How long does insomnia have to last before it counts? International guidance grades one treatment strong — 6 recommendations, 14 sleeping medicines, and the Hong Kong wait, 2026

TL;DR

The best-evidenced treatment for insomnia is not a sleeping pill but a behavioural therapy called CBT-I — of the six recommendations in the international guideline, it is the only one graded "strong". But the Hospital Authority's fee schedule carries no service published under that name, and the median wait for a stable new psychiatric case runs from 19 weeks to 69 weeks. The first step the official documents name is not psychiatry. It is your family doctor.

---

#How long before insomnia is "chronic"? Two numbers, and Hong Kong's official pages carry neither

The conclusion first: at least three months, at least three times a week. Both conditions have to be met before it is chronic insomnia.

That threshold comes from the International Classification of Sleep Disorders, Third Edition, as restated by the American Academy of Sleep Medicine (AASM) in its 2017 pharmacological guideline [Note 1]. The same passage gives two useful proportions: occasional, short-term insomnia affects 30% to 50% of the population; the prevalence of chronic insomnia in industrialised nations is estimated at at least 5% to 10%. And in medically and psychiatrically ill populations, and in older age groups, prevalence is significantly higher [Note 2].

Hong Kong's official pages do not carry that threshold. The Centre for Health Protection gives a different number: if insomnia persists for more than a month, see a doctor [Note 3].

Both numbers are current and they do different jobs — three months is a diagnostic threshold, one month is a prompt to seek help. A prompt arriving earlier than a diagnosis is sensible. But if you read it as "under three months, ignore it", you have read it backwards.

There are in fact more than two thresholds, because each document works to its own classification:

That cohort study covered 1,611 local Chinese children, not adults. What the three have in common is the frequency: three times a week or more. The durations run from one month to twelve, so any duration figure has to be read together with where it came from.

#The medicine you are already taking may be the cause

Few people raise this, but the Department of Health's Drug Office lists it on its own page [Note 4]:

This table is not telling you to stop anything. It has one use: when you see a doctor, list everything you are taking, over-the-counter medicines and supplements included. Because if the cause is on this table, adding a sleeping pill only adds a medicine and leaves the cause where it was. The same page says it too: tell your doctor about all your medications and medical conditions.

---

#"CBT-I is first line" — who actually said that?

The most direct source is the AASM's 2021 behavioural and psychological treatment guideline. Of its six recommendations, only CBT-I is graded "strong".

"Strong" has a definition in that guideline: clinicians should follow it under most circumstances; the full text puts it harder still — should be followed for almost all patients [Note 5]. A "conditional" recommendation, by contrast, has to be decided patient by patient.

The six recommendations and the evidence behind them:

Not one of the six reaches high quality of evidence; even the strongest is only moderate. Worth remembering: "strong" describes the force of the recommendation, not the grade of the evidence.

Three further things belong with this, or it gets cited wrongly.

"The only one" is true only of this guideline. The background section of the same AASM 2017 guideline records that the American College of Physicians made a strong recommendation of its own in 2016, on moderate quality evidence, and with wider scope — all patients with chronic insomnia, as the initial treatment intervention [Note 6]. So there is more than one CBT-I recommendation carrying a "strong" grade.

The guideline makes no recommendation about how it is delivered. The wording is explicit: there was insufficient evidence to make recommendations for specific delivery methods — individual, group, internet, self-help, video — for any of the treatments [Note 7]. Taking "AASM strongly recommends CBT-I" over to support a particular phone app goes outside the boundary the guideline drew for itself.

A difference in strength is not a difference in effect. Every recommendation was compared against a wait-list, a minimal intervention or a placebo, not against another treatment; the guideline's review expressly excluded comparisons between interventions [Note 8]. So the difference between recommendation 1 and recommendations 3 to 5 is a difference in the quantity and quality of evidence.

#In the UK: NICE describes the pathway and attributes it to another body

NICE's MTG70 (20 May 2022) does not hand down "CBT-I is first line" on NICE's own authority; it describes current best practice and attributes that clinical pathway to the British Association for Psychopharmacology's consensus statement [Note 9]. The same passage says two more things: face-to-face CBT-I has limited availability in the UK; and people who may be at higher risk of other sleep disorders, sleep apnoea among them, should have a medical assessment before referral.

The committee also recorded a positive finding and a price: clinical evidence shows Sleepio reduces insomnia symptoms compared with sleep hygiene and sleeping pills; at £45 per person it is cost saving, a conclusion based on resource use data before and after Sleepio was introduced in 9 GP practices [Note 10]. This is the only price for digital CBT-I published by an institution anywhere in the sources cited here, and it is a 2022 UK primary care price.

---

#Why is sleep hygiene on its own not a treatment?

Because among those six recommendations, sleep hygiene is the only one the guideline recommends against using on its own.

But under that same recommendation the guideline adds a sentence of its own: sleep hygiene may be included in multicomponent interventions [Note 11]. The two have to be read together. The correct reading is "do not do only this", not "this does nothing".

If you have followed the standard sleep hygiene advice for months and still cannot sleep, you are not doing it badly — you may be doing the one thing the guideline itself says not to do alone.

Nor is CBT-I "go to bed earlier and stay off your phone". The guideline treats multicomponent CBT-I and several single-component therapies separately:

⚠️ Two of the components carry safety limits, and those sentences are not inside the recommendation, so an article that quotes only the recommendation never shows them.

• Sleep restriction therapy may be contraindicated in certain populations: people working in high-risk occupations (heavy machinery operators or drivers), those predisposed to mania or hypomania, those with poorly controlled seizure disorders, or those with excessive daytime sleepiness [Note 12].
• Stimulus control may need adapting for safety in people at high risk of falls, with mobility issues, or using sedative-hypnotics [Note 13].

Think about what those two sentences mean. Stimulus control means getting up and leaving the bedroom when you cannot sleep — and if you are also taking a sleeping pill, getting up in the middle of the night is exactly when the risk of falling is highest. Sleep restriction therapy increases daytime sleepiness at first — and if you drive for a living, the word the guideline uses is "contraindicated", not "harder going". These are for a doctor to decide on your situation.

Hong Kong official documents do mention cognitive behavioural therapy. The Drug Office states that a doctor can identify underlying causes and recommend appropriate treatment, which may include cognitive behavioural therapy [Note 14]; and the Centre for Health Protection's own publication states that most insomnia can be effectively treated through medication or cognitive behavioural interventions, and that insomnia is often under-recognised and under-treated [Note 15].

So what is missing is not the concept. It is a public service pathway published under that name that you can join a queue for. Those are two different things and belong apart.

Incidentally, of the seven sleep hygiene items on that Centre for Health Protection page, the last one — if you cannot sleep, do not force it; get up and do something else — is stimulus control as the guideline defines it [Note 16]. The same act is a lifestyle tip on a consumer page and, in the international guideline, a single-component therapy carrying a falls-safety note. And in the original that list comes after "determine and treat the cause of insomnia", not instead of it.

---

#Can an app treat insomnia? The five caveats on a 708-person trial

One trial says it can, and its five caveats are worth more than its conclusion.

The trial: 708 participants randomised to app-based CBT-I (354) or app-based health education (354); Chinese youth aged 15 to 25 with insomnia disorder and subclinical depression, recruited in mainland China and Hong Kong from 9 September 2019 to 25 November 2022 [Note 17]. 407 were female (57%), mean age 22.1 years. 41 participants (6%) were from Hong Kong and 667 (94%) from the mainland.

