Lifespan and healthspan
Lifespan is how long you live. Healthspan is how long you live in good health, free of serious disease and disability. The goal of this guide is to extend both, but healthspan is the one people feel.
Life expectancy is recovering from the pandemic. In the US it reached 79.0 years in 2024 (76.5 for men, 81.4 for women), up 0.6 years from 2023, and a 65-year-old could expect about 19.7 more years. Worldwide, life expectancy returned to pre-pandemic levels in 2023: 76.3 years for women and 71.5 for men, according to the Global Burden of Disease study.
The gains are slowing. A 2024 analysis of the longest-lived countries found that improvements in life expectancy have decelerated since 1990, and estimated that in these populations survival to 100 is unlikely to exceed 15% for women and 5% for men this century, unless biological aging itself can be slowed.
The added years are not all healthy ones. A 2024 study of 183 countries found the gap between life expectancy and healthy life expectancy has widened to 9.6 years globally. The US had the largest gap of all, 12.4 years, driven by rising chronic disease.
- 79.0 years US life expectancy at birth in 2024 (NCHS)
- 12.4 years gap between US life expectancy and healthy life expectancy, the widest measured (JAMA Network Open 2024)
- ~500× higher lung cancer rate at 70 to 74 than at 25 to 29 (SEER)
Why age dominates risk
An aging-associated disease is one where chronological age is the dominant risk factor. Atherosclerosis, most cancers, type 2 diabetes, and the neurodegenerative diseases all qualify.
Lung cancer shows how lopsided this is. Smoking raises the risk about 25 times. Age does far more: US lung cancer incidence rises from 0.6 per 100,000 people at ages 25 to 29 to 327.6 per 100,000 at ages 70 to 74, a roughly 500-fold increase. Smoking still matters enormously, because it multiplies a risk that age has already made large.
This explains a strange fact: eliminating every cancer death in the US would add only about three years to life expectancy, and eliminating heart disease almost four, because the other diseases of aging catch up. That is the core argument of geroscience, the field that studies aging as the shared cause of these diseases: slow aging itself and you delay all of them at once.
The four horsemen
Physician Peter Attia's book Outlive groups the chronic diseases of aging into "four horsemen": atherosclerotic cardiovascular disease, cancer, neurodegenerative disease, and metabolic dysfunction. The framing is useful because the four share upstream causes, so the same habits and medicines push back on several at once.
The official numbers bear this out. In 2024, of the 3,072,666 deaths in the US, heart disease, stroke, cancer, Alzheimer disease, and diabetes alone accounted for about 55%. Worldwide, the WHO's latest data (for 2023) show the top 10 causes made up 54% of 61 million deaths, and eight of the 10 were noncommunicable diseases.
| Horseman | US deaths, 2024 | World deaths, 2023 | Where to read more |
|---|---|---|---|
| Heart disease and stroke | 683,491 heart disease; 166,852 stroke (CDC) | 9.5 million ischemic heart disease; 6.8 million stroke (WHO) | Heart and blood vessels |
| Cancer | 619,876 (CDC) | 1.8 million from lung cancer alone (WHO) | Cancer |
| Neurodegenerative disease | 116,022 Alzheimer disease (CDC) | 2.1 million Alzheimer's and other dementias (WHO) | Brain health |
| Metabolic disease | 94,445 diabetes; 55,081 kidney disease; 52,274 chronic liver disease (CDC) | 2.0 million diabetes (WHO) | Metabolic health |
Death certificates undercount metabolic disease, which drives many deaths recorded as heart attacks and strokes. The table also leaves out what kills younger people: unintentional injuries were the third leading cause of US death in 2024, with 197,449 deaths, and suicide was 10th. Those are covered in Avoiding harm and Mind and connection.
What aging is
The simplest working definition: aging is the progressive loss of function over time, which shows up as a risk of disease and death that rises with every year. Underneath is a tangle of molecular and cellular damage that accumulates faster than the body repairs it. For a long, careful treatment, Nintil's essay What is aging? remains one of the best.
Damage or program
Theories of why organisms age fall into two broad camps.
- Damage accumulation Aging is wear and tear: mutations, misfolded proteins, cross-linked tissues, and other damage pile up. Evolutionary theory explains why repair is imperfect: natural selection weakens after reproductive age, so genes that help early in life but harm later can spread (Williams 1957), and bodies invest in reproduction over maintenance (Kirkwood 1977).
- Programmed or runaway processes Aging partly reflects developmental programs that never switch off, or loss of the epigenetic information that tells cells what to be. This camp is behind the push to reset cells with reprogramming factors, covered in Longevity drugs.
Most researchers accept elements of both, and both predict the same thing: slowing the underlying processes should delay many diseases together.
