Biological age vs chronological age: what changes it and what does not

Biological age vs chronological age: the second is set by your birth certificate and advances one year every twelve months for everyone; the first is an estimate of how your systems are working, summarized in years so it can be understood without being a doctor. Apps and laboratories increasingly calculate it from a routine blood test with the PhenoAge formula, which combines nine markers with age. In what biological age is and how it is calculated we explain where the number comes from. This text answers the next question: what changes biological age for real in three to six months, which of those nine markers change slowly or do not depend on your habits, why a single test during a cold “ages” you several years when nothing has happened, and what the difference is between this blood clock and the epigenetic DNA clocks you see advertised.

What separates biological age from chronological age

The underlying idea has solid data behind it. The Dunedin Study has followed 1,037 people born in New Zealand in 1972 and 1973 since birth; in 954 of them, 18 biomarkers were measured at ages 26, 32 and 38. At a chronological age of 38, estimated biological age ranged from 28 to 61 years, and the pace of aging varied from almost zero years of physiological change per calendar year to close to three.^[4] Those aging faster performed worse on tests of physical function, declined on cognitive tests relative to their earlier level and, in the eyes of independent raters, had faces that looked older.^[4] People with the same date of birth carry bodies decades apart.

PhenoAge, the formula most blood-based tools use, was born in the US NHANES III survey. A statistical survival model (Cox with elastic net) selected the nine laboratory markers that best predicted mortality together with chronological age: albumin, creatinine, glucose, C-reactive protein on a log scale, lymphocyte percentage, mean corpuscular volume, red cell distribution width, alkaline phosphatase and white blood cell count.^[2] It was then validated in 11,432 adults aged 20 to 84 from NHANES IV, with 1,012 deaths over 12.6 years of follow-up: adjusting for chronological age, each year of phenotypic age above it was associated with a 9 % higher risk of dying from any cause, and the association held (8 % per year) in people with no diagnosed disease and normal weight.^[2] Participants with more diseases had an older phenotypic age, but the effect was modest: in young adults, one disease added about 0.2 years and two or three, about 0.6.^[2]

The same formula later served as the basis for a DNA clock: Morgan Levine’s team trained, in two steps, a DNA methylation marker on that phenotypic age, and the result, DNAm PhenoAge, was associated with mortality, cancers, disease-free years of life, physical functioning and Alzheimer’s disease.^[1] The blood version is not the “cheap” version of the DNA one; it is the one that defines it.

The nine PhenoAge markers: what moves each one and how fast

The table gathers what each marker reflects, what pushes it out of its usual range and how quickly it moves. The last column is practical guidance, not a guideline timeframe: it helps decide when it makes sense to repeat the test, not to promise results.

Grouped by pace, the rows become a strategy. The fast ones: CRP and white blood cells (with lymphocyte percentage) follow inflammation almost in real time, and glucose responds to what you ate and how you slept.^[7][9] The slow ones: MCV, RDW, albumin, creatinine and alkaline phosphatase depend on red blood cell turnover, muscle mass, liver and kidney, and do not notice a good week or a bad one.^[8][10] That is why a biological age that jumps between two tests is almost always explained by the fast markers, and one that rises slowly over years, by the slow ones.

What really moves it in three to six months

The question everyone asks is whether they can lower their biological age. The honest answer has two parts. First: yes, to the extent that you bring the fast markers closer to your healthy zone, and that is easier than it seems when you start from a test taken under poor conditions. Second: the slow ones will barely move in that time, and a plan that promises to “take off ten years in ninety days” is counting on the noise of the fast ones, not on real change.

Inflammation: CRP and white blood cells

CRP is the marker with the most leverage and, at the same time, the most treacherous. MedlinePlus considers a value of 0.8 to 1.0 mg/dL or lower healthy (8 to 10 mg/L; many laboratories in Mexico report it in mg/L) and reminds you that the result tells you how much inflammation there is, not where or why.^[7] What raises it in a sustained way and does depend partly on you: smoking, polluted air and obesity.^[7] What raises it temporarily and does not depend on you: any infection, a flare of an autoimmune or inflammatory disease, an injury that is healing.^[7] White blood cells follow along: they rise with infection, inflammation, allergy, tissue damage, smoking, stress and pregnancy, and as a reaction to medicines.^[9] A three-to-six-month plan that cuts smoking, improves sleep and brings the waist down has a real effect on these markers; seeing it requires repeating the test with no recent infection.

