Your NAD Levels Went Up.
Did Your Health Improve?
What human trials of NR, NMN and NAD actually show about energy, exercise, metabolic health and the brain.
Evidence sources & editorial review
A selective narrative review of human trials, with animal research used for context. Prepared by PED Evidence with AI assistance; independent clinician review has not been completed. Research exposures describe trials, not a personal dosing protocol. Review standards.
A supplement can change your biology without changing how you feel, perform or age. NAD research is a particularly useful example: biochemical effects are real, but the health outcomes depend on the trial, the population and the endpoint.
There are reasons for genuine interest. Some studies report better walking performance, muscle insulin sensitivity or exercise capacity under specific conditions. Others show a substantial rise in NAD with no clear improvement in the outcome participants would actually care about. The useful question is therefore more specific than “Does NAD work?” It is: which intervention, for whom, and for what result?[3][6][9][11]
Three different questions
Blood or tissue NAD and related metabolites.
Demonstrated in multiple trialsWalking, cognition, insulin sensitivity or exercise capacity.
Mixed; population-specificLess disability, fewer major diseases or longer survival.
Not established by these trialsWhat does an NAD increase actually mean?
Why the biology attracted attention
Nicotinamide adenine dinucleotide is involved in energy metabolism and serves as a substrate for enzymes involved in cellular maintenance. NR, or nicotinamide riboside, and NMN, or nicotinamide mononucleotide, are precursors used in NAD biosynthesis. They are related interventions, not interchangeable names for the same product.[1][2][3]
In a long-term mouse experiment, NMN improved several age-associated physiological measures, including aspects of energy metabolism and insulin sensitivity. Those findings helped justify human testing. They did not establish that supplementing a healthy human would extend life. An animal model can identify a plausible pathway while leaving the size, durability and clinical relevance of the human effect unresolved.[1]
The pathway is plausible. The outcome still needs testing.
More measured NAD must translate into a useful change in a relevant tissue—and then into a meaningful outcome.
A blood result is not a whole-body readout
Consider a crossover trial in 12 older men given oral NR, 1 g daily, for 21 days. Whole-blood NAD was 47.75 μM during NR versus 20.90 μM during placebo. Muscle NAD-related metabolites changed, but muscle NAD+ itself did not significantly increase. Mitochondrial bioenergetics and handgrip did not improve.[2]
This is why “boosted the NAD metabolome” requires careful reading. It may refer to precursors, breakdown products or a sample compartment rather than a uniform rise in NAD inside all cells. A striking blood result is evidence of biological exposure; it cannot substitute for a muscle-performance test. Read the aged-muscle study card.
One trial, two very different messages
Does it improve energy, exercise or mobility?
A positive signal in peripheral artery disease
The NICE trial randomized 90 people with peripheral artery disease to NR, NR plus resveratrol, or placebo for six months. The tested NR exposure was 1,000 mg daily; the combination added 125 mg resveratrol. Six-minute walking distance favored NR alone by 17.6 metres over placebo.[6]
Walking distance: the comparison matters
Distance declined
Distance increased
Advantage for NR
This is a functional finding in people with impaired circulation, not proof of an endurance advantage in healthy athletes. Resveratrol did not add benefit. The modest sample and statistical threshold make replication important; the larger 31-m estimate from adherent participants was post hoc, not the primary randomized comparison. Diarrhea was reported in 39.3% with NR alone versus 27.6% with placebo.[6][17]
Runners: thresholds are not VO₂max
A six-week study randomized 48 amateur runners to placebo or NMN at 300, 600 or 1,200 mg daily, alongside exercise training. Some ventilatory-threshold measures improved more at the medium and higher exposures, but VO₂max, peak power and body composition did not show corresponding benefits. No adverse events were reported.[5]
That distinction matters. A change in the workload at which breathing patterns shift does not mean maximal aerobic capacity increased or race times improved. With 12 people per group, this is an interesting performance signal that needs a larger, longer replication. Read the runner study card.
Older adults: read the primary endpoint first
In 60 older adults, 250 mg NMN daily for 12 weeks increased blood NAD and favored selected walking and sleep measures. But the prespecified primary stepping test showed no significant between-group difference. No treatment-related adverse effects were reported in the abstract.[7]
A positive title can sit above a null primary outcome
No significant difference
Selected results favored NMN
Increased
Read the walking and sleep study card. Secondary outcomes are not worthless. They simply answer a different evidentiary question, especially when many outcomes were tested.
Insulin sensitivity is promising. Fat loss is another claim.
The NMN trial in prediabetic women
Twenty-five postmenopausal women with overweight or obesity and prediabetes completed a 10-week randomized trial of 250 mg NMN daily or placebo. Insulin-stimulated glucose disposal increased by 25 ± 7% within the NMN group, with no corresponding change after placebo. The trial reported a significant group-by-time-by-condition interaction.[3]
What the 25% number does—and does not—mean
Measured using a metabolic clamp after 10 weeks.
