The Longevity Spending Curve

Peter Attia sounded genuinely irritated. Someone had asked him about follistatin gene therapy -- the one where you fly to Honduras, pay $25,000, and get injected with a treatment a Stanford geneticist publicly called "a scam."

"If it was $7, I wouldn't do it. Why? Because there is zero evidence that this works."

Then he listed what actually has near-100% probability of extending your life: high aerobic capacity, muscle mass, physical strength, sleep, social connection. Total cost: roughly nothing.

I've been staring at longevity intervention data for a long time now. And there's a pattern nobody in the longevity industry wants to talk about, because it's bad for business.

The ROI curve on longevity spending is logarithmic. The first $50 a month captures the vast majority of the evidence. Everything after that buys rapidly diminishing returns -- and past a certain point, you're not buying health outcomes at all. You're buying a feeling.

The $0 tier (the one people skip)

The most effective longevity interventions cost nothing. That sounds like generic health advice. It's not. The effect sizes here are larger than any drug or supplement on the market.

Moving from the bottom 25% of cardiovascular fitness to the 50th percentile reduces your all-cause mortality risk by roughly 50%. That's not a typo. A 2025 meta-analysis confirmed it -- and the relationship is dose-dependent. More fitness, less death, all the way up the curve.

Resistance training specifically: a systematic review and meta-analysis found that ~60 minutes per week of strength training was independently associated with lower rates of all-cause, cardiovascular, and cancer-specific mortality. When combined with cardio, the mortality reduction hit 40%.

Sleep is even more striking. A 2025 OHSU study found that sleep predicted life expectancy better than diet, exercise, or loneliness -- every factor except smoking. And sleep regularity mattered more than duration: the least regular sleepers had 20-88% higher all-cause mortality, independent of how many hours they got.

None of this requires a longevity clinic. None of it requires a subscription. None of it requires flying to a Caribbean island.

The $50/month tier (where the evidence lives)

Once you've nailed the free tier -- and I mean actually nailed it, not "I try to get 7 hours" -- there are a handful of pharmacological interventions with real data behind them. They're all generic. They're all cheap.

Rapamycin (~$30-60/month from a compounding pharmacy). 47 independent animal studies. Median 12% lifespan extension in mice. Zero studies showing harm. The NIA Interventions Testing Program validated it. The PEARL trial showed it's well-tolerated in healthy humans at longevity doses. I wrote about this in detail -- it's the most consistent dataset in longevity research.

Metformin (~$4/month generic). The TAME trial -- 3,000 non-diabetic adults, 14 centers, 6 years of metformin vs placebo -- is still underway. We don't have definitive human longevity data yet. But observational studies consistently show diabetics on metformin outliving non-diabetics not on it, which is strange enough to warrant investigation. Metformin decelerates multiple epigenetic aging clocks. It's been prescribed to hundreds of millions of people since the 1950s. The safety profile is one of the most established in medicine.

A daily multivitamin (~$12/month). I know. It sounds absurd. But a Nature Medicine paper published in March 2026 -- the COSMOS trial -- showed that daily Centrum Silver slowed two epigenetic aging clocks (PCGrimAge and PCPhenoAge) by about 4 months over 2 years compared to placebo. Steve Horvath, who developed one of the clocks, called the consistency "exactly what you want to see." He also cautioned it's "not a fountain of youth." Fair. But it's $12.

SGLT2 inhibitors (~$15-30/month generic). Canagliflozin extended mouse lifespan by 14% in the ITP. Human trials show 32% all-cause mortality reduction. A 2024 Nature Aging paper found they clear senescent cells through immune checkpoint stripping. The most human mortality data of any candidate longevity drug. I wrote the full breakdown -- this is the compound the longevity community is sleeping on.

Basic bloodwork (~$32/month amortized). Annual full panels -- ApoB, Lp(a), fasting insulin, hs-CRP, HbA1c, full lipids. Catching a problem early is worth more than any intervention that tries to fix it late. You can get these for $380/year through direct-to-consumer lab services. Your doctor probably isn't ordering half of them. We wrote a full guide to the longevity blood panel with optimal ranges for 25+ markers.

