What is the New Definition of Health?

Dr. Bomi Joseph defines health as the body’s measurable ability to detect, respond, and recover rapidly from stressors—biological, chemical, or physical.

Deep Health® · Updated after publication of the foundational study

What Is Health? From Biological Phenomenon to the Deep Health® Unit

Medicine measures thousands of variables about the human body. The harder question is more fundamental: what biological quantity do we mean when we say “health,” and what unit should express it?

UPDATED OCTOBER 2026

What has changed since I first wrote this article? The central idea has not changed. What has changed is the evidence base. The foundational human study has now been published, allowing the biological reference, the measurement hierarchy, and the distinction between a direct reference result and a Device-derived estimate to be stated much more precisely.

First: define the quantity

A unit is meaningless until the thing being measured has been defined. Length is a quantity. The inch is a unit. A ruler is an instrument. These are three different things.

Health is the body’s ability to detect, respond, and recover rapidly from stressors—biological, chemical, or physical.

Figure 1. The scientific distinction is quantity → unit → instrument. For health, the biological quantity is adaptive health, the unit is the Deep Health Unit (DHU), and the practical measuring instrument is the Deep Health® Device.

This is why biomarkers, diagnostic tests, blood pressure, glucose, VO₂ max, body composition, “biological age,” and the absence of diagnosed disease cannot themselves be definitions of health. They may describe important parts of physiology, but they are not the underlying quantity.

Figure 2. Health is treated here as a dynamic biological capacity, not as a static list of normal values.

The published biological foundation

The original question was not “Which biomarker is best?”

The original work asked something more direct: when the organism is challenged by a foreign biological stimulus, how much challenge is required before a meaningful response appears, and how long does that response take?

Rabbit serum albumin (RSA) was used as a benign foreign protein. Serial anti-RSA IgG measurements made the adaptive response visible over time. The important observation was not simply that antibody concentrations differed. The amount of challenge required and the time required to respond differed by orders of magnitude between health groups.

Figure 3. Illustrative representation of the original RSA challenge observations. The key biological variables are challenge burden and response time.

The paper now adds a second biological dimension: the immune-response phenotype was examined alongside ³¹P magnetic-resonance spectroscopy measures of cellular energy metabolism. This matters because the proposed quantity is not merely “antibody response.” It is a broader adaptive biological efficiency.

Published cohort	n	Resting metabolic index	τPCr recovery
Healthy	1,323	204.61 ± 21.87	16.21 ± 1.42 s
Average	1,148	43.11 ± 5.04	40.07 ± 4.20 s
Poor	1,365	11.19 ± 2.40	63.10 ± 7.57 s

Across the 3,836-person analyzed sample, the groups separated not only in immune-response behavior but also in resting energetic state and phosphocreatine recovery. The healthier group showed substantially greater resting metabolic capacity and much faster recovery; the poor-health group showed the opposite pattern.

One of the more interesting findings was the relationship between resting and maximal metabolic indices: the Healthy and Average groups retained extremely strong coupling, while that coupling was essentially absent in the Poor cohort. That raises a biologically important possibility: declining health may involve not only lower capacity, but loss of coordination between basal energy turnover and stimulated metabolic response.

Adaptive Biological Efficiency

Health is not just response. It is response efficiency.

Two organisms can eventually reach a response and still be biologically very different. One may require very little provocation and respond rapidly. Another may require an enormous burden and many additional days.

AE = 1 / (S₅₀ × T₅₀)

In this conceptual expression, S₅₀ is the standardized challenge level associated with 50% of the defined response, and T₅₀ is the time required to reach it. Lower burden and shorter time produce greater Adaptive Efficiency.

Figure 4. Combining challenge burden and response time produces differences in Adaptive Biological Efficiency that span orders of magnitude.

The system, not one biomarker

Four linked components of adaptive response

The adaptive response can be organized into four functional stages. These are not four independent scores. A serious weakness in one stage can limit the performance of the entire system.

Recognize

How rapidly does the organism detect a disturbance and initiate a response?

Produce

How rapidly and adequately can an appropriate biological response be generated?

Deliver

How effectively can biological resources be transported to the tissue where they are needed?

Eliminate & Recover

Can the disturbance be overcome, and how rapidly can the organism return toward its prior functional state?

AE = R₁ × R₂ × R₃ × R₄

This expression is a systems model: recognition, response production, transport, and elimination/recovery all contribute to the observable adaptive phenotype. It also prevents a common mistake—equating one sensor or one biomarker with “health.”

Figure 5. The four-stage model. Device signals should be described as physiological correlates calibrated to these higher-order functions, not as direct measurements of antibody production or immune recognition.

The unit

The Deep Health Unit: a logarithmic expression of biological efficiency

Because Adaptive Biological Efficiency spans orders of magnitude, a linear scale is cumbersome. The DHU therefore uses a logarithmic ratio to a defined biological reference.

