Hi {{ first_name| strip | default: there}},
In the last four months, I've had my VO2 max tested three ways.
My Apple Watch says 37.4
A treadmill test (mask on, running until I couldn't) says 44.8
A blood test says 47.3
Same person, ten points apart!
On most charts, that's the gap between "fine" and "excellent." So depending on which one you believe, I'm either doing fine or I'm kind of an athlete.
I care about this number more than most. Fitness is one of the strongest predictors of how long you live. In the largest study we have, being unfit carried more risk than any of the heart risk factors the researchers looked at.
So which number is mine? My treadmill result is up almost 5 points since 2024. Did I actually get fitter, and will it help me live longer?

At my October treadmill test. The mask measures the oxygen you breathe.
The short version
Did I get fitter? Yes. About three quarters of my gain was real; the rest was the scale.
Will it help me live longer? Probably. Fitter people live longer, but no study can say what my extra points are worth.
Which number is real? The treadmill test. My watch and blood test are estimates.
How should you track yours? Pick one test and only compare it with itself.
How do you raise yours? It depends on where you're starting. Here's my guide.
Why this number is worth caring about
VO2 max is the most oxygen your body can use when you're going all out. To score high, your lungs, blood, heart, vessels, and muscles all have to do their jobs, which is probably why it says so much about your health.
The numbers are pretty wild. In a study of 750,302 U.S. veterans followed for about ten years, the least fit fifth had four times the risk of dying of the fittest group. And there was no point where being fitter stopped helping.
You may have seen the chart from the other big one, because it gets shared a lot. The Cleveland Clinic followed 122,007 people after a treadmill test. The least fit had five times the risk of dying of the fittest. Smoking raised the risk 1.4 times. (Please don't take that as a case for smoking.)

And across 33 earlier studies, every 1 MET of fitness (about 3.5 points of VO2 max) went with 13% lower mortality.
Important to know: these studies compare different people. They can't tell me how many years my 44.8 is worth, or what I'd gain from three more points. When researchers actually assign people to exercise, the results get messier. In Generation 100, five years of supervised exercise for adults in their 70s didn't significantly lower deaths. (To be fair, the comparison group was told to follow the exercise guidelines, and mostly did.)
What exercise trials have shown is something I care about just as much: staying able to move. In the LIFE trial, sedentary adults aged 70 to 89 who did a program of walking, strength, and flexibility work were less likely to lose the ability to walk 400 meters (30.1% vs 35.5% over 2.6 years).
So I treat my VO2 max as a read on what my body can do today. It can't tell me how long I've got.
Wondering where yours sits? Here's a chart to save. These are treadmill-test numbers, so expect your watch to read a few points lower.

One test measures. Two guess.

