"NAD+" on a supplement shelf almost never means NAD+. The molecule itself is large, breaks down in the gut and is sold mainly as an infusion or an injection. What people swallow is one of four precursors the body turns into NAD+: nicotinamide riboside (the Niagen ingredient), nicotinamide mononucleotide (NMN), nicotinamide (also called niacinamide), and niacin (nicotinic acid). Each has its own trials, and their side-effect records are not interchangeable. This page sets them out one by one from the trials themselves, and names what none of them measured. It reports; it does not advise, and a decision about any of these, or about a symptom, belongs with a clinician.
What happens during an NAD+ drip or after an NAD+ injection is a separate record, covered on our NAD+ injection side effects page; how the injected and swallowed forms compare on what they do to blood levels is on our NAD+ injection vs oral page. The amounts sources describe are on our NAD+ dosage page. How brands appear on this site is on our disclosure page.
The record at a glance
| Form | Largest trial read | Duration | What the trial reported on side effects |
|---|---|---|---|
| Nicotinamide riboside | NOPARK, 410 people with early Parkinson's disease, 1,000 mg a day | 52 weeks | Serious adverse events 8.3% vs 14.1% on placebo; clinical scores slightly worse than placebo |
| Nicotinamide riboside | Conze 2019, 140 overweight adults, 100–1,000 mg a day | 8 weeks | No flushing; no significant difference from placebo |
| NMN | Three small trials, 250–2,000 mg a day | 2–24 weeks | No safety signal reported (detail on our NAD+ vs NMN page) |
| Nicotinamide | ENDIT, 552 relatives of people with type 1 diabetes, 1.2 g/m² a day | 5 years | Serious adverse events did not differ from placebo |
| Niacin (extended release) | HPS2-THRIVE, 25,673 adults, 2 g a day with laropiprant | Median 3.9 years | More serious diabetes, infection, bleeding, gut, muscle and skin events than placebo |
Nicotinamide riboside
Nicotinamide riboside is the precursor with the most modern trial record, much of it paid for by the company that sells it.
The safety trial. In 2019, an 8-week trial randomised 140 healthy, overweight adults aged 40 to 60 to placebo or 100, 300 or 1,000 mg of nicotinamide riboside a day (Conze et al., Scientific Reports 2019). Blood NAD+ rose by 22%, 51% and 142% at the three doses. The authors reported no flushing and no significant difference in adverse events between any group and placebo, and no rise in LDL cholesterol. Participants recorded adverse events in a diary. The paper states that ChromaDex funded the study and that two of its authors were ChromaDex employees.
Older adults. A crossover trial gave 30 healthy adults aged 55 to 79 (NCT02921659) nicotinamide riboside 500 mg twice a day and placebo for six weeks each (Martens et al., Nature Communications 2018). Seven of the 30 reported 14 adverse events between them, all mild. During the supplement phase they were nausea, flushing, leg cramps and increased bruising; during the placebo phase, headache, skin rash, flushing, fainting and drowsiness. Two people left the study over side effects, both while on placebo. The supplement was supplied by ChromaDex.
Higher dose, men with obesity. Forty men aged 40 to 70 with a body-mass index above 30 received 1,000 mg twice a day (2,000 mg a day) or placebo for 12 weeks (Dollerup et al., American Journal of Clinical Nutrition 2018). The authors observed no serious adverse events due to the supplement and normal safety blood tests. Nor did it improve insulin sensitivity, the outcome the trial was built to test.
The longest trial, and the one that went the wrong way. NOPARK, published in JAMA on 8 October 2026, randomised 410 people in Norway with Parkinson's disease diagnosed in the previous two years to nicotinamide riboside 500 mg twice daily or placebo for 52 weeks (Brakedal et al., JAMA 2026; NCT03568968). It followed a 30-patient, 30-day phase 1 study in which the supplement was well tolerated and raised brain NAD+ (NADPARK, Cell Metabolism 2022). The year-long result, as the abstract reports it:
- the total Parkinson's rating score (MDS-UPDRS, 0 to 236, higher is worse) rose by 2.45 points on nicotinamide riboside and fell by 0.27 on placebo — a difference of 2.72 points favouring placebo (95% CI 0.47 to 4.98; P = .02). The trial's own threshold for a clinically important worsening was 5.2 points;
- non-motor symptom burden worsened more on nicotinamide riboside (NMSS difference 4.22 points; P = .009);
- serious adverse events occurred in 17 of 205 people (8.3%) on nicotinamide riboside and 29 of 205 (14.1%) on placebo.
