Summary of the Report

Change for physicians and their craft arrives as a mosaic rather than by gradual evolution or revolution. The unfamiliar creeps quietly into view, easy to ignore, until one day a quantum shift disrupts everything seemingly at once. The distinction matters, because an evolutionary picture lets a group assume it can absorb change on the fly. Physicians are pragmatists by temperament and have never been adept at peripheral vision, least of all about the future of their own profession. That must now change. A technological society in accelerating flux, complexified healthcare ecosystems across the globe, and information-empowered patients make it so. Someone will redefine the physician's role regardless of whether the profession takes part, and deferral has consequences: scope gets written into reimbursement codes by people who will never meet a patient, and identity emerges by default. This de facto translates into further erosion of authority and talent flight. Over the centuries, physicianship has serially transmuted from a guild of tradesmen (Hippocrates 1.0) to a priesthood of craftsmen (Hippocrates 2.0) to an association of professionals (Hippocrates 3.0). The vision now is one of a collective of experts (Hippocrates 4.0)practicing technoscientific, N-of-1 medicine in a rehumanized way. The early 20^th century transition to the Flexnerian physician was chosen, which is itself a strong argument for designing the 21^st century transition deliberately. The Hippocrates 3.0 scientific physician of the past century must give way to a Hippocrates 4.0 paradigm in which identity becomes fluid and plural across a career, mapping to roles that are irreducibly human in an age of human + humanoid medicine. At the core is a fundamentally new relationship to knowledge.

A seismic transfiguration of the biomedical sciences is unfolding. Our grasp of biological mechanism is deepening at spatiotemporal and biostructural levels, and our technical powers to modulate that mechanism are expanding dramatically. New biologies are crystallizing knowledge domains with considerable translational potential. Technical power now reaches far enough so that what can be detected and measured can be modulated, and what can be modulated can be redesigned. The unit of fluency becomes a combination of fields rather than any one field in isolation, and generative transdisciplinary unions materialize apace, with this convergence spawning entirely new knowledge domains: emergent bioengineering at the nexus of complexity science·synthetic biology, by example. The spark is the insightful question, with competing answers unleashed by new instrumentation and unbounded by disciplinary gospel. Complementing discovery traditionally organized around patient and disease entity, new lines of investigation reorganize around mechanism and system, so that the core biomedical insight becomes a pattern running across diseases, pathways, and interventions rather than any single condition. Machine learning extracts such patterns already present in training data, as well as relationships not yet known to anyone. Computational medicine gels as a generative field, as it deciphers patterns and relationships in raw gene and protein sequence data and converts them into actionable items. Next-wave theranostics couple high-dimension diagnostic data to ever more creative therapeutic interventions, as the interval between knowing and designing is narrowed. Yet as the space of the possible expands, the clinical translational gap widens faster still—a gap that is overwhelmingly a made problem rather than a found one. The phase 1-3 clinical trial continuum warrants top-to-bottom reform, and all-important clinical validation need not be relegated to second-class status. Frontier subjects—editing the germline, deciphering the brain, extending lifespan, and conquering space—now enter the discourse in a realistic way. For these reaches in uncharted territory, whoever bears the consequence sits far from whoever decides, which means much of the ethical work must per force shift to the early design phase. These developments call for a fundamental rethinking of the discovery process itself. AI is emerging as an investigative tool, and on the horizon: closed-loop AI and automated robotic laboratories with minimal human intervention.

