Melatonin Is a Hormone, Not a Vitamin

Melatonin Is a Hormone, Not a Vitamin

A heart-failure signal, a dosing puzzle and unreliable labels: how to think about nightly melatonin in 2026.

Key takeaways

  • Timing may matter as much as milligrams. A 2024 dose-response meta-analysis of 26 trials found the sleep benefit of melatonin peaked around 4 mg/day, and that taking it roughly three hours before bedtime may work better than the usual small dose just before lights-out. The pooled effect is modest and the trials are heterogeneous.
  • A new safety signal is a question, not a verdict. A November 2025 conference abstract linked at least one year of prescribed melatonin for insomnia to more heart failure over five years. It is observational, not yet peer-reviewed, and cannot show that melatonin caused the difference.
  • What is in the bottle is not always what the label says. A 2023 analysis of 25 US melatonin gummies found most were outside 90-110% of the labeled dose.

01 / Why a hormone?

Melatonin is made by the pineal gland when light fades. It does not knock you out. It tells your body that biological night has begun, shifting body temperature, alertness and the timing of other hormones such as cortisol. That is why a hormone sold next to vitamins deserves more care than a vitamin: you are adding a signal to a system that already runs on timing.

02 / Does it work?

For some people, modestly. Cruz-Sanabria and colleagues (2024) pooled double-blind randomized trials in people with insomnia and healthy volunteers. Melatonin gradually shortened the time to fall asleep and lengthened sleep, with the effect peaking near 4 mg/day. The size of the benefit for sleep latency depended on whether participants had insomnia and on how long before sleep the dose was taken. The authors suggest about 3 hours before desired bedtime may outperform the common regimen of 2 mg taken 30 minutes before bed.

Two cautions. First, the timing finding comes from comparing across trials, not from head-to-head testing, so it is a hypothesis worth testing rather than a rule. Second, a 2026 scoping review of systematic reviews (Iyer et al.) exists for readers who want the wider picture; we cite it for completeness and did not extract its figures for this post.

03 / The heart-failure signal

At the American Heart Association Scientific Sessions in November 2025, researchers presented a retrospective analysis of the TriNetX electronic-records network: 130,828 adults with insomnia, split into 65,414 with documented melatonin use for 12 months or more and a matched comparison group. Over five years, new heart failure was recorded in about 4.6% of the melatonin group versus 2.7% of controls (reported hazard ratio about 1.9), heart-failure hospitalization in 19.0% versus 6.6%, and all-cause death in 7.8% versus 4.3%.

The authors themselves list the problems. Over-the-counter melatonin was not captured, so many "unexposed" patients probably took it. The database mixes countries with different prescribing rules. Insomnia severity, depression and anxiety were not measured, and any of them could drive both melatonin use and heart risk. Confidence intervals were not reported in the coverage we reviewed. An independent sleep-health expert quoted by the AHA noted that melatonin is not an approved long-term insomnia treatment in the US. Our reading: worth following up with proper trials, not a reason to panic.

04 / The label problem

In a 2023 JAMA research letter, Cohen and colleagues measured 25 gummy products. Melatonin content ranged from 74% to 347% of the label among the 24 that contained it; 22 of 25 products fell outside 90-110%. One had no detectable melatonin but contained a substantial amount of CBD. The sample was small, one batch per brand, and gummies only, so tablets and capsules may differ. Still, if you are studying "4 mg", you cannot be sure you are taking 4 mg.

