Your Organs May Age at Different Speeds. Lifestyle Matters.

Your liver, heart-lung system, and metabolism may not be aging on the same schedule. A 2025 study found that lifestyle patterns tracked differently with biological-age acceleration across those systems—but it did not produce a personal instruction manual for fixing one organ at a time.

That distinction matters. The study offers a useful hypothesis: smoking, alcohol, diet, activity, and sedentary time may leave different signatures in different organ systems. But it was an observational study in Southwest China, and its organ-specific rankings need replication. The sensible conclusion is less glamorous than an organ-hacking checklist: improve the habits with the strongest overall evidence, then manage known disease risks with a clinician.

What the main study actually found

In a 2025 eLife study, Yuan Zhang and colleagues followed 8,396 adults aged 30 to 79 in Southwest China across an approximately two-year repeat-survey window from 2018 to 2021. The participants came from a 99,556-person baseline cohort. Researchers used routine clinical biomarkers and the Klemera-Doubal Method to estimate biological age for five systems: cardiopulmonary, metabolic, liver, renal, and immune.

They then used quantile G-computation, a method designed to assess correlated exposures together, to estimate the relative contributions of smoking, alcohol use, dietary quality, physical activity, and sedentary behavior. This is more informative than testing each habit in isolation, because people who eat well often exercise more, smoke less, and differ in many other ways.

Only three clocks—cardiopulmonary, metabolic, and liver—passed the study's disease-association validation. The renal and immune clocks did not, so they should not support claims about which habit matters most for kidney or immune aging. The framework also did not construct brain or musculoskeletal biological ages. It therefore cannot tell you which behavior best protects cognition, dementia risk, muscle, or bone.

The study's results support organ-specific differences in lifestyle attribution for the three validated clocks. They do not establish that changing one behavior causes a particular organ to become biologically younger. The publication's available quantitative details also do not permit a reliable reader-facing table of exact percentages or definitive organ-by-habit winners.

A biological-age clock is a risk marker, not an organ biopsy

Clinical-biomarker clocks summarize measurements that tend to shift with age. They can be useful, but they do not directly measure every cellular process occurring in an organ. A 2024 UK Biobank cohort study linked accelerated biological age with adverse cardiac structure and function and with incident heart failure, independent of conventional risk factors. That gives cardiopulmonary biological-age acceleration clinical relevance, but it does not prove that the clock itself causes heart failure or that lowering a clock score prevents it. Read the cardiac study.

Other cohort evidence puts the issue in a broader health context. In 332,012 UK Biobank participants followed from 2006 to 2022, faster aging measured by KDM biological age and PhenoAge was associated with more physical, psychological, and cognitive multimorbidity and shorter life expectancy. Participants in the highest aging-speed quartile had a 16% to 17% higher risk of developing one disorder and a 41% to 44% higher risk of multimorbidity. Those are associations, not proof that manipulating a biological-age score changes outcomes. See the UK Biobank multimorbidity analysis.

Diet has the most consistent cross-population signal

Dietary quality has the most consistent support across the evidence reviewed here. In a cross-sectional U.S. NHANES analysis spanning 1999 to 2018, higher scores on five diet-quality measures—AHEI, alternate Mediterranean diet, HEI-2020, DASH, and Dietary Inflammatory Index—were associated with lower biological age across clinical-biomarker and epigenetic measures. The study also examined organ-specific aging metrics and mortality risk. The NHANES analysis supports the direction of the China findings in a different population and with different measures.

Still, cross-sectional diet research has a familiar problem: healthy-user confounding. People with higher diet scores often differ in income, healthcare use, sleep, exercise, smoking, and many less measurable habits. Statistical adjustment helps, but it cannot fully isolate the effect of food from the rest of a health-conscious life.

A prospective UK Biobank study adds another piece of directional evidence. Greater adherence to the EAT-Lancet diet was associated with slower biological aging and longer life expectancy. The population, dietary assessment, and biomarker methods differed from the China study, so the studies cannot yield a shared organ-specific effect size. They do point in the same direction: diet quality tracks with slower biological aging. Read the UK Biobank EAT-Lancet study.

Why the organ rankings may change with the measuring tool

The China study used clinical laboratory markers. Proteomic clocks use many circulating proteins with tissue-specific expression patterns and may detect molecular changes before standard labs move outside normal ranges. A 2026 plasma-proteomics analysis reported dynamic protein patterns associated with biological aging, while a 2025 proteomic organ-aging study linked protein-based organ aging with cardiometabolic risk in people at risk for heart failure.

These findings reinforce the idea that organs age asynchronously. They also limit confidence in a universal ranking of habits by organ. A smoking signal may rank highly in a clinical-lab metabolic clock because smoking changes the biomarkers in that clock particularly strongly. A protein-based clock could rank exposures differently. No published study described here has compared lifestyle-attribution rankings from clinical-biomarker and proteomic organ clocks in the same people.

Population averages can miss the person in front of you

Biological aging varies within the same diagnosis. In people with type 2 diabetes, a 2024 GeroScience study found substantial heterogeneity in biological age that related to tissue-specific DNA methylation, clinical variables, comorbidities, and medication use. Someone taking drugs that alter glucose, lipids, inflammation, blood pressure, or kidney markers may not resemble the average participant in an organ-clock study.

That is especially relevant for diabetes. In 12,828 UK Biobank participants with diabetes, accelerated biological aging was associated with incident cardiovascular disease and life-expectancy loss over a median 13.1 years. The result identifies a higher-risk group; it does not show that a lifestyle intervention aimed at a clock will eliminate that risk. See the diabetes cohort study.

Useful hypothesis: organs may respond differently to lifestyle exposures. Established prescription: stop smoking, eat a high-quality dietary pattern, move regularly, limit prolonged sitting, and address alcohol use when it harms health. The first claim needs replication; the second does not depend on a biological-age clock.

What to do with this now

Use organ-specific biological-age claims as context, not as a substitute for prevention. If you have elevated cardiovascular risk, the cardiopulmonary clock evidence is clinically relevant because accelerated biological age has predicted adverse cardiac changes and heart failure in a prospective cohort. If you have diabetes or several chronic conditions, prioritize standard risk-factor treatment and do not assume a population lifestyle ranking identifies your most responsive organ.

For most people, dietary quality is the lifestyle target with the broadest, most consistent association with slower biological aging across China, the United States, and the United Kingdom. Smoking cessation and regular activity remain sensible targets, but this evidence cannot assign a precise biological-age payoff to either one for your liver, metabolism, or heart.

The missing experiment is straightforward to describe and hard to run: randomized lifestyle interventions, organ-specific biological-age endpoints, multiple populations, and both clinical-biomarker and proteomic clocks. Until then, organ-specific lifestyle maps are promising observational evidence—not a calibrated navigation system for an individual body.