The primary outcome was the prevention of depression. Over 12 months, 37 participants (10%) in the intervention group and 62 (18%) in the control group developed new-onset major depressive disorder; hazard ratio 0.58 (95% CI 0.38–0.87; p = 0.008), with a number needed to treat at 1 year of 10.9 (6.8–26.6) [Note 18].

Insomnia remission was a secondary outcome:

The effect is real, and it narrows over time. The gap between the groups goes from 24 percentage points at the end of the intervention to 9 percentage points at 12 months. Quoting only the 1.17 at 12 months understates the immediate effect; quoting only the 1.83 overstates how long it lasts.

The five caveats:

1. 57%／48% is a secondary outcome. The trial was designed to answer whether depression could be prevented [Note 19].
2. The 12-month result is only just statistically significant. The lower bound of the confidence interval is 1.01; a little lower and it crosses 1.
3. The control group was not doing nothing — it did 6 weeks of app-based health education. So that 48% is not the natural course of insomnia.
4. "Remission" is a composite of three conditions, medication among them: no functional impairment or distress with sleep; insomnia symptoms fewer than three times a week for at least a month; and sleep-promoting medication used less than once a week in the past month [Note 20].
5. Nearly half the participants had a history of depression — 48% (339/708) reported a prior history of major depressive disorder, so a substantial part of "preventing new-onset depression" is preventing recurrence [Note 21].

And one more thing, if you are looking this up for a secondary school student: of the 708 participants only 10 were under 18, and the mean age was 22.1. The paper states plainly that the modest sample size of adolescents limits how far the findings generalise to that age group [Note 22]. This is in substance a trial of people around twenty, and the intervention is a phone app, not face-to-face therapy.

On safety, the paper states that no adverse events related to the interventions were reported.

#While we are here, a denominator that gets cited wrongly

You may have seen "89 studies support CBT-I". Not so.

The systematic review underlying the guideline identified 1,244 studies, of which 124 met the inclusion criteria and 89 provided data suitable for statistical analysis — but those 89 are spread across ten interventions [Note 23].

The actual denominator for the CBT-I recommendation is 66 randomised controlled trials, 49 of them analysable [Note 24].

And the result of sleep hygiene's single analysable study deserves recording too: it showed the response rate in the sleep hygiene group to be clinically significantly higher than the control. Recommendation 6 is against using it alone; it does not say the thing did nothing in the research.

---

#The evidence for sleeping pills is graded "weak" throughout — which does not mean they do not work

The AASM's 2017 pharmacological guideline makes 14 recommendations: 8 for and 6 against, and every one of them is graded "weak". The same academy graded CBT-I "strong".

But before those recommendations the guideline sets out a passage designed to prevent exactly the misreading, and that passage is the most important thing in this section: readers will note that all specific recommendations fall in the "weak" classification, and this should not be construed to mean that no sleep-promoting medications are clearly efficacious or indicated [Note 25]. The guideline separately states that "weak" reflects a lower degree of certainty and should not be construed as an indication of ineffectiveness.

The guideline also explains why they are all weak: the funding source of most pharmacological clinical trials and the attendant risk of publication bias; the relatively small number of eligible trials for each individual agent; and the observed heterogeneity in the data [Note 26]. "Weak" is about how clearly we know these medicines, not about how useless they are.

⚠️ If you are currently on a prescribed sleeping medicine, do not stop it because you have read this section. As set out below, several can produce rebound insomnia on withdrawal, and the page says in terms that chloral hydrate should not be stopped abruptly. Stopping is for the prescribing doctor to arrange.

#The 14 recommendations, with the doses the guideline itself footnotes

⚠️ Without the dose, "we suggest against melatonin" reads as a statement about every dose. Each recommendation carries a remark naming the dose the trials used. Every comparison is against no treatment, not against CBT-I.

Recommended for use (8, all WEAK):

Recommended against (6, all WEAK):

Two things often treated as the gentle option are both on the "against" side: diphenhydramine and melatonin.

⚠️ But readers over 55 must finish the melatonin entry. The evidence supporting that recommendation against is graded by the guideline itself as very low; the guideline also states that the task force judged the benefits to be approximately equal to harms, and considered that the majority of well-informed patients would use melatonin over no treatment. More to the point: the same guideline records that the British Association for Psychopharmacology recommended prolonged-release melatonin as a first-line treatment for insomnia in people over 55 [Note 29] — and the three melatonin studies AASM reviewed enrolled that same population.

Different formulation, different body, opposite direction. A reader over 55 who sees only "suggested against" is seeing a picture that is the reverse of the part most relevant to them. Whether to use it, and in which formulation, is for a doctor.

⚠️ Nor can the dose column be read as "other doses are fine". The guideline states expressly that its literature review, meta-analyses and recommendations are based only on FDA-approved doses, and that this should not be interpreted as a recommendation for any specific dose [Note 30].

---

#The treatment-duration limits in the official page: four weeks applies to two classes only

In Hong Kong most medicines used to treat insomnia are prescription-only [Note 31]. "Most" is not "all" — the same page also says not to self-medicate or take over-the-counter sleeping pills without professional advice.

The official framing of sleeping pills is unambiguous: these medicines do not address the root cause of insomnia, should be prescribed by a doctor, used at the lowest effective dose for the shortest possible duration, and are suitable for short-term management of severe and disabling insomnia [Note 32].

So the blanket line that "sleeping pills are generally not for more than four weeks" does not hold. Four weeks applies to benzodiazepines and zolpidem; zopiclone is 7 to 10 days; and two classes carry no figure at all.

⚠️ The right-hand column deserves as much attention: three rows name sleep apnoea explicitly.

Newer does not mean safer. The page states that there is still little evidence of any clinical advantage of zopiclone over benzodiazepines in tolerance, dependence or withdrawal symptoms; and equally little evidence that zolpidem has any advantage over short-acting benzodiazepines in "hangover feeling", tolerance, withdrawal or dependence [Note 35].

Watch for rebound on stopping. Benzodiazepines, zopiclone and zolpidem are commonly associated with tolerance, dependence and withdrawal symptoms; rebound insomnia may occur on withdrawal of zopiclone and zolpidem — which means stopping can be worse than before you started, and that is very easily misread as proof that you need to keep taking it.

⚠️ And one risk that has nothing to do with how alert you feel next morning: the page states for several of these medicines that complex sleep-related behaviours can occur — sleep driving, making phone calls, preparing and eating food while asleep — with amnesia for the event, and that alcohol or other CNS depressants appear to increase the risk.

#How long before you can drive?

⚠️ The rows without a figure are not the looser ones — the page simply tells you not to drive. The page's general advice also carries "do not drive or operate machinery after taking the medicine", applying to all sleeping medicines, with no hours attached.

The page's six items of general advice [Note 36]: use only under medical supervision; always take the lowest effective dose for the shortest duration; be aware of the medicine name and dosage; avoid alcohol while on sleeping medicines; do not drive or operate machinery after taking the medicine; and if symptoms persist, consult your doctor for alternative treatments.

⚠️ Two risks deserve particular attention: for DORA, worsening of depression and suicidal thoughts and actions, including completed suicides, have been reported in primarily depressed patients treated with hypnotics. For chloral hydrate, elderly people may be more susceptible to side effects such as ataxia and confusion, which may lead to falls and injury, and fatalities from cardiovascular collapse may rarely occur. If mood worsens or suicidal thoughts appear while taking these, do not wait for the next appointment.

