choice for residency following the internship. (That’s the way things were done back then.) I became a resident at the Peter Bent Brigham in Boston in 1960. It then turned out that Octo Barnett and I were residents together. Following the residency, I began a fellowship in cardiology as Octo opened his first lab at the Brigham. I was heavily influenced by him. I began taking courses in computer science during its infancy at MIT. Octo and I collaborated on a major project during the second year of my fellowship that I won’t go into, but his influence followed.
In 1963 I returned to Pitt as Jack’s chief resident. That allowed me to get into the mind of Jack who was by then acknowledged as one of the finest diagnosticians in the nation. He was serving as the President of the American College of Physicians. I was, of course, present each morning as he held resident conferences. He and I picked patients for presentation at grand rounds every week. Additionally, by agreement, he and I would spend Friday afternoons together, seeing the most interesting admissions of the week. I really had the opportunity to experience the way Jack thought about disease.
In the spring of 1964, for grand rounds, we picked a young man with a persistent congenital heart disease. I told him I would like to be the one to discuss the case. He agreed. I went over to Children’s Hospital, and picked up records of 10 children with congenital heart disease. From Octo I got Homer Warner’s congenital heart disease program. I then abstracted the data from the 10 cases, gave the data to 10 senior medical and surgical residents, 5 practicing internists, 5 surgeons, and fed it into Homer’s program.
Outcome: (I think I remember all of this, but, if not, I’m close.) The computer program correctly diagnosed 8 of the 10 cases; two persons – the chair of surgery and the chief resident in surgery correctly diagnosed 6 or 7; no one else came close. At the end of my presentation, Jack did something unprecedented. Grand rounds always consisted of two cases, each to be discussed for an hour. That morning, he said that there would be no second case, and began to talk about what had just been presented. In retrospect, he was talking about his cognitive approach to medical diagnosis.
From the chief residency I went into the Air Force at the Aerospace Medical Research laboratory where I became chief of the physiology branch doing research on hazardous environments – particularly weightlessness – in anticipation of the forthcoming Apollo missions. Two years later I returned to Pitt where I continued doing work for the Air Force under contract. But my interest shifted dramatically when Jack asked me to take responsibility for the medical outpatient services in the clinics and the emergency room. I became committed to issues in health care delivery.
Four years later, in 1970, I was asked to join the faculty at the George Washington University to establish a new Division of General Medicine. It was later that year that Jack stepped down as chair of medicine, and began working with Harry Pople from Carnegie Mellon, on the computer program that became “Internist-1,” and then “QMR.” That same year, jack became Chairman of the National Board of Medical Examiners.
As fate would have it, Jack and I began working together again when I was asked to serve on an ad hoc “Goals and Priorities Committee” of the Board, and later on the Board itself. It’s too complicated to go into in detail in an e-mail, but I then served 28 years on one or another committee of the Board. I was part of the widespread use of “patient management problems” that became tools of testing among all of the medical specialty boards. I later was the first chair of the committee on computer based testing – creating case simulations that candidates were asked to manage.
“PMPs” and computer simulations had the same goal – eliciting responses that would reliably identify those candidates that had the knowledge and management skills to practice medicine, while equally reliably identifying those who didn’t. As time passed, I moved from a role in creating the simulations to a role in scoring them. Therein lies the task that took me into cognitive territory.
I was a member of the PART III scoring committee working with new psychometricians that definitively ended the use of PMPs by the NBME and all specialty boards that used them. I then was part of a small committee that read more than 1000 computer printouts to try to accurately define the minimal data set that separated the wheat from the chaff.
During most of this time, of course, Jack was engaged in trying to identify the minimal data sets that would permit diagnosis by computer.
Dr. Piemme received his undergraduate (1954) and medical education (1958) at the University of Pittsburgh. He then trained in internal medicine and cardiology at Pitt, and at the Peter Bent Brigham Hospital in Boston. While at the Brigham, working with G. Octo Barnett, he pioneered in the use of high frequency response transducers to record pressure, flow, and sound from within the heart to document the precise timing of events of the cardiac cycle. Following two years in the United States Air Force at the Aerospace Medical Research Laboratory at Wright Patterson Air Force Base, working on environmental issues facing the forthcoming Apollo missions, he joined the faculty of the Department of Medicine at the University of Pittsburgh where he was a Scholar in Academic Medicine of the Markle Foundation.
In 1970, Dr. Piemme moved to the George Washington University as Professor of Medicine, where he became the founding director of the Division of General Medicine. Responsible for outpatient and emergency services at the Medical Center, he founded a Physician Assistant training program, and established a pre-paid health maintenance organization, the George Washington University Health Plan. Appointed to direct Continuing Medical Education at the Medical School in 1977, Dr. Piemme met with William Yamamoto and Helmuth Orthner, who had conducted two regional meetings of the fledgling Symposium on Computer Applications in Medical Care (SCAMC). Anticipating a wider interest, he invested the resources to undertake an international marketing effort that resulted in a tripling of attendance in the next year. He then saw to the incorporation of SCAMC, now recognized as the premier meeting of medical informatics in the United States, and became its Executive Director.
Within the next few years, Dr. Piemme was elected to the Board of Directors of the American Association for Medical Systems and Informatics (AAMSI), and appointed to the Biomedical Library Review Committee of the National Library of Medicine. Having been involved with the National Board of Medical Examiners for some years, he became the first Chair of the Computer Based Examination Test Committee, responsible for developing simulations to test a candidate’s ability to manage clinical problems. In 1983, together with co-author Marion J. Ball, he wrote the influential monograph, Executive Management of Computer Resources in the Academic Health Center, published, and widely distributed, by the Association of Academic Health Centers.
Dr. Piemme was one of five persons (with Scott Blois, Morris Collen, Don Lindberg, and Ted Shortliffe) who conceived and implemented the concept of the American College of Medical Informatics. Elected to Fellowship in the first year, Piemme served as the Founding Secretary.