Transformation on the horizon for healthcare ecosystems can be framed as a set of emerging modes of medicine, clustered into five thematic categories: paradigms of care (multimodal systems of care; remote medicine; forecasting-preemptive medicine; preventive-societal medicine); machine intelligence (data-aware medicine; AI-augmented medicine; physical AI medicine); the human factor (rehumanized medicine; holistic-integrative medicine); the discovery edge (technoscientific medicine); and bridging humanity (global medicine). These operate in parallel, intertwine, and are co-dependent. The structure, the place, the time, and the scale of care is changing, and in the process, the points of failure. If 20^th-century system failure was the patient who could not reach the right care at the right time, failure now before us could well be the patient monitored continuously, generating a tsunami of data nuggets, and yet never actually seen or heard in a meaningful way. Holistic-integrative medicine promises to integrate the parts of a person that specialization divided, and a rehumanized medicine would reconnect the physician and patient, a connection that the relentless drive to efficiency separated. Exposomics, the largest available lever for disease prevention, sits outside the patient encounter, distant from the hands of the physician. Machine intelligence is the decisive disruptor because it reaches into every other mode, and the constraint is ultimately less in cognitive capability and more in defining its effective use. Clinical AI is therefore an integration problem, to be solved at the human plane rather than inside the machine model. In the looming age of AI, the dominant failure mode on the physician side shifts from ignorance to miscalibrated reliance. Machines absorb and navigate the typical clinical case first, so what they leave behind is a select remainder, weighted toward conditions where statistical inference is weakest, human judgment developed, and physical skill irreplaceable. Physical AI (robotics) and remote medicine, exemplified by hospital-at-home, stand to profoundly disrupt the by whom, the how, and the where across the care spectrum. The learning health system, a key component of the discovery edge, is the organizational mechanism by which this matrix rewrites itself. Because the learning health system inherently optimizes toward whatever it can measure, it de facto selects against the unmeasured values and goods of medicine, not to mention the nuanced wisdom in human decision-making and the virtuous qualities of human connection. AI-era learning systems can in principle detect a signal, test it against population-scale records, and surface a revised recommendation within a clinical cycle. No single mode of medicine dominates, and a physician trained for just one is thus trained for but a narrow band of healthcare’s future. Further, any vision of the future physician built around a fixed picture of care will be obsolete before the trainees it describes complete residency.

Two streams of technoscience bear on biomedicine. One grows inside biomedicine; the other outside of it, in far-flung fields, the likes of quantum sensing, metamaterials, photonics, neuromorphic and edge computing, privacy-preserving computation, and ambient sensing. Clinical applications are set by industries that at times consult no clinician, and use cases often arise in an opportunistic and eclectic fashion. Disruption by emerging technologies is multilevel and multimodal, spanning specialties. Classical reverence for simplified models (think lac operon in biological mechanism) gives way to an era in which nonlinear dynamics and complexity are grudgingly admitted. Complexity science enters the picture as a formal discipline, complementing reductionism. It defines the limits of prediction, as it tells a physician in advance when average-based reasoning is about to mislead. In parallel, high-dimensional data assumes center stage, coupled to artificial intelligence tools for grappling with it. There is no single thing called AI. Frontier systems fragment into interoperable forms (predictive, linguistic, robotic, and ambient) that the physician must weigh and validate, and they operate in three modes (augmented, artificial, and emergent), whose demands on oversight differ. Pattern recognition is not understanding. Data quality is the critical constraint, and prioritizing data procurement over curation is a strategic error; data volume without provenance, harmonization, and maintenance produces confident models with no defensible relationship to reality. Biomedical data is layered: validation is required at every layer, not only at the visible output. Medicine may need an FDA-like function for health data and analytics to certify the substrate on which clinical AI rests. In the face of complex systems and big data, mathematics and statistics ascend in importance. In the physical AI space, robotic autonomy is bounded by perception, and machines are already beginning to sense what clinicians cannot. The physician's task thus acquires a new dimension: in validating machine perception and deciding when a machine-sensed reality is trustworthy enough to act upon. Viewed as a gradient of autonomy (robot assistance; task autonomy; conditional/supervised autonomy; high autonomy; full autonomy), the physical AI field is in its infancy, but with downstream implications for the preceduralist that are profound. As capabilities of cognitive and physical AI’s, in parallel with an ever-expanding suite of other game-changing technologies, detach from human specification, oversight must go beyond auditing inputs to interrogating behavior. Human intervention will remain critical: while a system may exceed any physician at inference and task execution, it holds no meaningful view about what is worth wanting.

