The Sleep-Aid Hierarchy: What Actually Works
CBT-I, melatonin, magnesium, L-theanine, apigenin, and prescription drugs, ranked by evidence
For occasional poor sleep, the fundamentals in Sections 8 and 9 are the right starting point. For persistent, chronic insomnia, there's a clear evidence-based hierarchy of what to try — and it's not the one implied by pharmacy shelf space. For the supplements below, set realistic expectations: their effects are modest and cumulative, not a dramatic difference on the first night.
CBT-I: The First-Line Treatment
Cognitive behavioural therapy for insomnia (CBT-I) is a structured, typically 4–8 session programme combining stimulus control (retraining the brain to associate bed with sleep rather than wakeful frustration), sleep restriction therapy (temporarily limiting time in bed to build sleep pressure and consolidate sleep, done under guidance), cognitive techniques (addressing the anxious thoughts that often accompany insomnia), and relaxation training. Clinical guidelines recommend it as first-line treatment ahead of medication for chronic insomnia in adults[14], with improvements that are clinically meaningful and well sustained over time — unlike medication, whose benefits typically fade once stopped[15]. It's available through therapists, some doctors, and increasingly through structured digital programmes.
Melatonin: Modest, Real, and Best Used for Timing Rather Than Sedation
A meta-analysis of 19 randomised, placebo-controlled trials covering over 1,600 subjects found that melatonin modestly but reliably reduces the time it takes to fall asleep, increases total sleep time, and improves overall sleep quality, with somewhat larger effects at higher doses and longer duration of use[21]. Its most well-supported use case is circadian, not sedative: taken at the right time, it's most useful for shifting the clock (jet lag, shift work, delayed sleep phase) rather than as a nightly sedative for otherwise well-timed insomnia. Doses are typically far lower than what's sold over the counter in many markets — 0.5–3mg is usually sufficient, and more isn't necessarily better.
Magnesium: Some Evidence, Real Caveats
A placebo-controlled trial in elderly adults with primary insomnia found that 500mg of magnesium supplementation over 8 weeks significantly improved sleep time, sleep efficiency, and sleep onset latency (moderate evidence — the trial used general magnesium, not glycinate specifically, so glycinate's popularity rests more on its absorption profile than on sleep-specific trial data)[22]. The effect is real but modest — a reasonable low-risk addition, not a primary fix for significant insomnia.
L-Theanine: Modest, Consistent, and Better Evidenced Than It Gets Credit For
L-theanine is an amino acid found in tea leaves, typically supplemented at 200mg taken shortly before bed. A 2025 meta-analysis pooling 18 trials found small but statistically significant improvements in sleep onset latency, daytime dysfunction, and overall sleep quality (moderate evidence — few trials tested pure L-theanine in isolation, so the isolated compound's effect is somewhat less pinned down than the headline result)[23]. It's non-sedating, so it doesn't carry next-morning grogginess risk the way a sedative might.
Apigenin: Plausible Mechanism, Thin Human Evidence
Apigenin is a flavonoid — the compound largely responsible for chamomile's traditional reputation as a sleep aid — that binds, in animal studies, to the same brain receptor site (the GABA-A receptor's benzodiazepine site, the target of sedative medications) as prescription sleep drugs. It's commonly supplemented in isolated form at around 50mg (weak evidence — no published human trial of isolated apigenin exists; the closest is a small trial of chamomile extract, which contains apigenin among other compounds, showing no significant improvement over placebo[24]). It's low-risk and reasonable to try, but rests on mechanism and animal data rather than demonstrated human benefit.
Prescription Sleep Medication
Z-drugs (such as zolpidem) and benzodiazepines are effective for short-term insomnia relief but carry real risks with extended use: tolerance, dependency, rebound insomnia on discontinuation, and — particularly in older adults — increased fall risk. The clinical guideline position is that these are appropriate for short-term or intermittent use, not as a long-term first-line strategy, precisely because CBT-I achieves comparable or better results without the dependency risk[14]. If you're currently on one, the message is not "stop" — it's "this is a conversation to have with your prescriber about tapering onto CBT-I," not something to change unilaterally.
| Option | Evidence | Best used for |
|---|---|---|
| CBT-I | Strong — first-line clinical recommendation | Chronic insomnia; durable improvement without dependency risk |
| Melatonin (0.5–3mg) | Moderate — consistent, modest effect | Shifting circadian timing (jet lag, shift work, delayed phase), more than nightly sedation |
| Magnesium (~300–500mg) | Moderate, one strong trial in older adults | A low-risk addition alongside fundamentals, not a primary fix |
| L-Theanine (~200mg) | Moderate — consistent small effect across pooled trials | A low-risk, non-sedating addition, ideally before bed |
| Apigenin (~50mg) | Weak — no isolated-compound human trials | Reasonable to try given low risk, but resting on mechanism more than evidence |
| Z-drugs / benzodiazepines | Effective short-term; risk rises with duration | Short-term or intermittent use under medical guidance |
Section takeaway
For chronic insomnia, the evidence-based order is CBT-I first, melatonin for circadian timing problems specifically, and magnesium or L-theanine as reasonable low-risk adjuncts — with apigenin a lower-confidence option resting more on mechanism than direct evidence. Prescription sedatives are reserved for short-term or intermittent use under medical guidance — the inverse of how most people instinctively reach for options.