Bloodwork
Your driving licence says one number. Your bloodwork often says another.
Phenotypic age is an attempt to measure the second one. It takes nine markers from a standard blood panel and produces a single figure: the age at which your results would be unremarkable. If you are 38 and your phenotypic age comes back at 31, your bloodwork looks like that of a typical 31-year-old. If it comes back at 46, it does not.
The gap between the two numbers is the part that matters.
The algorithm was published in 2018 by Morgan Levine and colleagues at UCLA. They were not trying to build a wellness gimmick. They were looking for a better way to predict mortality and disease risk than chronological age alone, and they built it from NHANES data: roughly ten thousand US adults with around two decades of follow-up, then validated on thousands more.
What they found was that a specific combination of nine routine markers predicted who died and who did not, considerably better than age by itself. Not perfectly. Better.
That is the entire claim, and it is worth being precise about it. PhenoAge is a risk estimate built from mortality data. It is not a measurement of how many years you have left, and it is not a diagnosis.
Every one of these appears on a standard panel. None of them is exotic, and none requires a specialist lab.
Chronological age goes in alongside them. The algorithm weights each marker, combines them, and converts the result back into a number expressed in years.
Most people are surprised that testosterone, cholesterol and thyroid markers are absent. They were not excluded because they do not matter. They were excluded because, in this dataset and with this statistical approach, they did not add predictive power once the other nine were accounted for. A marker only earns a place in a model like this if it tells you something the others do not.
That is a useful thing to understand about any risk score. It is not a list of everything important. It is the shortest list that still works.
You get two numbers: your chronological age and your phenotypic age. The gap between them is what you act on.
Phenotypic age below chronological age. Your markers look younger than your years. Good, and worth confirming on a second panel before believing it.
Roughly equal. Unremarkable, which in this context is a reasonable place to be.
Phenotypic age above chronological age. Something in the panel is pulling the number up. The useful next step is not to worry about the composite score but to find out which marker is responsible.
That last point is where most people go wrong. The single number is a headline. The individual markers are the story.
You can run your own numbers through the biological age calculator, which uses the published PhenoAge algorithm and shows which markers are driving your result rather than just handing you a figure. If you have a fuller report in hand, the bloodwork analyzer reads the rest of the panel in context, including the markers PhenoAge leaves out.
Several of the nine respond to things you control, and some respond faster than people expect.
Inflammation. C-reactive protein carries real weight in the algorithm, and it moves. Poor sleep, excess body fat, untreated infection, heavy alcohol intake and chronic stress all raise it. So does a hard training block, temporarily, which is one reason a single reading taken during a heavy week can mislead. Chronically raised CRP is also a cardiovascular risk signal, which the free guide The Armoured Heart covers alongside blood pressure, ApoB and long-term heart monitoring.
Glucose. Fasting glucose responds to bodyweight, carbohydrate intake, activity and sleep quality. A person who drops fifteen kilos and starts walking usually sees this marker move within months. If glucose is the marker dragging your result up, The Insulin Resistance Reversal System goes through the bloodwork, severity grading and the step-by-step fix.
Red cell distribution width. Often reflects iron, B12 or folate status. Correctable in many cases, and worth investigating rather than accepting.
Albumin and liver enzymes. Alcohol intake shows up here reliably. So do some medications.
What does not move it: supplements marketed at longevity, most of which have no effect on any of these nine markers and were never tested against them.
Three limits are worth stating plainly, because almost nobody selling a biological age test mentions them.
It is a population average, not a personal prediction. The algorithm was built from data on thousands of people. Applied to you individually, it carries far more uncertainty than the clean single number suggests.
A single panel is a snapshot. An acute infection, a recent hard training block, dehydration or a poor night's sleep can all move markers enough to change the result. A trend across three panels over a year tells you something. One panel tells you very little.
It says nothing about cause. A phenotypic age five years above your chronological age tells you the pattern is unfavourable. It does not tell you whether that is inflammation, kidney function, a nutritional deficiency or something else entirely. That is what reading the individual markers is for.
You will see three broad categories advertised, and they are not measuring the same thing.
Questionnaire-based calculators. Ask about sleep, smoking, exercise and diet, then produce a number. These measure your habits, not your biology. Free, instant, and largely a restatement of what you already told them.
Fitness-test calculators. Push-ups, grip strength, the sit-to-rise test, balance. These measure physical function, which genuinely correlates with longevity. Useful, and completely independent of your bloodwork.
Blood-based algorithms like PhenoAge. Measure what is actually happening in your blood. Requires a panel, which is the friction, and is also the reason the result carries more weight.
Epigenetic clocks are a fourth category, measuring DNA methylation. More expensive, arguably more accurate, and not something you get from a routine panel.
None of these replaces the others. They answer different questions.
You need all nine markers, and most standard packages include seven or eight of them. The one most often missing is C-reactive protein, which usually needs requesting by name.
In an Indian lab, ask for a complete blood count (CBC) with differential, a liver function test (LFT), a kidney function test (KFT), fasting glucose and hs-CRP. The CBC covers white cell count, lymphocyte percentage, mean cell volume and red cell distribution width. The LFT covers albumin and alkaline phosphatase. The KFT covers creatinine. Ordered individually, that usually costs less than a packaged full-body checkup.
Check the unit your report uses for CRP. Some labs report mg/L and others mg/dL, and entering one as the other shifts the result by years.
If you are unsure what to request, the blood test selector builds the list based on what you are trying to find out, and flags the markers that standard packages leave out.
Draw it fasted, in the morning, and use the same lab each time you retest. Inter-laboratory variation is larger than most of the real changes you are looking for.
Phenotypic age is a reasonable, published, evidence-backed way to compress nine blood markers into one number. It predicts mortality better than age alone. That is a real finding and it is not controversial.
What it is not is an oracle. The number is less interesting than the markers underneath it, and anyone treating a single result as a verdict on their health has misunderstood what the model does.
Run it. Look at which markers are pulling the result. Fix those. Retest in six months.
Lower than your chronological age is favourable, but the gap matters more than the number. A 45-year-old with a phenotypic age of 40 is in a better position than a 30-year-old with a phenotypic age of 34, even though the second number is lower.
Every six to twelve months. More frequently than that and you are mostly measuring noise, because normal biological variation between panels is larger than most real changes.
Yes. Several of the nine markers respond to bodyweight, sleep, alcohol intake and activity. Inflammation and fasting glucose in particular can move substantially within a few months.
No. Epigenetic clocks measure DNA methylation patterns and require a specialised test. PhenoAge uses nine markers from a routine blood panel. They correlate, but they are measuring different things.
No. All nine markers appear on standard panels. A CBC with differential, LFT, KFT, fasting glucose and hs-CRP will cover them in most labs.
Because most biological age calculators are questionnaires about lifestyle rather than algorithms applied to bloodwork. If one asks how many hours you sleep and the other asks for your albumin level, they are not measuring the same thing and there is no reason for the results to agree.
Levine ME, Lu AT, Quach A, et al. An epigenetic biomarker of aging for lifespan and healthspan. Aging (Albany NY). 2018;10(4):573–591. Read the paper
This article is educational and is not medical advice. See the medical disclaimer.
Want a programme built around your own numbers?
Work With Dinesh
Newsletter
One email when a new guide, calculator or breakdown goes live. Nothing else.
