Research methods · Putting it to work

Vetting the product

You did the hard part. You walked the headline back to the study, checked it was a real trial, confirmed the effect was solid and the dose was 5 g a day. The evidence holds. So you pick up a bottle — and it says "proprietary blend, 500 mg." You know almost nothing now. You can't tell how much of the active ingredient is in there (the blend hides it), and 500 mg total is already a fraction of the 5 g that was studied. The science was sound. The product is a guess. That gap is the whole lesson.

Why gate two is real

In the US, supplements need no pre-market approval for safety or effectiveness. There's no agency checking that the bottle contains what the label says before it reaches the shelf — the manufacturer is responsible for its own quality, and most problems surface only after the fact. So the label is a marketing surface, not a guarantee. Independent testing has repeatedly turned up products that were under-dosed, over-dosed, or contained ingredients that weren't on the label at all; in some analyses of a given category, only a minority of products actually matched their own label. A sound claim can still be a bad purchase.

1. Is it third-party tested?

Look for an independent seal: NSF Certified for Sport, USP Verified, Informed Sport. These mean an outside lab confirmed the bottle's contents match its label and screened for contaminants. No seal means the label is an unverified marketing claim — you're trusting the seller's own word that 500 mg of X is really in there. Given how often independent testing finds mismatches, that trust is unearned. The seal is the cheapest signal you have; start here.

2. Does the dose and form match the study?

Two products can both list a big number and deliver wildly different amounts of the thing that matters. Check the elemental content, not the compound weight. A label boasting "2,000 mg magnesium L-threonate" sounds generous, but the threonate is most of that weight — it delivers only about 144 mg of elemental magnesium. If the study used 300 mg of elemental magnesium, this bottle is at roughly half the dose and can't be expected to reproduce the effect. Same trap with the form: if a trial used a specific salt, ester, or extract standardized to a certain percentage, a cheaper form at the same milligram count isn't the same thing.

*Elemental percentages vary by salt; the point is to read past the headline milligrams to what reaches you.

3. Treat a proprietary blend as a red flag

A "proprietary blend" lists the total weight of several ingredients but hides how much of each. That's usually a sign the active is under-dosed — if a brand were putting in the studied amount, they'd want to show it. You can't vet what you can't see. Watch for the tell "clinically studied blend": it almost always means one ingredient in the mix was studied on its own, not the blend as sold. The study doesn't transfer to the product.

4. Screen for interactions

This is the gate that protects your body, not your wallet. Flag anything that shares a metabolic pathway with your medications — many drugs and supplements are processed by the same liver enzymes (e.g. CYP3A4) or transporters (e.g. P-gp), and competing for the same pathway can push a drug's level up or down. Also flag anything with an overlapping physiological effect: a blood-pressure-lowering supplement on top of a blood-pressure drug, or a blood-thinning one on top of an anticoagulant, where the effects add up. The risk is additive. Route anything that touches a prescription through your prescriber or pharmacist before you start.

5. Change one thing at a time

Start a single new product in isolation, with nothing else changing. If you begin three at once and feel better — or worse — you can't tell which one did it, or whether anything did. One change at a time turns your own body into a clean little experiment, which is exactly the n-of-1 idea: the only trial whose subject is you.

Work one, then finish one

Worked: "creatine monohydrate, 5 g per day." Run the gates. Gate one: the evidence for creatine is among the strongest in the field. Gate two: it's cheap and single-ingredient, so there's no blend to hide anything; look for a Creapure or third-party mark to confirm purity; the studied dose is about 5 g of elemental creatine and the bottle says 5 g of creatine monohydrate (which is nearly all creatine), so the dose matches; interactions are minimal for most people. It clears both gates. This is what an easy buy looks like.

Your turn: vet a nootropic "memory blend, 800 mg proprietary" with no third-party seal. Does it pass? (No — it fails gate two on three counts: the per-ingredient doses are hidden inside the blend, there's no independent verification of what's actually in the bottle, and "blend" means the mix as sold was never the thing studied. Skip it regardless of how good any single ingredient's evidence looks — gate one can't rescue a product that fails gate two.)

Why this matters

This is where good evidence dies on the shelf. You can do everything right appraising the science — walk the headline back, check the trial, confirm the dose — and still hand over money for an under-dosed, unverified, or interacting product, because the bottle was never bound by the study. Picture yourself holding two magnesium products: one is "2,000 mg L-threonate, proprietary sleep blend, no seal," the other is "200 mg elemental magnesium glycinate, USP Verified, single ingredient." The first sounds stronger and is the worse buy. The second is the one that can actually deliver what the research promised. Gate two is the difference between paying for a result and paying for a label.