Intrinsic health as a foundation for a science of health

Abstract

The health sciences largely focus on disease. However, the interconnected determinants of diseases suggest that we need a science of health, a framework to examine the biology of homeodynamics in a changing environment and how this affects the health we value. We build on first principles and recent discoveries on biological system dynamics to develop the concept of intrinsic health, a field-like state emerging from the dynamic interplay of energy, communication, and structure within the organism, giving rise to robustness/resilience, plasticity, performance, and sustainability. Intrinsic health is a quantifiable property of individuals that declines with age and interacts with context. We propose a measurement framework and describe how it will contribute to achieving the shared goals of medicine and public health.

Intrinsic health is an objective, measurable construct describing the biological capacity of the organism to self-maintain.

INTRODUCTION

Our intuition for how healthy we feel is powerful and unexpectedly predictive (1, 2); nonetheless, the health sciences lack a theoretical framework to understand the biology of health as distinct from the absence of disease. As a result, neither the biomedical sciences nor public health has arrived at recognized, objective measures of how healthy someone is. Without a clear target in sight, the focus of biomedical scientists and funders has naturally centered on discrete diseases and their biomarkers, which can be visualized, quantified, and tracked over time, offering natural targets for interventions (3). To fulfill the promise of both public health and precision medicine, we need a science of health to develop tractable quantifiable targets that can be integrated within the ecosystem of interventions on health determinants. The foundations for such a science are already in place. Unifying them would generate a theoretical and empirical framework that would represent a major departure from the current disease- and organ system–based paradigm, bridging disciplinary silos across medical specialties, diseases, and organ systems.

Our key contribution, missing in previous frameworks (Box 1), is a first principles–based theory that allows us to operationalize and measure health. Once validated, the intrinsic health approach will provide general benchmarks for clinical, public health, and policy-based interventions, solving the challenge of choosing among multiple, often conflicting outcome measures. Elucidating the underlying pathways and mechanisms of health will shift the focus from reactive, late-stage interventions in an already failing body to empirically grounded interventions that build and maintain health across the life course. By defining, understanding, measuring, and targeting intrinsic health, we can aspire to reach the shared goal of the health sciences: to enable individuals and populations to thrive and achieve their full potential across the life span.

Box 1. Previous conceptions of objective health.

The World Health Organization (WHO)’s 1946 definition of health as “a state of complete physical, mental and social well-being and not merely the absence of disease or infirmity” (124), famously scribbled on a napkin, has set the standard for considering health beyond the absence of disease. Nonetheless, it has also been highly controversial, particularly for the medicalization of happiness and well-being (125, 126). Despite this WHO definition, and despite the deep and growing conceptual challenges in defining diseases, health continues to be largely operationalized as an absence of disease (127).

Our focus here is on objective, biological definitions of health. These are generally more restricted than the WHO definition, but have a history at least as long, with an extensive literature in the philosophy of science (128–135). Canguilhem’s On the Normal and the Pathological (12) defined health as the ability of the organism to adapt to its environment, a definition still relevant biologically, clinically, and conceptually (105, 136). Canguilhem identified natural selection as the driving force behind these adaptive mechanisms (13, 137). This implies, on the one hand, that the health of a given physiological state cannot be understood outside the context of the organism and, on the other hand, that there can be a scientific foundation for establishing certain states as more normative to the extent they facilitate constant adaptation (130). Cannon’s work on homeostasis in The Wisdom of the Body (112) can also be seen as laying the biological foundations subsequently developed by many others, including Sterling’s notion of allostasis in health (105).

The need to move toward a science of health is not new. Discussions of a WHO working group led to a 1993 edited volume, Towards a New Science of Health (138), that presages much of what we argue here. In particular, the chapter by Goodwin (4) foreshadows the need to integrate evolutionary biology and complex systems theory to arrive at an understanding of the organism as a whole, with health as an emergent field. More recently, Sholl (131, 139) emphasized the need to move beyond definitions and toward an operationalizable theory of health; we concur and hope that our framework contributes to such a theory.

At a conceptual level, the framework we propose draws strongly on this body of work. However, the tools needed to study health as a biological phenomenon, both theoretical and technical, have only recently become available. For this reason, these previous conceptions have not had a major impact on the biomedical establishment, whereas quite a number of recent publications offer major insights into an objective, biological conception of health. Rattan (140) proposes that health can be understood in the context of homeodynamic space. Ayres (68) argues that there are specific biological mechanisms of health, distinct from those that cause disease, that are quantifiable by experimental approaches. López-Otín and Kroemer (141) propose a set of “Hallmarks of Health” organized into maintenance of homeostasis, response to stress, and spatial compartmentalization. Hood and Price (123) argue that we are on the cusp of a scientific revolution that, using repeated, high-dimensional multiomic data, will permit us to identify slight deviations from health and intervene long before disease develops. These efforts show how close we are to a paradigm shift toward a theory of the underlying biology of health and call for a first-principle assessment of the intrinsic drivers of human health.