The ambition is N-of-1 medicine: ultra-personalized diagnosis, therapy, and disease preemption, delivered with a human touch. For most of medical history physicians had too little information about the patient; soon they may have far too much. The defining question becomes how to parse billions of data points and turn them into wisdom about one human being. Personalization inherently narrows the comparator subgroup, creating a barrier to arriving at N-of-1 care: the band of relevant published evidence collapses as tailored specification rises. This paradox may be addressed down the road by mining electronic health record databases for unpublished correlations and developing sophisticated inference models that delineate individuated biological and pathobiological mechanism. Tailored drug purposing becomes a real-time N-of-1 deliverable.Yet, a complete description of a person at the pathophysiological level would still not decide what to do about them, because acting requires judging what does not actually matter for this one individual, and assuring that the outcome sought is the one they would choose. Patients are as plural as their biology, since sociodemographic variables explain but a small percent of the variance in what a given patient wants. A treatment recommendation that individualizes the dose while standardizing the justification has personalized only half the encounter. The irreducible N-of-1 gap thus relates to tangible data and intangible human desires. The biomedical digital twin emerges as a promising instrument for addressing the reducible part of the N-of-1 gap: when an intervention can first be tested in silico on an individual patient’s simulacrum, that patient in effect serves as their own control, before the intervention reaches the bedside. Counterfactuals become available at no risk, and the reference-class problem is dissolved by making each person their own reference class. By definition, preemptive medicine acts on a trajectory before disease has manifested itself overtly. The burden of proof for intervening in someone healthy is greater than that for intervening when illness has already manifested; this points to a competence the profession has yet to explicitly articulate—existential counselling. As individualized, preemptive medicine gains traction, lifetime health trajectories come to the fore: long-horizon health pathways integrating biological, behavioral, environmental, and social influences, managed as a continuous object rather than as a sequence of episodes. The N-of-1 ambition, pursued honestly, forces the aperture open—both range of data and range of what may be done—and evidence-based, holistic-integrative medicine goes mainstream.

Even after machine superintelligence arrives, a compelling human niche will persist. Human advantages cluster around being a subject with stakes and standing, which is to say axiological grounding, answerability, practical wisdom, and first-person perspective, while machine advantages cluster around the mechanics of cognition. The former is meaning-based intelligence: understanding grounded in context, values, and lived experience, as against intelligence grounded in pattern and association. Machine advantages render a decision better; human advantages make it a decision at all. The dividing line between human and machine runs between kinds of question rather than between specialty interests, and it splits clinical judgment internally: diagnosis and cataloguing of therapeutic options moves toward the machine, while the goals-of-care conversation stays human. Any claim to authority resting on knowing more, remembering better, or seeing patterns faster is already lost, and the profession should surrender it without protest. That surrender has implications that are yet to be confronted, as admission, licensure, and certification still select and examine for exactly the capacities being conceded. What remains is the physician as Guardian of Health, and guardianship is a standing relation rather than a task, carrying obligations that cannot be delegated to a system that bears nothing. The physician is uniquely positioned to bridge the dimensions of a high-dimensional health ecosystem, and as knowledge itself is equalized, moves from keeper of scarce knowledge to appraiser of it, skilled in verification. The impactful physician is the one who converts shared knowledge into wise, patient-specific action, and who can distinguish genuine insight from persuasive recombination of incomplete or flawed prior information. The physician is reinforced as the trusted and accountable representative. Representation is a stronger relation than interpretation per se, since it implies an interest held on someone else's behalf, and patients will be seeking exactly that when the information itself is free-flowing. The current Hippocrates 3.0association of professionals gives way to a Hippocrates 4.0 collective of experts, in which an expert is not the member who knows the most but the one who can be held responsible for the decision at hand. This expertise will increasingly call upon the cultivation and instantiation of a new set of capacities and personas in physicians, designated here as identities. Four physician identities are delineated, each mapping to and leveraging defined human advantages: The technoscientific identity is fluency at the scientific and technological frontiers melded to a drive to discover. The consilient identity is capacity to bridge knowledge domains and reunify what specialization has divided. The transcendent identity speaks directly to human advantages—medical virtues, meaning-making, moral agency, and judgment at the edge of the unprecedented. The global identity calls for stewardship of populations and of the planet. Each identity is not only a description of what the physician becomes, but also serves as a beacon for the formative experiences that can cultivate it; an identity that cannot be cultivated and taught is a portrait and not a plan.

