Bryan Johnson's Best Result Is the One He Doesn't Lead With

LONGEVITY LATEST ISSUE 13 COMPANION · 3 JUNE 2026

LONGEVITY LATEST · DEEP DIVE

Bryan Johnson's Best Result Is the One He Doesn't Lead With

The DunedinPACE number is the headline. The bloods are the proof. Telling them apart is the whole game.

By Christian Thomsen · Companion to Issue 13 · 3 June 2026 · ~7-minute read

The most interesting thing about Bryan Johnson isn't the seventy pills. It's the spreadsheet.

He publishes his own data. The lipid panels, the inflammation markers, the sleep architecture, the DEXA scans, the epigenetic clock results, the body composition, the bloodwork over time — open, downloadable, time-stamped. That alone makes him almost unique in the wellness internet, where the people selling you a protocol usually decline to show you what it did to theirs. The question this piece settles is what the spreadsheet actually proves — and what it doesn't, even though the marketing has decided otherwise.

By the end you'll know which result he leads with on stage (the famous one), which result actually has the cleaner evidence underneath it (the boring one), and why the gap matters whether you ever swallow a single capsule.

The number on the t-shirt

The headline claim is a DunedinPACE score of about 0.69 — meaning, on the version of the clock used, Bryan's body ages roughly 8.3 months for every twelve months of calendar time. Repeated across the protocol's lifespan, that score has held up. The "Rejuvenation Olympics" league table built around exactly that metric ranks him near the top.

DunedinPACE is real science. Belsky et al. trained it on the Dunedin birth cohort — a thousand New Zealanders followed from childhood — to capture how fast a person is currently ageing, rather than how old their methylation looks at a snapshot. It correlates with handgrip, gait speed, and self-rated health. As a population tool, in a cohort, with thousands of samples, it does what it says on the tin.

What it has not been validated for is the use case Bryan and a growing industry are putting it to: a feedback signal for a single person's intervention. Higgins-Chen et al. (Nature Aging 2022) tested the canonical epigenetic clocks for replicate-to-replicate reliability and found technical noise of up to nine years between identical samples on the biggest clocks. The principal-component versions they introduced narrowed that to under 1.5 years — better, not perfect, and most commercial reports don't tell you which one they used. Read against that noise floor, a five-year "improvement" in an individual's reported epigenetic age is not obviously a signal. It might be the measurement breathing.

This is the structural ceiling on what a single person — even one as well-tracked as Bryan — can demonstrate with consumer epigenetic clocks. He can show a value. He cannot prove it moved because of his protocol, because no n-of-one design can separate intervention effect from measurement noise without a comparator. The most generous reading of the clock data is that it's consistent with slower ageing. It's a hypothesis with a t-shirt.

The clock numbers aren't fake. They're just doing work the clock can't actually carry on one person.

The 31-year claim, dismantled gently

The most-shared version of Bryan's story compresses to: "I've slowed my pace of ageing by 31 years." It's a sentence that travelled because it sounds quantitative, and it travelled wrong.

Pace of ageing isn't measured in years — it's a ratio. A score of 0.69 doesn't mean "I am 31 years younger." It means "for now, on this measure, I appear to be biologically accumulating damage about 30% slower than the cohort average." Multiply that across a lifetime and you do get a meaningful gap, in theory. But the unit conversion that turned a ratio into "31 years" wasn't validation — it was a sentence designed for a podcast clip, and the gap between those two things is where most of the wellness internet now lives.

Reviewers in the field have pointed this out — gently in the press, more sharply at conferences. Bryan is, to his credit, broadly transparent about uncertainty in his own technical write-ups. The 31-year line is the version that escaped, and once a number escapes it's hard to call back.

The win underneath the win

Now the part the headlines under-cover. The cardiometabolic data are exceptional. They are also the data Bryan doesn't open with, which is a shame, because they're the cleanest signal in the entire experiment.

Apolipoprotein B in the low double digits, well below the cardiology-society aggressive targets. High-sensitivity CRP consistently sub-0.3 mg/L — the floor that Lancet primary-prevention guidance now calls a residual-risk endpoint worth chasing. HbA1c bottom of the reference range. Resting heart rate that you would expect from a competitive endurance athlete. Lipid sub-fractions, oxidative-stress markers, and inflammatory cytokines that are, by any honest reading, exemplary for a man in his late forties.

These metrics share two features. First, they have been validated as targets — RCT-grade evidence supports lowering ApoB, hsCRP, HbA1c, and resting heart rate as drivers of reduced cardiovascular events and dementia risk in actual populations. Second, they are measured by assays whose precision and interpretation are not controversial. The signal is signal.

