Residency 2050: What Is the Future of Medical Training?

Ryan Syrek, MA


October 07, 2019

Brian Garibaldi, MD (Johns Hopkins University)

Healthcare delivery has changed dramatically in the 100 years since Sir William Osler established the modern residency training system at the Johns Hopkins Hospital in 1889. Medical and scientific knowledge is expanding at an unprecedented rate, nearly doubling every year. Yet Osler would still recognize the basic structure of graduate medical education with its mix of classroom-based teaching and experiential learning through activities such as bedside rounds and clinical service. As our healthcare system has evolved, so too must the way in which we train the next generation of physicians .

Predicting what residency training will look like in the next 25-30 years is difficult; however, current trends offer some insight into possible directions. The residency programs of the future must better prepare physicians to search for medical knowledge and to translate new information into bedside clinical medicine. Artificial intelligence (AI) will undoubtedly assist physicians in data synthesis, but physicians will need to use "human intelligence" to figure out how to apply that data to an individual patient.

Roy Ziegelstein, MD, coined the term "personomics" to refer to this specific field of medicine whereby knowing the patient as a person (eg, their goals, aspirations, personal relationships, fears) directly impacts their treatment plan and response. Osler predicted this more than 100 years ago when he said, "The good physician treats the disease. The great physician treats the patient who has the disease."

A pivot towards personomics will require a shift in the amount of time that trainees and physicians spend with patients. In current graduate medical training, some residents spend as little as 13% of their time in direct contact with patients and their families. This has led to a decline in bedside clinical skills and likely adversely affects the doctor-patient relationship. It also contributes to the alarming rise in physician and trainee burnout.

In the future, AI will help to unload the burden of electronic health record documentation and free up time to spend with patients. We must be more intentional about cultivating the necessary skills for trainees to connect with patients and navigate shared decision- making when faced with growing amounts of data. We must also prepare trainees and patients for the inevitable and paradoxical uncertainty that comes with information overload. A return to a broader focus on the humanities in undergraduate and graduate medical training has been advanced by some as a means to prepare physicians to navigate these complex relationships and deal with uncertainty.

In addition to focusing on personomics, residency training will likely have a renewed focus on traditional bedside clinical skills such as the physical examination. Groups such as the Society of Bedside Medicine and the Stanford Medicine 25 team are leading a growing effort to reinvigorate the practice of the bedside physical examination. This is a critical skill that has fallen by the wayside in recent years.

In order for the power of AI in diagnosis to be realized, we must ensure that the data provided to the system are accurate. In order to adequately train residents in the bedside physical examination, we must be more intentional about assessing and providing feedback on these skills to physicians in training. Formative assessment programs such as the Assessment of Physical Examination and Communication Skills (APECS) will likely play an important role in bedside skills training.

The current time-based model of graduate medical education will also likely change over the next 25 years. Interest is growing in competency-based training programs, whereby residents may be able to advance to the next stage of their career more rapidly if they demonstrate a certain level of proficiency. AI will create more discriminating metrics of clinical performance. Coupled with a renewed emphasis on direct observation and assessment, we will likely be able to individualize resident learning plans and tailor training in both content and time based on a specific resident's performance.

Just as we use the scientific method to advance medical knowledge, we must apply the same principles to study and develop the best approaches to residency training. Programs such as the Reimagining Residency initiative are an important start to gather the objective data needed to inform system wide changes in graduate medical education.

Our project will examine the impact of modifiable factors in the residency training environment (eg, time spent at the bedside, activities in the electronic health record, workload, patient complexity) on the important outcomes of resident wellness and clinical skill. The data we gather will allow us to develop and assess the impact of changes in residency program structure to improve these outcomes and, ultimately, the quality of patient care.

Johns Hopkins University School of Medicine, The University of Alabama at Birmingham School of Medicine, and Stanford University School of Medicine were awarded an AMA grant for their project: The Graduate Medical Training "Laboratory": An Innovative Program to Generate, Implement, and Evaluate Interventions to Improve Resident Burnout and Clinical Skill.


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.