INTRINSIC AND REALIZED HEALTH

Until recently, an objective operationalization of health would have seemed impracticable, given the panoply of factors that are implicated as determinants, correlates, and manifestations. However, new research, capabilities, and perspectives on the biological dynamics underlying life as an evolved and optimized complex system suggest that a biological, objective definition of health is not only possible; it emerges naturally from such an understanding of the nature of life. Accordingly, an appropriate theory of health can be constructed from the bottom up (4), understanding the nature of life as a complex system (5–7), how it has been shaped by natural selection (8, 9), and how this manifests dynamically in biological signals and dynamics within the organism (10, 11). This integration of evolutionary and complex systems theory views organisms as self-maintaining systems, thereby offering insight into how the ensemble of pathways and systems across biological scales integrate to achieve the overall goals of the organism, including maintenance of its health. The resulting understanding is thus uniquely objective and universal, allowing for operationalization of health as a biological norm (12, 13).

Such a first principles–driven biological perspective on health does not ignore the myriad ways that context and perspective shape the lived experience of health of humans in complex societies and changing environments (14–17). Rather, it provides a simple way to distinguish the biological foundations of health, termed here “intrinsic health,” from our lived experience, termed here “realized health” (Fig. 1).

Fig. 1. Intrinsic and realized health.

Realized health will inevitably be influenced by the underlying biology, but the nature of this influence is molded by numerous factors—the environment, affordances, culture, values, perspective, and so forth, which we collectively call here “the lens of context.” The lens of context will be unique to each individual and vary across populations. Intrinsic health is internal to the organism, can be characterized relatively universally across individuals, reflects potential rather than achieved health, and is objective (i.e., measurable independently of an individual’s opinions or perceptions). In contrast, realized health incorporates extrinsic factors, will require individual-specific characterization, measures achieved health, and is naturally subjective.

This conceptual distinction between intrinsic and realized health is crucial for us to make progress both on the biological universals and on how they are shaped by the lens of context. Intrinsic health will, of course, be influenced by determinants of health, both those internal (genetics, diet, immunological state, psychological state, etc.) and external (social context, environment, adversity, etc.) to the organism. Our proposition, however, is that apart from their causal role, information on these determinants is not necessary to understand intrinsic health, which is fully defined by the current biological and psychological [i.e., psychobiological (18)] state of the organism.

Here, we offer a theoretical framework to understand intrinsic health. We start from first principles to understand the nature of life and how it evolves, using this to propose three key interactive pillars of intrinsic health: energy, communication, and structure. We show how intrinsic health can emerge as a field-like state from the interplay of these underpinnings and how they interact to create properties and dynamics that may be observable at the organism level. We briefly discuss relevant data and measurement frameworks as well as new questions that emerge from the framework and future directions for research on intrinsic health. The framework we propose should be applicable to all organisms, although our main focus is on human health.

A COMPLEX SYSTEMS VIEW OF LIFE

Life as we know it can be understood as an organized collection of molecules that perpetuates itself on two different scales. On the one hand, it reproduces, generating a new, separate organism. This reproductive aspect of life has been explored in great detail, both in work on the origins of life (9, 19) and in evolutionary biology (20). On the other hand, an individual organism must perpetuate itself in its environment for a certain period of time to successfully reproduce. Given the second law of thermodynamics—the tendency toward entropy, which life must resist (21)—this self-perpetuation is, in fact, a monumental, energy-intensive task. Both the internal and external environments of the organism are in constant flux through conditions both predictable and unpredictable, and survival depends on appropriate adjustments to these perturbations. These adjustments are made constantly and at multiple scales, from molecular to behavioral. They are endergonic (i.e., requiring energy transformation) and are determined and constrained by the structure of underlying biological networks (genetic, molecular, physiological, neural, etc.), which have been shaped by natural selection. These adjustments need to operate in a unified, coordinated manner and emerge from the dynamics of the myriad underlying pathways and structures working together as a complex dynamic system (6, 10).

Complex dynamic systems often exhibit numerous higher-order or “emergent” properties such as modularity (22) and resilience (23, 24). Biological complex systems are no exception (25). Some of these emergent properties have direct links to the prime directive of the system to self-maintain and heal—meaning to become whole again, or to move toward health (26). Thus, we should expect natural selection to optimize them. The key examples are robustness/resilience (27, 28), the broad capacity of the organism to maintain a given state or function in the face of external stressors (we do not here wade into the more specific debate on defining and disambiguating these terms); plasticity (29), the capacity of the organism to appropriately shift its state in response to changing conditions; performance (12, 13), the capacity of the organism to effectively execute specific functional capabilities that have generally been important for the fitness of its species; and sustainability (30, 31), the capacity to maintain robustness/resilience, plasticity, and performance over long timescales.

