Healthspan versus lifespan
Lifespan is how long you live. Healthspan is how long you live in good function — able to walk, think, work, carry shopping, get off the floor, hear a conversation. The gap between the two has widened in most wealthy countries, which is the uncomfortable fact behind the whole field: medicine has become better at preventing death than at preventing decline, and the extra years are not always ones people would choose.
That has practical consequences for what you should aim at. Interventions that extend life without preserving function are a poor deal, and interventions that preserve function are worth pursuing whether or not they extend life at all. It also changes which measures matter. Grip strength, gait speed, the ability to rise from a chair without using your arms, balance, cognition and hearing are better markers of how the next decade will go than most of the things sold as longevity biomarkers. They are also free to measure and responsive to training, which is the opposite of most of this market.
What "biological age" tests can and cannot tell you
Biological age tests, most commonly based on patterns of DNA methylation, are genuine science. Epigenetic clocks are useful research instruments: across large populations, they correlate with mortality and disease risk better than chronological age alone. That is a real finding and it is why the field takes them seriously. The problem is what happens when a population-level research tool is sold to an individual as a number about them.
Different clocks give different answers on the same sample. Repeat testing on the same person can shift by years for reasons that have nothing to do with health, including how the sample was collected and processed. There is no standardisation across providers, no agreed reference range, and — most importantly — no validated action attached to the result. If your test says you are three years older than your birth certificate, nobody can tell you what to do that is different from what you should be doing anyway. Treat the number as entertainment with a plausible scientific backstory, and spend the money on a blood pressure monitor and a gym membership instead.
The levers with real evidence
Blood pressure is the highest-value number in the set: sustained elevation damages arteries, kidneys, heart and brain over decades, treatment is cheap and well studied, and control is strongly linked to lower stroke, heart attack and dementia risk. Lipids, particularly LDL cholesterol, accumulate risk over a lifetime, which is why duration of exposure matters as much as the level. Glucose regulation belongs in the same group — the metabolic conditions cluster, and preventing progression from prediabetes has good trial support.
Muscle mass and strength are the physical-function equivalent. Resistance training produces measurable gains into the eighties and nineties, and strength and gait speed predict independence, falls and mortality with unusual consistency. Two sessions a week is the standard target. Sleep, both duration and regularity, is repeatedly associated with cardiovascular and cognitive outcomes. Not smoking remains the single largest modifiable factor for most people. Social connection is not soft: isolation and loneliness carry measurable health risk. And hearing loss is one of the more actionable modifiable risk factors identified for dementia, which makes an audiology appointment a longevity intervention with better support than anything in a supplement aisle.
The unproven end of the market
Most longevity supplements have not been shown to extend life or healthspan in humans. NAD precursors such as nicotinamide riboside and nicotinamide mononucleotide reliably raise NAD levels in blood, which is the finding that gets marketed. Whether raising NAD produces any clinical benefit in humans is not established; the human trials have been small, short and focused on biomarkers rather than outcomes. Senolytics — drugs intended to clear senescent cells — are a serious scientific programme with promising animal data and only early human trials, and are not a proven human therapy.
Beyond supplements, several offerings deserve outright scepticism. Stem-cell clinics selling unapproved infusions for ageing operate largely outside regulatory oversight, and regulators have repeatedly issued warnings about harms including infection and vision loss. Hyperbaric oxygen has legitimate approved indications, and anti-ageing is not among them. Full-body imaging marketed as preventive screening in healthy people has a well-documented incidental-finding problem that leads to further tests, procedures and anxiety without demonstrated benefit. The general rule holds well: if a longevity product could show hard outcome data, it would, and the ones that can — blood pressure treatment, statins in the right people, exercise, vaccination — are boring and cheap.
Diet, fasting and caloric restriction
Caloric restriction extends lifespan in several model organisms, which is one of the more robust findings in ageing biology and the origin of much of the field's enthusiasm. Translating it to humans is a different matter. Human trials of sustained caloric restriction have shown improvements in cardiometabolic markers over a year or two; they cannot answer questions about lifespan, and the practical costs — bone density, muscle mass, adherence, and risk in people with a history of disordered eating — are real.
Intermittent fasting and time-restricted eating have been studied more than most popular interventions, and the fair summary is that they work about as well as other approaches that reduce overall intake, with no consistent metabolic advantage beyond the calories. Some people find the structure easier to follow, which is a legitimate reason to choose it. Dietary patterns with the strongest long-term evidence remain the unexciting ones: mostly plants, plenty of fibre, more unsaturated than saturated fat, limited ultra-processed food, moderate sodium. Our nutrition guide and heart-healthy eating guide cover this properly, and there is a longer treatment in longevity basics.
How to evaluate any longevity claim
Six questions handle almost everything. What species was this shown in? What outcome was measured — lifespan, healthspan, function, or a biomarker? Over what duration? Compared with what? Who paid for it and who is selling it? And what is the plausible harm, including the money and the opportunity cost of doing this instead of the things that work?
Two more are worth adding for anything marketed directly to consumers. Is the result being reported in the same population you belong to — much ageing research is done in people already at high risk, and benefits do not always transfer. And is anyone offering a testable prediction, or only a mechanism? Mechanisms are cheap; the history of ageing research is full of elegant mechanisms that produced nothing in people. The unglamorous programme — control blood pressure, get strong, sleep enough, do not smoke, stay connected, get your hearing checked, attend your screenings — has more evidence behind it than everything sold under the longevity label combined.