The hallmarks of aging
The most-used map of those processes is the hallmarks of aging: nine in the original 2013 paper, 12 in the 2023 update. Each must change with age, speed aging when accelerated in experiments, and slow aging when targeted.
| Hallmark | What goes wrong | Group |
|---|---|---|
| Genomic instability | DNA mutations and damage accumulate over a lifetime. | Primary |
| Telomere attrition | The protective caps on chromosomes shorten with each division until cells stop dividing. | Primary |
| Epigenetic alterations | The chemical marks that switch genes on and off drift out of tune. | Primary |
| Loss of proteostasis | Cells get worse at folding, maintaining, and clearing proteins. | Primary |
| Disabled macroautophagy | The cell's bulk recycling system slows down (added 2023). | Primary |
| Deregulated nutrient sensing | Pathways that read food and energy supply (insulin and IGF-1, mTOR, AMPK, sirtuins) stop responding properly. | Antagonistic |
| Mitochondrial dysfunction | The cell's power plants make less energy and more damaging byproducts. | Antagonistic |
| Cellular senescence | Damaged cells stop dividing, refuse to die, and secrete inflammatory signals. | Antagonistic |
| Stem cell exhaustion | The reserve of cells that repair tissue runs down. | Integrative |
| Altered intercellular communication | Hormonal, neural, and immune signaling between tissues becomes noisy. | Integrative |
| Chronic inflammation | Low-grade, body-wide inflammation ("inflammaging") rises (added 2023). | Integrative |
| Dysbiosis | The gut microbiome drifts away from a healthy mix (added 2023). | Integrative |
How the hallmarks fit together
The groups describe a rough causal chain. Primary hallmarks are the sources of damage. Antagonistic hallmarks are responses that protect in the short term but cause harm when they become chronic; senescence, for example, stops damaged cells from turning cancerous but poisons their neighbors when senescent cells accumulate. Integrative hallmarks are the system-wide failures that result, and they are what you experience as loss of function.
Biology does not respect the boxes. Caloric restriction, exercise, and rapamycin each touch several hallmarks at once. Researchers have proposed more candidates, such as stiffening of the extracellular matrix (the scaffolding between cells), and critics argue the list mixes causes with consequences and works better as a catalogue than a theory.
Genes, lifestyle, and centenarians
How much of lifespan is inherited is genuinely contested. Twin studies put the heritability of lifespan at about 20% to 25%, and a 2018 study of hundreds of millions of people in family trees concluded that the true figure for people born in the 1800s and early 1900s was "well below 10%" once spouses' similar lifespans were accounted for. A 2026 modeling study in Science (the same link) pushed back: it argued those estimates are diluted by deaths from accidents and infections, and that heritability of "intrinsic" lifespan, death from aging itself, is above 50%.
Both can be true. In historical populations, infections, accidents, and hardship decided much of who died when; as those causes recede, genetic differences in the pace of aging matter more. Longevity also clusters in families at the top end: in a 2019 study of 314,819 people, having relatives in the top 10% of survivors for their birth cohort was linked to up to a 31% survival advantage.
Studies of people who reach 100 point the same way:
- Centenarians often did not live especially healthy lives. Among 477 Ashkenazi Jewish adults aged 95 and older, body weight, smoking, drinking, exercise, and diet were no better than in their peers in the general population. Their advantage seems to lie more in their biology than their habits, which is not a strategy available to most people.
- They delay or escape disease. In a study of 424 centenarians, 87% of men and 83% of women had delayed heart disease, cancer, and stroke until after 80 or escaped them entirely.
- The oldest compress their illness into the end. Among 104 supercentenarians (people 110 and older), disease and disability arrived later and occupied a smaller share of life.
That last pattern is called compression of morbidity, an idea proposed in 1980: if the onset of chronic illness can be pushed later faster than lifespan grows, the sick years at the end shrink. It is the most realistic goal for most people, even though the widening healthspan gap shows populations moving the other way.
The practical lesson: you cannot choose your genes, but family history tells you where to look. A parent with a heart attack at 50, or a family pattern of a specific cancer, should change your screening and treatment, as covered in Building your care team, Heart and blood vessels, and Cancer.
Measuring biological age
Biological age is an estimate of how old your body behaves, as opposed to the years since birth. The leading tools are epigenetic clocks, algorithms that read methylation marks on DNA (chemical tags that change with age) from a blood or saliva sample.
| Clock | What it estimates | Notes |
|---|---|---|
| Horvath (2013) | Chronological age across tissues | The first multi-tissue clock; predicts age well, health less well |
| PhenoAge (2018) | A blood-chemistry measure of physiological age | Trained to predict disease and death |
| GrimAge (2019) and GrimAge2 (2022) | Mortality risk | Among the strongest predictors of death in cohorts |
| DunedinPACE (2022) | Pace of aging, in biological years per calendar year | Built from repeated measures of organ function over two decades |
| Proteomic clock (2024) | Age from 204 blood proteins | Predicted 18 chronic diseases and death in 45,441 UK adults |
| Organ-age clocks (2023) | Separate ages for 11 organs from blood proteins | About 20% of people had one strongly "older" organ |
Organ clocks are the newest idea. In the 2023 study of 5,676 adults, an older-than-expected organ carried a 20% to 50% higher risk of death, and an "old" heart a 250% higher risk of heart failure. A 2025 analysis of 44,498 people found an especially aged brain raised Alzheimer's risk about as much as carrying one copy of APOE4, while a youthful brain and immune system together were linked to 56% lower mortality.
