Melatonin Is a 0.3 mg Supplement, Not a 10 mg One: The Dose-Response Curve the Industry Walked Past

Pick up three bottles of melatonin off the same shelf. The first is a 3 mg tablet. The second is a 5 mg gummy. The third is a 10 mg fast-release lozenge. The bottles all market themselves on the same claims — sleep onset, sleep quality, jet-lag recovery, shift-work adjustment. None of the three bottles is at the dose the original human-trial work demonstrated produces a clean physiologic sleep signal. That dose, established in MIT-conducted dose-response studies in the mid-1990s and replicated in independent labs since, is roughly 0.3 mg. The 3, 5, and 10 mg products are 10-33x the trial-validated physiologic dose. The labels are not wrong about the milligrams in the capsule; they are wrong about the implied claim that more milligrams produce more benefit.

The mechanism is not subtle. Endogenous melatonin secretion from the pineal gland in a healthy adult produces plasma concentrations of roughly 10-60 picograms per milliliter at the nighttime peak, with substantial individual variation. A 0.3 mg oral dose raises plasma melatonin to the upper end of this physiologic range — around 60-100 pg/mL — and the increase tracks the natural circadian rhythm closely. A 3 mg dose raises plasma melatonin to roughly 1,000 pg/mL. A 10 mg dose pushes plasma melatonin into the 3,000-10,000 pg/mL range, depending on individual absorption and clearance kinetics. The bottle dose is producing plasma melatonin concentrations 30-100x what the body would ever produce endogenously, and the supraphysiologic concentration persists into the morning when endogenous melatonin should have fallen to near-undetectable levels.

The downstream problems with the supraphysiologic dose are predictable and reproducible. Next-morning plasma melatonin remains elevated, producing the residual grogginess that 5-10 mg users routinely report. Core body temperature regulation, which depends on falling melatonin in the morning, is blunted, producing the slow-to-warm sensation. Melatonin receptor density and sensitivity downregulate with sustained supraphysiologic exposure, producing the tolerance pattern where users find the same dose less effective over time and escalate. And the dose-response curve for sleep-onset latency and sleep architecture peaks somewhere in the 0.3-1 mg range, with no additional benefit — and some measurable cost — at higher doses.

The industry standardized on the wrong end of the dose-response curve because the marketing math is simpler at higher doses. More milligrams looks like more product. The consumer expectation that "more is stronger" maps onto a non-monotonic dose-response that physiology does not honor. This post is the dose-response math, the supraphysiologic-overshoot problem, the trial doses that actually moved sleep onset and quality, and the label-and-form patterns that distinguish a physiologic-dose melatonin product from the gummies and 10 mg tablets that dominate the aisle.

The endogenous baseline the supplement is supposed to mimic

Melatonin is secreted by the pineal gland in response to darkness, with a clear circadian rhythm. Plasma melatonin concentrations are low or undetectable during daylight hours, begin to rise 2-3 hours before habitual bedtime, peak in the middle of the night, and fall to baseline before morning. The nighttime peak in a healthy adult is typically 10-60 pg/mL, with substantial variation by age, season, light exposure history, and individual physiology. Children and adolescents produce higher nighttime peaks (often 60-150 pg/mL). Older adults produce lower peaks (sometimes 5-20 pg/mL by age 70+), which is part of the rationale for the age-related decline in sleep consolidation that melatonin supplementation is often pitched against.

The supplement question is what plasma melatonin concentration produces the sleep signal. The answer the trial literature converges on is the physiologic range — restoring plasma melatonin to what a healthy young adult produces endogenously — not the supraphysiologic range. The 0.3 mg dose, taken 30-60 minutes before intended sleep, produces a plasma rise that lands inside the physiologic range and tracks the natural circadian fall-off through the night. The 3-10 mg doses produce a plasma rise that overshoots the physiologic range by 1-2 orders of magnitude and persists into the morning hours when endogenous melatonin should have fallen.

The dose-response question is not "how much melatonin do I need to take to fall asleep faster." It is "how much exogenous melatonin do I need to restore plasma melatonin to the concentration the body would have produced if the endogenous signal were intact." The 0.3 mg answer is the answer to the second question. The 5-10 mg products are answering the wrong question.

The dose-response trials the industry walked past

The dose-response work that established the physiologic-dose paradigm is largely from Richard Wurtman's MIT lab from the mid-1990s through the early 2000s, with replication and extension by independent investigators since. The key findings, condensed:

0.3 mg produces the cleanest plasma-concentration match to physiologic nighttime melatonin and improves sleep-onset latency and sleep efficiency in older adults with documented age-related melatonin decline. The 0.3 mg dose taken 30-60 minutes before intended sleep restores plasma melatonin to the upper end of the young-adult physiologic range and produces measurable improvements on sleep-onset latency, sleep efficiency, and self-reported sleep quality in trials lasting 2-12 weeks.

