The Weekly Briefing
The Everhealth Weekly
Evidence-based intelligence for preventing chronic disease and extending healthspan.
Good morning,
Everything in this issue is about a single question: how old are you, really? Not the number on your driver’s license — the number written in your arteries, your metabolism, and your brain. Three weeks of new evidence gave us three very different answers, and they don’t entirely agree.
A 26-year study of 12,409 adults found that three ordinary midlife measurements separated people by roughly 13 dementia-free years. A cross-sectional study of 1,283 twenty-three-year-olds found measurable artery thickening in a population most of us would call young and healthy. And a major Nature Medicine analysis of 51 human trials delivered an uncomfortable verdict on the commercial “biological age” tests now being sold to answer this exact question: most of them barely move.
Then, Friday, cardiology’s largest meeting opens in Munich with 59 trials — including one that has spent eleven years asking whether healthy adults over 70 should be on a statin at all. Our preview is in Section 2.
— The Everhealth Institute team
This Week in the Evidence
Three midlife numbers separated people by 13 dementia-free years
Researchers followed 12,409 dementia-free adults from the Atherosclerosis Risk in Communities study for an average of 26 years — from their mid-fifties into their eighties. They asked a deceptively simple question: how many dementia-free years remain, depending on how many of three vascular risk factors you carry in midlife? The three were high blood pressure, diabetes, and smoking.
Adults with none of the three went on to live roughly 30 dementia-free years after baseline. Adults with all three managed about 17. That is a gap of nearly 13 years of cognitively intact life, predicted by measurements taken decades before any symptom. Over follow-up, 3,008 participants developed dementia and 5,238 died without it. The gap was not evenly distributed: among participants carrying all three risk factors, Black adults averaged 16.0 dementia-free years versus 19.6 for white adults, and men averaged 16.6 versus 18.1 for women.
Source: Hu J, Weiss J, Smith J, Sharrett AR, Selvin E, Fang M, Gottesman RF, Lutsey PL, Mosley TH, Coresh J. Midlife Vascular Risk Burden and Dementia-Free Survival Years: The ARIC Neurocognitive Study. Neurology Open Access. Published online August 5, 2026. Read the study → · Plain-language summary →
Artery damage was already measurable at age 23
Cardiovascular-kidney-metabolic (CKM) syndrome is the framework the American Heart Association introduced to describe how obesity, blood sugar, blood pressure, lipids, and kidney function interact as one disease process rather than five. Stage 0 means no risk factors. This analysis applied the staging to 1,283 adults with a mean age of 23, and the distribution is the finding: only about 21% were Stage 0. Roughly 80% already met criteria for Stage 1, 2, or 3.
More importantly, the stages weren’t just labels. Participants at higher CKM stages had measurably thicker carotid artery walls — an early, imaging-based sign of arterial injury — and lower scores on Life’s Essential 8, the AHA’s composite of cardiovascular health behaviors. In a group of twenty-three-year-olds, the biology was already diverging.
Source: Krishnan V, Lloyd-Jones DM, et al. Cardiovascular-kidney-metabolic syndrome staging and subclinical atherosclerosis in young adults (FF-CHAYA). Circulation: Population Health and Outcomes. August 20, 2026. American Heart Association announcement →
Most “biological age” tests barely moved when people actually intervened
Epigenetic clocks estimate “biological age” from DNA methylation patterns in a blood sample. They are the engine behind a fast-growing consumer testing market — and until now, nobody had systematically asked whether they respond to the interventions people buy them to track. This team assembled every longitudinal human intervention study they could find with methylation data: 51 studies, 3,128 samples, harmonized against 16 clocks and 94 additional methylation biomarkers.
The result is sobering. Of the 51 interventions, only 19 significantly decreased epigenetic age — and only 13 survived correction for multiple testing. Five interventions increased it. The remaining 26 did nothing measurable at all. Among the clocks, DunedinPACE — which estimates the current rate of aging rather than a cumulative age — was the most responsive (mean effect −0.089, P=0.0012), with PCGrimAge the most statistically robust. Pharmacologic interventions produced substantially larger shifts than lifestyle interventions, and clocks moved considerably more in people who were already sick than in healthy people — for PCPhenoAge, a mean difference of 0.28 between disease and healthy populations (P<0.0001).
Source: Sehgal R, Borrus D, Armstrong JF, Gonzalez J, Kasamoto J, Markov Y, Priyanka A, Smith R, Carreras-Gallo N, Lasky-Su J, Dwaraka VB, Corley MJ, Higgins-Chen AT, et al. Responsiveness of epigenetic aging biomarkers to longevity interventions in humans. Nature Medicine. Published online August 21, 2026. Read the study →
A Mediterranean lifestyle — not just the diet — tracked with half the cardiovascular mortality in type 2 diabetes
Most “Mediterranean diet” research scores food. This analysis used MEDLIFE, a 25-item index that also scores how people live — physical activity, sleep, napping, socializing, eating with others, limiting screen time — applied to 3,612 UK Biobank adults aged 40–70 with type 2 diabetes and followed for more than a decade. Over follow-up there were 737 major adverse cardiovascular events, 645 deaths, and 171 cardiovascular deaths.
