Normal Is Not Always the Whole Story: What Prefrailty Teaches Us About Biomarker Patterns

Most laboratory reports are designed around a simple question: is the result inside or outside the reference interval?

That distinction is clinically useful. A markedly abnormal glucose level, inflammatory marker, liver enzyme, or renal marker deserves attention. But biology does not suddenly change at the exact boundary printed beside a laboratory result. Physiological function exists on a continuum.

A new study in GeroScience provides an unusually clear example of why this matters.

Stürmer and colleagues examined 117,163 middle-aged adults from the UK Biobank who were free of known chronic disease. Their question was straightforward:

can frailty-related physiological differences already be detected before people develop recognized chronic disease?

The answer was yes.

About 32.8% were classified as prefrail and 1.1% as frail, despite the cohort being deliberately selected to exclude known chronic conditions. More importantly, prefrail and frail participants differed from robust participants across numerous physiological systems—not through one distinctive biomarker, but through a broad pattern involving body composition, lung function, glucose regulation, inflammation, liver-related markers, renal and muscle-related measures, endocrine signaling, and white blood cells.

Perhaps the most important observation was also the easiest to overlook: most average biomarker values remained within conventional clinical ranges.

That has major implications for how we think about early physiological vulnerability.

A reference interval is not a map of physiological function

Reference intervals serve an important purpose. They help identify values that are unusual relative to a defined reference population.

But “within range” does not necessarily mean that every position inside that range carries identical physiological meaning.

The Stürmer study analyzed many biomarkers as continuous variables, examining how the odds of prefrailty and frailty changed with a one-standard-deviation difference in each measure.

The results showed consistent directionality.

Higher body fat, waist-to-hip ratio, hsCRP, cystatin C, leukocyte count, neutrophil count, alkaline phosphatase and triglycerides were associated with greater odds of frailty. In contrast, higher creatinine-to-cystatin C ratio, lung function, IGF-1 and bilirubin were associated with lower odds.

The message is not that a high-normal or low-normal result should automatically be considered abnormal.

It is subtler—and more useful.

Position within a biological range may contain information.

A value may still satisfy a conventional laboratory definition of “normal” while sitting in a direction that, together with other measures, reflects a different physiological state.

That is very different from creating arbitrary “optimal ranges.”

The more scientifically defensible question is:

Does relative biomarker position carry information about physiological function when interpreted in biological context?

This study suggests that sometimes it does.

Position becomes more meaningful when direction is considered

A biomarker number by itself rarely tells the whole story.

Consider hsCRP. A somewhat higher value may reflect infection, adiposity, tissue stress, immune activity, recent exercise, or many other processes.

Creatinine presents the opposite problem. Lower creatinine can appear reassuring if it is considered only as a renal marker. But in the context of frailty, lower creatinine may partly reflect lower muscle mass. That is why the study found the creatinine-to-cystatin C ratio to be particularly informative: a higher ratio was strongly associated with lower odds of frailty.

So interpretation requires more than asking whether a result is “high” or “low.”

It requires asking:

In which biological direction is this marker moving, and what might that direction mean in this physiological context?

The same numerical direction can carry different meanings for different biomarkers.

For inflammatory markers, higher values may suggest greater stress-response activity.

For muscle-related or anabolic markers, lower values may suggest reduced structural or maintenance reserve.

For metabolic markers, direction may indicate altered substrate handling.

For body-composition measures, greater fat storage can coexist with declining muscle or functional reserve.

This is why directionality must be biologically defined rather than numerically assumed.

From individual markers to physiological domains

The Stürmer study becomes even more interesting when the markers are viewed as domains rather than as isolated laboratory tests.

The frailty-associated pattern extended across several systems:

body composition → metabolic regulation → inflammation and immunity → respiratory function → renal/muscle-related physiology → endocrine regulation → hepatic physiology.

No single one of these domains defined frailty.

But together they described a physiological configuration.

This matters because conventional medicine often divides physiology by organ:

liver tests belong to the liver, glucose belongs to diabetes risk, creatinine belongs to the kidneys, CRP belongs to inflammation.

That organization is necessary for diagnosing disease.

But physiological adaptation does not respect those boundaries so neatly.

Stress, immune activation, nutrient allocation, metabolism, tissue repair and recovery involve multiple systems simultaneously.

A small shift in one marker may mean very little.

Several related shifts within a domain begin to mean more.

And coordinated shifts across multiple domains may begin to describe an underlying physiological state.

