NAD+ Benefits: Graded by Organism and Tier

Lists of NAD+ benefits are unusually easy to assemble and unusually hard to read, because the underlying literature is large, active and mostly about something other than the product being sold. Every claim on this page therefore carries two labels rather than one.

The first label is the organism: human, rat, mouse or cell culture. Every article on this site applies it. The second label is specific to this compound and is the one that decides most arguments: which intervention was administered, an oral precursor or NAD+ itself given parenterally.

NAD+ benefits: every claim needs two labels, not one

Consider the sentence “NAD+ has been studied extensively in humans.” It is true of the abbreviation and misleading about the vial. The extensive human work overwhelmingly administered nicotinamide riboside or nicotinamide mononucleotide by mouth. Those are precursor molecules that the body converts. The subject of that sentence, as a reader will understand it, was not the subject of those studies.

Now consider the same problem in reverse. Saying the compound has no human evidence would also be wrong, because precursor trials in humans genuinely exist and genuinely report results. The accurate statement is narrower and more useful: a substantial human literature exists on oral precursors, and a thin one exists on the injected coenzyme.

The column most benefit lists leave out

Adding the intervention column changes what a list looks like. Claims that appeared to rest on human trials turn out to rest on human trials of a different molecule given by a different route. Claims that appeared to rest on mechanism turn out to rest on mechanism plus rodent work.

Neither correction makes the underlying science bad. Precursor trials are real trials, often well conducted, and the biochemistry of NAD metabolism is genuinely interesting. The correction only stops the evidence from being credited to the wrong product.

The reason the column goes missing is structural rather than conspiratorial. Reviews and press summaries write about NAD biology, because that is the coherent scientific subject. Vendors write about a vial. A reader moving between the two encounters the same abbreviation on both sides and has no prompt to ask which molecule went into which person by which route. The label has to be added deliberately or it will not appear.

The age decline observation, and what it licenses

One claim underwrites most of the others, so it is worth isolating. Concentrations of NAD+ have been reported to fall with age, measured in rodent tissues and in human tissue and blood samples. Taken at face value, that is a descriptive finding about natural history in the species measured.

It licenses a hypothesis: that restoring the pool might reverse something associated with the decline. It does not license the conclusion, and the distinction matters because the hypothesis is often presented as though the observation had already tested it. Many biological quantities change with age without being the cause of anything, and reversing a marker does not automatically reverse whatever the marker accompanies. Only an interventional study with a clinical outcome can settle which of those descriptions is right, and the observational data cannot adjudicate it in either direction.

Raising a blood marker is a pharmacodynamic result

The most reproducible human finding in this field is that oral nicotinamide riboside raises NAD+ concentrations measured in blood. Multiple trials in humans have reported it, and it is about as replicated as anything in the supplement literature.

It is worth being exact about what that establishes. It establishes that the compound is absorbed, that it reaches the measured compartment, and that the assay can detect the change. It is a pharmacodynamic result: the intervention did something to a marker.

It does not establish that concentrations rose in muscle, liver, brain or any other tissue where a benefit is proposed, because blood is not those tissues. It does not establish that a higher pool produces more sirtuin activity, or that more sirtuin activity produces an outcome a person would notice. Each of those links is a separate empirical claim, and the chain from a blood measurement to a clinical benefit is longer than most summaries admit.

Grading the individual claims

Two features of that table deserve attention. The rows with a human organism almost all have a precursor in the intervention column. The rows naming the injected product almost all sit in the uncontrolled report tier. That pattern is the honest summary of this compound’s evidence base, and no rearrangement of the claims changes it.

Where the injected route sits on this ladder

Infusion services report that clients feel better afterwards, and some of those reports are undoubtedly sincere. They are also collected under conditions that guarantee a favourable answer: the client paid, waited in a chair for an hour or more, expected an effect, and was asked afterwards by the party that supplied it. No comparison group exists, no blinding is possible, and nobody records the clients who felt nothing and did not return.

There is a further mechanistic caution specific to the route. Extracellular enzymes degrade NAD+ outside cells, and the fragments produced are themselves salvage substrates. Whether an infusion delivers intact coenzyme into cells, or functions as a rapid precursor delivery route, is unresolved in the published literature. If the latter is closer to the truth, the injected product’s distinguishing claim over an oral precursor is weaker than advertised while the evidence supporting it remains thinner.

What would let a row change columns

Moving a claim from the lower tiers to the top requires a specific and unglamorous design. Participants assigned at random to injected NAD+ or to a matched placebo infusion. Outcomes fixed in writing before enrolment, measured with instruments validated for the outcome rather than with a satisfaction question. Assessors blinded. Dropouts accounted for in the analysis rather than quietly excluded. Results published where a reader can retrieve them.

Nothing in that list is exotic. It is the ordinary apparatus of clinical research, and for the injected form of this compound it has largely not been applied. Until it is, the benefits attributed to an infusion remain claims about a marker, about a precursor, about a rodent, or about how someone felt in a chair.

It is worth noting what would not move a row. More infusion clinics opening does not. A larger volume of client testimonials does not, because adding uncontrolled observations to uncontrolled observations produces a bigger uncontrolled dataset rather than a controlled one. Another mouse study does not move a human row, however elegant its design, since it can only strengthen the animal row it already belongs to. And a further precursor trial, whatever it finds, strengthens or weakens the precursor rows and leaves the injected ones exactly where they were.