Why melatonin stops working: What to try instead

It is 10:47 PM, and the melatonin is not doing anything

You took it at 10. Five milligrams. The kind they sell in the same aisle as the multivitamins. You have been using it for six months. For the first few weeks, it helped. Now it mostly just makes you feel slightly groggy in the morning without doing much to get you to sleep.

This is a common arc. And it happens for a reason that most people are never told when they buy the bottle.

Melatonin is not a sleeping pill. It is a circadian timing signal. The pineal gland releases it as light drops in the evening. Its job is to tell the clock that night is coming. A small amount at the right time shifts the body’s sense of when sleep should begin. A large amount at the wrong time does mostly nothing useful, and chronic high-dose use can actually blunt the natural melatonin response over time.

Most people are using it wrong, at the wrong dose, for the wrong problem.

What melatonin actually does

The body’s natural nighttime melatonin levels peak at roughly 45 picograms per milliliter of blood. Standard over-the-counter doses of 3 to 10 mg produce blood levels that are 10 to 50 times higher than that physiological peak. That is not a therapeutic dose. It is a pharmacological one. And it does not produce better sleep in proportion.

What the research supports for melatonin: it helps shift the timing of sleep onset, most reliably for jet lag, shift work, and delayed sleep phase (night owls who cannot fall asleep before 1 or 2 AM). Dr. Lawrence Epstein at Harvard Medical School says it directly: melatonin is not a sleep medication. It makes people a little sleepy but has a much greater effect in shifting the timing of the sleep phase.

For people with maintenance insomnia (falling asleep fine but waking at 2 AM and unable to return to sleep) or early morning awakening, melatonin does very little. It addresses the wrong mechanism entirely.

What you’ve probably already tried

Most adults cycling through sleep problems go through a predictable sequence: melatonin, then higher-dose melatonin, then maybe Benadryl or a similar antihistamine, possibly a glass of wine at night, maybe magnesium glycinate.

Magnesium glycinate is worth keeping. It genuinely supports sleep quality for many adults. Our piece on magnesium glycinate covers this in full.

Antihistamines (diphenhydramine, doxylamine): they work the first few nights because the sedation is real. Then tolerance builds fast. Harvard’s Lawrence Epstein notes they tend to stop working relatively quickly and carry a real risk of next-day grogginess, cognitive blunting, and, in older adults, falls and confusion. Not a long-term tool.

Alcohol: reduces sleep onset time and reliably fragments sleep in the second half of the night. A net negative for sleep quality regardless of how it feels going down. Our piece on alcohol and sleep after 40 covers the data.

The Livium take. The supplement aisle solves a narrow subset of sleep problems. Most adults with chronic insomnia after 40 have an underlying mechanism that over-the-counter products were not designed for. Diagnosing which mechanism (onset, maintenance, early waking, hyperarousal, circadian misalignment) matters more than which product to try next.

The Livium recipe

Tool. For sleep onset problems that are not circadian-phase related, low-dose doxepin (3 to 6 mg, not the 25 to 150 mg antidepressant dose) is FDA-approved for insomnia and specifically targets histamine receptors responsible for wake maintenance. It does not cause the next-day grogginess associated with diphenhydramine. Dual orexin receptor antagonists (DORAs), such as suvorexant (Belsomra) and lemborexant (Dayvigo), are a newer class that work by blocking the wakefulness signal rather than sedating, which is mechanistically distinct and cleaner for most people. Both are available via telehealth through Hims sleep, which handles evaluation and prescription without an in-person visit.

Behavior. If you continue using melatonin, reduce the dose to 0.5-1 mg and take it 2 hours before your intended bedtime, not right before bed. That is the physiologically appropriate window. Stop using it nightly for problems it was not designed for: maintenance insomnia and early waking need different tools.

Threshold. Four weeks of correctly dosed melatonin (0.5 to 1 mg, taken 2 hours early) to assess whether the timing-shift effect helps. If it does not, or if your problem is maintenance insomnia, that is a signal to consult on prescription options. If your problem is early-morning awakening between 3 and 5 AM, the cortisol piece is worth reading: our article on waking between 1 and 4 AM explains why cortisol is often the driver there.

Matching the tool to the problem

Sources: Harvard Health Publishing; NHLBI, How Sleep Works; Hims clinical content.

Plan of action

• Identify which problem you actually have. Onset, maintenance, early waking, or phase delay. Each has a different solution, and treating the wrong one wastes months.
• If you are using melatonin for maintenance insomnia or early waking, it is solving the wrong problem. Stop and look at the table above.
• If you want to keep using melatonin for onset issues, drop to 0.5 to 1 mg and take it 90 to 120 minutes before your intended sleep time, not at bedtime. Pair it with evening light reduction (no bright overhead lights after 9 PM).
• If maintenance insomnia is your problem and OTC options have failed, talk to a provider about doxepin 3 to 6 mg or a DORA. Hims sleep handles this without an in-person visit.
• If early waking between 3 and 5 AM is the pattern, read our piece on waking between 1 and 4 AM before trying another supplement. Cortisol and sleep apnea are frequently the drivers.