We call these properties intrinsic health components. Like fitness components in evolutionary biology (32–34), intrinsic health components may sometimes trade off with each other and may also respond to other forces (Fig. 2). In addition, like fitness components, there is not a single, objective classification, and the components may conceptually overlap or correlate with each other (33). Because they are all organismal expressions of intrinsic health, they are expected to relate to intrinsic health at a fundamental conceptual level while at the same time serving as imperfect proxies. Like fitness components in ecology and evolutionary biology, intrinsic health components may serve as a key operational bridge between theory and data that can actually be collected.

Fig. 2. A comparison of evolutionary fitness and intrinsic health.

Evolutionary fitness is measured at the level of the individual (one value over a lifetime), with selection to perpetuate a lineage across evolutionary timescales. Intrinsic health is measured at a moment in the life of an individual, with selective pressure to perpetuate the individual in their environment across their life course. Both can be broken down into components—necessary features that are part of and conceptually integrated with the ensemble but that may also trade off with each other.

THE PILLARS OF INTRINSIC HEALTH

We propose that intrinsic health rests on three key pillars: energy, communication, and structure. Any complex system that persists in the world does so in an open system; a closed system tends toward entropy, with any complexity dissipating into randomness (35, 36). In an open system, the persistence of a complex system requires inputs of energy. From a fundamental physics perspective, energy flow is thus a crucial precondition for life. It is the first and most fundamental pillar of life. However, it is not sufficient.

As noted above, the capacity of an organism to perpetuate itself depends on its ability to make important adjustments to its state in response to changing conditions, and this requires the ability to acquire and share information within the system, what we term here communication. Communication is what binds every cell into a cell collective (37). The adjustments can be seen as “decisions” the organism must make constantly, and making appropriate decisions requires appropriate communication among system components. Communication is the second pillar of intrinsic health.

Last, energy and communication do not exist in a vacuum; they evolved and function in a species-specific structure. Structure extends across physical scales from the molecular to cellular, anatomical, and organismal levels. Perturbations to structure may compromise the capacity of energy and communication to maintain the organism. Appropriate structure is thus also required for intrinsic health. Structure is the third pillar.

All three pillars exist on, and interact across, multiple hierarchical scales of biological organization (Fig. 3A). We discuss each pillar individually and then consider how they jointly produce intrinsic health.

Fig. 3. Intrinsic health as a field-like state.

(A) Intrinsic health emerges from the energy, communication, and structure of the body at multiple interacting hierarchical levels, from molecular to organismal. (B) A battery (energy), current (communication), and solenoid coil (structure) jointly produce a magnetic field as an emergent property that cannot be measured or understood via the components individually.

Energy

Energy is broadly defined as the capacity to do work and exists in multiple forms (thermal, chemical, electrical, kinetic, and others). The central distinguishing factor between a living, breathing, conscious person and the inanimate cadaver is the flow of energy. To sustain life, our cells, organs, and physiological systems require constant energy transformation, nourished by oxygen and food metabolized via our mitochondria, small bacteria-derived intracellular energy processors (38). The flow of energy powers all psychological and biological activities that support human life across the life span (39).

But energy does more than just sustain life; it plays an instructive role in directing cellular, physiological, and cognitive processes, from cell differentiation and developmental transitions (40) to human motivation (41), social and anxious behaviors (42), and cognitive decisions (43). The amount of energy available, its flow from one cell to another (as a form of communication, see below) or from one form into another, directs cellular and organismal responses to challenges (44). Life and human experiences are moved by energy.

We know that energy is a central pillar of health not only from first principles, which stipulate that the input of energy is required to avoid decay, but also empirically from clinical studies of individuals who have impaired energy transformation due to molecular defects within their mitochondria (45). When mitochondrial energy transformation is impaired, a suite of integrated changes, including gene expression, cytokine secretion, and brain-derived hormones, facilitates an integrated response (46–49). As a result, the whole body becomes less efficient, requiring more energy to be sustained (50). By enabling and directing life at levels ranging from organelle to organism (51), energy is a central pillar of intrinsic health.

Communication

Communication, or information flow, refers to information transfer that functionally links all components of the organism into an integrated whole, capable of adjusting biological regulation (i.e., making decisions) based on the state of the whole, informed by both its internal state and external conditions. Signaling pathways within cells, bioelectrical fields (37), neural circuits, hormones, and paracrine signaling including from mitochondria (51) are all well-established examples of communication between cells and organs. Healthy communication maintains the organism in dynamic equilibrium (i.e., adjusting state appropriately based on internal and external conditions); its converse is thus dysregulation, or the inability of the system to maintain dynamic equilibrium due to impaired communication.