The authors concluded that the findings do not support nicotinamide riboside as a disease-modifying treatment for Parkinson's disease. The abstract does not explain the worse scores, and this page has not read the full paper; the result belongs to people with Parkinson's disease and is not a measurement in healthy users. It is, however, the largest and longest randomised record of the supplement among the trials read for this page.
NMN
NMN's trials are small and short: the three set out on our NAD+ vs NMN page ran two to 24 weeks at 250 to 2,000 mg a day, and several were funded or staffed by manufacturers. None reported a safety signal. The trial-by-trial table is on that page and is not repeated here; the one trial that gave NMN and nicotinamide riboside to the same study population is set out on NMN vs NR. In openFDA's adverse-event database (data updated 28 October 2026), a search on 12 November 2026 returned 12 reports with "NMN" in the product name, against 8 on 30 September.
Nicotinamide
Nicotinamide is the form of vitamin B3 found in most multivitamins and in skin-cancer prevention trials. It does not cause niacin's flush.
- ONTRAC randomised 386 people who had had at least two non-melanoma skin cancers in five years to 500 mg twice daily or placebo for 12 months (Chen et al., NEJM 2015). New skin cancers fell by 23%, and the authors found no noteworthy between-group differences in the number or types of adverse events.
- The transplant trial used the same dose in 158 organ-transplant recipients and was stopped early for poor recruitment (Allen et al., NEJM 2023). It found no effect on skin cancers (207 vs 210), and adverse events and blood and urine tests were similar in the two groups.
- ENDIT gave 552 relatives of people with type 1 diabetes a modified-release dose of 1.2 g per square metre of body surface or placebo for five years (Gale et al., Lancet 2004). It did not prevent diabetes; serious adverse events did not differ, and growth in children was unaffected.
The caution in this record comes from very high doses. A 2000 review in Diabetologia reported reversible liver toxicity in animals and humans at very high doses, infrequent minor liver-enzyme abnormalities at the doses used for diabetes prevention, and some reports of mild insulin resistance, and said high-dose nicotinamide "should still, however, be considered as a drug with toxic potential at adult doses in excess of 3 gm/day" (Knip et al., Diabetologia 2000).
Niacin
Niacin is also an FDA-approved prescription drug, for cholesterol, which makes it the one precursor on this page with a labelled adverse-reaction table. A current extended-release niacin label (Amneal, DailyMed, published 18 September 2026) reports:
- flushing — warmth, redness, itching or tingling — as the most common reaction, "reported by as many as 88% of patients" in placebo-controlled trials; 6% (14 of 245) discontinued because of it in the pivotal studies. Spontaneous reports describe flushing with dizziness, a fast heartbeat, shortness of breath, sweating or chills, which "in rare cases may lead to syncope";
- diarrhoea, nausea, vomiting, increased cough and itching as the other reactions above 5% and above placebo, in a trial database of 402 patients;
- liver-enzyme rises, with fewer than 1% (2 of 245) stopping for transaminase elevations above twice the upper limit of normal, and liver tests required throughout treatment.
The largest outcome trial of niacin went further. In HPS2-THRIVE, 25,673 adults with vascular disease took 2 g of extended-release niacin with laropiprant or placebo for a median of 3.9 years (HPS2-THRIVE Collaborative Group, NEJM 2014). Major vascular events did not fall (13.2% vs 13.7%). Serious adverse events rose: disturbances of diabetes control by an absolute 3.7 percentage points, new diabetes by 1.3, infections by 1.4, gastrointestinal events by 1.0, bleeding by 0.7 and skin events by 0.3. Those doses are drug doses in people with heart disease, many times what a vitamin pill holds; the result is the clearest evidence that an NAD+ precursor at pharmacological doses carries measurable harm.