The metabolic-clamp result is biologically meaningful, but body composition and routine metabolic measures did not improve relative to placebo. No adverse events were reported. This small, selected population does not establish a general anti-obesity effect, and improved muscle insulin signaling should not be relabeled as muscle hypertrophy.[3]
NR did not reproduce the same metabolic pattern
In a different trial, 40 sedentary men with obesity received NR at 1,000 mg twice daily or placebo for 12 weeks. NR did not improve insulin sensitivity, resting energy expenditure, lipid oxidation or body composition.[4]
It would be tempting to conclude “NMN beats NR.” These were not head-to-head trials: sex, metabolic state, compound, exposure and study methods differed. The correct conclusion is narrower—one NMN trial found a specific metabolic benefit that a separate NR trial did not show. A comparative trial is needed to attribute that difference to the molecule.
Brain health: target engagement versus clinical improvement
Parkinson’s disease: a rationale worth pursuing
NADPARK randomized 30 newly diagnosed, untreated patients to 1,000 mg NR daily or placebo for 30 days. Brain NAD rose variably, with metabolic changes and mild clinical improvement associated with participants who showed an increase. NR was well tolerated.[8]
A small phase I study can show that an intervention reaches its biological target. It cannot establish that Parkinson’s progression slows. The larger NOPARK registry was checked for this article; posted efficacy results were not located. Read the NADPARK card.[16]
Cognition: the newer negative results matter
A 2026 randomized pilot in amnestic mild cognitive impairment provides a direct test of the article’s central question. Forty-two participants completed 12 weeks. Blood NAD doubled in the NR group, but the primary cognitive outcome did not improve; neither did total cerebral blood flow or blood pressure. Exploratory regional blood-flow signals should not be presented as demonstrated memory improvement.[9]
Similarly, a 58-person long-COVID trial tested 2,000 mg NR daily with a placebo lead-in design. NAD rose approximately 2.6–3.1-fold after 5–10 weeks in the initial NR group, but between-group differences in cognition, fatigue, sleep and mood were not significant. Favorable post-hoc changes from baseline were exploratory, unadjusted for multiple testing, and complicated by substantial dropout.[10]
The biomarker–benefit gap
Primary cognition: no improvement demonstrated over 12 weeks.
Cognition and symptoms: no significant advantage over placebo.
A 2026 exercise result deserves attention, too
In Friedreich’s ataxia, 66 participants were randomized to NR, exercise, both, or control for 12 weeks. The NR exposure was weight-based, 300–900 mg daily. Combination therapy improved peak oxygen uptake versus control by 0.21 L/min (95% CI 0.05–0.36; adjusted P=.0299). NR alone did not show a significant advantage.[11]
This is encouraging in a specific neurological disease. The combination-versus-control result does not, by itself, prove that NR adds benefit beyond exercise or establishes synergy. It also does not supply a performance forecast for healthy lifters. The comparison must match the claim.
Oral supplements and NAD infusions have different evidence
“NAD therapy” can describe oral NR, oral NMN, oral formulations of NAD, intravenous NAD, intravenous NR or other injected preparations. Changing the compound and route changes the intervention. An oral precursor trial cannot establish the benefits of a wellness infusion.
Whose evidence belongs to which intervention?
Repeated biomarker effects; mixed functional and metabolic outcomes.
Pharmacokinetic and tolerability studies, plus disease-specific trials.
Do not assume subcutaneous, intramuscular or proprietary oral products reproduce another intervention’s results.
What infusion studies actually tested
A 2019 pilot tracked metabolites during a six-hour, 750-mg NAD infusion: eight men received NAD and three saline. That design examined disposition and metabolism, not whether treatment improves fatigue, memory or longevity.[12]
It would nevertheless be inaccurate to say direct NAD has no randomized human research. A single-centre trial in 180 people with ischemic cardiomyopathy tested 10 mg IV NAD daily for seven days alongside heart-failure treatment. At one month, mean left-ventricular ejection fraction was 45.44% versus 42.44% with placebo (P=.024). The six-month composite event result—14.6% versus 24.7%, P=.089—was not statistically significant.[13]
That is a disease-specific cardiac finding at a particular exposure. It cannot validate a much larger wellness drip in healthy people or establish improved survival. The study was not powered to settle clinical-event differences.[13]
Tolerability is part of the story
A 2026 retrospective comparison of commercial IV NAD and IV NR reported moderate-to-severe gastrointestinal symptoms, increased heart rate and chest pressure during NAD infusions. Symptoms resolved after infusion completion, but the observational design and short follow-up limit what it can establish. Its exploratory biomarker findings are not proof of rejuvenation.[14]
New oral NAD formulations also need outcome testing. A 2026 five-day randomized trial of LNAD+ enrolled 60 adults, with 50 in the primary analysis. It demonstrated biochemical activity, but no secondary clinical, wellbeing or wearable endpoint survived multiplicity correction. Five days can inform early pharmacology; it cannot establish healthy aging.[15]
What does this mean for healthy aging?
