Home > Articles > Does NAD+ decline with age? What human measurements show

Does NAD+ decline with age? What human measurements show

October 22, 2026

Prepared by: Research Desk

Prepared by: Research Desk. Sources are linked for verification.

"NAD+ levels halve by middle age" is one of the most repeated lines in longevity marketing. The idea has a real scientific basis, but the evidence is narrower than the slogan. Most of the clearest data come from rodents. Human measurements exist, but they come from different tissues, use different assays and are often small. This explainer sets out what has been measured in people, what has been inferred from animals, and why a single headline number doesn't hold up.

For the wider biochemistry of NAD+, including how precursors feed into it, see our NAD+ guide.

The short answer

Reviews of the field agree that NAD+ levels fall with age in several tissues in animal models. Human data point in the same direction for some tissues, but how much NAD+ falls depends heavily on which tissue was sampled and how it was measured [1][2]. A 2020 review devoted to age-related NAD+ decline described the human evidence as more variable than the rodent evidence. It noted that the size of the reported change differs between studies and tissues [1].

So the accurate statement is: age-associated NAD+ decline has been reported in some human tissues; its size and its consequences are still being worked out.

What has actually been measured in human tissue

Skin

One of the most cited human datasets is a 2012 study of donated human skin tissue. Researchers measured NAD+ alongside markers of oxidative damage and NAD+-consuming enzyme activity across donors of different ages [3]. They reported age-associated changes in NAD+ metabolism. Lower NAD+ was seen with increasing age, together with greater oxidative damage and higher activity of PARP, an NAD+-consuming DNA-repair enzyme [3].

This study is useful because it measured human tissue directly rather than extrapolating from mice. It also has clear limits:

It's cross-sectional. It compares different people of different ages, not the same people over time.
It covers one tissue. Skin results can't be assumed to apply to muscle, brain or liver.
Sample size is modest. That limits how precisely the slope of decline can be estimated.

Other tissues

Reviews summarising human studies describe reports of lower NAD+ with age in tissues such as skeletal muscle and brain. They also describe other measurements, particularly in blood, where age effects are smaller or inconsistent [1][4]. A review of cardiac NAD+ metabolism reports altered NAD+ handling in aged and diseased hearts. Much of that mechanistic evidence comes from animal models rather than direct human heart measurements [7].

The pattern across reviews is consistent: the decline is tissue-specific, not a single body-wide number [1][2][8].

Blood: a 2026 multi-cohort study that found no age difference

Blood is where the human picture is most contested, and a 2026 study added a significant data point. Trętowicz and colleagues, a group based mainly at Amsterdam UMC, measured NAD+ in whole blood across seven independent human cohorts [14]. They used an ultra-high-performance liquid chromatography and high-resolution mass spectrometry method that the authors describe as rigorously validated to account for real-world analytical variability [14]. They reported that whole-blood NAD+ stayed remarkably stable with age and across the lifestyle interventions they examined. Levels did change in response to nicotinamide riboside, which the authors note was expected [14]. Their conclusion is that blood NAD+ is of questionable use as a biomarker of ageing or lifestyle factors [14].

This sits alongside earlier blood work that pointed the other way. A 2019 cross-sectional study of plasma from healthy adults aged 20 to 87, measured by liquid chromatography and tandem mass spectrometry, reported lower plasma NAD+, NADP+ and NAAD at older ages, with higher NADH, NADPH and nicotinamide [15].

The two findings are not directly comparable, and neither settles the question on its own:

Different compartments. The 2026 study measured whole blood, which is dominated by red blood cells. The 2019 study measured plasma, the fluid outside the cells [14][15].
Different scale. The 2026 study drew on seven cohorts; the 2019 study was a single cross-sectional sample [14][15].
Blood is not tissue. A stable whole-blood value does not rule out age-related changes in skin, muscle or brain, where declines have been reported [1][3].

What the 2026 result does challenge is the habit of reading a blood NAD+ number as a measure of "biological age". It fits the cautious reading set out above: any age-related decline appears to depend heavily on tissue, compartment and method.

Why measurements disagree

When two studies report different "NAD+ levels", they often aren't measuring the same thing. Several technical factors drive the disagreement [1][4]:

Tissue and cell type. Whole blood is dominated by red blood cells, whose NAD+ pool may not reflect what happens in muscle or liver. Tissue biopsies are more direct but invasive, so human tissue datasets tend to be small.
What is counted. Some assays report NAD+ alone. Others report total NAD (NAD+ plus NADH), the NAD+/NADH ratio, or related metabolites. These numbers can't be compared directly.
Sample handling. NAD+ and NADH can interconvert and degrade after collection. The time to freezing and the extraction method can shift results.
Assay method. Enzymatic cycling assays and mass-spectrometry methods differ in specificity. Values from one platform can't simply be read against reference ranges from another.
Study design. Cross-sectional comparisons between young and old groups can be confounded by health status, activity and diet. Long-term studies following the same people over years are scarce.

This is why our site handles any single "NAD+ level" figure with caution, whether it appears in a paper, an advert or a test report.

Why NAD+ might fall: the proposed mechanisms

Researchers describe NAD+ concentration as a balance between synthesis and consumption [2][4]. Age-related decline could come from either side.

