NAD+ Benefits: The Claims, Against the Numbers

Sorting by intervention rather than by claim changes the picture more than any argument about study quality does.

One row on that table has genuine human data behind it, and it is the least dramatic one. Oral nicotinamide riboside has been shown in human trials to raise blood NAD levels. That is a biomarker moving in people who swallowed a precursor. It is not the same statement as an injected vial producing an outcome someone can feel, and the distance between those two sentences is where most of the marketing lives.

Here the site's usual method earns its place. A benefit claim that names no mass cannot be checked at all, and a benefit claim that names a mass can be checked for internal consistency even when the biology is unsettled.

Take two claims of the form that circulates, using assumed durations so the arithmetic can run. Neither is a recommendation and neither is quoted from a real seller.

The first two rows differ by a factor of two in total mass and are identical in rate. The third differs from the first by a factor of twenty in mass and by orders of magnitude in rate, and it is routinely discussed under the same benefit heading as the other two. When a page says NAD+ helped with something, the word is covering three arithmetically unrelated exposures.

That is a checkable defect in a claim, and it needs no biology to spot. It is also the reason a report with no mass, no concentration and no duration cannot be compared with anything, including itself a month later.

It is worth writing down what a benefit claim would have to contain before it could even be disagreed with, because the list is short and almost nothing in circulation meets it.

It would name the intervention exactly: injected NAD+, or oral nicotinamide riboside, or oral nicotinamide mononucleotide, since those are three different things and the human evidence is distributed very unevenly across them. It would name the organism, because a result in mice and a result in people are not interchangeable and the distinction is free to state. It would name a mass, which for an injected preparation means a labelled vial mass and a fill volume, since without both the mass is unrecoverable. It would name what was measured and who measured it, separating a blood level from a reported feeling. And it would name a comparator, because an outcome with nothing to compare against is a description rather than a finding.

Run the table at the top of this page against those five requirements. The energy and clarity row fails on all five. The mitochondrial row names an organism and fails on intervention transfer, since mouse and cell culture work does not carry to an injected human amount. The blood level row is the only one that passes, and what it passes with is a biomarker in humans given an oral precursor.

That is the honest summary of the benefits question, and it is a much smaller statement than the size of the literature suggests.

The strongest human result in the area is a measurement of a level, not a measurement of a life. Blood NAD rising after an oral precursor is a real, replicated observation in people. Whether that rise produces more energy, better cognition or slower ageing is a separate question with a separate evidence requirement, and the marketing habit of treating the first as proof of the second is the single most common error in this topic.

The distinction has a practical edge. A biomarker result gives no dose response curve for a different route, so it cannot be used to argue that more injected material means more benefit. It licenses one sentence: an oral precursor raised a level in humans. Everything past that sentence is a further claim that needs its own support.

Arithmetic can settle a small, genuinely useful set of things here, and it should be honest about the size of that set.

It can compute concentration from a label and a fill volume. On a 500 mg vial in 5 mL that is 100 mg per mL, so one unit on a U-100 syringe holds 1 mg and a 50 mg portion is 50 units, half a barrel. It can compute cost per 100 mg: for a vial priced at P, that is P divided by five, and for a 1000 mg vial at Q it is Q divided by ten, so the larger container wins on shelf price only when Q is below twice P. It can compute how many portions a vial yields, which for 500 mg drawn in 50 mg portions is ten.

It cannot compute whether any of those portions does anything. Not because the arithmetic is weak, but because the input it would need, an established relationship between an injected mass and an outcome in people, does not exist to be divided by. Nor is there an approved product to fall back on: no regulator anywhere has issued a marketing authorisation for injected NAD+, so there is no assessed label carrying a benefit statement, and the price arithmetic above is priced against a labelled mass that is a seller's claim rather than an assay result.

Most compounds covered on this site have thin human evidence, so the honest summary is that little is known. NAD+ is the opposite shape of problem. A great deal is known, and most of it is about something else: an oral precursor, in a capsule, converted internally, studied in humans under conditions that share a molecule with the vial and share almost nothing else.

That means a reader can be misled by accurate sentences. Every individual statement in a marketing paragraph can be true of nicotinamide riboside in humans and none of it true of the injected product being sold. The defence is mechanical rather than clever: for every claim, ask which intervention was tested and in which organism, and write both down before the claim is allowed to count. Most of the list does not survive that step, and the part that does turns out to be a level in the blood rather than a benefit in a person.