The treadmill (44.8) measures. At VOSK I ran with a mask on while a machine measured the oxygen I was using, and it kept getting harder until I couldn't keep going. I was super wiped by the end. This is the gold standard, and the only one of the three that measures oxygen directly.
The watch (37.4) guesses. It looks at my heart rate against my pace on outdoor walks and runs and fills in the rest. The only independent check I could find is a 2025 study of 30 people, where the Apple Watch read about 6 points below a lab treadmill test on average, with a wide range from person to person. Mine read 7.4 below. Which is pretty much what you'd expect.
The blood test (47.3) also guesses. VO Health measures proteins in a blood sample and predicts what a lab test would say. I think this is a really interesting idea! Research in 650 adults and then more than 14,000 shows that blood proteins do track fitness and how people respond to training.
To be fair to all three, they weren't taken on the same day. The blood sample was from July, the watch reading from September, and the treadmill test from October. I can't say how much of the spread is the method and how much is me.
(VOSK and VO Health both provided my tests for free. I have no other financial relationship with either company.)
Then I pulled my old results
Okay, it gets better. I've been testing this since 2022, so I went back through everything.
September 2022, treadmill test: 54.5
February 2024, treadmill test at a different lab: 40.0
October 2025, a blood test (a different model from this year's): 34.3
July 2026, blood test: 47.3
October 2026, treadmill test: 44.8
My watch the whole time: about 32 in 2023, 36 in 2024 and 2025, and 37 to 38 this year
Read that as one story and here's what happened to me. In 2022 I was apparently an athlete. By 2024 I'd lost a quarter of my fitness, and I kept sliding in 2025. Then I staged a 13-point comeback in nine months.
None of that happened! (I would remember.)
The 2022 number never felt right to me. Something about that test day was off, and I've stopped using it as a baseline.
The two blood tests are my favorite part. Different models, nine months apart: 34.3 and 47.3. If I believed both, I got 38% fitter between October and July. I did not.
Now what if I trended the measurements against others using the same method?
Watch: up about 6 points from 2023 to this year
Treadmill (leaving out 2022): 40.0 to 44.8
Even the treadmill was done with different testing labs, so not a perfect trend. But I believe the synthesis tells the real story: I'm fitter than I was three years ago.
You can raise your VO2 max without getting fitter
Okay, now where can VO2 max go wrong?
VO2 max is a fraction. (Stay with me.)
The top is how much oxygen your body uses per minute when you're going all out. The bottom is your body weight. That's the "per kilogram" in mL/kg/min.
So there are two ways to raise it: use more oxygen, or weigh less. Lose 5% of your weight while your heart, lungs, and muscles do exactly what they did before, and your VO2 max goes up about 5%. But in that case, you didn't get any fitter!
I ran this on myself. Between my 2024 and 2026 treadmill tests, I went from 40.0 to 44.8. I also weighed about 3% less at the second test.
If only my weight had changed, I'd have gone from 40.0 to about 41.3
The other 3.5 points are my body actually using more oxygen (about 8% more)
About a quarter of my "improvement" was the scale, and three quarters was the engine. I'll take it! But it's why I look at the raw oxygen number on the report too. And it's why I'd never cut weight (or lose muscle) just to make this number look better.
All this contextual data matters for accuracy! The blood model in my report uses my age, sex, height, and weight on top of the protein markers. Your watch also works from what's in your health profile, so keep that up to date.
The tests disagree by more than training can change
How much can training move your VO2 max? In the classic studies, eight weeks of hard 4-minute intervals, three times a week, raised it about 7% in moderately trained men and in well-trained women. For me, that's about 3 points.
Now look at how far apart the tests are:
Watch vs. treadmill: 6 to 7 points
Blood vs. a lab test: about 10% average error in the company's early study (done on a bike or rower, not a treadmill), which for me is 4 to 5 points
My two blood tests: 13 points
My three treadmill tests, at three labs: 14 points
See the problem? The tests disagree with each other by more than I could gain from months of hard work. If I did the treadmill in October and a blood test in March, a 3-point "gain" could be nothing at all.
So think of your wearable or whichever measurement test as a compass. It can tell you which way you're heading. It can't tell you where you are with great precision.
Pick one test, and only compare it with itself.
My blood says I'm in the 99th percentile. Also the 60th.
Both rankings came out of the same sample, analysis by VO Health.
The 99th is for my estimated VO2 max, compared with women my age in general. The 60th is for a "molecular fitness score," compared with the company's other customers, who are a very fit crowd. (I also couldn't get to the 99th from the reference table the report cites, so I'm treating that one as the company's claim.)
When you see a percentile, ask: compared with whom?
The report also says my "fitness age" is 20. Fun to read (I am indeed no longer 20 years old, chronologically)!
And it gave me ten pathway scores, all somewhat more esoteric and hard to interpret given the research nature of this test.
Before I'd use a blood test to track my training, I'd want to see people tested in a lab and by blood, before and after training, with the two numbers moving together. The validation I could find is one company-funded abstract in 50 unusually fit adults, using a 10-protein model. My report used about 50 proteins, so already it's a different version that can't be completely validated by the published evidence.
I really do want this to work. Not everyone can get to an exercise lab, and a decent estimate from a blood draw would be so useful. I'm rooting for the data.
Which test should you get?
Not exercising regularly? Skip the test. You already know the answer, and the biggest drop in risk comes from getting out of the least fit group. Start walking briskly and build from there.
Have a wearable? Use it as your compass. Don't compare it with a lab number or with your friend's watch. Do notice whether it drifts up or down over months. It only updates from outdoor walks, runs, and hikes, so you have to give it some data to work with.
Want a free test? See how far you can go in 12 minutes on a track, and repeat it every few months. Same track, same conditions.
Get an exercise test if it'll change what you do. Maybe you want a true baseline, or training zones built from your own breathing. Then go back to the same lab for the retest. (Learn from me.)
Blood tests: interesting, and I'll keep following them. If you get one, just know that it's still early although much easier to do than the exercise version.