More consumption

CD38. CD38 is an NAD+-degrading enzyme found on immune cells. A 2020 study, mainly in mice, reported that senescent cells promote the build-up of CD38-positive macrophages in tissues, and that this was linked to tissue NAD+ decline during ageing [5]. This is a mechanistic finding from animal work. It hasn't been directly established as the main driver in humans.
PARPs. Accumulating DNA damage activates PARP enzymes, which consume NAD+. The human skin study reported higher PARP activity alongside lower NAD+ in older tissue [3].

Less synthesis

The salvage pathway. Most NAD+ is recycled from nicotinamide through the salvage pathway. Reviews describe evidence, largely from animal studies, that activity in this pathway falls in some aged tissues [6][2].

Downstream consequences

NAD+ is required by sirtuins, enzymes that regulate metabolism and gene expression. A long-standing hypothesis holds that falling NAD+ reduces sirtuin activity and contributes to age-related changes [6]. This idea comes largely from animal and cell studies. Whether it explains specific features of human ageing is still an open research question [2][8].

Precursor trials: what they measured, and what they didn't

Precursor trials are often cited as indirect evidence about age-related decline, so it's worth being precise about what they show.

A randomized, placebo-controlled crossover study of nicotinamide riboside (NR) in healthy middle-aged and older adults reported that NR was well tolerated and raised NAD+ in blood [9].
A randomized, placebo-controlled trial of nicotinamide mononucleotide (NMN) in women with prediabetes reported an increase in muscle insulin sensitivity. This was a small, single-population study [10].

Both studies measured specific biomarkers in defined groups. Neither establishes that age-related NAD+ decline causes a given outcome in humans. Neither shows that raising a blood NAD+ number changes ageing. Reviews of NMN and NR biology emphasise this gap between rodent data and human trials [11]. A 2026 systematic review assessed the preclinical and clinical evidence for NAD+ supplementation separately, and the two bodies of evidence answer different questions [12]. A 2024 overview of human trials of candidate anti-ageing compounds likewise frames NAD+ precursors as an area of ongoing testing, not settled findings [13].

The key principle is that a biomarker change is not the same as a health outcome. For how these products are classified in the U.S., see our page on the legal status of NMN, NR and IV NAD+.

How to read an "NAD+ declines with age" claim

When a claim about age-related NAD+ decline appears in an article, advert or study summary, these questions help place it:

Which tissue? Skin, muscle, brain and blood can behave differently.
Human or animal? Many widely quoted percentages come from rodent studies.
Cross-sectional or longitudinal? Comparing age groups is weaker than following the same people over time.
Which analyte and assay? NAD+, total NAD and the NAD+/NADH ratio are different measurements.
Who was studied? Results from older adults with a specific condition may not generalise.
Is a mechanism being presented as an outcome? Lower NAD+ in a tissue sample is a measurement, not a demonstrated cause of ageing.

Tools for comparing NAD+ tests

At-home NAD+ blood tests, usually based on a finger-prick dried blood spot, are sold to people curious about this marker. They can be a reasonable way to track a number over time on a consistent method. Their value depends on transparency. When comparing kits, buyers can weigh these criteria:

Analyte reported: NAD+ alone, total NAD, or a ratio, stated clearly on the report.
Sample type: whole blood dried spot versus other formats, and whether results reflect blood only (they do).
Method disclosure: whether the lab names its assay platform and publishes validation or precision data.
Reference ranges: where they come from, and whether they are age-stratified.
Laboratory accreditation: whether the testing lab states recognised quality accreditation.
Price transparency: the full cost of a single test versus repeat-test bundles, with no hidden subscription.
Data privacy: how results and samples are stored, shared or retained.

Broader biomarker panels can add context such as standard metabolic markers, but they don't measure NAD+ unless stated.

A blood NAD+ result describes one blood compartment on one day using one method. It doesn't measure NAD+ in muscle, brain or skin, and it isn't a diagnosis of anything.

Frequently asked questions

Does NAD+ decline with age in every tissue?

Not uniformly. Human studies have reported lower NAD+ with age in some tissues, such as skin [1][3]. In blood the picture is mixed: a 2019 plasma study reported lower NAD+ at older ages [15], while a 2026 study of seven cohorts found whole-blood NAD+ stable across age [14]. Reviews describe the decline as tissue-specific rather than body-wide [2][4].

Is the "NAD+ halves by middle age" figure from human studies?

Widely repeated percentages often trace back to animal data or to single tissues, not to a consistent human-wide finding [1]. The size of reported change in people varies with tissue, analyte and assay, so no single percentage is well supported.

Does a home NAD+ test show NAD+ levels throughout the body?

No. These tests measure NAD+ (or related metabolites) in a blood sample, which is dominated by red blood cells and may not mirror other tissues [1][4]. A 2026 multi-cohort study also found that whole-blood NAD+ did not differ with age [14]. Results are most interpretable when the same method is used repeatedly and the lab discloses what it measures.

How we prepared this page

This explainer summarises peer-reviewed studies and reviews only. It reports what each study measured, without making recommendations. Our sourcing and disclosure approach is described on the About page.

Last updated: 20 October 2026.

Related on this site

Not medical advice. This page reports published research and regulatory records for information only.

Research-only notice: Content on this site is provided for research and educational purposes only and is not intended for human use. It is not medical advice.

Translation notice: Some content on this site may have been machine-translated using Google Translate and may contain errors.