---

#Where in Hong Kong can you actually get CBT-I?

In the Hospital Authority material this article examined, no service is published under the name "cognitive behavioural therapy for insomnia" or CBT-I.

Specifically: the Hospital Authority fee schedule (effective 1 January 2026) does not carry such an item. That page lists a series of named services, and the strings CBT, Cognitive and insomnia each appear 0 times on it [Note 37].

⚠️ That sentence has a firm boundary. This article can say "the Hospital Authority has not published a service under that name"; it cannot say "nobody in the public system provides CBT-I". A clinical psychologist may deliver CBT-I content within a consultation without there being a separately published service line.

#How long is the wait?

(Data period 1 July 2025 to 30 June 2026. Urgent new cases run from "less than 1 week" to "1 week", semi-urgent 2 to 3 weeks.)

These are not marginal numbers. The same dataset shows psychiatric new case bookings split into urgent 2,142 (4%), semi-urgent 9,342 (17%) and stable 42,169 (79%), out of 53,664 in total.

Put the two numbers side by side: the Centre for Health Protection tells you to see a doctor if insomnia lasts more than a month. The New Territories East median wait for a stable new psychiatric case is 69 weeks ÷ 4.3 ≈ 16 months; the longest, 104 weeks, is two years. Depending on where you live, that wait runs from roughly 4 times to roughly 16 times the prompt (the shortest median being Kowloon West at 19 weeks).

⚠️ This table is psychiatry as a whole, not insomnia. No public document states which triage category insomnia cases fall into, and this article does not infer one.

#The charges: $20 for a drug item is actually four weeks

"$20 per drug item" does not buy one dispensing; it buys four weeks' supply (fee schedule note 7). And the duration limit for benzodiazepines and zolpidem, in the Drug Office's own words, is also four weeks — so one charging unit is roughly one complete officially-stated course. (That is two documents' figures set side by side, not something either document says.)

The fee schedule also sets out two stacked safety nets: the means-tested medical fee waiver, and, from 1 January 2026, an annual cap of HK$10,000 on public healthcare fees requiring no financial assessment [Note 38]. The fee schedule this article examined does not define the scope of "chronic illness", so this article does not say insomnia is included.

#The official first step is your family doctor

The Centre for Health Protection's publication is explicit: if insomnia has been interfering with your daytime functioning, consult your family doctor [Note 39]; and be sure to discuss with your family doctor before taking any sleeping medication.

So the official pathway starts with a family doctor, not with psychiatry. The waiting table above, measured in years, is the downstream situation after a referral and a triage into stable new cases. The first step is something you can do within days.

#At community level: the one published pathway this article found

The Jockey Club Sleep Well Programme, run by the Department of Psychiatry of the Faculty of Medicine at the Chinese University of Hong Kong, is the only Hong Kong CBT-I route within this article's search that publishes its service content and eligibility on the operator's own website. All of the following comes from that website:

• Ages 18 to 70, with the course conducted mainly in Cantonese.
• A screening questionnaire comes first, followed up by staff — not a service you can start on the spot.
• Support is tiered: all members get sleep knowledge resources; only the tier for "members troubled by insomnia" mentions individual guidance from a dedicated sleep coach, and the website's word is that there is a chance of being matched.
• There is a waiting period, which the programme acknowledges, but does not say how long.
• Participation carries research questionnaire obligations: one before the course starts, one on completion, and one each at 3 and 12 months afterwards.
• No end date is published. It is funded by a charitable trust, and the website footer carries a copyright year of 2023.

On charges: the operator has written two first-hand sentences — the participant consent form says that participation in the study's assessments and course is entirely free of charge, and the terms of use say that the website is free. ⚠️ But each sentence has its own scope, and the operator publishes no schedule of service charges and says nothing about arrangements outside the research, so whether this programme is free across the board is something this article cannot state [Note 40]. Ask the programme directly.

⚠️ Naming this programme is description, not recommendation. This article does not compare Hong Kong service providers.

The programme has written an assessment of its own: "support in the community targeting sleep problems is very limited" (our translation from the Chinese original) [Note 41]. Set beside NICE on limited availability in the UK, a shared shape appears: in two different healthcare systems, the treatment the guidelines most recommend is the hardest one to find.

---

#Is "nearly half of Hong Kong has insomnia" true?

48.0% is true, and it uses the loosest definition available. In the same survey, reported on the same page, the proportion meeting "three or more times a week" is only 6.0% to 7.3%.

The Department of Health's Population Health Survey 2014/15 interviewed over 12,000 non-institutionalised people aged 15 and above [Note 42]:

The full frequency distribution runs like this:

The middle two rows are the stretch between "48.0%" and "6% to 7%". Once a week or more comes to 15.9%, 17.7% and 13.9% respectively; having experienced the problem at all (100 minus "no such problem") comes to 33.2%, 35.2% and 29.4%. (The first column adds to 33.2 by group and to 33.1 as 100 minus 66.9; that 0.1 is the rounding the figure itself notes.)

⚠️ 48.0% is not any cell in the table above. It is the combined figure for any of the three disturbances at any frequency, and since all three can occur in the same person, this article does not try to derive 48.0% from the three numbers.

The same publication cites international literature explaining why figures differ so much between places [Note 43]: insomnia prevalence ranges from as low as 6% under stringent diagnostic criteria, to 10–15% among people reporting insomnia symptoms with daytime consequences, and up to 48% among those with any insomnia symptoms.

From 6% to 48% is a factor of eight, and it is not because the surveys are inaccurate. It is because the question differs.

#How many actually seek help

From the same publication: only about 10% of children and adolescents, and 40% of adults, with insomnia reported having sought treatment for it [Note 44].

What did the rest do? A substantial proportion sought treatment through complementary and alternative medicine such as Chinese herbal medicine and acupuncture, some turned to over-the-counter sleeping aids, and some to alcohol for relief.

⚠️ "A drink helps me sleep" is a misconception the Centre for Health Protection names — the publication states that alcohol disrupts normal sleep patterns, depriving the drinker of sound and restful sleep. That joins up with the medicines section: one of the six general advice items is to avoid alcohol while on sleeping medicines, and for both zopiclone and zolpidem alcohol increases the risk of doing things while not fully awake.

⚠️ Year limitation: this is a 2014/15 survey, now eleven to twelve years old.

---

#When should you seek help immediately?

The Centre for Health Protection's prompt is one month. It differs from the diagnostic threshold (at least three months, at least three times a week) because the two do different jobs — where two authorities differ, take the more cautious one, which means seeking help earlier.

⚠️ What follows is not a discriminating list and not a self-check. It is an entrance, not an exit — not one line of it can be used to conclude "so I am fine".

• Loud, persistent snoring at night with daytime sleepiness. The Centre for Health Protection states these are the main symptoms of sleep apnoea; another symptom is frequent long pauses in breathing during sleep followed by choking and gasping [Note 45]. The same page states that sleep apnoea may lead to hypertension, heart disease, heart attack and stroke, and that a patient who is treated can live a normal life.
• Falling asleep uncontrollably during the day, at any time and in any situation, which may be narcolepsy [Note 46].
• Tingling sensations in the legs, and sometimes the arms, while sitting or lying still, especially at night, with a need to keep stretching or moving them, which may be restless legs syndrome [Note 47].

The same Centre for Health Protection quiz page describes three sleep disorders by symptom, not only sleep apnoea.

