Revisiting Lifelong Learning: The Adaptable Physician

The century-old sequence of medical school, then residency and fellowship, then episodic continuing education is no longer the right structure for a forty-year or more career in a changing technoscience realm. The continuum should be reconfigured in several specific ways.

Rebalancing across the phases. Some graduate medical education content can be off-loaded to undergraduate medical education and to continuing education, particularly for pathways on the long side. Where a pathway runs seven or eight years post-graduation, the question is whether all of the content must sit in the most intense and least flexible phase. Whether it is necessary is a separate matter.

Curated cross-training beyond biomedicine. Graduate medical education today is generic within a specialty and subspecialty-driven framework, shaped almost entirely by biomedicine. The alternative is a different axis, not less rigour: training that is personalized and curated, reaching deliberately outside biomedicine.

Two properties define the domains for which to reach. They are specialty-agnostic, useful to a pathologist and a pediatrician alike, which is why no board has taken ownership of them; and they are technology-intensive, shaped by the paradigms of Part I, which cut across every specialty and are owned by none. In practice that means structured engagement with data science and AI, complexity, health economics, regulatory frameworks, engineering, design, and the human sciences, placed inside the formal continuum and credentialed there. Left to a trainee's spare time, it happens only for those who already have time, self-motivation, and connections.

Decompressing the middle. The structural move that makes room is to stop treating residency and fellowship as the container for everything. Content can be pulled earlier, into undergraduate medical education through earlier specialization, and pushed later, into continuing education reconceived as continuing technoscientific education, a genuine vehicle for deep expertise in selected domains across a career rather than a compliance exercise. Decompressing the middle serves several ends at once: more efficient learning, the cultivation of master adaptive learners, a pathway for producing physicians with deep technoscientific expertise, and training suited to the more diverse set of roles this report anticipates.

Personalized storyboards. Tracking a formation of this kind requires more than logged procedures and completed modules. The proposal is a personalized storyboard of professional growth: app-based tracking of a trainee's experiences, simulations, and didactics, following the triad of biomedical competencies, other knowledge-domain proficiencies, and human qualities development, assembled over time into a narrative of who this physician is becoming.

The distinction from a learning portfolio is not cosmetic, since a portfolio is an archive of completions whereas a storyboard is a longitudinal account of development, including experience accumulated beyond the walls of the academic medical center. Aggregated across large numbers of trainees, such data would power a level of GME outcomes analysis that does not currently exist, beginning to answer questions the profession has been unable to answer about which formative experiences produce which capabilities. The privacy and governance requirements are the same ones Chapter IV sets out for clinical data.⁷⁷

Converting continuing education into a vehicle for growth. This is perhaps the largest single opportunity and the most neglected. Continuing medical education currently functions as an informal check that a physician remains minimally current. It should instead be the mechanism for genuine lifelong learning: opportunities to branch a career path, enter new arenas, and develop transdisciplinary insight, through sub-specialization within specialties, meaning continuous structured engagement in a focused domain across a career, replacing fragmented, eclectic continuing education with something that accumulates. The conventional placement of the research doctorate in the middle of undergraduate medical education should be revisited on the same logic, and Chapter VII sets out the case for siting it during fellowship instead.

Credentialing should evolve in step. The present model certifies expertise once, around a body of knowledge assumed to be stable, and then monitors for decay. What is required is an apparatus built for a career in which the science turns over several times: learning portfolios that accumulate instead of resetting; role-based credentials recognizing what a physician actually does, not only the specialty they entered; periodic capability renewal in domains that have genuinely changed; and formal recognition of cross-disciplinary reskilling.

The cultural change matters more than the mechanism. At present a physician who steps out to acquire a new domain is often seen as having lost their way, and the incentives run heavily toward uninterrupted linear advancement. In a forty-year career spanning several technological eras that reading is inverted, and the profession should normalize reskilling as a marker of excellence rather than of deficiency.