Key figures

For clinicians

  • Dose-response meta-analysis (Cruz-Sanabria 2024): 26 double-blind RCTs (1987-2020), 1,689 observations, standardized mean differences for sleep onset latency and total sleep time; peak effect near 4 mg/day. Meta-regression: insomnia status and dosing-to-bedtime interval predicted latency effects; time of day predicted total sleep time. Limitations: heterogeneous populations, cross-trial timing comparisons, effect reported in standardized units, older trials.
  • AHA 2025 abstract (Nnadi et al.): retrospective cohort, TriNetX, propensity-matched 1:1, n = 65,414 per arm, five-year follow-up, mean age about 56, 61% women. Reported HR about 1.89 for incident heart failure (1.82 in a sensitivity analysis requiring two or more fills at least 90 days apart) and about 2.09 for all-cause mortality; CIs not reported in the sources we saw. Limitations: not peer-reviewed, unmeasured confounding by indication, exposure misclassification from OTC use, mixed jurisdictions, no insomnia-severity or psychiatric data, no dose or timing data. Hospitalization rates exceeded diagnosis rates, which suggests coding artifacts.
  • Label accuracy (Cohen 2023): 25 gummy products, one sample each; 22 of 25 outside 90-110% of label.
  • Bottom line: the efficacy evidence supports short-term, modest benefit. Long-term cardiovascular safety is unanswered; a prospective or trial-emulation analysis with active comparators would help. Consider asking patients about non-prescribed melatonin products and doses.

What's new in regulation

Melatonin sold as a supplement in the US is not pre-approved by FDA for efficacy, which is part of why label accuracy varies; in many other jurisdictions it is a prescription medicine. (This general regulatory status is from background knowledge and should be verified before publication.) On the device side, FDA issued a revised General Wellness guidance on January 6, 2026. Per law-firm summaries, it clarifies that non-invasive wearables reporting physiologic values such as heart rate and heart rate variability can fall under the wellness policy if they make no disease claims, do not mimic clinical-grade values, and do not prompt clinical action. It does not cover diagnosing or managing disease. We did not find a melatonin-specific FDA or EU MDR/IVDR update in this check.

What this means for patients

  • Melatonin may help you fall asleep a little faster, particularly if you have trouble with timing, such as jet lag or shift work. It is not a sedative.
  • Chronic insomnia deserves a proper evaluation. Cognitive behavioral therapy for insomnia remains the usual first-line treatment.
  • If you use it nightly for months, tell your doctor, especially if you have heart disease or take other medicines.
  • Choose products that carry independent third-party testing marks, and keep gummies away from children.
  • Do not stop prescribed medicines because of a headline. Talk to your clinician first.

Educational content only. This is not medical advice. Consult a physician about your own health.

About 6Letters

At 6Letters we are building non-invasive, wearable tools aimed at making hormone-related rhythms visible. MOJO KING and MOJO QUEEN are not described here as cleared or approved for any medical use. Learn more at https://6letters.co.kr.

References

  1. Cruz-Sanabria F, Bruno S, Crippa A, et al. Optimizing the time and dose of melatonin as a sleep-promoting drug: a systematic review of randomized controlled trials and dose-response meta-analysis. J Pineal Res. 2024;76(5):e12985. DOI: 10.1111/jpi.12985. PMID: 38888087.
  2. Nnadi E, et al. Effect of long-term melatonin supplementation on incidence of heart failure in patients with insomnia. Abstract presented at AHA Scientific Sessions, November 2025 (AHA abstract MP2306; Circulation 2025;152(Suppl 3):4371606). AHA newsroom; DOI listed: 10.1161/circ.152.suppl_3.4371606.
  3. American College of Cardiology. Research suggests long-term melatonin use for insomnia increases HF risk. November 2025. ACC
  4. Cohen PA, Avula B, Wang YH, Katragunta K, Khan I. Quantity of melatonin and CBD in melatonin gummies sold in the US. JAMA. 2023;329(16):1401-1402. DOI: 10.1001/jama.2023.2296.
  5. Iyer et al. Exogenous melatonin and sleep quality: a scoping review of systematic reviews. J Clin Pharmacol. 2026. DOI: 10.1002/jcph.70115. (Cited for context; figures not extracted.)
  6. Covington & Burling. FDA issues revised guidance on general wellness products. January 2026. Covington