The Flexnerian reform transformed the physician into a practitioner of scientific medicine. Since then, wide-ranging technological innovation has accelerated dramatically, with a profound impact on base biomedical knowledge, as well as the discovery process itself. Technoscience names a situation in which science and technology have stopped developing in parallel paths; the technological instrument enables the science, and the science exists in the form the instrument permits. The technoscientific physician works at points of convergence: where instrument and question are co-instantiated, and where those two meet the patient—so uniquely positioned to reify clinical application optimally. Discovery can no longer be reserved for an eclectic subset of physician-scientists; every physician should see themselves as a creative, and to be empowered to do so. That creativity runs along two lines. The explorer recognizes that a method built for one problem answers another, and the designer notices an arrangement everyone had stopped seeing. The designer's half tends to be undervalued because it often does not involve dazzling technologies, requiring only curiosity, a willingness to challenge what is assumed, and a synthetic aptitude for connecting and rearranging. Machines are inherently incurious, since nothing surprises a system that holds no expectations of its own; therefore, as machines take over answering questions, asking them becomes the seminal act. Interactional fluency, not deep dual expertise, is the requisite cross-disciplinary skill; it is the ability to state a clinical problem in terms an outsider can act on, and to appreciate a compelling answer from the outside when it arrives. Though the physician is aptly suited to creative discovery, there has paradoxically been an ever-widening divergence between accelerating science and the Flexnerian scientific physician’s clinical practice. The two are evolving along disconnected paths, leaving physicians progressively less able to meaningfully engage with, critically evaluate, or efficiently apply what is being discovered. The biomedical research enterprise must be reimagined to remedy this divergence and to instantiate the technoscientific physician of the future.

Breakthroughs increasingly happen at disciplinary interfaces, yet too often, specialty and health profession boundaries obstruct the crosstalk predicate. Machine intelligence changes that equation. The challenge of keeping up is what forced specialization, but current AI systems already match specialists at bounded interpretive tasks while progressing far more slowly toward the generalist's discipline-spanning intuition. The specialist's territory turns out to be the more automatable one. While organ system identity and disease biology have typically determined where specialty divides fall, transdisciplinary technology applications, shared computational tractability, and procedural overlap argue for more permeable boundaries. The consilient physician operates across knowledge domains, both within and outside of biomedicine, and integrates what they offer, in Whewell's original sense of independent lines of evidence jumping together. As knowledge domain streams converge within each physician, what the work demands is not necessarily fluency per se—enough to simply know to which stream a problem belongs, and then, with sufficient grounding and interactional expertise, to solicit what is needed collaboratively. Redrawing specialty boundaries is consilience applied to practice structure: grouping knowledge domains so the combinations produce more coherent and creative webs of explanation than those inherited from an earlier century. The logic for restructuring the specialty landscape revolves around interlaced themes: converging biologies, unifying the care continuum, recognizing platform-defined practice, stewarding new capabilities, and reclaiming coherent domains.Because redrawing specialties confronts specialty fiefdom vetoes, one tractable, short-term path is to create overlay specialties, which limits taking of territory and establishes dedicated homes for emerging theranostic and operationally demanding technological domains currently orphaned. Data sciences, complexity studies, and clinical robotics are obvious early candidates, being specialty-agnostic in application and owned by no board. The singular physician meanwhile dissolves into distributed cognition, where remembering and reasoning are carried by a system of people and instruments rather than by one mind. The physician’s authority migrates from credentialed knowledge toward interpretation, judgment, and rendering final decisions amidst complexity. The horizontal care team ensemble replaces the singular physician, as it embraces the full spectrum of medical fields. The team widens laterally to complementary professions and forward to professions not yet existing, covering tasks such as digital health stewardship, longitudinal patient navigation, and virtual care coordination.A point of view on how scope of practice will increase for other healthcare professionals, in terms of cognitive and procedural activities, is a predicate for delineating the roles of the future physician. Interprofessional education must no longer be treated as merely a scenic detour in undergraduate medical education and instead should advance to higher rungs of the interprofessional training ladder. The notion of trans-professional education arises, extending formation beyond the walls of the health professions and putting physicians into genuine working interdependence with fields outside biomedicine altogether, on real problems that would benefit from complementary expertise and shared accountability.

