Wearable aging clocks — how the Apple Watch became a biomarker for longevity

The same green light an Apple Watch uses to measure your pulse can now estimate your biological age with 2.4 years of error. The clock is powered by deep learning, not guesswork. A study of 213,593 people published in Nature Communications proved it. The sensor is already on millions of wrists. The software to turn that signal into a longevity biomarker is what didn't exist — until now.

Aging clocks have existed for years. The problem was they required a blood draw, a lab, and a week of waiting. Epigenetic methylation clocks like GrimAge and DunedinPACE remain the gold standard for measuring biological age — they predict all-cause mortality with correlations above r = 0.90. But they are single-point measurements, taken in a clinic, expensive to run, and impossible to repeat daily.

Wrist-worn photoplethysmography (PPG) changes that. The sensor costs pennies. The data streams continuously.

It works.

And now, for the first time, the signal has been validated against clinical outcomes across a population larger than most drug trials.

The science is already on your wrist

The PpgAge algorithm, developed by researchers at Duke and Apple, uses deep learning to extract aging signals from raw PPG waveforms. The model does not look at heart rate alone. It analyzes the full morphology of the pulse wave — how fast the blood volume rises, how long the peak holds, how the waveform decays between beats. These features change with arterial stiffness, endothelial function, and autonomic regulation, all of which degrade with age.

Participants with a PpgAge gap greater than six years had heart disease diagnosis rates of 3.6%, versus 1.0% for those with a gap below -2 years. The signal held after controlling for age, sex, BMI, and comorbidities. The same gap predicted incident cardiac events in previously healthy people.

The study also tracked behavioral associations. Smokers had elevated PpgAge. People who exercised regularly had lower PpgAge. Pregnancy triggered a sharp, transient increase. So did bypass surgery.

The sensor is commodity hardware. What is new is the proof — 213,000 people wearing it, four years of data, and a deep-learning model that turns a pulse wave into a clinically meaningful aging signal.

This matters because 149 million participant-days is more data than any aging clock study has ever accumulated by orders of magnitude. Blood-based clocks from the Berlin Aging Study II or the UK Biobank measure thousands of participants at one or two time points. The wearable equivalent measures hundreds of thousands of people across years of daily life, catching transient events — a pregnancy, a surgery recovery, a change in smoking habits — that single-point tests miss entirely.

What is falling — the old model of reactive biomarker testing

The traditional approach to biological age testing is a single blood draw sent to a lab. The patient waits days for results. The measurement is a snapshot — useful but static. Annual physicals and cholesterol panels tell you almost nothing about aging velocity.

That model is losing ground on every dimension. Whole-blood proteomics costs have fallen 18% annually since 2022. Sequencing a genome that cost $1,000 a decade ago is now under $200. The price of generating a longitudinal biomarker dataset — the kind Fountain Life produces per member — is dropping faster than the cost of running a single batch of standard lab tests five years ago.

What is dying is the idea that annual snapshots are sufficient. Continuous monitoring from wearables, combined with AI analytics, produces a curve instead of a point. That shift makes older diagnostic models look expensive and low-resolution in one stroke.

Forty-seven longevity clinics now operate across the United States, each deploying four to eight biomarker panels per patient visit. Corporate wellness programs are adopting biological age testing to reduce absenteeism. Life insurance underwriters are piloting biological age tests to refine actuarial models. The one-time test is being replaced by a subscription — continuous measurement, continuous analysis, continuous adjustment.

What is growing — from research to commercial platform

Fountain Life has already commercialized the concept. The company's ZORI platform integrates advanced imaging, blood biomarkers, genetic analysis, and cognitive assessments with continuous health tracking. As of January 2026, it had processed over 15 billion longitudinal data points from 8,000+ members and detected early cancer signals in 3.4% of them — compared to roughly 1% for standard screening.

Fountain Life is not alone. InsideTracker and ZOE have built AI engines that fuse continuous glucose data, sleep metrics from Oura rings, activity data, and blood panels into unified biological age scores that update in near real-time. Function Health acquired GetLabs in 2026 to add at-home blood draw capability to its AI biomarker platform. YOU(th), a Berlin-based startup, raised $4.5 million for a smartphone screening tool that analyzes face videos and voice recordings to assess 50 digital biomarkers in under two minutes.

Swiss startup Xsensio is developing a nanotech skin patch for continuous metabolite and inflammatory-marker tracking. In China, the AmiAge study published in Nature Communications in May 2026 demonstrated that an 8-amino-acid blood panel predicts biological age across 270,000+ samples — a simpler, cheaper alternative to full epigenetic sequencing.

The longevity biomarker diagnostics market hit $2.4 billion in 2025 and is projected to reach $8.2 billion by 2034. Epigenetic methylation clock testing alone accounts for $821 million — 34% of that market. The growth is driven by falling sequencing costs, AI-powered analytics, and the expansion of longevity clinics beyond ultra-high-net-worth clientele into corporate wellness and insurance.

The government is treating aging as a treatable condition

In March 2026, ARPA-H committed $144 million through its PROSPR program to fund clinical trials that treat aging as a medical condition. Cambrian BioPharma received $30.8 million for trials of a next-generation rapamycin analog. Linnaeus Therapeutics received $22 million for oncology-derived drugs with age-related benefits. The program also funds biomarker development and clinical trial infrastructure.

This is the same agency that funds moon-shot biology and high-risk medical technology. Treating aging as a fundable target changes the regulatory math for every company in the space. ARPA-H accepting intermediate biomarkers as surrogate clinical trial endpoints — which it did in 2026 — means a company can show it slows epigenetic aging in a six-month trial and get that counted as evidence, rather than waiting for a 20-year mortality study.

As we wrote in June, NewLimit raised $435 million for epigenetic reprogramming — reversing the cellular age of tissues without dedifferentiation. The longevity sector is no longer a collection of supplement companies. It is a capital-intensive, clinically-driven industry with $13.1 billion in cumulative venture funding and 47 dedicated longevity clinics operating across the United States as of 2026. The convergence of wearable sensors, AI analytics, and regulatory acceptance of aging biomarkers is creating the infrastructure for a continuous health monitoring system that did not exist two years ago.