Sleep apnoea recurs throughout the sources cited here, always pointing the same way: the Non-Communicable Diseases Watch states that insomnia readily becomes a chronic problem when it co-exists with other sleep disorders such as obstructive sleep apnoea and restless legs syndrome [Note 48]; the Drug Office lists three commonly used sleeping medicines as each contraindicated in sleep apnoea; and NICE says people at higher risk should have a medical assessment before referral.

Two things this article cannot do, said plainly:

1. It carries no list that can be used to "rule out" another condition. Describing symptoms can send you to a doctor; nothing here can tell you not to go.
2. The side effects listed here are not a complete list. The page's lists use "including" and "such as" and are illustrative; nowhere on it is there an incidence rate or a denominator. So "I do not have anything on this list" does not lead to "I am fine".

---

#What to do next

1. See a family doctor first, rather than queueing for psychiatry. That is the first step the official documents name, and it is one you can take within days.
2. Before you go, list every medicine you take — over-the-counter medicines and supplements included. The four classes above may be the cause.
3. If you are doing sleep hygiene and not improving, do not do it harder; change the question. The guideline recommends against using it alone.
4. Ask one question: is what is offered here multicomponent CBT-I? That is what the guideline graded "strong" — not a single component, and not any app.
5. If you are on a sleeping medicine, leave the stopping to the prescriber. Rebound insomnia will make you think you cannot do without it.
6. Snoring plus daytime sleepiness: take it to a doctor. Together they are the main symptoms of sleep apnoea, and several sleeping medicines are contraindicated in exactly those patients.

---

#Frequently asked questions

If I cannot sleep long-term, do I have to take a sleeping pill?

Of the six recommendations in the AASM's 2021 guideline, the only "strong" one is multicomponent CBT-I; all 14 pharmacological recommendations in the same academy's 2017 guideline are graded "weak". Individual treatment choices are for a doctor to make on your situation, and this article makes no recommendation.

"All 14 drug recommendations are weak" — does that mean sleeping pills do not work?

No, and this is not this article's judgement but the guideline's own: all specific recommendations fall in GRADE's "weak" classification, and this should not be construed to mean that no sleep-promoting medications are clearly efficacious or indicated; a weak recommendation should also not be construed as an indication of ineffectiveness. If you have a prescribed sleeping medicine, do not stop it because you read "all weak" — several can produce rebound insomnia on withdrawal, and prolonged users of chloral hydrate should not stop abruptly. Stopping is for the prescribing doctor to arrange.

I have been going to bed early, keeping the room dark and staying off my phone for ages. Why has it not worked?

Recommendation 6 of the AASM 2021 guideline is against using sleep hygiene as a single-component therapy (conditional grade), but the remark under that same recommendation says sleep hygiene may be included in multicomponent interventions. The correct reading is "do not do only this", not "this does nothing".

Melatonin is natural. Is it safer?

The AASM 2017 guideline puts melatonin and diphenhydramine both on the "suggested against" side, at WEAK strength. The melatonin recommendation rests on trials of 2 mg, and those three studies enrolled only people over 55; the diphenhydramine one rests on 50 mg. The quality of evidence for the melatonin recommendation is very low, and the guideline states that the task force judged benefits approximately equal to harms and considered that the majority of well-informed patients would use melatonin over no treatment. The same guideline records that the British Association for Psychopharmacology recommended prolonged-release melatonin as a first-line treatment for insomnia in people over 55 — two bodies pointing opposite ways about different formulations, inside one document. The guideline also states that its review is based only on FDA-approved doses. This article makes no recommendation about any product.

Can I drive the day after taking a sleeping pill?

The Drug Office page states that patients should wait at least 12 hours after zopiclone before driving, and recommends not driving until 8 hours after zolpidem. But you cannot infer that other sleeping medicines have no limit — for sedating antihistamines and chloral hydrate the same page says patients should not drive or operate machinery after taking them, with no hours attached. The page also states for several of these medicines that complex sleep behaviours such as sleep driving can occur, with amnesia for the event. Ask your doctor or pharmacist about your own situation.

I snore very loudly and I am sleepy all day. Is that insomnia?

This article cannot answer that, and no source cited here can. Loud persistent snoring at night together with daytime sleepiness is itself what the Centre for Health Protection describes as the main symptoms of sleep apnoea — that alone is worth taking to a doctor. Not having these does not rule anything out either.

Do public hospitals offer CBT-I?

The Hospital Authority fee schedule this article examined carries no service named CBT-I. That is "not published under this name", which is not the same as the public system not providing it at all. Median waits for stable new psychiatric cases run from 19 weeks (Kowloon West) to 69 weeks (New Territories East), with the longest from 69 to 104 weeks; stable new cases account for 79% of Hospital Authority psychiatric new case bookings (42,169 of 53,664).

Is $20 per drug item at a public clinic one dispensing or a whole course?

Note 7 of the fee schedule states that each chargeable unit covers four weeks per drug item. So for an eligible person at a specialist out-patient clinic the drug element of one medicine is $20 every four weeks ($5 at a family medicine clinic). The fee schedule also sets out the fee waiver mechanism and the HK$10,000 annual cap requiring no financial assessment from 1 January 2026 as two stacked safety nets; the fee schedule this article examined does not define the scope of "chronic illness".

Is the Jockey Club Sleep Well Programme free?

The operator has written two first-hand sentences, but each has its own scope: the consent form sentence is about the assessments and course of that study, and the terms of use sentence is about the website. The operator publishes no schedule of service charges and says nothing about arrangements outside the research or after it ends. So whether the programme is free across the board is something this article cannot state. Ask the programme directly.

Is app-based CBT-I as effective as seeing a therapist?

No source cited here has made that comparison. The trial compared app-based CBT-I against app-based health education, not against face-to-face therapy. The AASM 2021 guideline states expressly that there was insufficient evidence to make recommendations about delivery method for any of the treatments, and NICE MTG70 likewise states there is no direct evidence of Sleepio's effectiveness compared with face-to-face CBT-I.

If it has been less than three months, can I ignore it?

No. Three months is a diagnostic threshold and the Centre for Health Protection's one month is a prompt to seek help; they do different jobs. Among the sources cited here the duration thresholds run from one month to twelve, and the only thing they share is the frequency: three times a week or more. Where two authorities differ, take the more cautious one and seek help earlier.

---

#What this article does not state

• Whether the Hong Kong public system actually provides CBT-I. The only material this article relies on is the Hospital Authority fee schedule; it does not include Legislative Council papers or Audit Commission reports. All this article can say is that the Hospital Authority has not published a service under that name.
• Charges for the Jockey Club Sleep Well Programme outside the research, and the programme's end date. Neither is published on the operator's website.
• How many CBT-I providers there are in Hong Kong. The sources cited here cover one organisation's own website.
• Hong Kong prescribing volumes for sleeping medicines, dependence rates and side-effect rates. The Drug Office page is qualitative, and carries no denominator or incidence rate anywhere.
• The wording of the British Association for Psychopharmacology consensus statement. This article has not obtained that document, and everything it says about it is limited to how NICE MTG70 and the AASM 2017 guideline each describe it.
• The original first-line wording used in UK primary care. This article cites only MTG70.
• Whether the AASM guidelines have been updated. This article cites the 2021 and 2017 documents.
• The text of the American College of Physicians' 2016 guideline is not among the sources cited here. That does not mean it lacks a "strong" CBT-I recommendation — the background section of AASM 2017 states that it has one, and with wider scope.
• The analysable number of participants per group at each time point in that trial. The paper reports percentages from a weighted model and does not publish denominators, so headcounts cannot be worked back from the percentages.
• The phrasing "42% lower incidence of depression than controls" is not used here. The paper reports a hazard ratio of 0.58; "42% lower" is a conversion of 1 − 0.58 and is not a figure the paper itself publishes.
• Whether the Hospital Authority's HK$10,000 cap covers chronic insomnia. The fee schedule does not define the scope of "chronic illness".
• The Hospital Authority's psychiatric triage criteria. What the Authority publishes is waiting times and case numbers by triage category, not the criteria.
• The diagnosis, testing indications or treatment of sleep apnoea. This article reports only the parts directly bearing on insomnia and sleeping medicines; the subject has its own article on this site.

