The mechanism case for NAD+ is one of the strongest in modern aging biology. The outcomes case, the part that matters when a real person asks whether this will do anything for them, is much thinner. Most NAD+ articles cite mechanism findings as if they were outcomes. They are not. This article reads the actual human trial record, names the studies, and is honest about where the data is solid, where it’s mixed, and where the field is still extrapolating from rodents.
For the mechanism walkthrough (sirtuins, PARP, CD38, mitochondrial energy), see the NAD+ guide. For the route-of-administration comparison (subcutaneous vs IV vs oral), see NAD+ injections. This page is about what the published human outcome data does and does not show.
The clinical-outcomes question most NAD+ articles dodge
Read ten popular NAD+ articles and you will see the same claims: more energy, better cognition, slower aging, faster recovery. You will rarely see those claims tied to a specific randomized controlled trial in humans. That is the gap this article exists to close.
The mechanism story is real. NAD+ levels fall with age in human tissue. Sirtuins, PARPs, and CD38 all consume NAD+. Mitochondrial output depends on NAD+ cycling. Animal models show that raising NAD+ can rescue age-related dysfunction in mice. None of that is in dispute.
The harder question: when researchers have given NAD+ precursors (or direct NAD+) to humans in controlled trials, measured clinical or near-clinical endpoints, and compared against placebo, what have they found? The answer is more nuanced than the marketing suggests.
- Mechanism is not outcome. “Sirtuin activation extends lifespan in yeast” is a mechanism finding. “Twelve weeks of NR improved walking distance in adults over 65” would be an outcome finding. Most NAD+ marketing blends the two.
- Pharmacokinetics is not biology. Raising blood NAD+ metabolites confirms a precursor was absorbed. It does not, by itself, prove tissue NAD+ rose, or that rising tissue NAD+ produced clinical change.
- Most published human NAD+ trials are small and short. The bulk of the data sits in 8 to 12 week trials with 20 to 140 participants. Larger, longer trials are underway.
Energy and fatigue: what controlled human trials measured
More energy is the most common claim made for NAD+ therapy, and it is the area where the gap between mechanism enthusiasm and outcome data is widest. Three trials anchor what we currently know.
Martens 2018: NR pharmacokinetics and vascular function in older adults
Martens and colleagues, in Nature Communications (2018; 9:1286), reported a randomized, double-blind, placebo-controlled crossover trial of nicotinamide riboside (NR) at 1,000 mg per day for six weeks in 24 healthy middle-aged and older adults (ages 55 to 79). The headline result: NR was well tolerated, and it raised whole blood NAD+ by roughly 60%. That is a clean pharmacokinetic finding.
What it did not show was a robust energy or fatigue benefit. The trial was not powered to find one. It measured biomarkers (NAD+ metabolome, blood pressure, arterial stiffness) rather than patient-reported energy. Subgroup analysis suggested a possible reduction in systolic blood pressure and aortic stiffness in participants with elevated baseline values, a real signal worth replicating, but not a fatigue endpoint.
Dollerup 2018: NR in obese, insulin-resistant men
Dollerup and colleagues, in the American Journal of Clinical Nutrition(2018; 108(2):343–353), reported a 12-week randomized, placebo-controlled trial of NR at 1,000 mg twice daily in 40 obese, insulin-resistant men. Endpoints were clinically meaningful: insulin sensitivity (gold-standard hyperinsulinemic-euglycemic clamp), body composition, and lipid mobilization.
The result, honestly, was largely null. NR raised NAD+ metabolites but did not improve insulin sensitivity, beta-cell function, hepatic fat, or body composition compared to placebo. This is one of the most carefully designed NAD+ precursor trials ever conducted, and it did not find the metabolic benefits the mechanism story predicted. A truthful read of the field has to weight Dollerup heavily.
Conze 2019: long-term NR safety and metabolism
Conze, Brenner, and Kruger, in Scientific Reports (2019; 9:9772), reported an eight-week trial of NR (NIAGEN) at 100, 300, or 1,000 mg per day in 140 healthy overweight adults. This is the largest published RCT of an NAD+ precursor to date, and its primary aim was safety and dose-response pharmacokinetics. NR was well tolerated at all three doses, raised whole blood NAD+ in a dose-dependent fashion, and produced no serious adverse events. Conze 2019 is important precisely because it does not over-claim: it establishes safety and predictable PK so larger efficacy trials can proceed on solid footing.