Pattern matters more than any single biomarker

This is probably the most important lesson from the study.

There was no “frailty biomarker.”

Instead, robust, prefrail and frail participants tended to occupy progressively different positions across a collection of physiological variables.

For many measures, the pattern looked approximately like:

robust → intermediate prefrail state → more pronounced frail state.

Prefrail individuals commonly showed values between those of robust and frail individuals.

This suggests that early functional vulnerability may be better recognized through pattern recognition than by waiting for a single biomarker to cross an abnormal threshold.

Think of the difference between these two interpretations.

The first asks:

Is hsCRP abnormal?

The second asks:

Is inflammatory activity trending upward while glucose regulation worsens, muscle-related reserve declines, body composition shifts, recovery deteriorates and symptoms increase?

The second question is much closer to how biological systems actually operate.

The meaningful signal may reside in the configuration, not in any single number.

A useful hierarchy: position, direction, domain, pattern

This study fits well with a framework we have been developing in our own research.

Instead of treating laboratory data simply as normal or abnormal, interpretation can proceed through four layers.

Position asks where the value sits within its available reference or physiological framework.

Direction asks whether movement toward the higher or lower end has biological relevance for that particular marker.

Domain asks whether several biologically related variables tell a consistent story.

Pattern asks whether those domains converge into a broader physiological configuration associated with symptoms, body composition, function or recovery.

Importantly:

position is not pathology.

direction is not diagnosis.

a pattern is not proof of causation.

They are ways of preserving biological information that a binary normal/abnormal classification can otherwise discard.

The Stürmer study is especially useful because it provides independent population-level evidence for this general methodological principle. The authors did not use our positioning system, nor did they study Exposure-Related Malnutrition. But they showed that continuous biomarker differences across multiple systems were associated with functional vulnerability even when conventional laboratory values were often still clinically normal.

Prefrailty may represent compensated vulnerability

There is another intriguing interpretation.

Frailty is usually associated with older age and established disease. Yet this study deliberately examined middle-aged adults without known chronic disease.

Still, nearly one-third were prefrail.

That raises the possibility that prefrailty is not simply “mild frailty.” It may sometimes represent a compensated state in which conventional homeostasis remains largely intact but maintaining that stability is becoming more costly.

The organism may still preserve blood chemistry within acceptable limits while sacrificing functional reserve elsewhere.

Muscle maintenance may decline.

Inflammatory demand may rise.

Metabolic regulation may become less efficient.

Respiratory reserve may fall.

Anabolic signaling may become less permissive.

The laboratory report may still look reassuring when each value is read separately.

The physiological pattern may be telling a different story.

This interpretation is consistent with—but does not prove—the concept of allostatic triage: when biological demands persist, resources may be preferentially directed toward immediate survival and stress-response functions while maintenance, repair and long-term reserve receive less investment.

Within our Exposure-Related Malnutrition framework, this is what we mean by functional malnourishment. It is not necessarily starvation or inadequate calorie intake. A person may have adequate calories, or even excess adiposity, while still having inadequate usable resources relative to the demands of adaptation, maintenance and recovery.

The Stürmer study did not test this mechanism directly. But its observation of early multisystem vulnerability before recognized chronic disease is highly compatible with such a model.

The next frontier is not another biomarker

The logical next step is not to search for one magical test for frailty, aging or physiological resilience.

It is to ask better questions of the data we already have.

Does biomarker position contain information before conventional thresholds are crossed?

Do several markers move in biologically consistent directions?

Do those markers form recognizable physiological domains?

Do those domains converge into reproducible patterns?

And most importantly, do those patterns correspond to function—fatigue, strength, cognition, body composition, exercise tolerance, symptom burden and the ability to recover after stress?

That last question may ultimately matter most.

Health is not simply the ability to maintain a laboratory value at rest.

It is the ability to respond, adapt and then recover.

The Stürmer study suggests that declining functional reserve may already leave a detectable physiological footprint long before conventional disease becomes obvious.

The challenge is learning how to read that footprint without turning every normal-range variation into disease.

That requires moving beyond a single biomarker—and toward position, direction, domain and pattern.

Reference

Stürmer, P., Silva, G. C., Fayosse, A., Pichon, R., Sabia, S., Lieb, W., & Landré, B. (2026). Frailty before chronic disease: multisystem physiological differences in middle-aged adults from the UK Biobank. GeroScience. https://doi.org/10.1007/s11357-026-02450-1