---

#Notes: the official wording

Where a source publishes in English, the wording below is the source's own. Where a source has no English edition, the rendering is ours and is marked as such, with the Chinese original set out alongside so it can be checked.

[Note 1] The ICSD-3 definition of chronic insomnia, as restated in the AASM 2017 guideline:

Insomnia disorder is defined in the International Classification of Sleep Disorders, Third Edition as a complaint of trouble initiating or maintaining sleep which is associated with daytime consequences and is not attributable to environmental circumstances or inadequate opportunity to sleep. The disorder is identified as chronic when it has persisted for at least three months at a frequency of at least three times per week. When the disorder meets the symptom criteria but has persisted for less than three months, it is considered short-term insomnia. Occasional, short-term insomnia affects 30% to 50% of the population. The prevalence of chronic insomnia disorder in industrialized nations is estimated to be at least 5% to 10%.

[Note 2] The same source, the next sentence:

In medically and psychiatrically ill populations, as well as in older age groups, the prevalence is significantly higher.

[Note 3] Centre for Health Protection, "Men's mental health — insomnia", English edition:

Because insomnia can become a chronic problem, it is important to get it diagnosed and treated if it persists for more than a month.

[Note 4] The Drug Office page uses "such as" and "including" in its list of causes; the list is illustrative rather than exhaustive, and medicines beyond those named may also be involved.

[Note 5] The AASM 2021 guideline's definition of recommendation strength (summary and full text):

Each recommendation statement is assigned a strength ("strong" or "conditional"). A "strong" recommendation (ie, "We recommend…") is one that clinicians should follow under most circumstances. A "conditional" recommendation is one that requires that the clinician use clinical knowledge and experience, and to strongly consider the patient's values and preferences to determine the best course of action.

A "strong" recommendation is one that clinicians should follow for almost all patients (ie, something that might qualify as a quality measure).

[Note 6] The AASM 2017 guideline's background section on the ACP 2016 guideline:

In May 2016, the American College of Physicians published its own clinical practice guideline for the management of chronic insomnia. This guideline makes two major recommendations. The first is that all patients with chronic insomnia receive CBT-I as the initial treatment intervention. This is a strong recommendation based on moderate quality evidence.

[Note 7] The AASM 2021 guideline's scope statement on delivery methods:

There was insufficient evidence to make recommendations for specific delivery methods (eg, individual, group, internet, self-help, video) for any of the treatments.

[Note 8] The same source, methods section:

The review did not include comparisons of different interventions or combinations of pharmacotherapy with behavioral and psychological therapy.

The only head-to-head statement anywhere in the guideline is an observation within a single study, under the sleep hygiene recommendation (within the same study CBT-I was superior to sleep hygiene alone), and it comes from that one study.

[Note 9] NICE MTG70:

Usual treatment for people with sleep problems is advice about sleep hygiene. Sleeping pills may also be considered if insomnia symptoms are likely to resolve soon. If insomnia symptoms are not likely to resolve soon, best practice is to refer for face-to-face CBT‑I, although its availability in the UK is limited. This clinical pathway is outlined in the British Association for Psychopharmacology (BAP) consensus statement on insomnia. People who may be at higher risk of other sleep disorders including sleep apnoea should have a medical assessment before referral to Sleepio.

MTG70 recommendations 1.1 to 1.3:

1.1 Sleepio is recommended as a cost saving option for treating insomnia and insomnia symptoms in primary care for people who would otherwise be offered sleep hygiene or sleeping pills.

1.2 For people who may be at higher risk of other sleep disorder conditions, such as in pregnancy, or in people with comorbidities, a medical assessment should be done before referral to Sleepio.

1.3 More research or data collection is recommended on Sleepio for people who are eligible for face-to-face cognitive behavioural therapy for insomnia (CBT‑I) in primary care. This is because there is limited clinical evidence to show how effective Sleepio is compared with face-to-face CBT‑I.

[Note 10] MTG70, "Why the committee made these recommendations":

Sleepio is a digital self-help programme that includes CBT‑I. It could therefore increase patients' access to CBT‑I. It also increases the options available to GPs treating insomnia.

Clinical evidence shows that Sleepio reduces insomnia symptoms compared with sleep hygiene and sleeping pills. There is no direct evidence of its effectiveness compared with face-to-face CBT‑I, so further research is recommended in this context.

At a price of £45 per person, Sleepio is cost saving compared with usual treatment in primary care. This is based on an analysis of primary care resource use data before and after Sleepio was introduced in 9 GP practices. Healthcare costs were lower at 1 year, mostly because of fewer GP appointments and sleeping pills prescribed.

[Note 11] Recommendation 1 and its remarks:

We recommend that clinicians use multicomponent cognitive behavioral therapy for insomnia for the treatment of chronic insomnia disorder in adults.

Remarks: This recommendation is based primarily on studies in which CBT-I was delivered by a trained professional to patients with and without comorbid conditions.

Recommendation 6 and its remarks:

We suggest that clinicians not use sleep hygiene as a single-component therapy for the treatment of chronic insomnia disorder in adults.

Remarks: Although sleep hygiene is not recommended as a single-component approach (ie, the only treatment) for patients with chronic insomnia disorder, sleep hygiene may be included in multicomponent interventions.

[Note 12] The safety note on sleep restriction therapy:

Clinicians should note that this treatment may be contraindicated in certain populations such as those working in high risk occupations (eg, heavy machinery operators or drivers) or those predisposed to mania/hypomania poorly controlled seizure disorders or excessive daytime sleepiness.

(A comma is missing in the source sentence; it is reproduced as printed.)

[Note 13] The safety note on stimulus control:

The TF noted that stimulus control may need to be adapted for safety in some patient populations, such as those at high risk for falls, with mobility issues, or using sedative-hypnotics.

[Note 14] The Drug Office page, treatment section, English edition:

If insomnia persists despite these measures and affects daily functioning, consult your doctor. Your doctor can identify underlying causes and recommend appropriate treatment, which may include cognitive behavioural therapy, sleep hygiene reinforcement, or medications.

[Note 15] Non-Communicable Diseases Watch, March 2018:

Most insomnia can be effectively treated through medications, or cognitive behavioural interventions to identify and correct maladaptive behaviours or thoughts about sleep.

It can also be prevented from recurrence by paying attention to 'sleep hygiene' (Box 1).

However, insomnia is often under-recognised and under-treated in both children and adult populations despite its impact on daily functioning and quality of life.

[Note 16] The seven items under "How can I deal with insomnia?", Centre for Health Protection, English edition:

Get into a daily routine, especially in relation to sleep

Stop smoking

Get regular exercise, but avoid strenuous activity immediately before going to bed.