Comparing the highest to the lowest quartile of adherence: major cardiovascular events were 27% lower (HR 0.73), all-cause mortality 26% lower (HR 0.74), and cardiovascular mortality 51% lower (HR 0.49). The dose–response is the more useful number for patients — each 2-point increase in the MEDLIFE score tracked with 12% lower cardiovascular event risk, 9% lower all-cause mortality, and 21% lower cardiovascular mortality.
Source: Aznar de la Riera MC, Díaz-Gutiérrez J, Kales SN, et al. Mediterranean lifestyle and risk of adverse cardiovascular events, all-cause and cardiovascular mortality among individuals with type 2 diabetes. eClinicalMedicine. 2026;98:104134. Read the study →
On the Horizon: ESC Congress 2026
The European Society of Cardiology Congress — the largest cardiovascular meeting in the world — opens Friday in Munich. The program includes 59 trials across 12 Hot Line sessions plus four new clinical practice guidelines: heart failure, cardiac rehabilitation, cardiovascular disease in chronic kidney disease, and the Fifth Universal Definition of Myocardial Infarction (jointly with the ACC, AHA, and World Heart Federation). Here is what a prevention-minded reader should watch for.
- STAREEShould healthy adults over 70 take a statin? Nearly 10,000 community-dwelling adults aged 70+ with no prior cardiovascular disease, diabetes, or dementia, randomized to atorvastatin 40 mg or placebo and followed for years. It is publicly funded, not industry-funded, and its primary endpoint is disability-free survival — survival free of dementia and persistent physical disability — alongside cardiovascular events. This is the single most consequential unanswered question in primary prevention for older adults, and after more than a decade we finally get an answer. Saturday, August 29.
- REACTHow common is silent atherosclerosis across the adult lifespan? A prevalence study of subclinical plaque in people without symptoms — the natural companion to the CKM finding above, and directly relevant to who should be imaged and when. Saturday, August 29.
- CRHCPDoes long-term blood pressure control lower dementia risk? Long-term follow-up of a large blood-pressure intervention, reporting cognitive outcomes. Given the ARIC data above, this is the randomized test of the same hypothesis. Sunday, August 30.
- TRANQUILITYCan blocking inflammation directly reduce cardiovascular risk? Pacibekitug, an interleukin-6 antibody, in patients with chronic kidney disease at high cardiovascular risk. The inflammation hypothesis — the reason hs-CRP is worth measuring — gets another direct test. Saturday, August 29.
- PRESC1SE-MIFaster, safer chest-pain triage. Roughly 68,000 emergency department presentations across four continents evaluating a rapid rule-in/rule-out troponin algorithm. Not prevention, but it will change what happens the day something goes wrong. Saturday, August 29.
Sources: ESC Hot Line announcement → · ESC Congress 2026 → · STAREE trial registration →
The Long View
The Long View
You have three ages. Only one of them is worth paying for right now.
Everyone in longevity medicine talks about biological age as though it were one thing. It isn’t. This month’s evidence pulls it apart into three distinct objects, and the difference between them is the difference between a useful clinic and an expensive one.
The first is the age of your tissue, and it can be photographed. In 1,283 adults with an average age of twenty-three, roughly eighty percent already carried at least one cardiovascular-kidney-metabolic risk factor, and the ones further along had thicker carotid walls to show for it. Nobody in that study felt sick. Nobody had symptoms. But the artery had already begun keeping a record, and an ultrasound could read it. That is biological age you can see.
The second is the age of your trajectory — where the tissue is heading, inferred from the forces acting on it. This is what the ARIC study measured, and it is the most quantitatively striking result of the month: three ordinary midlife numbers, and a 13-year difference in dementia-free life. Not 13 years of survival. Thirteen years of remaining yourself. Blood pressure, glucose, and tobacco are not exotic markers. They are on every chart in America, usually collected and then filed.
The third is the age on the test you paid for — the epigenetic clock, the number that arrives by email and feels like an answer. And here is where this month gets uncomfortable. Across 51 human intervention studies, only 19 produced a significant reduction in epigenetic age, and only 13 survived statistical correction. Five interventions made the number worse. Twenty-six did nothing at all. The clocks moved most in people who were already ill, and least in exactly the healthy, motivated population that buys them.
The right conclusion is not that epigenetic clocks are a scam. They aren’t — the biology underneath them is real, DunedinPACE in particular is measuring something meaningful about rate of aging, and this field is five years from being genuinely clinically useful. The right conclusion is about sequencing. There is a hierarchy of certainty in preventive medicine, and it currently runs in an unglamorous direction: what can be imaged beats what can be measured in blood, what can be measured in blood beats what can be modeled, and what can be modeled beats what can be sold.