Metabolism: glucose, waist and what is not in the formula

Fasting glucose is the metabolic marker of the formula and the one that moves most with habits over weeks. But it is a single number from one morning, and it also moves with an incomplete fast, the time of the draw, stress and sleep; we wrote about that in high fasting glucose and in HOMA-IR. What is not in PhenoAge, and should be in your log, is the context: waist, weight, blood pressure, triglycerides and HDL. The formula includes glucose because in NHANES it predicted mortality together with the other markers,^[2] but your doctor needs the full picture to decide whether that number is noise or pattern.

Liver, kidney, muscle and hydration: albumin, creatinine and alkaline phosphatase

These three move slowly, with one exception that confuses many people: water. Albumin rises with dehydration and falls after 24 to 48 hours without eating, in pregnancy and with birth control pills; steroids, insulin and hormones can raise it.^[8] Creatinine depends on how much muscle you have and how your kidneys filter, and a dehydrated morning or hard training the day before moves it without the kidney having changed; the details are in high creatinine. Gaining muscle over six months can raise creatinine without that being bad news: the number is interpreted, not chased.

Red blood cells: MCV and RDW

MCV rises with a lack of vitamin B12 or folic acid and with liver disease, and falls with iron deficiency and thalassemia.^[10] RDW grows while a deficiency appears or is corrected. Neither responds to a week of changes, because the body renews its red blood cells over months; they are the markers that say the most about your underlying trend and the least about your last two weeks. In complete blood count we explain how to read each line.

Why a single test with a cold “ages” you years, and why it is not a verdict

A typical case. You had your blood count and chemistry panel done the week you were getting over a cold. CRP is high because there was an infection, white blood cells too, and lymphocyte percentage moved with them; you also drank little water and albumin and creatinine came out more concentrated.^[7][8][9] None of those markers speaks of your aging: they say you were sick and dehydrated. But the formula does not know that. It takes the nine values as they are and returns a phenotypic age several years above what the same blood would give three weeks later.

That is not a flaw of PhenoAge; it is the consequence of its having been designed and validated in populations, where individual noise averages out, and not for one person on one particular morning.^[2] The reverse holds too: a “young” result on a calm day is not a certificate either. What counts is the direction across several tests under similar conditions: same laboratory, fasting, no infection in the previous two weeks, no intense training the day before and well hydrated. On how long the body takes to return to its level after a cold, a trip or a race, see how many days the body takes to recover.

What does not move it (or not the way you think)

Longevity advertising promises to move biological age with things that, looking at the formula, have no way to do so. The second table sorts out the most frequent myths.

Blood clock or DNA clock: which one serves you at home

There are two families of clocks. Clinical ones, like PhenoAge, use routine laboratory markers. Epigenetic ones measure DNA methylation at hundreds of points in the genome. The first and best known, published by Steve Horvath in 2013, uses 353 CpG sites, was trained on 7,844 samples from 51 tissues and cell types, and in test data reached a correlation of 0.96 with chronological age with a median error of 3.6 years.^[3] Read that twice: the reference epigenetic clock departs from the calendar, in half of all cases, by more than three and a half years. A five-year gap between your biological and chronological age falls close to that margin.