Do not go to bed with a full or empty stomach. Avoid caffeine-containing drinks such as coffee, cola or tea.

Make your sleeping place comfortable: dark, quiet, not too warm or too cold.

Do something to relax, such as meditation, listen to soothing music, or have a warm bath.

If you can't sleep, get up and do something else until you feel sleepy again.

Two sentences qualify that list before it begins: the page first says "Determine and treat the cause of insomnia", and only then that "If insomnia is only temporary, it can be resolved by methods, other than medication".

Immediately after the list the same page adds:

Persistent insomnia needs professional support. Do not take over the-counter treatment. Consult your doctor who will look out for any underlying cause of insomnia and prescribe appropriate treatment for you.

[Note 17] The participants and recruitment of the Chen et al. 2025 trial:

This was a randomized, assessor-blind, parallel group-controlled trial conducted in Chinese youth (aged 15−25 years) with insomnia disorder and subclinical depression. Potential participants were recruited from universities, high schools, and communities in mainland China and Hong Kong from September 9, 2019 to November 25, 2022.

Between September 9, 2019, and November 25, 2022, 708 participants (407 females [57%]; mean age, 22.1 years [SD = 1.9]) were randomly allocated to app-based CBT-I group (n = 354) or app-based HE group (n = 354).

Participants were randomly assigned (1:1) to 6-week app-based CBT-I or 6-week app-based health education (HE) delivered through smartphones.

[Note 18] The primary outcome:

Thirty-seven participants (10%) in the intervention group and 62 participants (18%) in the control group developed new-onset MDD throughout the 12-month follow-up, with a hazard ratio of 0.58 (95% confidence interval 0.38–0.87; p = 0.008). The number needed to treat to prevent MDD at 1 year was 10.9 (6.8–26.6).

[Note 19] How primary and secondary outcomes were defined:

The primary outcome was time to onset of MDD. The secondary outcomes included depressive symptoms and insomnia at both symptom and disorder levels.

The three time points for insomnia remission, in the original:

At post-intervention, the weighted percentage of remitters based on clinical interviews in app-based CBT-I group was higher than that of the app-based HE group (52% versus 28%; relative risk [RR] 1.83 [95% CI 1.49–2.24]; p < 0.001) after controlling for strata variables and accounting for the missing data patterns.

Improvements achieved in the app-based CBT-I group were well sustained throughout the follow-up period, with higher remission rates than the controls at both 6-month (56% versus 44%; RR 1.27 [1.08–1.48]; p = 0.003) and 12-month follow-ups (57% versus 48%; RR 1.17 [1.01–1.35]; p = 0.03).

[Note 20] The definition of remission:

Remission of insomnia disorder was determined by the following criteria: (1) no functional impairments or distress with sleep, (2) insomnia symptoms less than three times per week for at least 1 month, and (3) sleep-promoting medication use less than once per week in the past month. Although the use of sleep-promoting medication is not a standard criterion for diagnosing insomnia disorder, it may mask the underlying symptoms and was thus taken into account in the current study.

Remission rates of insomnia were analyzed using a weighted generalized estimating equations model.

[Note 21] The paper's limitations section:

First, participants in our study included a mix of individuals with and without a prior history of MDD, with 48% (339/708) of participants reporting a history of prior MDD.

[Note 22] The same section, last sentence:

Last, the modest sample size of adolescents (<18 years, n = 10) limits the generalizability of the findings to this age group.

Safety:

No adverse events related to the interventions were reported.

The paper's declarations of interest: YKW has received lecture and consultancy fees from Eisai Co, Ltd and sponsorship from Lundbeck HK Ltd and Aculys Pharma, Inc (unrelated to this study); CMM has received research support from Idorsia, Eisai and Lallemand Health Solutions, personal fees from Idorsia, Eisai and Haleon, and royalties from Mapi Research Trust (unrelated to this study); JHZ has provided consultancy to BestCare & SuMian BioTech Co, Ltd; JWYC has received personal fees from Eisai Co., Ltd and travel support from Lundbeck HK Ltd for overseas meetings outside the scope of this study.

[Note 23] The systematic review's denominators:

The literature search identified 1,244 studies; 124 studies met the inclusion criteria, and 89 studies provided data suitable for statistical analyses. Evidence for the following interventions is presented in this review: cognitive-behavioral therapy for insomnia, brief therapies for insomnia, stimulus control, sleep restriction therapy, relaxation training, sleep hygiene, biofeedback, paradoxical intention, intensive sleep retraining, and mindfulness.

[Note 24] The denominator under recommendation 1:

The TF identified 66 randomized controlled trials (RCTs) in adult patients diagnosed with chronic insomnia disorder that compared CBT-I to wait-list, minimal interventions, or placebo therapies. Forty-nine of the studies provided data suitable for meta-analyses for at least one critical outcome.

AASM 2017 guideline:

GRADE recommendation strengths do not refer to the magnitude of treatment effects in a particular patient, but rather, to the strength of evidence in published data.

[Note 25] The AASM 2017 guideline's passage against misreading:

It is essential that the recommendations which follow be interpreted within the appropriate context of clinical practice. Readers will note that all specific recommendations fall within the "weak" (for or against) classification of the GRADE system. This should not be construed to mean that no sleep-promoting medications are clearly efficacious or indicated in the treatment of chronic insomnia. Hypnotic medications, along with management of comorbidities and non-pharmacological interventions such as CBT, are an important therapeutic option for chronic insomnia.

The same source, on recommendation strength:

A WEAK recommendation reflects a lower degree of certainty in the outcome and appropriateness of the patient-care strategy for all patients, but should not be construed as an indication of ineffectiveness.

The ultimate judgment regarding propriety of any specific care must be made by the clinician in light of the individual circumstances presented by the patient, available diagnostic tools, accessible treatment options, and resources.

[Note 26] The same source, on why the evidence was downgraded:

Downgrading the quality of evidence for these treatments is predictable in GRADE, due to the funding source for most pharmacological clinical trials and the attendant risk of publication bias; the relatively small number of eligible trials for each individual agent; and the observed heterogeneity in the data.

[Note 27] The remark on zolpidem is the only one of the fourteen that runs to more than a sentence, and the only one carrying a regulator's subsequent action; reproduced in full:

Remarks: This recommendation is based on trials of 10 mg doses of zolpidem. N.B. Although 10 mg. was the recommended starting dosage for adults at the time of initial approval, the FDA has subsequently lowered the recommended starting dosage of immediate-release zolpidem products to 5 mg. Further, the FDA has recommended a reduction of starting dosage for extended-release forms of zolpidem from 12.5 mg to 6.25 mg.

[Note 28] The remark on melatonin gives the dose as 2 mg, and the three relevant studies enrolled people over 55 only — the original reads only older adults (> 55 years). The same summary carries three further statements: the quality of evidence is very low; the task force judged the benefits to be approximately equal to harms; and the majority of well-informed patients would use melatonin over no treatment.

[Note 29] The AASM 2017 guideline's background section on the BAP consensus statement:

Limited evidence and toxicity concerns were cited for other prescription and non-prescription agents, although prolonged-release melatonin was recommended as a first-line treatment for insomnia in persons over 55 years.

[Note 30] The same source, dose scope statement:

Finally, the literature review, meta-analyses, and recommendations are based only on FDA-approved doses. This should not be interpreted as a recommendation for the use of a specific dose in clinical practice.