So a rational 2026 prevention plan spends its first dollars on the boring end of that hierarchy — blood pressure, A1c, lipids including Lp(a) and hs-CRP, body composition, cardiorespiratory fitness, and, where indicated, imaging that shows the artery itself. It adds genetic risk where it changes a decision. And it treats the epigenetic clock as what it honestly is today: a promising longitudinal research marker, tracked over years, interpreted alongside everything else — not the headline number, and never the only one.
The Mediterranean-lifestyle data make the final point almost too neatly. The strongest signal in that cohort came not from a molecule but from a way of living — moving, sleeping, and eating with other people — and it tracked with half the cardiovascular mortality. The most powerful longevity intervention we have is still the one that generates no test result at all. Measure what changes decisions. Then go live in a way that makes the measurements better.
Do This Week
- 1Write down your three ARIC numbers — blood pressure, A1c, and smoking status — and put a date on the next check. These are the three that separated 30 dementia-free years from 17. Most people have all three sitting in a patient portal right now and have never seen them together on one line. If your blood pressure is above 130/80 or your A1c is above 5.7%, you have found your highest-yield project for the next twelve months — and midlife is when moving them pays the largest dividend.
- 2If you are under 35, stop treating that as a reason to wait. Eighty percent of the twenty-three-year-olds in the CKM analysis already had at least one risk factor, and the arteries showed it. You do not need a scan at 25. You do need to know your blood pressure, your lipids, your waist-to-height ratio, and whether you smoke or vape — because the thirty-year silent phase of atherosclerosis is happening now, and it is the cheapest window you will ever have to change the slope.
- 3Before you buy a biological-age test, ask what decision it will change. Fair questions for any provider: which specific clock is this, has it been shown to respond to intervention, what change would count as meaningful, and how will it alter my plan? After this month’s Nature Medicine analysis, “it went down 1.2 years” is not an answer — most interventions studied produced no measurable change at all. Tracked serially over years alongside real biomarkers, an epigenetic panel can add signal. Bought once, in isolation, it mostly adds a number.
- 4Add one non-food element of the Mediterranean pattern this week. The MEDLIFE index scores sleep, physical activity, and eating with other people — not just what is on the plate — and a 2-point improvement tracked with 21% lower cardiovascular mortality. Pick the one you are worst at. For most people that is a consistent sleep window or a shared, unhurried meal, and both are free.
Questions We Got This Week
Are epigenetic age tests worth the money?
It depends entirely on how they are used. The August 2026 Nature Medicine analysis of 51 human intervention studies found that fewer than half produced a significant reduction in epigenetic age, and the clocks responded least in healthy people — the group most likely to buy them. There is also no agreed definition yet of how much change is clinically meaningful. That said, the biology is real, second-generation clocks like DunedinPACE do respond, and a test tracked serially over years alongside standard biomarkers can add genuine information about the rate at which you are aging. Our position: it is a reasonable component of a comprehensive advanced testing program and a poor standalone purchase. Spend your first dollars on the measurements that change decisions today — blood pressure, A1c, a full lipid panel with Lp(a), hs-CRP, body composition, and fitness.
I saw a headline that speaking several languages makes your brain 13 years younger. Is that real?
Partly, and it is worth understanding why the caveat matters. Researchers at the Basque Center on Cognition, Brain and Language built an AI “brain age” model from brain-activity recordings in 728 people, then tested it in 144 adults. Bilingual speakers’ brains looked about 6 years younger than their chronological age, trilingual about 7 years, and people speaking four or more languages about 13 years — with stronger effects for higher proficiency and earlier acquisition. Two important limits: this was presented at a scientific meeting (the FENS Forum) and has not yet been published or peer-reviewed, and the researchers themselves note they cannot rule out that multilingual people simply lead more socially and cognitively engaged lives. The defensible version of the finding is the one we already act on — sustained, effortful cognitive and social engagement tracks with better brain aging. Learning a language is an excellent way to get it. Just don’t expect the 13 years.
Should my 70-year-old parent be on a statin if they’ve never had a heart problem?
Honestly, this is the question medicine has not been able to answer well — which is exactly why the STAREE trial presenting in Munich this Saturday matters. Most statin evidence in primary prevention comes from adults under 70, and extrapolating it upward has been an assumption rather than a finding. STAREE randomized nearly 10,000 healthy community-dwelling adults over 70 to atorvastatin or placebo, and it measures disability-free survival — living free of dementia and physical disability — rather than cholesterol alone. Until those results are public, the reasonable approach is individualized: a coronary calcium score can convert an ambiguous decision into a clear one in either direction, and a score of zero in an older adult is meaningful information. We will cover the STAREE result in next week’s issue.