One caution. If you get chest pain, unusual breathlessness, or lightheadedness when you exert yourself, talk to your doctor before any all-out test.
Want the physiology behind all of this? My VO2 max guide goes deeper.
What I'm doing with mine
Honestly, I'm happy with 44.8. My plan is to keep it, not chase a bigger one.
A confession first. Back in 2024, the lab that measured my 40.0 told me to do two HIIT sessions a week. I did them on and off, and never reliably. My number went up anyway, so I can't even tell you the intervals were the reason!
→ The treadmill is my reference. I'll retest at the same lab, on the same protocol, in about a year. (I've now learned this one three times.)
→ My watch is my compass. If 37 becomes 39 over the winter, great. I still won't believe the 39.
→ I'm committing to one hard interval session a week on top of my lifting and walking. One I'll actually do beats two I won't.
→ I'm not tracking the blood score yet, and I'm not changing anything based on its pathway scores.
What about you? I put together a short guide on my Curator page: how to raise your VO2 max depending on where you're starting. Following is free, and it's where I post shorter notes between Letters. (I'm the co-founder and CEO of Curator.) Get the guide.
My fitness is "excellent." My LDL was still 192.
Those two results came from the same body, and one did nothing for the other.
Fit people have heart attacks too. My cholesterol is mostly genetic, I'm on lipid-lowering medication for it, and no amount of treadmill time was going to do that job. You can't outrun your genes. (Believe me, I've now run with a mask on.)
A great VO2 max tells you what your body can do. It doesn't tell you about your ApoB, your Lp(a), your blood pressure, or what's in your arteries.
If you know your fitness number but not those, check those next. My free heart-test guide walks through which test answers which question.
Quick hits
💪 On tirzepatide, everyone lost about 22 kg. Only the lifters got stronger. In a 40-week trial of 120 adults, people who lifted three times a week added almost 6 kg to their knee extension, while those who didn't lost more than 4 kg. The scale and the scan looked nearly the same in both groups.
🔥 Lowering inflammation didn't lower heart attacks. In the ZEUS trial, an IL-6 blocker cut hsCRP sharply in 6,376 people with heart and kidney disease, and heart events came out at 17.6% vs 17.7%. Serious infections were more common on the drug.
👁️ GLP-1 drugs are getting a warning for a rare cause of sudden vision loss. The FDA found 15 probable cases among 134,253 trial participants, 10 of them on a GLP-1. It's very rare, but sudden, painless vision loss in one eye means an eye doctor that day.
🛑 "Safe to stop your statin at 70" needs the fine print. The STREAM trial called stopping "noninferior," but 12.7% of people who stopped had a heart attack, a stroke, or died, vs 9.2% of those who continued. I wouldn't stop one on this.
🧪 Your NAD may not be falling with age after all. In 1,068 adults, whole-blood NAD showed no clear drop with age in healthy men and rose with age in women. A low result on a blood test doesn't tell you your muscle or brain is running short.
💊 A longevity supplement improved a blood-vessel test. Its maker helped pay for the trial. In a six-month placebo-controlled study, NOVOS Core improved how well arteries widen by 2.9 points over placebo, and cholesterol didn't budge. Only 43 of the 61 people who started were in the main analysis, so I'd want a replication before buying it.
Last week's poll
Last Sunday I asked what you'd do if your calcium score went from 120 to 140 on a statin with your ApoB at goal.
Here's how you voted:
Stay the course: 68%
Get a CCTA to look at the soft plaque: 32%
Add another cholesterol drug: 0%
Rescan next year to check: 0%
Honestly, I'm kind of with the CCTA group, if there was a baseline CCTA to compare with. A score that rises while the drivers are on target usually means the plaque is hardening, which is what we want, so I wouldn't change the treatment. But a second CCTA next to the first one can show whether the soft plaque is shrinking. A repeat calcium score can't.
Without a baseline? I'd stay the course.
Your turn this week.
Which VO2 max number do you actually have?
Talk soon, Hillary Lin, MD
From The Longevity Show
Brooks Leitner is from VO Health, the company behind my blood estimate. We get into what VO2 max is, why strength matters alongside it, and where the blood science stands.
Where to find me
December 12, Las Vegas: I'll be on an AI panel at A4M's LongevityFest, showing what's possible when a clinician actually uses AI every day.
February 2027, Scottsdale: I'll join the faculty at the Science of Skin Longevity Summit for AI in diagnostics and the peptides panel.
March 5–6, 2027, San Francisco: I'll be back at the Livelong Women's Health Summit.
Forwarded this email? Get The Longevity Letter.
Educational content only, not personal medical advice.