So the honest summary of Blueprint, judged by the data Bryan himself publishes, looks like this. The protocol almost certainly drove down ApoB, CRP, and visceral fat to levels associated in large cohorts with substantially lower long-term disease risk. That isn't speculative. It's what extreme dietary discipline, daily exercise, sleep optimisation, and aggressive lipid management do in any clinical trial you'd care to name. The pills are a distant third in that causal chain — and probably some of them are doing something at the margins, which is exactly what an A-grade for olive oil and a B− for lithium in this week's issue would suggest. The structure is sound. The drama is the parts you can't actually prove.

Why the n-of-one problem doesn't go away

You can imagine a more rigorous version. More clocks, more retests, more biomarkers, a year's worth of measurements at each timepoint. Bryan is moving in that direction, and the discipline is admirable. It still can't beat the structural problem.

Without a comparator — without a second Bryan Johnson on a £4 multivitamin, identical genetics, identical sleep, identical bank balance, identical stress — the question "what did the protocol do?" has no clean answer. Lifestyle, finance, attention, and time are confounders that no measurement strategy retires. The reason RCTs were invented in the 1940s is precisely that the smart, motivated, well-resourced individual is the worst possible test of "does this work?" — they tend to do several things right at once and attribute the result to whichever one they're selling.

This is the line the Hype Check in the newsletter pointed at. The consumer-clock industry is monetising a methodological boundary, not crossing it. Even Bryan, with the cleanest possible setup, can't.

Why the boring number is the one that matters

There is a temptation to call the cardiometabolic data a consolation prize. It isn't. It's the actual prize, dressed in less compelling language.

An ApoB sitting near the floor of the cardiology-society aggressive targets is the single most defensible predictor of not having a heart attack in your sixties that modern medicine has produced. A high-sensitivity CRP below 0.3 mg/L is the residual-risk signal The Lancet keeps writing about. Lowering both across two decades is the cleanest predictor we have of more years where your body keeps up with you — the years where you carry your own bag through the airport, the years where you can still get down on the floor with grandchildren and back up again, the years before the diagnoses start arriving in clusters.

Translated out of the journal language: lower lifetime disease-risk burden, later onset of the things that finally slow you down, better odds of compressing the bad years rather than extending them. That isn't reversing ageing. It's the part of healthspan that responds to evidence-based effort, dressed in less compelling language than a t-shirt slogan — and almost certainly a more durable outcome than a clock score.

The takeaway

So, three honest sentences to leave with.

The Blueprint protocol is best read as one extremely well-executed cardiometabolic optimisation programme — and on that endpoint the data are impressive, with about £20 of supermarket olive oil, sleep, lifting, and discipline carrying most of the load.

The slowed-ageing claim, on current evidence and current clocks, is an unproven hypothesis dressed in a number — interesting, possibly directionally right, not something to plan a stack around.

And the bit of Blueprint worth copying is the spreadsheet. Cheap bloods, a wearable, a notebook, a habit of looking at the trend. That part is free, it's been validated in clinical trials for decades, and it moves endpoints that — unlike DunedinPACE — your future self will actually feel.

This is the Issue 13 Deep Dive. Longevity Latest runs one every week — the long version of the argument the newsletter only has room to start. Next week we open up cold exposure: ice baths, plunge studios, and the one cold protocol with surprisingly tight RCT data underneath the hype.

Sources and further reading

1. Belsky DW, Caspi A, Corcoran DL, et al. DunedinPACE, a DNA methylation biomarker of the pace of ageing. eLife. 2022;11:e73420.

2. Higgins-Chen AT, Thrush KL, Wang Y, et al. A computational solution for bolstering reliability of epigenetic clocks: implications for clinical trials and longitudinal tracking. Nature Aging. 2022;2(7):644–661.

3. Tessier AJ, Cortese M, Yuan C, et al. Consumption of olive oil and diet quality and risk of dementia-related death. JAMA Network Open. 2024;7(5):e2410021.

4. Aron L, Ngian ZK, Qiu C, et al. Lithium deficiency and the onset of Alzheimer's disease. Nature. 2025;643:1077–1086.

5. Baker LD, Manson JE, Rapp SR, et al. Effects of cocoa extract and a multivitamin on cognitive function (COSMOS-Mind). Alzheimer's & Dementia. 2022;19:1308–1319.

6. Yeung LK, Alschuler DM, Wall M, et al. Multivitamin supplementation improves memory in older adults: a randomized clinical trial (COSMOS-Web). American Journal of Clinical Nutrition. 2023;118(1):273–282.

7. Vyas C

© 2026 FrontWave Media Ltd · Longevity Latest