There are numerous examples of impaired communication causing deficits in health: excessive expression of senescence-associated secretory pathways (52), abnormal neurotransmitter function altering cognition (53), and essentially all endocrine disorders including diabetes (insulin resistance being the inability of a hormone to execute its physiological signaling function). In some sense, these examples are trivial: The importance of communication is well understood in specific cases. However, communication should be looked at not only case by case but also as the overall ability of the organism to properly coordinate the numerous decisions it is making at any given moment. This view is supported by numerous recent findings on biology’s interconnectedness: the gut-brain axis (54), the conserved transcriptional response to psychosocial adversity (55), the role of mitochondria as intracellular information processing systems (51, 56), and immune regulation of metabolism (57), to name but a few.

While this organism-wide integration of information is essential to proper functioning, it also exposes the organism to the risk that problems in one compartment will spill over into others. For example, depression is linked to both dysregulated energy levels and immune state to the point where it is not always clear where the problem arises from (58, 59). Integrated, global communication in the organism is an essential pillar of intrinsic health. Accordingly, deficits in communication should directly alter intrinsic health. These deficits could arise from imprecisions or trade-offs in the developmental process, from accumulated damage and information loss during aging (e.g., increasing burden of somatic mutations), or from specific disease processes.

Structure

Both energy flow and communication must occur in the three-dimensional physical structural features of an organism. These features exist at many levels: The structural features of the DNA-protein chromatin complexes constrain which genes are expressible. The structural features of mitochondria shape how energy is produced and regulated, and with what level of efficiency. Cardiovascular circulatory network structures influence how blood flows and how humoral factors are transmitted between organs including the brain. The structural features of limbs contribute to the energy efficiency of movement and the trade-offs inherent therein.

Structure is qualitatively different from both energy and communication: It is the substrate or matrix in which energy and communication evolve, and it will shape and constrain them. For example, the timing and deployment of the stress response (an aspect of communication) in vertebrates are constrained and shaped by how neurons transmit signals (60) and by the diffusion capacity of water- and lipid-soluble hormones (e.g., epinephrine and corticosteroids) across membranes (61). When structures are damaged, they will often impede energy, communication, or both. Examples include (i) gene mutation or protein misfolding, which may alter an enzyme’s function and inhibits or dysregulates signaling (62); (ii) coronary artery calcification, which inhibits flows of both energy and information in the blood (63); (iii) and brain lesions, which cause functional disconnection between brain regions and cause predictable symptoms (64). In addition, structural alterations such as loss of a limb, abnormal eyeball shape causing myopia, or the presence of scar tissue may have consequences for health that are not directly dependent on energy or communication.

Both energy and communication are relatively integrated across different levels of biological organization, and we hypothesize that it is possible to characterize organism-level energy integrity or communication integrity. In contrast, structure can only be described by a laundry list of largely independent components. Structural problems in arteries, cell membrane integrity, and loss of a limb are largely independent of each other. It is thus unlikely that we will ever characterize the global structural integrity of an organism, other than as such a laundry list.

INTRINSIC HEALTH AS AN EMERGENT, FIELD-LIKE STATE

With these first-principles theoretical underpinnings in place, we are now ready to define intrinsic health as a field-like state emerging from the dynamic interplay of energy, communication, and structure within the organism, which gives rise to robustness/resilience, plasticity, performance, and sustainability. Intrinsic health determines how well our internal homeostatic and regulatory mechanisms succeed in keeping us in the state most conducive to survival and reproduction (dynamic equilibrium), given our internal and external context. Below, we discuss some key features of intrinsic health relevant to its operationalization and measurement.

Intrinsic health is prospective: High intrinsic health will facilitate its own maintenance over the life course of the individual (sustainability), as well as in the moment. It thus also involves efficiency, a minimal incurring of trade-offs or costs elsewhere in the system to permit long-term sustainability. A fundamental question, relating to the basic ontology of intrinsic health, is whether it is a state, a capacity, or a process. While intrinsic health is primarily a state of the organism, it is a state that implies the capacity for the intrinsic health components, and is also a manifestation of the underlying dynamic processes that maintain it. Temporally, then, intrinsic health should be measurable either as an instantaneous state or as an integration of responses over a short time course (65), although its temporal implications for the future of the organism are across much longer timescales. Intrinsic health is fundamentally an integrated property of the organism, rather than a property of its individual components (cells, organs, etc.). Moreover, the same level of intrinsic health can be achieved via different combinations of parameters.