The second generation changed the question. As its authors explain, the clocks estimate “the progress of aging up to a cross-section in time,” whereas DunedinPACE, published in 2022, estimates the pace: how much change you accumulate per year. It was built by following 19 organ-system biomarkers for 20 years, from age 26 to 45, in the same Dunedin Study, and compressing that pace into a single-timepoint blood methylation test with 173 CpG sites and excellent test-retest reliability (ICC 0.96).^[5]

So which one serves you? A 2017 review in EBioMedicine sorts predictors into six families (epigenetic clocks, telomere length, transcriptomic, proteomic, metabolomic and composite clinical biomarkers) and concludes that “the various biological age predictors tend to reflect different aspects of the aging process,” that one type correlates weakly with the others and that no marker or combination yet meets the criteria for a true biomarker of aging.^[6] The authors consider the epigenetic clock the best predictor available for its correlation with age and with mortality, but note that its true value requires longitudinal confirmation.^[6] For following yourself at home the conclusion is simple: the blood clock uses tests you already have, can be repeated every few months and tells you which marker moved; the DNA one is done once, costs more and does not tell you what to change. To see your trend, blood is enough.

What to do with the result and when to repeat it

• If it is your first calculation: save it with its date and conditions (fasting, water, sleep, illness in the previous two weeks, training, medicines). Without that, the number cannot be interpreted later.
• If it came out several years above and there was a cold, stress or dehydration: do not conclude anything. Repeat a complete blood count, chemistry panel and CRP in 4 to 8 weeks, once recovered, at the same laboratory.^[7][9]
• If it came out above with nothing to explain it: look at which marker is out of range. High CRP without infection, glucose in the prediabetes zone, high creatinine or low albumin are each a concrete question for your doctor, not an “age.” In a reference range is not your personal norm we explain why your own baseline matters.
• If it came out below: that is not a certificate either. The formula looks at nine things and there is much it does not look at; it was built to predict risk in groups.^[2]
• When to repeat: the guidelines give no frequency. The reasonable approach for a trend is every 6 to 12 months with your check-up, under the same conditions; sooner only if the first calculation was taken under poor conditions.
• What to change: what moves the fast markers (smoking, weight, sleep, activity) and what corrects a confirmed deficiency; how much and how is a decision with your doctor, with your data in front of you.^[7][10]

Warning signs

An abrupt change in the markers of the formula is a reason to see your doctor, not to look for how to “get younger.” These combinations do not wait for the next check-up:

• High CRP that does not come down weeks after an infection, or high with no known infection: it may reflect autoimmune or inflammatory diseases that need to be studied.^[7] See your doctor in the coming weeks.
• Very high or very low white blood cells with no infection to explain them: a low count may point to autoimmune or bone marrow disorders, and a high one to infection, inflammation or a reaction to a medicine.^[9] See your doctor in the coming days.
• Creatinine that rises between two tests or albumin that falls without a long fast or pregnancy: they may reflect kidney or liver, and are checked soon, not in a year.^[8]
• Symptoms, not markers: unintended loss of weight and appetite, swelling in the legs or eyelids, dark-colored or foamy urine, yellow skin or eyes, persistent itching, nausea or vomiting are signs of liver or kidney for which albumin is ordered; fever, chills and body aches, those of infection for which white blood cells are ordered.^[8][9] They come before any calculation. With a high fever that will not break, confusion, trouble breathing or severe abdominal pain, go to the emergency room.

What to bring to the doctor

The biological age figure helps little in the appointment; the specific marker that moved it helps a lot. Arrive with the marker, not with the age.

1. The complete tests with date and laboratory: complete blood count with differential, chemistry panel with albumin, creatinine, glucose and alkaline phosphatase, and CRP. If you have more than one, bring them all: the direction says more than the value.^[2]
2. The conditions of each draw: fasting, water, recent infection, sleep, training, stress.^[7][8][9]
3. The marker or markers out of range, flagged, with the laboratory range next to them.
4. Recent weight, waist and blood pressure: the metabolic context the formula does not include.
5. Medicines and supplements, including steroids, birth control pills, hormone therapy and insulin, which move albumin, CRP or white blood cells.^[7][8][9]
6. History: anemia or iron, B12 or folate deficiency, liver, kidney or thyroid disease, autoimmune disease.^[8][10]
7. Three questions: whether any marker deserves repeating or expanding (ferritin or B12 if MCV is out, HbA1c if glucose is high), when to repeat the panel and whether any medicine explains a value.