In research, clocks serve as early signals in trials: two years of caloric restriction in the CALERIE trial slowed DunedinPACE slightly, and in a post hoc analysis of the DO-HEALTH trial, omega-3s slowed several clocks by about three months over three years. Neither result shows that the people involved will live longer.
How aging is studied
Aging interventions are tested in organisms of increasing complexity. Yeast, roundworms (C. elegans), and fruit flies live days to weeks and are ideal for finding genes. Mice and rats, living two to three years, are the workhorses for drug testing. Naked mole rats, which barely seem to age, are studied for their resistance to cancer. Pet dogs, which share human environments, are now in real longevity drug trials, and monkeys have been used in decades-long caloric restriction studies. Human trials come last and take longest, because aging takes decades.
Lifespan results are usually shown as Kaplan-Meier survival curves, which plot the share of animals still alive over time. A drug that works shifts the curve to the right.
The most rigorous mouse test is the National Institute on Aging's Interventions Testing Program (ITP). Each compound is tested in genetically diverse mice of both sexes, at three independent sites, with every result published, positive or not. Rapamycin is its most robust result, followed by acarbose, 17α-estradiol (in males), and the diabetes drug canagliflozin (in males). Many heavily marketed supplements, including resveratrol, nicotinamide riboside, and fisetin, did nothing. Most positive results are male-only, and several did not hold up when retested at other doses or ages in 2026. The full ITP results are in Longevity drugs; how to weigh mouse data is in Reading the evidence.
The field in 2026
- No drug is approved to slow aging in people. The TAME trial, proposed a decade ago to test metformin in 3,000 adults aged 65 to 79, has produced no results, and public reports on its funding and enrollment conflict as of October 2026.
- The first rapamycin trial in healthy adults was modest. PEARL, run by the telehealth company AgelessRx, randomized adults 50 to 85 to low-dose rapamycin or placebo for 48 weeks. It was about as safe as placebo but missed its primary endpoint. Details are in Longevity drugs.
- Dogs may get the first lifespan drug. The FDA has accepted Loyal's evidence of effectiveness (February 2025) and safety (January 2026) for LOY-002, a drug intended to extend the lifespan of older dogs. The company expects to apply for conditional approval in 2027; nothing is approved yet.
- Cellular reprogramming has reached humans. In October 2026 Life Biosciences reported first data from a Phase 1 trial of ER-100, a gene therapy that switches on three of the four Yamanaka factors (OCT4, SOX2, and KLF4) in the eye for eight weeks. Three people with glaucoma, no serious side effects through day 56, and visual-field gains in two. Open-label, uncontrolled, and very early.
- A $101 million prize is under way. XPRIZE Healthspan will reward a treatment that restores at least 10 years of muscle, cognitive, and immune function in people aged 50 to 80. Its 20 finalists were named in August 2026, and the prize runs to 2030.
- The biggest "longevity drugs" so far were built for other diseases. Semaglutide cut major cardiovascular events by 20% in 17,604 adults with obesity and heart disease, and a veterans' cohort of 215,970 GLP-1 users with diabetes linked the drugs to lower rates of dementia, addiction, and infections. But when semaglutide was tested in 3,808 people with early Alzheimer's, it did not slow the disease. Observational signals are not proof.
- A vaccine may protect the brain. A natural experiment in Wales found that the shingles vaccine cut new dementia diagnoses by 3.5 percentage points over seven years, a 20% relative reduction. See Infections and vaccines.
- Hype corrects itself, slowly. A 2023 paper proposed taurine deficiency as a driver of aging; a 2025 study found that taurine levels in people often rise or stay flat with age, undercutting the idea.
Two good primers for going deeper are Laura Deming's Longevity FAQ (a 2017 snapshot, still a clear introduction) and Nintil's Longevity FAQ (updated through 2023). The ethics of extending life are outside this guide; Lifespan.io summarizes the common debates.
What this means for you
Aging is the background risk you cannot yet change; the four horsemen are the foreground risks you can. Nearly everything that delays them works through a short list: not smoking, controlling blood pressure, lowering ApoB, keeping glucose and weight in check, staying fit and strong, sleeping enough, and staying connected.
Three implications follow.
- Start early. Atherosclerosis, insulin resistance, and the brain changes of Alzheimer's begin decades before symptoms. In a Spanish study of 1,779 middle-aged adults with no conventional risk factors, half already had plaque, and LDL predicted it even at levels considered normal. The damage from LDL tracks both how high it is and for how long.
- Measure what predicts outcomes, every year, against optimal targets. ApoB under about 60 mg/dL, blood pressure under 120/80, HbA1c under about 5.5%, fitness, and strength, rather than a biological-age score. Guideline minimum: most guidelines start treatment when a 10-year risk calculator crosses a threshold, which in practice waits until middle age; see Heart and blood vessels.
- Treat longevity drugs as experiments, to consider only after proven medicines and habits are in place.
The essentials gives the one-page version of that plan.