The same source, on the guideline's own scope:

The purpose of this guideline is to establish clinical practice recommendations for the pharmacologic treatment of chronic insomnia in adults, when such treatment is clinically indicated. Unlike previous meta-analyses, which focused on broad classes of drugs, this guideline focuses on individual drugs commonly used to treat insomnia. It includes drugs that are FDA-approved for the treatment of insomnia, as well as several drugs commonly used to treat insomnia without an FDA indication for this condition.

[Note 31] The Drug Office page, English edition:

In Hong Kong, most medications used to treat insomnia are prescription-only medicines and should be taken under a doctor's supervision.

Do not self-medicate or take over-the-counter sleeping pills without professional advice.

[Note 32] The same source, prescribing principles:

Hypnotics are medications that act on the central nervous system (CNS) to induce sleep. These medications do not address the root cause of insomnia and should be prescribed by a doctor, used at the lowest effective dose, and taken for the shortest possible duration. They are suitable for short-term management of severe and disabling insomnia.

The same page also carries a positive statement:

Benzodiazepines shorten the time to fall asleep, decrease awakenings during the night and increase total sleeping time.

The threshold for use and the duration limit for benzodiazepines, two consecutive sentences:

Due to the development of tolerance, dependence and withdrawal symptoms, benzodiazepines should be used to treat insomnia only when it is severe, disabling, or causing the patient extreme distress. Benzodiazepines, when used as hypnotics, should be administered at the lowest effective dose for the shortest possible duration, which should not exceed 4 weeks.

[Note 33] ⚠️ This row cannot be read in halves, and the two language editions differ on where it ends. The English edition sets the cap out as a sentence of its own:

The use of chloral hydrate in children and adolescents is not generally recommended, and if used should be under the supervision of a medical specialist. Treatment should be for the shortest duration possible and should not exceed the maximum treatment period of 2 weeks.

The Chinese edition runs the whole thing as a single sentence whose subject is children and adolescents from beginning to end, so the two-week cap reads there as bounded to that group, where the English reads as a general rule. This article follows the Chinese edition, which is the narrower reading, and marks the difference here.

[Note 34] The caution list for sedating antihistamines, and where the two editions diverge. The Chinese edition carries all six items — prostatic hyperplasia, narrow-angle glaucoma, asthma, bronchitis, liver or renal dysfunction, myasthenia gravis — in both its prose and its side-effects table. The English edition carries all six in its prose:

Sedating antihistamines should be used with caution in patients with prostatic hyperplasia, narrow-angle glaucoma, asthma, bronchitis, liver or renal dysfunction, myasthenia gravis.

but only four in its side-effects table:

Caution in patients with asthma, bronchitis, narrow angle glaucoma, prostatic hypertrophy

So the divergence is between that page's own prose and its own table, not between the languages. This article uses the six-item list, which both editions' prose supports.

[Note 35] Zopiclone and zolpidem, English edition:

However, there is little evidence to show any clinical advantages of zopiclone over benzodiazepines in terms of tolerance, dependence or withdrawal symptoms

Zolpidem tartrate is used as a hypnotic in the short-term management of insomnia. It is reported to have similar sedative properties as the benzodiazepines, but very little anxiolytic and muscle relaxant properties in comparison. Zolpidem has a rapid onset and short duration of hypnotic action, but there is little evidence to show any advantage over short-acting benzodiazepines in terms of 'hangover feeling', or its potential to induce tolerance or withdrawal symptoms or dependence.

DORA:

Dual orexin receptor antagonists (DORA) are central nervous system (CNS) depressants that can impair daytime wakefulness even when used as prescribed. CNS depressant effects may persist in some patients for up to several days after discontinuing the medicine.

Stopping chloral hydrate:

Abrupt discontinuation should not be undertaken in patients receiving prolonged treatment with chloral hydrate. Slowly withdraw chloral hydrate.

[Note 36] "General Advice on Taking Sleeping Medicines", the six items in the English edition:

Use only under medical supervision.

Always take the lowest effective dose for the shortest duration.

Be aware of the medicine name and dosage.

Avoid alcohol while on sleeping medicines.

Do not drive or operate machinery after taking the medicine.

If symptoms persist, consult your doctor for alternative treatments.

The two editions order these differently: the last two items appear in the opposite order in Chinese, which is why this article never identifies them by position.

[Note 37] The Hospital Authority fee schedule lists a series of named service items, and the strings CBT, Cognitive and insomnia each appear 0 times on that page, while "Psychiatric day hospital" appears 2 times and the footnote "N7" 7 times — which demonstrates that this article did in fact read the page's service names.

⚠️ But the specialist out-patient waiting time open dataset cannot be used to make the same judgement, and this article does not use it for that. That dataset has 288 rows and only the fields date_id_from, date_id_to, cluster, specialty, Category, Description and Value; specialty takes only eight values across the whole file and Description only five strings. It has no "service name" field at all — no service's name would ever appear in it. Even if every cluster were delivering CBT-I, this dataset would look exactly the same. A precise figure like "288 rows" makes an empty check look like a search that was carried out, so this article says here that it was not.

[Note 38] The Hospital Authority fee schedule, two consecutive sentences:

People who have financial difficulties in paying medical expenses at the public sector may apply for medical fee waiver.

In addition to the medical fee waiving mechanism, the HA establishes a second safety net by introducing a HK$10,000 cap on public medical fees and charges without requiring financial assessment with effect from 1 January 2026. The new measure aims at easing the financial burden faced by patients suffering from sudden severe illnesses or chronic illnesses.

Footnote N7:

N7 Each chargeable unit covers a duration of 4 weeks per drug item.

[Note 39] Non-Communicable Diseases Watch, key messages and body:

Most insomnia can be effectively treated and prevented from recurrence by paying attention to good 'sleep hygiene'. If insomnia has been interfering with your daytime functioning, please consult your family doctor.

Thus, never use alcohol as a sleeping aid and be sure to discuss with your family doctor before taking any sleeping medications.

[Note 40] ⚠️ On whether these two sentences can be checked. That website is server-side generated HTML and most of its content is readable in the rendered text layer of its own page; but these two sentences are not. Checking the eighteen paths reachable from the site's navigation, neither sentence appears in the rendered text layer of any of them (0/18); they appear only in the bytes the server sends, inside a site-wide string dictionary that every one of the eighteen paths carries. So this article cannot give a URL a reader could open to check them. The consent form sentence sits under researchConsentForm in that dictionary (the document titles itself a participant information and consent form of the Department of Psychiatry, Faculty of Medicine, CUHK, subtitled on validating the effectiveness of cognitive behavioural therapy for insomnia in the community); the terms of use sentence sits under terms. The navigation's "disclaimer" link goes to /disclaimer, which in fact displays a list of organiser and co-organiser disclaimer links rather than those terms of use. To check them, ask the programme for the two documents directly.

The site has two other descriptions containing the word "free": one offering booking of a free nap and light-exposure experience, and one mentioning a chance of receiving a free small gift at programme activities. The programme's site publishes in Chinese only; those renderings are ours.

[Note 41] The programme's website. The site publishes in Chinese only and no English edition exists, so the renderings below are ours, with the Chinese original given so they can be checked.