If machines absorb the routine cognitive and technical core of medicine, the profession's center of gravity shifts to what a machine can perhaps simulate but not genuinely possess. Some human-machine gaps are irreducible, rooted in theorems, in physical principles, and in the first-person character of experience, and no engineering closes them. Inside those limits—the zone where the human transcends the machine—one frames a functional space for the physician not explicitly delineated heretofore: as decipherer, navigator, and engineer of complex adaptive systems; as consultant to the patient from a place of standing and situated judgment; as comforter to the patient, manifesting medical virtue in its manifold dimensions; as bioethical guidepost for patient and community alike, with a honed capacity for ethical judgment.; and as a creative, leveraging deep biomedical grounding to explore, pioneer, and prioritize technosciencientic advances and new modes of medicine. Divide any encounter into work that depends on knowing something and work that depends on standing in a particular relation to the patient. Automation reduces the first and leaves the second untouched; thus, the human factor grows more consequential as the technical is automated. The physician concentrates at the non-routine edge, governing AI-native systems through supervised autonomy and owning the clinical decision. What emerges is a physician-patient-AI triad.Complexity sets a permanent ceiling on prediction, which gives human judgment a permanent home at the edge of the unprecedented, where the physician answers what should be done rather than what can be done for this particular patient. Ethical judgment is the connective tissue of the human advantage, running through rehumanization, a grasp of complexity and ambiguity, evaluation of truth, and unbounded curiosity that are tailored to the individual patient. Caring, the physician·patient I-Thou, becomes a core human differentiator, with medical virtue as an encompassing descriptor of the settled dispositions a physician must hold to achieve the ends of medicine: comprising fidelity to trust, compassion, prudence, justice, courage, temperance, self-effacement, and integrity. Empathy alone is an insufficient ground on which to stake the human claim, because it is a perceptual capacity machines can simulate; the stronger framing is medical virtue. Every résumé virtue of medicine is in principle machine-attainable; not one eulogy virtue is. Virtues form through habituation, and habituation is subject to environments, so a system that prices out authentic physician·patient engagement misses the mark. Indeed, a system that allows medical virtue to lapse, let alone one that intrudes on the physician's capacity to enact actions they know to be ethically correct, inflicts moral injury to the physician. An assemblage of complementary physician intelligences come to the fore—cognitive, emotional, moral, spiritual, social, and aesthetic; the profession has tended to cultivate and assess only the first. These other intelligences, and the clinician attitudes and behaviors that go with them, merit formal research centers devoted to them. Compassion science and relational healing enter the picture—investigating mechanism, dose, and durability with the seriousness applied to any other intervention—and compassion is to be hardwired into medical education and into the research enterprise.In a twist, machine intelligence turns out to be that which clears the forest and makes rehumanization possible, by relieving physicians of busy work that interferes with human connection. Thus, through machines, Hippocrates 2.0 sensibility is resurrected in the Hippocrates 4.0 physician.

Health is managed at the level of populations, not just individuals, and so physician·community commitment parallels physician·patient relations. The lineage runs from Virchow, who traced typhus in Upper Silesia to poverty and disenfranchisement, to Lown, who argued that a profession organized around prevention must prevent what it cannot treat, by example, nuclear proliferation. Social responsibility therefore extends to sectors that determine health, which is not mission creep: if the obligation is to prevent morbidity and mortality, it should follow the causes wherever they sit. Algorithmic systems in organizations and societies are now determinants of health in the same sense as housing and air quality, which makes procurement and threshold-setting public health decisions. Rose's distinction between the causes of cases versus causes of incidence explains why clinical instincts can mislead at population scale, and the efficacy-to-effectiveness gap and inverse care law convey how extending care to populations frequently fails. While N-of-1 precision seeks to tailor care to one person, population health heads in the other direction as it strives to make what is already known reach everyone with the same reliability. The population health frontier is by construction transdisciplinary: the consilient physician of Chapter VIII and the global physician described here are therefore the same person viewed from two angles.The frontier has moved upstream toward structural, commercial, and political determinants, exposomics, planetary health, and implementation science, and none of these high-impact areas are restricted to a single discipline. Care without an address, remote monitoring and treating, and networked expertise and data analytics extend care across borders, narrowing the gap to low- and middle-income countries and turning equity and access into foundational design questions.Technological leapfrogging for low- and middle-income countries in the health sector offers intriguing possibilities. Governance of frontier biotechnologies is normative without being binding, but the profession still has levers: refusal to participate, publication norms, and international specialty standards that travel with the practitioner. While the physician is not to be a social engineer digressing beyond the health sphere in controversial, politically charged ways, the physician does have essential roles to play in global health governance, security, and ethics. Climate change, and its projected impact on health, is a cause that beckons physician global action. Yet other population threats that have global health consequence, and hence merit physician engagement, are nuclear weapons, biological weapons, and artificial intelligence itself.