The honest read on energy
Across the controlled human trial record, NAD+ precursors have not produced large, replicated improvements in measured energy expenditure, fatigue scales, or physical performance in healthy adults. They have produced reliable increases in blood NAD+ metabolites and an acceptable safety profile. Anyone telling you the literature shows a clean “NAD+ supplementation reduces fatigue” result is not reading the same papers.
Cognitive function: where mechanism is strong but outcomes are scarce
NAD+ and cognition is a textbook mechanism-outcome gap. Neurons are highly metabolically active. Mitochondrial dysfunction is tightly linked to age-related cognitive decline. The mechanistic case is excellent. The clinical-outcome data in healthy adults is, in 2026, quite thin.
- No large, completed RCT has shown NAD+ precursor supplementation improves validated cognitive testing scores in healthy middle-aged or older adults.
- Parkinson’s disease is one neurological area where preliminary human work is more developed (the NR-SAFE trial; the larger N-DOSE program). Early data suggest NR raises cerebral NAD+ in patients and is well tolerated. Whether that translates to symptom or progression effects is what larger trials are designed to answer.
- Healthy adult cognition remains the area where the literature is weakest. The intuitive “NAD+ for mental clarity” pitch is exactly the population where rigorous outcome data is hardest to find.
The translation gap is not a hidden conspiracy. It reflects how hard it is to demonstrate cognitive benefit in healthy people: small effect sizes, ceiling effects on testing, high placebo response, and expensive trial durations. Meaningful outcome data on NAD+ and healthy-adult cognition will likely take another five to ten years to mature.
Cardiometabolic findings: the most defensible benefits area
If there is a single area where the human outcome data lines up with the mechanism story most cleanly, it is cardiometabolic markers in middle-aged and older adults. The signals are not large, and they are not consistent across every trial, but the area is where evidence is least disappointing relative to mechanistic prediction.
Vascular function and arterial stiffness
The Martens 2018 trial is the most cited single finding. In the subgroup with elevated baseline systolic blood pressure or aortic pulse-wave velocity, NR appeared to reduce both modestly. The mechanism, NAD+-dependent support of endothelial function and SIRT1-mediated nitric oxide signaling, is biologically plausible. Replication in a larger trial is the key open question, because subgroup findings in small RCTs are well known to inflate.
Insulin sensitivity: the Yoshino versus Dollerup contrast
Yoshino and colleagues, in Science (2021), reported that 10 weeks of NMN at 250 mg per day improved muscle insulin sensitivity in prediabetic postmenopausal women. The effect was real and statistically significant, though muscle NAD+ content did not measurably change with supplementation. Dollerup 2018, on the other hand, found no insulin-sensitivity benefit of NR in obese men, despite a longer trial and higher dose. The discrepancy could reflect the precursor used (NMN vs NR), the population, the dose, or the assay. The point is that the insulin-sensitivity story is mixed, not uniformly positive.
Why this category is the most defensible
Cardiometabolic markers earn the most-defensible label because three things line up: the mechanism is well-characterized (SIRT1, eNOS, mitochondrial efficiency), the trial endpoints are objective (blood pressure, pulse-wave velocity, insulin sensitivity), and the population in which signals appear (middle-aged and older adults with mildly elevated baseline values) matches the population where age-related NAD+ decline is real. Even here, effect sizes are modest. The literature does not support NAD+ as a substitute for blood-pressure medications, statins, or metformin in patients with established disease.
Athletic performance and recovery: hype versus evidence
If a podcast convinced you NAD+ is a performance-enhancing intervention for athletes, the published trial record does not back that up.
- Dolopikou and colleagues (2020) reported that NR raised NAD+ metabolites in young and older recreationally active adults but did not improve maximal exercise performance or VO2max.
- Stocks and colleagues (2021) found that NR did not enhance training adaptations in young men doing endurance training, despite raising blood NAD+ metabolites.
- Older-adult mobility (gait speed, six-minute walk distance) is the area where the rationale is strongest and where ongoing trials are positioned. Data is preliminary.