Total for the evidence tier: ~$78-108 per month. That's the whole stack. Everything with published human data or large-scale animal validation supporting a longevity effect.

The $10,000/month tier (where the evidence thins)

This is where longevity medicine turns into longevity theater.

NAD+ IV drips ($750-1,000 per session, typically weekly). NAD+ precursors reliably raise NAD+ levels in blood. That part is real. What's not established: any clinical evidence that higher NAD+ levels translate to longer life or slower aging in humans. A 2026 Genetic Literacy Project review put it bluntly: "Don't waste your money." NMN supplements ($200/month oral) are the cheaper version of the same unproven bet.

Exosome injections ($3,000-15,000 per treatment). No standardization. No quality control. No published human longevity data. The longevity clinics selling these aren't citing evidence -- they're citing demand.

Hyperbaric oxygen therapy ($6,000-15,000/year). One interesting Tel Aviv study showed telomere lengthening. Animal data on cellular repair is promising. But the longevity-specific evidence in humans is thin, and the cost-per-unit-of-evidence is orders of magnitude worse than anything in the $50/month tier.

Follistatin gene therapy ($25,000 one-time). Bryan Johnson flew to Honduras for this. The company's own trial data hasn't been published in a peer-reviewed journal. The unpublished pre-print claimed 11 years of epigenetic age reversal. A Stanford geneticist reviewed it and used the word "scam." Animal data shows muscle mass increases. Human longevity data: zero.

And then there's the full Bryan Johnson experience. Blueprint Immortals: $1 million per year. Includes a 30-person medical team, continuous monitoring, gene therapies, stem cells, every experimental intervention available. An independent analysis concluded you could capture 90% of the protocol's value for under $2,000 a year -- the lifestyle changes, basic bloodwork, and a handful of generic prescriptions.

The other $998,000 buys the medical team and the experimental stuff with limited evidence.

Why the curve is logarithmic

There's a structural reason the returns diminish so fast.

The interventions with the strongest evidence are the ones that have been around the longest, studied the most, and prescribed to the most people. They're generic. They're boring. They're cheap. Exercise has been studied in millions of subjects. Metformin has been prescribed to hundreds of millions. Rapamycin has 47 independent animal studies and a growing body of human data.

The expensive interventions are expensive precisely because they're new and unproven. Gene therapy costs $25,000 because there's one company doing it in Honduras. Exosomes cost $10,000 because there's no generic market and no standardization. The price reflects the novelty, not the evidence.

This is the same pattern you see in every immature market. The premium goes to the novel, not the effective. And the people paying the premium convince themselves the price signals quality. In finance, we call this adverse selection. In longevity, we call it a wellness retreat.

The uncomfortable question

Here's what the $100K/year longevity optimizers don't want to hear: the gap between their protocol and the $50/month protocol, in terms of expected lifespan impact, is probably small. Maybe very small.

Exercise, sleep, and basic pharmacological interventions address the primary drivers of aging-related mortality: cardiovascular disease, metabolic dysfunction, cancer risk, and chronic inflammation. The expensive stuff mostly targets secondary and tertiary mechanisms with much smaller effect sizes -- when it has demonstrated effect sizes at all.

The honest framework looks like this:

The table tells the story. The evidence clusters at the bottom of the cost curve and thins out as prices climb.

What this actually means for you

If you're spending $200/month on NMN and you don't strength train three times a week, you've got the hierarchy backwards.

If you're considering a $15,000 longevity clinic membership and you haven't had your ApoB checked, you're optimizing the wrong variable.

If you're impressed by someone's longevity protocol because it costs $2 million a year -- ask what the evidence says, not what the price tag says. Those are different questions, and in this market, the answers point in opposite directions.

The boring stuff works. The expensive stuff might. Sequence accordingly.