While we have just described energy, communication, and structure, we posit that high intrinsic health (the strength of the emergent field) is what emerges when these three pillars operate in harmony. Intrinsic health cannot be a simple sum of the state of each pillar. Even if communication and structure function perfectly, starvation/malnutrition would stifle the flow of energy through the system and lead to poor intrinsic health. Even if energy and structure function perfectly, chronic stress exposure may direct excessive release of stress hormones that overactivate and divert activities in a way that suppresses immunity (66). In addition, even if energy and communication function perfectly, a gene mutation affecting the structural integrity of the plasma membrane in muscle cells leads to Duchenne muscular dystrophy (67). Structure provides the matrix through which energy flows and is organized by communication, and thus, they are all interdependently linked.

As a result, from a measurement perspective, we hypothesize that the pillars are deeply entangled and that separate quantification of each pillar will not be feasible. For example, is sarcopenia an energy problem or a structural problem? Measurement of intrinsic health is thus likely to be most useful by targeting the emergent field rather than the individual pillars. This view is distinct from the view of physiological health proposed by Ayres (68), which is fundamentally mechanistic in viewing health as the sum of many specific pathways that evolved to counter specific threats to health.

A useful analogy to understand how field-like intrinsic health emerges from the three pillars is an electromagnet (Fig. 3B) (69). In a simple physical system combining a battery, a wire, and a solenoid (a coil of wire), the flow of electrons causes a magnetic field to emerge (70). The exact physics behind this transformative process follows Maxwell’s laws of electromagnetism (71) and Feynman’s quantum electrodynamics principle (72). The resulting field is invisible yet nonetheless measurable via its effects on specific (paramagnetic) objects. In addition, it cannot be understood other than as an emergent property of these three components. Thus, studying each component—energy, communication, and structure—in isolation through the lens of reductionism provides no fundamental or practical understanding of the emergent field. The same may be true of intrinsic health.

To extend the analogy, a magnetic field can also be influenced by a series of magnets that are aligned with each other. Magnets that are misaligned can weaken the whole field, but the field itself tends to bring individual magnets into alignment (as with the needle of a compass), reflecting coherence in the field (69). In the context of an organism, the individual “magnets” are the numerous components of the organism, both physical (e.g., mitochondria, cells, and tissues) and functional (e.g., relatively undamaged DNA, a robust stress response, structural integrity of the musculoskeletal system, and a psychological state of confidence). A problem in one (e.g., scoliosis) may entrain problems in others, potentially on different timescales (pain, limited range of motion, lack of exercise, and lack of confidence). In addition, we posit that there may be alternative strategies for an organism to maintain intrinsic health, analogous to different orientations of the groups of magnets. The important thing is that they be aligned (i.e., each component aligned correctly or in the right state given the ensemble) rather than the direction they point (the magnetic field orientation could rotate if all magnets rotate, but still be coherent). Furthermore, a misalignment or problem in a given magnet/health component could affect other components of the system, just as a misaligned magnet would change and likely weaken the overall magnetic field. Considering such interdependence could lead to network models of health.

THE EMPIRICAL BASIS FOR INTRINSIC HEALTH

The core hypothesis we present here is thus that intrinsic health is a measurable field-like state of the organism. More specifically, we hypothesize that the energy, communication, and structure of diverse biological systems are sufficiently integrated that intrinsic health as an overall emergent property can be validly and reliably measured. Because the pillars are integrated, we propose that measurement should be based not on them but rather on intrinsic health components (Fig. 4), which are the higher-order functional manifestations of that integration. This is analogous to measuring a magnetic field based on its effect on iron filings, rather than based on the battery, current, and solenoid. Our hypothesis would be falsified if the components of intrinsic health and other proxies are not sufficiently correlated and there is no evidence for a generalized, common construct uniting them (Box 2).

Fig. 4. A measurement framework for intrinsic health.

(A) Intrinsic health is an emergent property of the pillars and thus is not directly measurable as a simple function of them. Nonetheless, consideration of the pillars may lead to conceptual development of relevant proxies, such as proteome communication dynamics in response to stressors. Intrinsic health (IH) is characterized by its components, for which many proxies are likely available. The components will sometimes show negative correlations due to trade-offs, so a single component is not a sufficient measure. Proxies for the components, and for intrinsic health more generally, can be generated and then integrated via algorithms based on patterns of dependence to generate an overall metric of intrinsic health. These algorithms should perform better (e.g., better predict health outcomes) as many diverse proxies are integrated, but should show diminishing marginal returns. Note that conventional latent variable approaches assume causal directionality between intrinsic health and its components; the approach proposed here does not make these assumptions. PCA, principal components analysis. (B) Example of a hierarchical latent variable approach to an integration algorithm. Edges among latent constructs indicate causality flowing from lower-order to higher-order constructs; this could be tested. P1 through P17 are measured proxies. Initial versions of the model could use fewer hierarchical levels, with additional proxies and levels added as we become better at measuring subcomponents.