1 mg produces a plasma rise slightly above the physiologic range, with sleep effects similar to 0.3 mg. The 1 mg dose is also commonly cited in the physiologic-dose literature, and the trade-off between 0.3 and 1 mg is small. Both fall within the range where the dose-response curve is rising; both produce the sleep signal; both clear by morning.

3-10 mg doses produce supraphysiologic plasma concentrations (1,000-10,000 pg/mL) that do not produce additional sleep-onset benefit and that persist into the morning. Trials directly comparing 0.3 mg vs higher doses show no statistically significant additional benefit at higher doses on sleep-onset latency or sleep efficiency. Higher doses do produce measurable increases in next-morning residual sleepiness, slower morning core-temperature rise, and (in some users) headache and vivid dreams.

The 0.3 mg dose is most effective in populations with documented endogenous melatonin deficiency. The strongest sleep-onset effects appear in older adults with low endogenous nighttime melatonin, in shift workers with disrupted circadian rhythm, and in jet-lag scenarios where the goal is to phase-shift the circadian system in a new time zone. The dose-response signal is weakest in young adults with intact endogenous melatonin secretion — the supplement is restoring what the body is already producing, and the marginal benefit is small.

What 3-10 mg products actually do

The supraphysiologic dose produces a different pharmacologic effect than the physiologic dose, and the difference matters.

Sedation, not circadian phase shift. A 5-10 mg dose can produce a sedative effect through direct activation of melatonin MT1 and MT2 receptors at concentrations above the physiologic range, and through some GABA-system interaction at the highest doses. The user falls asleep because the supplement is acting as a sedative, not because the supplement restored the natural circadian melatonin signal. The next-morning grogginess is the predictable consequence of the sedative effect that has not cleared.

Persistence into morning hours. Melatonin half-life is roughly 40-60 minutes in healthy adults. A 0.3 mg dose taken at 10 PM produces a plasma peak around 11 PM and falls back to near-endogenous levels by 4-5 AM, when natural cortisol rise should begin. A 10 mg dose taken at 10 PM produces a plasma peak in the same hour but at 30-100x the concentration, and the half-life clearance leaves plasma melatonin still well above physiologic levels at 6-8 AM. The next-morning core-temperature rise is blunted, alertness is impaired, and the user reports feeling like she "took something."

Receptor downregulation with sustained use. Melatonin receptors, like most G-protein-coupled receptors, downregulate in density and signaling efficiency in response to sustained supraphysiologic ligand exposure. Users who take 5-10 mg nightly for months often report that the dose stops working the way it did initially, and the practical response is to escalate the dose — which deepens the downregulation. The physiologic 0.3 mg dose does not produce this pattern because the receptor exposure is at concentrations the system is calibrated to handle.

Dose-response inversion in some sleep architecture endpoints. A handful of trials have shown that supraphysiologic doses can disrupt REM architecture, fragment slow-wave sleep, or produce vivid dreams that wake the user — effects that the physiologic dose does not produce. The dose-response curve is not just flat above 1 mg; for some endpoints it inverts.

The label patterns that distinguish a physiologic-dose product

Most of the U.S. melatonin aisle is at 1-10 mg per dose. The 0.3 mg products exist but are rare and require label literacy to find. The patterns to look for:

Numerical dose printed on the front, ideally in milligrams with a decimal. A bottle labeled "0.3 mg" or "300 mcg" is at the physiologic-dose range. A bottle labeled "3 mg," "5 mg," or "10 mg" is at 10-33x the physiologic dose.

Microgram (mcg) labeling on the front. 300 mcg = 0.3 mg. Some products use the microgram unit to make the dose look larger; the math is the same.

No "extra strength," "maximum strength," or "fast-acting high-dose" marketing language. These phrases signal a supraphysiologic-dose product. The physiologic-dose product has nothing to gain from this marketing because more milligrams is not the value proposition.

Sublingual or fast-release formulation at the 0.3-1 mg dose range. A 0.3 mg sublingual or fast-dissolve tablet reaches peak plasma concentrations faster, mimics the endogenous rise more closely, and clears more cleanly. Slow-release formulations at the same physiologic dose are also reasonable for users with sleep-maintenance issues, where the goal is to extend the elevated-plasma window through more of the night.

No combination with proprietary "sleep blends," GABA, theanine bound into the same capsule, valerian, hops, or other co-ingredients. A clean melatonin product at the physiologic dose is the test. Once the bottle is a blend, the melatonin dose becomes harder to isolate and the formulation usually drifts toward the supraphysiologic dose to compensate for the muddied dose-response signal.