In young, healthy, well-trained adults, the published outcome data does not support NAD+ as a performance enhancer. In older adults with declining mobility, the signal is more plausible but the data is early. The takeaway is not that NAD+ does nothing for athletes. It is that the high-volume claim space outruns the published evidence by a wide margin.
The biological-age question
Epigenetic age clocks are one of the more fraught endpoints in the NAD+ literature. The Horvath clock, introduced in Genome Biology (2013), estimates biological age from DNA methylation patterns at defined CpG sites. Second-generation clocks (PhenoAge, GrimAge, DunedinPACE) correlate more strongly with mortality and age-related disease incidence than first-generation clocks.
A small number of NAD+ precursor trials have included clock endpoints, mostly as exploratory secondary outcomes. The general picture: short trials at standard doses have not produced consistent, large reductions in biological age compared to placebo. Multi-component interventions that include NAD+ precursors alongside other ingredients have reported clock-age reductions, but isolating the NAD+ contribution is generally not possible in those designs.
Methodological caveats matter. Blood-derived methylation reflects immune-cell composition, not systemic biological age. Test-retest noise is real: a 1.5-year clock-age reduction in 12 weeks may sit inside the assay’s noise floor. Reducing a clock estimate is not the same as slowing the underlying aging biology. Clocks are correlational tools, not direct measures of mortality risk in an individual. Anyone framing NAD+ therapy as proven to reverse biological age on the basis of small-n clock trials is overstating what the methodology supports.
The pharmacokinetic ceiling: why delivery method changes the outcomes question
Most published human NAD+ outcome trials use oral precursors (NR or NMN). Oral precursors raise blood NAD+ metabolites reliably. Whether they raise tissue NAD+ enough to produce the effects predicted by mechanism work is the rate-limiting question for the field.
- Tissue uptake is the bottleneck. Oral NR and NMN are absorbed and metabolized rapidly. Most of the rise in blood metabolites reflects intermediates rather than direct delivery of NAD+ into the cells where it works. Studies that have measured intracellular NAD+ in muscle (Yoshino 2021) show tissue-level repletion is partial, slow, and variable across tissues.
- Subcutaneous and IV NAD+ have different PK profiles. Direct administration bypasses first-pass metabolism. The key open question is whether that translates to higher or more durable intracellular NAD+ in target tissues. Outcome trials directly comparing oral precursors to subcutaneous or IV NAD+ in the same population are still rare.
Two practical implications. First, a null result in an oral NR trial does not necessarily mean “NAD+ doesn’t work for this endpoint.” It may mean “1,000 mg of oral NR for 12 weeks did not raise tissue NAD+ enough to produce the predicted effect.” Those are different statements, and the field has not always distinguished them carefully. Second, when patient reports under direct NAD+ protocols describe effects that the published precursor trials have not demonstrated, the discrepancy is not automatically marketing. It can reflect a real PK difference. It can also reflect placebo, expectancy, or unmeasured co-interventions.
Where to set expectations
If you are considering NAD+ therapy, the published outcomes literature supports a specific set of expectations.
Strongest signals. Cardiometabolic markers in middle-aged and older adults: vascular function, blood pressure in mildly hypertensive subgroups, possibly muscle insulin sensitivity in selected populations.
Plausible signals, awaiting larger trials. Mobility and functional capacity in older adults. Adjunctive support in early Parkinson’s. Subjective recovery and sleep quality under supervised protocols (lower-quality evidence, but consistent enough to track).
Weak signals. Cognitive enhancement in healthy adults. Performance enhancement in young trained athletes. Large measurable reductions in epigenetic age over short trial durations. Lipid and inflammatory marker improvements as primary outcomes.
A reasonable framing for someone considering NAD+ therapy: the mechanism justifies clinical interest, particularly for adults over 45 with cardiometabolic risk markers. The outcomes data is real but modest, and it does not yet support the more aggressive longevity claims that dominate consumer marketing. A clinician-supervised protocol that names this honestly and tracks objective markers (blood pressure, fasting glucose, lipid panel, sleep quality, energy) over 8 to 12 weeks gives you the best chance of distinguishing real response from placebo in your own case.
Compounded NAD+ formulations are not FDA-approved drugs. Through PepScribe, they are prepared in the USA by licensed 503A pharmacies under clinician prescription, not 503B, not international. No hidden overseas supply chain.