Sleep Optimization: the Food & Supplements That Actually Help

L-tryptophan

Tryptophan is the amino-acid precursor your body turns into serotonin and then melatonin. It does reach the brain — it's carried across the blood–brain barrier — but it has to compete with other amino acids for that transporter, so it works best taken with some carbohydrate and away from a high-protein meal. ^{2 3} A meta-analysis of randomized trials found doses of 1 g or more shortened the time spent awake after falling asleep (WASO), but did not significantly change how fast people fell asleep or their total sleep time. ¹ So: a real but narrow effect, honestly stated.

Magnesium (glycinate or bisglycinate)

Magnesium supports the GABA system that calms the nervous system, and many people are mildly low in it. An 8-week double-blind trial in older adults with insomnia found improvements in sleep efficiency, sleep onset and insomnia severity, ⁷ and a meta-analysis estimated roughly 17 minutes faster sleep onset versus placebo — but rated the underlying trials as low quality and at moderate-to-high risk of bias. ⁸ So: plausible and low-risk, not proven. The glycinate form is gentle on the stomach.

Glycine

An amino acid studied in small randomized trials: 3 g before bed improved subjective sleep quality and next-day fatigue, ⁵ and a polysomnography study found it shortened the time to fall asleep and reduced daytime sleepiness, partly by gently lowering core body temperature. ⁴ Inexpensive and well tolerated, though the trials are small.

L-theanine

An amino acid from green tea that promotes calm without sedation. In a randomized controlled trial, 200 mg/day for 4 weeks lowered stress scores and improved several sleep-quality measures (sleep latency, disturbance, use of sleep medication). ⁶ Most useful for people whose sleep problem is really a racing, anxious mind at bedtime. Pairs well with magnesium.

Melatonin

Melatonin is a timing signal, not a sleeping pill. It shines for jet lag and shift work, and for people whose body clock runs late. A dose-response meta-analysis of randomized trials found low physiological doses (0.5–1 mg) effective, with timing — about 2–3 hours before target bedtime — mattering as much as the dose; very large 5–10 mg tablets add little and cause more next-day grogginess. ⁹

Microdoses can be enough. A classic controlled study found that a tiny physiological dose of 0.3 mg restored sleep efficiency in adults over 50 — performing as well as, or better than, a 3 mg dose, because it lifts melatonin only into its natural night-time range. Higher doses left melatonin elevated into the morning, which can cause daytime grogginess. For many people, less really is more. ¹⁶

Is long-term use safe? For adults, the evidence is broadly reassuring: systematic reviews find side effects are mild and similar to placebo (most often headache, dizziness, daytime drowsiness), and — unlike sleeping pills — melatonin shows very little tolerance or dependence. The honest caveat is that good data mostly cover use up to about two years; truly long-term (5+ years) safety is under-studied. ¹⁴ In children and teenagers the picture is less settled — a 2023 review raised a very-low-certainty signal of possible delayed puberty with prolonged use, so it should only be used under medical guidance. ¹⁵

Herbal options: valerian & passionflower

Valerian: meta-analyses of randomized trials suggest it can improve subjective sleep quality (about 1.8× more likely to report better sleep than placebo), but the benefit has not been confirmed on objective measures, and the trials vary widely in dose and quality. ^{10 11}

Passionflower (Passiflora incarnata): acts on the same calming GABA-A system. Small double-blind trials found better subjective sleep quality from a low-dose tea, ¹² and a more recent placebo-controlled trial in people with stress and poor sleep reported increased total sleep time and lower stress. ¹³ Promising but still small studies.

Other names you'll see

Tart cherry extract (a natural melatonin source) and apigenin (a flavonoid found in chamomile and also sold as a pure supplement) also have a following, but the human evidence is thinner. Our apigenin guide explains why chamomile studies cannot be treated as trials of isolated apigenin. Worth considering only after the better-studied options above.