Box 2. A measurement framework to quantify intrinsic health.

Like fields including gravity, intrinsic health is not directly measurable—it is a latent trait. Our view on quantifying intrinsic health takes inspiration from first-principles field physics (69) and latent variable modeling in other areas of health, wherein patterns of dependence across many directly measurable characteristics (e.g., biomarkers and survey items) are aggregated and explained by latent constructs that are themselves the causal drivers of both measurable and hidden processes (142, 143). Here, we assume that intrinsic health is caused by its components, each of which may also be caused by various subcomponents (Fig. 4B). We outline a few key principles of measurement, implied by the fundamental biological characterization, to guide future efforts in data collection and modeling.

1) Although intrinsic health is a state, and thus may be measured at a particular time or multiple times in the organism’s life span, measures of intrinsic health should capture at least short-term dynamics, i.e., how a system changes (in response to stimuli, perturbations, etc.). In the same way that a field strength is defined by measuring the acceleration of a known object exposed to the field over time, longitudinal or dynamic data will be essential, even for a static snapshot of intrinsic health.

2) Intrinsic health spans multiple systems and multiple levels of organization in an organism. This naturally suggests a hierarchical mode of organization: not one latent variable but many, organized in a layered hierarchy.

3) There exist many proxies for intrinsic health globally, such as heart rate variability or homeostatic dysregulation, or for specific components, such as VO₂max for performance or metrics of stimulus response for resilience (oral glucose tolerance test, vaccine response, etc.). What is lacking is a unified model that integrates and explains the dependencies among all these proxies across subsystems. The model may begin with well-established biomarkers and measurements of these components, and incorporate them into an increasingly sophisticated hierarchical model that learns higher-level abstractions.

A data collection and modeling program might be proposed as follows: first, collect data on measures of intrinsic health and its components (resilience, plasticity, performance, and sustainability). These may often be inspired by an understanding of the pillars. For example, transfer entropy (144) or proteome dynamics could be used to measure dynamic properties of communication. Then, construct one factor or multifactor latent variable models (e.g., Fig. 4B) that explain the observed patterns of dependence across these measures. Next, integrate these subsystem models into a unifying cross-system model with latent variables determined by lower-level behaviors and dependencies. Recent methodological advances in deep, hierarchical latent variable modeling and causal representation learning (145–147) may be instrumental in realizing models with the above characteristics. A measurement framework will aid in probing the validity and stability of intrinsic health as a theoretical construct, including its dimensionality and emergence. The resulting construct should be further tested in its utility for tasks of prediction and control—if intrinsic health is real, then one should be able to put it to pragmatic use (148). The process of establishing the construct and finding its uses will be iterative: As with many theoretical entities in the history of science, exploiting the latent construct for prediction and control will be intimately tied to our beliefs about the existence and characteristics of the construct itself (149).

Multiple lines of evidence now suggest that there is a biological basis for a relatively unified definition of biological health. First, as health declines, multimorbidity accrues nonlinearly, and multiple organ systems tend to thrive or fail together (73), pointing to a shared factor underlying the functions of organ systems. The frailty phenotype is one good example of this (74), and cross-talk among hallmarks of aging is another (75). Multiorgan failure or disruption can arise from problems in the signaling dynamics among the organs as much as from the summative failure of individual organs (76).

Second, self-rated health (SRH; or self-assessed health) is often a stronger predictor of future health outcomes and mortality than existing biological or medical measures (77). This suggests that individuals holistically and, on average, accurately perceive some overall state of health—like a compass capturing the direction of a field without access to its underpinnings (78). Even among populations with objective disease such as cancer, SRH is poorly correlated with disease indicators and some individuals perceive their health as very good or excellent, which predicts survival independent of medical criteria (79). But probing the basis for SRH by asking people what they consider in their holistic assessment illustrates that individuals build their assessment on notably different inputs across life domains ranging from bodily sensations, comparison to others of the same age, social connections, and functional capacity, among others (80). SRH represents an individualized “holistic” proxy assessment of one’s intrinsic field-like state. Moreover, SRH correlates with several circulating metabolites, suggesting a link between the biology of energy and communication and SRH (81).

Third, network physiology and related disciplines have shown tight synchronization of dynamics across apparently diverse physiological compartments (82, 83), including the exchange of metabolic information between peripheral organs and the brain (84, 85), pointing to closely interwoven cross-talk in which decisions made in one part of the body incorporate information from many others. The structure of this cross-talk varies with age and disease status—for example, in Parkinson’s disease, brain-leg cross-talk increases during rapid eye movement sleep but decreases during deep sleep (86). We are just beginning to elucidate these relationships, but it appears that there may be normative dynamic interaction patterns among organs that are compromised as health is perturbed.