DHU = log₁₀(AE / AE_reference)

One Deep Health Unit (1 DHU) represents a tenfold difference in Adaptive Biological Efficiency relative to the frozen reference. Therefore +1 DHU = 10× reference, +2 DHU = 100×, −1 DHU = one-tenth, and a twofold difference is approximately 0.301 DHU.

Figure 6. The DHU is a logarithmic unit. Its meaning is fixed by the reference definition; the instrument or algorithm may improve without changing the unit itself.

DHU-R and DHU-D

Publication also makes it useful to distinguish the direct reference result from the practical Device estimate.

DHU-R

Reference-derived DHU. Adaptive Biological Efficiency obtained from the defined challenge-response reference framework.

DHU-D

Device-derived DHU. The non-invasive Deep Health® Device estimate of that same reference quantity, obtained from validated physiological measurements and calibration models.

This distinction is important because the algorithm is not allowed to become the definition of health. The biology defines the quantity. The Device estimates it.

Reference method versus practical method

Direct biological observation is powerful. It is not practical for routine use.

Direct reference pathway

Standardized biological challenge exposureApply the defined reference challenge.

Serial blood samplingFollow the anti-RSA IgG response over time.

Laboratory analysisQuantify the dynamic response trajectory.

Extended observationResolve challenge magnitude and response time.

Practical Device pathway

Standardized Deep Health® Device measurementAcquire multimodal physiological signals non-invasively.

Physiological feature extractionMeasure validated tissue, vascular, cardiac, neuromuscular, metabolic and related features.

Calibration modelEstimate Adaptive Biological Efficiency relative to the biological reference.

DHU-DExpress the Device-derived estimate in Deep Health Units.

Figure 7. Direct reference measurement provides greater biological directness; the Device provides rapid, repeatable, non-invasive population-scale estimation.

Measurement hierarchy

The biology comes first. The Device comes later.

The most important change in the updated framework is precision about what is directly measured, what is derived, and what is estimated.

HealthThe biological capacity to detect, respond, deliver, resolve and recover from stress.

Adaptive Biological EfficiencyThe quantitative construct describing how efficiently that adaptation occurs.

Biological referenceThe challenge-response framework that defines the target quantity and yields DHU-R.

Physiological measurementIndependent tissue, vascular, neuromuscular, mechanical, acoustic, optical, metabolic and bioelectrical features.

Deep Health® DeviceThe practical multimodal instrument used to acquire those measurements.

Calibration and estimationValidated models map Device measurements back to the reference quantity.

DHU-DThe Device-derived estimate expressed in the same fixed unit.

Figure 8. Updated hierarchy. The reference pathway defines and measures the target quantity; the Device pathway estimates that same quantity non-invasively. DHU-R and DHU-D should remain distinguishable.

Why the published paper matters

The DHU is not intended to be another wellness score

A wellness score can be redesigned whenever its creator chooses. A scientific unit cannot. If the Device hardware improves, the DHU cannot change. If a new sensor reduces error, the DHU cannot change. If a better model improves estimation, the DHU cannot change.

The ruler may improve. The inch must not change.

That is why the reference must be fixed, why the measurement hierarchy matters, and why a consumer display should remain downstream of the scientific unit. The DHU expresses Adaptive Biological Efficiency. It is not, by itself, a diagnosis, a disease-risk score, or a cosmetic 0–100 wellness number.

What the paper changed in this article

“Adaptive Efficiency” is now better described as Adaptive Biological Efficiency to make clear that the target is a whole biological response construct.
The direct reference and Device estimate are explicitly separated: DHU-R versus DHU-D.
The Device is no longer described as directly measuring immune functions. It measures physiological signals and validated correlates that are used to estimate the higher-order construct.
The RSA reference pathway is described as a standardized biological challenge exposure, not simply as an injection.
The 3,836-person foundational dataset adds cellular bioenergetic evidence to the adaptive-response observations, including large group differences in resting metabolic index and phosphocreatine recovery.
The unit is fixed to a biological reference. Technology may improve; the meaning of one DHU must not drift.

Scientific note: the foundational paper establishes the biological reference framework and associated adaptive/metabolic observations. Device-derived DHU-D should be presented as an estimate calibrated to that reference framework, not as identical to the direct reference measurement.

Ask DrBomi® — Health Intelligence Built on Deep Health®

The Deep Health® Device does more than collect physiological measurements. Since 2017, it has incorporated our own artificial intelligence engine to analyze the data generated by the Device, identify patterns, and provide new insights into human health.

That work ultimately led to Ask DrBomi®, the Health AI engine available on this website.

Ask DrBomi® is designed specifically for health. It brings together the principles of Deep Health®, decades of health research, and artificial intelligence to help you explore questions about your body, physiology, nutrition, exercise, recovery, longevity, and health.

The Deep Health Unit asks:

How healthy is the organism?

Ask DrBomi® lets you take the next step:

What can I do with that knowledge

Ask a Health Question

Have a question about your health, nutrition, exercise, recovery, or longevity?