What FDA's database holds
A search of openFDA's adverse-event database on 12 November 2026 (data updated 28 October 2026) returned 42 reports with "nicotinamide riboside" in the product name, 18 of them naming "nicotinamide riboside chloride", and 12 naming "Niagen". These searches overlap and are not added together. The database does not establish that the product caused the event, and supplement users rarely report, so the counts are not rates. On ClinicalTrials.gov the same day, 126 studies listed nicotinamide riboside as an intervention; 17 had posted results.
What is not established
- Long-term use in healthy people. Apart from NOPARK (Parkinson's disease, one year) and the nicotinamide trials, the precursor trials last 2 to 12 weeks.
- Why NOPARK went the way it did. The abstract reports the outcome, not the cause; the full paper and any follow-up analysis should settle more than this page can.
- Who paid. Several nicotinamide riboside and NMN trials were funded or supplied by the ingredient's maker, which the papers state; independent replication is thinner than the number of trials suggests.
- What the capsule contains. Trials used specified ingredients. A supplement's label is not verified by FDA before sale, and none of these trials tested a retail product.
- Cancer. No human trial on this page was designed or long enough to measure cancer risk; the nicotinamide skin-cancer trials measured prevention, not harm.
Who makes the decisions
Niacin at cholesterol-lowering doses is a prescription drug that needs liver monitoring. The other precursors are sold as supplements, but people with liver disease, diabetes, Parkinson's disease or a history of cancer, and anyone taking other medicines, have specific reasons to take the question to a clinician or pharmacist. This page reports what trials recorded; it does not tell anyone what to use.
Sources and dates
- Brakedal B et al. Nicotinamide riboside for patients with early-stage Parkinson disease: the NOPARK randomized clinical trial. JAMA. Published online 8 October 2026. PMID 42848371, doi 10.1001/jama.2026.18828. Abstract read 12 November 2026.
- Brakedal B et al. The NADPARK study. Cell Metab. 2022. PMID 35235774. Abstract read 12 November 2026.
- Conze D, Brenner C, Kruger CL. Safety and metabolism of long-term administration of NIAGEN in a randomized, double-blind, placebo-controlled clinical trial of healthy overweight adults. Sci Rep. 2019. PMID 31278280; full text (PMC6611812) read 12 November 2026, including the competing-interests statement.
- Martens CR et al. Chronic nicotinamide riboside supplementation is well-tolerated and elevates NAD+ in healthy middle-aged and older adults. Nat Commun. 2018. PMID 29599478; full text (PMC5876407) read 12 November 2026.
- Dollerup OL et al. Am J Clin Nutr. 2018. PMID 29992272. Abstract read 12 November 2026.
- Chen AC et al. A phase 3 randomized trial of nicotinamide for skin-cancer chemoprevention. N Engl J Med. 2015. PMID 26488693. Abstract read 12 November 2026.
- Allen et al. Nicotinamide for skin-cancer chemoprevention in transplant recipients. N Engl J Med. 2023. PMID 36856616. Abstract read 12 November 2026.
- Gale EAM et al. (ENDIT Group). Lancet. 2004. PMID 15043959. Abstract read 12 November 2026.
- Knip M et al. Safety of high-dose nicotinamide: a review. Diabetologia. 2000. PMID 11126400. Abstract read 12 November 2026.
- HPS2-THRIVE Collaborative Group. Effects of extended-release niacin with laropiprant in high-risk patients. N Engl J Med. 2014. PMID 25014686. Abstract read 12 November 2026.
- Niacin extended-release tablets, Amneal Pharmaceuticals, US label, DailyMed, published 18 September 2026. Sections 6.1 and 5 read 12 November 2026.
- openFDA drug adverse-event endpoint,
patient.drug.medicinalproduct= "NICOTINAMIDE RIBOSIDE" (42), "NICOTINAMIDE RIBOSIDE CHLORIDE" (18), "NIAGEN" (12), "NMN" (12); data last updated 2026-10-28; queried 12 November 2026. - ClinicalTrials.gov API v2, intervention "nicotinamide riboside": 126 studies, 17 with results; queried 12 November 2026.