A credential is not primarily a document about a physician. It is a signal to the people who must rely on one. Patients read it to decide whom to trust, hospitals to decide whom to privilege, insurers to decide whom to pay, and courts to decide what the standard of care required. Each of those parties wants a reliable answer to what this physician can be trusted to do now, and a board certificate supplies that answer crudely but legibly.

A portfolio supplies a richer answer and a far less legible one, and a profession that asks for the flexibility without solving the legibility problem is asking the public to accept less assurance for the profession's convenience. The answer is a summary layer: a small set of currently maintained, independently verified capability statements, each dated and each carrying an expiry, sitting above the accumulated portfolio. The portfolio serves the physician and their development. The summary serves everyone else, and it has to be as easy to read as the certificate it replaces.

Behind every mechanism set out here stands a single human quality. Rebalancing the phases, curating cross-training, decompressing the middle, storyboarding formation, reconceiving continuing education, re-sequencing the doctorate, and rebuilding credentials are all of them scaffolding, and what they scaffold is adaptability: the capacity to keep functioning well, and to keep becoming competent, as the ground moves. In an age of technoscientific disruption, adaptability is not a soft complement to expertise but the property that determines whether expertise survives contact with a changed world.

Adaptive Expertise as a Foundational Quality

Three strands of a substantial literature bear directly on adaptive expertise of the physician. The first is the distinction drawn by Hatano and Inagaki between routine and adaptive expertise. Both rest on accumulated domain experience and both perform accurately in familiar situations. The difference appears at the edge of the familiar, where the routine expert is fast but does not extend, while the adaptive expert holds the conceptual understanding that allows a procedure to be modified or invented.

Their account of what produces adaptive expertise poses a challenge, since it requires environments with enough variability that skills must be adjusted, settings in which reward does not hang on immediate performance, and cultures that value understanding over efficiency; conventional clinical training is optimized against all three.

The second is medicine's own translation of that work, the master adaptive learner model developed by Cutrer and colleagues, which makes adaptability procedural through a repeating cycle of planning what to learn, learning it, assessing whether the gap closed, and adjusting practice.

The third is the broader discourse on rethinking, of which Adam Grant's Think Again is the most influential recent statement.⁷⁸

Training is optimized that way for reasons that should be understood before they are overridden. The current design was built to guarantee a floor: that every graduate meets a minimum standard; that the variability between them is small enough for a system to rely on; and that a patient meeting a trainee is protected by supervision and protocol. Those aims are legitimate and were hard won, and each is served by the features that suppress adaptive expertise, which are low variability, immediate performance pressure, and efficiency valued above exploration. The trade-off is real rather than a failure of imagination, and the levers proposed here ask program directors to accept more variability and slower early performance in exchange for graduates who can still function when the ground moves. That bargain will be accepted only by people who are told plainly what they are being asked to give up.

Three features of the future physician's situation make the quality decisive. The half-life of knowledge is shortening while career length is not. Roles will change and not only knowledge, since the four identities are a repertoire moved among rather than a sequence passed through, which means adaptability is the capacity to change what one is for. And the machine partner itself changes on a release cycle rather than a generational one, so the division of labor between physician and system will be renegotiated repeatedly within a single career.

The levers are clear enough. Design for variability, so that formation includes unfamiliar problems encountered before the procedure is taught. Use simulation as the safe randomizer, rehearsing the unprecedented and not only drilling the routine. Treat cross-domain rotation as adaptability practice, whose durable product is the experience of being a novice again in mid-career. Assess revision and not only accumulation. And model it visibly, since a mentor who says aloud that they have revised a long-held position teaches more than a curriculum on rethinking.

One caution belongs with all of this. Adaptability is not agreeableness, nor is it infinite plasticity. A physician who revises every position under pressure has become unmoored, not adaptive, and the patient loses the one party who was supposed to hold steady. The quality to cultivate pairs openness to revision with the stable commitments of Chapter IX, the fixed points of honesty, fidelity, and answerability that give revision its meaning.⁷⁹

Education is the lever. But no medical school can pull it alone, which is the subject of the final chapter.