Gummies are almost universally supraphysiologic-dose. Gummies are formulated for shelf appeal and require enough melatonin per gummy to justify the format. The 0.3 mg gummy is rare in the market. The 5-10 mg gummy is the common form. If you are reaching for the gummy, you are almost certainly reaching for the supraphysiologic dose.

See also how to read a supplement label in 60 seconds for the broader label-reading framework, and proprietary blends are where underdosing hides for the proprietary-blend pattern that obscures dose-response on every ingredient.

When to use melatonin at all

Even at the physiologic dose, melatonin is not a universal sleep aid. The use cases where the trial literature supports it most clearly:

Age-related sleep-onset issues in adults 50+ with documented or suspected endogenous melatonin decline. The physiologic-dose evidence base is strongest here. A 0.3 mg sublingual at 30-60 minutes before intended sleep is the trial-supported protocol.

Jet lag and circadian phase shifts. Taking 0.3-1 mg at the destination's bedtime for 2-5 nights after travel helps reset the circadian rhythm. The mechanism is the phase-shift effect, not the sedative effect.

Shift work circadian misalignment. A 0.3-1 mg dose at the intended sleep time for a night-shift worker, taken consistently, supports circadian re-entrainment to the shifted sleep window.

Delayed sleep phase syndrome. A 0.3 mg dose taken several hours before habitual sleep onset can phase-advance the circadian rhythm. This is a specific clinical use case and the timing matters more than the dose.

The use cases where melatonin is over-prescribed and under-effective:

General sleep aid in young or middle-aged adults with intact endogenous melatonin secretion. The supplement is restoring what the body is already producing. The marginal sleep-onset benefit is small. Other interventions (sleep hygiene, light exposure timing, caffeine pharmacokinetics, alcohol elimination) usually have higher leverage.

Chronic insomnia management. Cognitive behavioral therapy for insomnia (CBT-I) is the trial-validated first-line treatment. Melatonin at any dose is not a substitute.

As a sedative for anxious or wired adults at bedtime. The supraphysiologic-dose sedative effect users are reaching for in this case is the wrong tool. The 5-10 mg dose is producing the sedation through pharmacologic overshoot, not through restoration of the physiologic sleep signal, and the next-morning cost is reproducible.

For children. Pediatric melatonin use is poorly regulated, frequently supraphysiologic-dosed, and not adequately studied for long-term safety. The American Academy of Pediatrics has issued cautionary guidance. Pediatric melatonin use should be a pediatrician-supervised decision, not a parent-driven supplement purchase.

The dose-response math, condensed

| Dose | Plasma peak (approx.) | Multiple of physiologic | Clearance by morning | |---|---|---|---| | Endogenous nighttime peak | 10-60 pg/mL | 1x | yes (by sunrise) | | 0.3 mg oral | 50-100 pg/mL | 1-2x | yes | | 1 mg oral | 200-500 pg/mL | 5-10x | mostly | | 3 mg oral | 1,000-2,000 pg/mL | 20-40x | partial | | 5 mg oral | 1,500-3,500 pg/mL | 30-70x | partial | | 10 mg oral | 3,000-10,000 pg/mL | 60-200x | persistent into AM |

The table is the answer. The dose-response signal for sleep onset peaks somewhere between 0.3 and 1 mg. Above that, plasma melatonin is supraphysiologic, the clearance does not finish by morning, and the next-day effects degrade. The marketing-driven 5-10 mg dose is not where the dose-response curve is rising; it is well past the peak.

The bottom line

Melatonin is one of the clearest examples in the supplement aisle of a market that walked past the dose-response curve in favor of a marketing math that does not match the physiology. The 0.3 mg dose, taken 30-60 minutes before intended sleep, is the dose the human-trial work supports for the use cases where melatonin actually has signal — age-related sleep-onset issues, jet lag, shift work, delayed sleep phase. The 5-10 mg products that dominate the shelf are at 15-30x the trial-supported dose, produce plasma melatonin concentrations 30-100x physiologic, persist into the morning, blunt natural core-temperature regulation, downregulate receptor sensitivity over time, and produce the next-day grogginess that the physiologic dose does not.

The label test is short. If the bottle says 3 mg, 5 mg, or 10 mg, the product is on the wrong end of the dose-response curve regardless of how clean the rest of the label looks. If the bottle says 0.3 mg or 300 mcg and is a clean single-ingredient melatonin product (no proprietary sleep blend, no co-formulated GABA or valerian), it is at the dose the trial literature supports. The dose written on the front of the bottle is the dose the formulator chose to commit to. Choosing the physiologic dose is choosing to make a product that does the thing melatonin can do, instead of a product that does the thing a low-grade sedative does and calls it melatonin.

Read the milligrams. Read the form. Read the formulation. Buy melatonin for the use case the molecule has evidence for, at the dose the molecule has evidence at. The shelf full of 10 mg gummies is what happens when label aesthetics replace dose-response physiology as the formulation logic.