The studies reviewed here support a real biological effect and several worthwhile therapeutic hypotheses. They do not establish longer human lifespan, broad rejuvenation or a reliable increase in everyday energy for healthy users. Nor does an unsuccessful short trial eliminate every possible benefit in a different condition.
The distinction is between a research hypothesis and a dependable expectation. A person with a specific metabolic or neurological disorder may respond differently from a healthy athlete. A blood biomarker may move before a functional outcome does—or move without that outcome ever improving. Both possibilities require testing.
The human-evidence scorecard
Short-term tolerance is not lifetime safety
Several small oral trials reported good short-term tolerability. That is useful information, but the sample sizes and durations cannot exclude uncommon harms or establish the safety of years of use. Infusion tolerability must be evaluated separately. “No adverse events reported” in a brief trial should retain its exact scope.[3][5][7][8][14]
How to read the next NAD headline
Start with the endpoint before looking at the percentage. Was it blood NAD, a metabolite, a functional test or a clinical event? Then ask whether the improvement exceeded placebo, whether it was primary or exploratory, and whether the tested population resembles the person the advertisement is addressing.
Four checks before accepting the claim
- Same intervention? Match molecule, formulation and route.
- Same population? A disease-specific result is not a healthy-athlete result.
- Right comparison? Improvement from baseline is not automatically an advantage over placebo.
- Meaningful outcome? A biomarker change is a starting point, not the finish line.
The most defensible position is interested but specific. NAD biology is a serious research area. The clearest benefits should be described at the level at which they were demonstrated, while broader anti-aging promises remain questions for future trials.
Sources & assessment notes
Evidence checked October 6, 2026. This is a selective narrative review, not an exhaustive systematic review. Original human trial reports and indexed abstracts were prioritized; full-text access is identified below. Where full text was unavailable, claims are limited to the abstract or accessible indexed primary-report material. Funding and product-linked interests occur in this literature and should be considered alongside design and replication; inclusion here is not an endorsement. Figures are original explanatory graphics based on cited results. No unreported error bars or inferred clinical benefits have been added.
- Mills et al. (2016). Long-term NMN administration and age-associated physiological decline in mice.
Primary animal research; contextual evidence. - Elhassan et al. (2019). NR and the aged human skeletal-muscle NAD metabolome.
Full-text assessment; PMID 31412242. - Yoshino et al. (2021). NMN and muscle insulin sensitivity in prediabetic women.
Full-text assessment; PMID 33888596. - Dollerup et al. (2018). NR in men with obesity: insulin sensitivity and lipid metabolism.
Abstract assessment. - Liao et al. (2021). NMN supplementation and aerobic capacity in amateur runners.
Abstract and primary-report assessment; DOI 10.1186/s12970-021-00442-4. - McDermott et al. (2024). The NICE randomized trial in peripheral artery disease.
Primary report and indexed results; PMID 38871717. - Morifuji et al. (2024). NMN, walking function and sleep in older adults.
Abstract assessment; PMID 38789831. - Brakedal et al. (2022). NADPARK: a randomized phase I trial in Parkinson’s disease.
Abstract assessment; DOI 10.1016/j.cmet.2022.02.001. - Martens et al. (2026). NR in amnestic mild cognitive impairment.
Abstract assessment; DOI 10.1002/alz.71605. - Wu et al. (2025). NR, cognition and symptom recovery in long COVID.
Abstract assessment; DOI 10.1016/j.eclinm.2025.103633. - Exercise and NR in Friedreich’s ataxia (2026). Randomized factorial trial.
Abstract and indexed methods; DOI 10.1016/S1474-4422(26)00082-7. - Grant et al. (2019). Plasma and urine NAD metabolome during a six-hour IV NAD infusion.
Primary pharmacokinetic pilot; PMID 31572171. - Yu et al. (online 2025; issue 2026). IV NAD in ischemic cardiomyopathy.
Full-text assessment; PMID 40954388. - Reyna et al. (2026). IV NAD versus IV NR: retrospective tolerability pilot.
Abstract assessment; DOI 10.3389/fragi.2026.1652582. - RENEWAL-NAD+ (2026). Oral LNAD+ and whole-blood intracellular NAD.
Abstract assessment; DOI 10.1007/s11357-026-02399-1. - NOPARK study registry.
Registry checked October 6, 2026; no posted results located. - Publisher correction to NICE (2024).
Restored the omitted trial-registration identifier in the abstract; no outcome correction.