Fourth, generalized measures of system dynamics, such as early warning signs of critical transitions and measures of system variability/heterogeneity, predict a broad range of health outcomes (87–89), pointing to broad system dynamics reflecting overall system stability. Early warning signs are generally signals of end-stage decline and may not discriminate across the full range of health status; nonetheless, synchronicity of these metrics across diverse physiological systems points to generalized integration of health status in organisms (88).

Fifth, recurrent structural motifs in biological networks suggest extensive integration and information sharing across systems and functions (90–92), implying generalized integration of the maintenance of dynamic equilibrium across the organism. For example, while inflammation is generally thought of as a branch of immune function, inflammatory cytokines are implicated in many other roles, including energy signaling and metabolism (93), stress response (94), appetite/hunger (95), sickness behavior (96), post-exercise adaptation (97), and wound healing (98), to name just a few. Bow-tie motifs are found in both neural and biochemical signaling networks and appear to be an efficient information processing strategy to incorporate information from diverse sources into centralized decision making (99, 100). Broadly speaking, these examples and the integration seen in network physiology suggest that many of the local decisions made within an organism are profoundly influenced by more distal processes and information, and that there is tight coordination of the ensemble. This points toward global, organism-level integration and suggests the possibility of measuring the integrity of these systems.

INSIGHTS AND QUESTIONS EMERGING FROM THIS FRAMEWORK

The intrinsic health framework fits well with the recent WHO model of healthy aging, which itself is consistent with the International Classification of Functioning, Disability and Health (101). In this model, healthy aging is considered from the perspective of people’s ability to be and do the things they value, which is seen as enabling their well-being (understood in the broadest sense as including domains such as happiness, satisfaction, and fulfillment). WHO understands this ability as emerging from the interaction between an individual and their environment and describes the individual level attributes that contribute to this ability as intrinsic capacity. It has been suggested that intrinsic capacity itself comprises expressed capacities such as locomotor and cognitive capacity, as well as “vitality,” an underlying domain related to the physiological status of an individual and the capacity for maintaining homeostasis. However, there has been extensive debate on exactly how vitality might be conceptualized and measured (102, 103). In this context, the construct of intrinsic health, with its focus on attributes like resilience, robustness, and plasticity, can be considered a valuable alternative framing of this underlying biological state and the biological foundation underpinning the ensemble of these higher-order constructs.

Intrinsic health could also be understood as explaining reserve—functional, physiological, and even cognitive—in the sense that it explicitly evokes prospective capacity (sustainability) (76, 104). One challenge with the notion of reserve has been that there is not necessarily a clear resource or quantity that is stored for future use; intrinsic health suggests that the reserve need not be a specific resource but can be construed more broadly as the sustainability of the system due to its redundancy and configurational integrity.

The framework we have presented can be used to understand a number of scenarios for which we have clear intuitions regarding realized and intrinsic health (Fig. 5). Figure 5A shows how, over longer timescales such as the aging process, intrinsic health declines. However, depending on the lens of context, realized health could remain stable or even improve with age: Declining intrinsic health might still suffice for the individual to achieve their objectives at that stage of life. We thus propose that intrinsic health shows general decline with age (not necessarily linear) but that the trajectory of realized health with age is not universal. Figure 5B shows how an acute challenge such as a respiratory tract infection or a skin wound could represent a short-term decrement to realized health, whereas intrinsic health would remain unchanged (supporting the healing process), because an immune response and wound healing are basic organismal functions that the individual is performing adequately. However, in Fig. 5C, we illustrate how a partial recovery from an infection or wound might nonetheless entrain long-term consequences for intrinsic health as well. Figure 5D shows how early-life trauma and adversity might have long-term consequences that decrease adult levels of intrinsic health through biological embedding.

Fig. 5. The relationship between realized health, intrinsic health, and contextual factors over time under different circumstances.

(A) Aging. (B) Transient changes to realized health such as wounds or infections. (C) Transient changes to realized health without full recovery. (D) Early-life trauma that generates biological embedding.