Having crystallized four identities of the Hippocrates 4.0 physician, the question becomes how to instantiate them. Medical education assumes center stage. The four identities are meant to inhere in every physician, albeit in varying proportions depending on chosen role, and so the call is for establishing a common base of human qualities and proficiencies to be cultivated over the course of the training continuum. The notion of competencies takes on a different meaning—as a specification for formation that assimilates a spectrum of the human advantage arc. The architecture of physician training can be schematized as three concentric competency cores. The inner core holds compassion, ethics and moral agency, judgment and meaning-making—in other words, the human elements that make for a transcendent physician. The middle core holds technoscience and discovery, and so, is directed towards the technoscientific physician. The outer core holds transdisciplinary fluency and is part of what inheres in the notion of a consilient physician. All three cores are essential for cultivating the socially responsible global physician, highlighting how the four identities are interwined, in formation and in practice. Notably, the cultivators of the transcendent physician are intentionally positioned centrally as the inner core because they are the least examinable, the least fundable, the slowest to prove their value, and the first thing a productivity metric removes. A medical curriculum that builds the outer core without the inner produces a competent operator of systems, and one that builds the inner without the outer produces a humane practitioner who cannot evaluate the tools that deliver the care. As a next step in this deliberate educational design, one defines foundational elements that give the cores substance, as well as enablers, the forefront mechanics of curricular layout, didactic modality, assessment, and progression, without which none of the elements install themselves. The practical on-the-ground obstacle in realizing this vision for physician training is time allocation, since a curriculum is a fixed volume, and content that is not built in, examined, and resourced will be pushed out however strongly a mission statement affirms it. To offset this constraint, it will be essential to structure physician education as lifelong, morphing it into something far richer than the narrow continuing medical education of today. Ideally, this is to be accompanied by individualized longitudinal mentoring that spans a physician lifetime. As an essential part of this restructuring, the UME-GME-CME divide must be rethought from ground up and more deeply integrated longitudinally, with a common view towards an expanded slate of competencies. In this way, the notion of fluid and plural physician identities comes to life.

Fragmentation across knowledge domains, technologies, specialties, institutions, and national and international research enterprises is the single greatest obstacle to the future described herein. An increasingly atomized and corporatized ecosystem places the academic research enterprise at a disadvantage, and increasingly at risk. The maze of bodies that each partly controls or impacts the profession is itself the diagnosis, since each acts rationally within its own remit, and yet, no actor can authorize a change while every actor can veto it. What is needed is not a new supervening authority, which realistically will not be created and in any case would not be trusted, but rather coordinated evolution beginning with exploratory pilots. Precedents from engineering, aviation, and architecture show the mechanism: a small coalition of credible institutions demonstrates an alternative, measures it publicly, and lets the accrediting bodies follow the evidence. The stakes are real. A biomedical digital twin built on one institution's data models grounded in one institution's patients, and a learning health system that cannot see past its own walls, learns but a local artefact masquerading as solid evidence. Federation is the remedy, and standing arrangements across Japan, the United States, and Europe are illustrative and share a four-part design: each is a federation rather than a merger; each has a neutral convener no participant controls; each defines in advance what is shared; and each publishes. The physician is a natural orchestrator for such arrangements, with standing recognized across institutions, and at once, the person the patient believes is theirs. There is also a financial case to be made: the twelve foundational elements set forth as predicates for an identity-oriented curriculum would require an armada of a new genre of faculty; few schools could afford more than a handful, but ten schools sharing them makes the enterprise tractable.

Consolidated, these yield a single vision. The physician of the mid-twenty-first century is a member of a collective of experts, holding four fluid identities overlaid on specialty, working at the human plane of an intelligence ecosystem, filling the irreducible gaps that no machine closes, and formed continuously across a career rather than minted at graduation. The message is an optimistic one, as there is a definable constellation of human advantages that are not readily displaced, and quite the contrary, are likely to ramify and flourish as routine cognitive and procedural work is off-loaded to machines. Themes of convergence, transdisciplinarity, complexity, creation and virtue will be ascendent, as they infuse a robust new set of physician roles and the training thereof. The physician’s palette expands beyond the physician·patient I-Thou, to surrounding communities and mankind across the globe. Unleashing the physician in this way demands systemic changes at governmental and institutional levels. Yet, animating this vision starts per force with the physician’s embrace of self-valuation, self-actualization, peripheral vision, and a bold sense of agency. Notwithstanding dramatic advances in diagnostics and therapeutics in the Flexnerian era, medicine has steadily lost part of the best of the past—humanistically, socially, and relationally, and at the same time, failing to be sufficiently imaginative about a radically different future looming before us. The remedy starts with physicians themselves.

PART I

The Emerging Reality