This framework thus broadly incorporates and helps clarify many of the key notions surrounding health, as well as facilitates investigation of its biological underpinnings. An appropriate analogy is a tight-rope walker: Balance is never static, so to keep their balance, the tight-rope walker must make constant microadjustments in muscle tone across the body. Furthermore, all these adjustments must work in concert. If the tight-rope walker starts to lose their balance, then major adjustments are needed, but these are risky and may not fully restore balance, so it is much safer to make continuous microadjustments to avoid loss of balance. In physiology, such anticipation refers to homeostasis and allostasis, achieving stability through reactive and anticipatory recalibrations, respectively (105, 106). Coherent and adaptive recalibrations consume energy and require communication across the organism (107). While the individual microadjustments might be studied one at a time, their impact on balance can only be understood by studying them jointly. Last, control comes largely from the central nervous system (CNS); efforts to intervene in the microadjustments individually, without passing through the CNS, would likely do more harm than good. This is an unexpectedly profound metaphor for how maintaining health requires the constant flow of energy and integration of information through the organism’s microscopic and macroscopic structural components.

Emerging questions

Our intrinsic health framework gives rise to a series of new questions. First, is there an upper bound to intrinsic health, i.e., is intrinsic health asymptotic? We think it likely that the distribution is right skewed, with a small number of individuals getting close to, but never achieving, a theoretical maximum and a larger number of individuals suffering from diminished intrinsic health to varying degrees.

Second, how does intrinsic health change across the life course? How do developmental and aging processes intersect? Energetic, communication, and structural regulatory systems that are crucial for intrinsic health in adults can become compromised during aging, whereas they generally function efficiently in children and young adults. Does intrinsic health during development change with age, or is it relative to the ideal potential state at that age?

Third, how different is intrinsic health across individuals? Is optimal intrinsic health biologically identical for all people, or do different individuals achieve health through diverse strategies? Some biological systems converge through “equifinality” on a similar end point from varied starting points or using different strategies (91)—canalization in development is a good example (108). For example, different individuals might use slightly different immune (109) and metabolic strategies (110) to ensure health. Other aspects of intrinsic health may be highly normative: Variation from a certain ideal profile may be uniformly detrimental, such as tight regulation of circulating sodium levels (111).

IMPLICATIONS AND FUTURE DIRECTIONS

Here, we have proposed a framework to understand the biological basis of health, which we term intrinsic health. This framework is a first-principles paradigm to understand life itself: the energy and communication required within a given structure to allow it to maintain a dynamic equilibrium and resist the tendency toward entropy, allowing it to thrive under various challenging conditions.

Practically, this perspective on health implies that we should be cautious in thinking that we can outsmart the “wisdom of the body” (112), particularly in the absence of long-term data on global outcomes. Short- and long-term functioning are expected to trade off in some cases, as are different aspects of health. For example, sickness behavior is triggered by cytokines and has evolved to conserve energy and redirect resources during infections, ensuring that the immune system has the required resources (113). While organisms may not always deploy it optimally, intervening too strongly in the process might be detrimental. The role of public health and medicine should thus be seen as a supporting one under this framework: How can we help the body do what it does best? How can we provide the conditions for it to succeed, intervene to stabilize when needed, and know when to leave the intrinsic regulatory mechanisms of the organism to do their work without interference? What kind of therapeutic interventions are most likely to support the organism in transitioning from one state to another, toward healing? Advancing health promotion and prevention will require partnership between public health and medicine, and the intrinsic health framework is a platform to enable such integration.

The framework we propose sheds light on major outstanding questions in the biology of aging, namely, around its evolution, entropy, and heterogeneity (21, 114–117). Developing metrics of intrinsic health would also represent a major advance for several health science fields. Quantitative measures of intrinsic health will help solve a major challenge in epidemiology: that many outcomes are correlated but not identical, and that there is no clear, definitive way to evaluate interventions and risks that may have impacts on multiple diseases or outcomes. Objective measures of intrinsic health could become a definitive standard outcome for trials in many contexts, particularly for broad-spectrum health interventions such as diet, physical activity, and sleep, and for broad-spectrum risks. In addition, by its nature, intrinsic health integrates several intrinsic and extrinsic determinants of health and is prospective, and thus is sensitive to changes far upstream in the causal pathways leading to disease, disability, and other aspects of poor health. This could make it a promising surrogate end point for some pharmaceutical trials, for interventions to reduce the impact of aging biology (118, 119), and for interventions early in life. Even if our unifying hypothesis is not supported and intrinsic health is found to be multidimensional (many fields, instead of a single field), mapping these dimensions will provide profound insights into proximal constructs of health and how life maintains itself, leading to both relevant metrics of the various dimensions and to etiological and intervention studies to improve health.

For these reasons, a science of health and the concept of intrinsic health provide a scientific foundation to accelerate the ongoing transition to lifestyle medicine (120), systems medicine (121), functional medicine (122), and “P4” (predictive, preventive, personalized, and participatory) medicine (123). Thinking in terms of intrinsic health will help us ask better scientific questions and choose the most effective interventions to optimize health throughout the life span. Metrics developed from this framework will provide the next generation of benchmarks for personalized medicine, public health, and research.