Most insomnia problems can be managed and cured. However, support in the community targeting sleep problems is very limited, and the public's help-seeking behaviour and understanding of sleep problems are also very limited. (our translation from the Chinese original)

Chinese original:

大多數的失眠問題都是有方法可以處理及可治癒的。可是，社區上針對睡眠問題的支援卻十分有限，同時，市民對於睡眠問題的求助行為及認識亦很少。

The programme also describes itself:

The Jockey Club Sleep Well Programme is funded by The Hong Kong Jockey Club Charities Trust and is Hong Kong's first community promotion programme targeting sleep problems. (our translation from the Chinese original)

Chinese original:

賽馬會樂眠無憂計劃由香港賽馬會慈善信託基金捐助，是香港首個針對睡眠問題的社區推廣計劃。

This platform uses cognitive behavioural therapy for insomnia (CBT-I). CBT-I is a recommended first-choice treatment for insomnia, and its effect lasts even longer than that of medication. (our translation from the Chinese original)

Chinese original:

本平台採用「失眠的認知行為治療」(Cognitive Behavioral Therapy for Insomnia, CBT-I)，CBT-I 是一套治療失眠的建議首選方案，療效甚至比藥物更加持久。

Source: the programme website, cited in the sources block below. The site serves lang="zh-hant" and has no English edition; the paths that would carry one are absent, and the served bytes contain no language switcher.

[Note 42] Non-Communicable Diseases Watch, March 2018:

In Hong Kong, the Population Health Survey 2014/15 of the Department of Health interviewed over 12 000 non-institutionalised persons aged 15 and above and found that close to half (48.0%) of them had experienced sleep disturbance, namely 'difficulty in falling asleep within 30 minutes', 'intermittent awakenings or difficulty in maintaining sleep during the night', and 'early morning awakening and unable to sleep again', during the 30 days preceding the survey.

As shown in Figure 3, the proportions of having experienced these sleep disturbances three or more times a week in the 30 days preceding the survey were 6.8%, 7.3% and 6.0% respectively. Overall, females and people aged 65 and above were more likely to have experienced sleep disturbance.

The twelve values in Figure 3 are the data labels printed on the figure, read cell by cell at 600 dpi rather than estimated from bar lengths. The third bar in that figure is captioned with the plural Early morning awakenings while the body text on the same page uses the singular early morning awakening for the same item; the table above follows the figure as printed.

[Note 43] The same source, the international gradient (this passage precedes the 48.0% passage in the original):

…prevalence of insomnia varied from as low as 6% with stringent criteria to diagnose insomnia, to 10-15% in people reported insomnia symptoms with daytime consequences (e.g. because of fatigue), and further up to 48% in those having any insomnia symptoms (Figure 2).

The publication notes beneath Figure 2 that this gradient is "Modified from Ohayon 2002"; this article reports it as the publication states, and has not obtained the underlying paper.

[Note 44] The same source, on help-seeking:

In Hong Kong, studies reported that only about 10% of children and adolescents and 40% of adults with insomnia reported having sought treatment for insomnia.

While a substantial proportion of insomniacs had sought treatment using complementary and alternative medicine (such as Chinese herbal medicine and acupuncture), some turned to over-the-counter sleeping aids or even alcohol for relief. Of note, alcohol aids sleep is a common misconception. In fact, alcohol disrupts normal sleep pattern depriving the drinker of a sound and restful sleep.

[Note 45] Centre for Health Protection sleep quiz, English edition:

True. Persistent loud snoring at night and daytime sleepiness are the main symptoms of a common and serious sleep disorder, sleep apnea. Another symptom of sleep apnea is frequent long pauses in breathing during sleep, followed by choking and gasping for breath. People with sleep apnea don't get enough restful sleep, and their daytime performance is often seriously affected. Sleep apnea may also lead to hypertension, heart disease, heart attack, and stroke. However, it can be treated, and the sleep apnea patient can live a normal life.

True. Many people doze off unintentionally during the day despite getting their usual night of sleep. This could be a sign of a sleep disorder. Approximately 40 million Americans suffer from sleep disorders, including sleep apnea, insomnia, narcolepsy, and restless legs syndrome. An untreated sleep disorder can reduce your daytime productivity, increase your risk of accidents, and put you at risk for illness and even early death.

[Note 46] The same source:

True. People with narcolepsy fall asleep uncontrollably --at any time of the day, in all types of situations-- regardless of the amount or quality of sleep they've had the night before. Narcolepsy is characterized by these 'sleep attacks,' as well as by daytime sleepiness, episodes of muscle weakness or paralysis, and disrupted nighttime sleep.

[Note 47] The same source:

True. Restless legs syndrome (RLS) is a medical condition distinguished by tingling sensations in the legs--and sometimes the arms--while sitting or lying still, especially at bedtime. The person with RLS needs to constantly stretch or move the legs to try to relieve these uncomfortable or painful symptoms. As a result, he or she has difficulty falling asleep or staying asleep and usually feels extremely sleepy and unable to function fully during the day.

[Note 48] Non-Communicable Diseases Watch:

However, insomnia can become a chronic problem with cycles of relapse and remission or persistent symptoms, especially when it co-exists with chronic diseases or mental illnesses, substance abuse, chronic pain, or other sleep disorders (such as obstructive sleep apnoea and restless leg syndrome).

On quotation convention: quotation marks in this article do two jobs — reproducing a source's wording, and marking this article's own terms of art (such as "weak", "conditional", "suggested against"). Block quotations set on their own lines are always the source's wording, except where marked as our translation.

---

#Sources and dates checked

• AASM 2017 pharmacological guideline (chronic insomnia frequency and duration thresholds, prevalence, the 14 recommendations and the doses in their remarks, the passage against misreading, the FDA-approved-doses statement, and the background descriptions of the BAP consensus statement and the ACP 2016 guideline): Sateia MJ, Buysse DJ, Krystal AD, Neubauer DN, Heald JL. Clinical Practice Guideline for the Pharmacologic Treatment of Chronic Insomnia in Adults. J Clin Sleep Med. 2017;13(2):307–349 (PMID 27998379; DOI 10.5664/jcsm.6470). Full text: https://pmc.ncbi.nlm.nih.gov/articles/PMC5263087/. Published 15 February 2017. Retrieved 3 August 2026.
• AASM 2021 behavioural and psychological treatment guideline (the six recommendations, their remarks, the strength definitions, the delivery-method statement, the safety notes, trial counts): Edinger JD, Arnedt JT, Bertisch SM, et al. J Clin Sleep Med. 2021;17(2):255–262 (PMID 33164742; DOI 10.5664/jcsm.8986). Full text: https://pmc.ncbi.nlm.nih.gov/articles/PMC7853203/. Published 1 February 2021. Retrieved 3 August 2026.
• The systematic review (1,244 → 124 → 89 and the ten interventions): Edinger JD, et al. J Clin Sleep Med. 2021;17(2):263–298 (PMID 33164741; DOI 10.5664/jcsm.8988). Full text: https://pmc.ncbi.nlm.nih.gov/articles/PMC7853211/. Retrieved 3 August 2026.
• The app-based CBT-I trial (recruitment sites, primary and secondary outcomes, the nature of the control arm, the definition of remission, the weighted model, the four limitations, declarations of interest): Chen SJ, Que JY, Chan NY, et al. PLoS Med. 2025;22(1):e1004510 (PMID 39836656; DOI 10.1371/journal.pmed.1004510; ClinicalTrials.gov NCT04069247). Full text: https://pmc.ncbi.nlm.nih.gov/articles/PMC11750088/. Published 21 January 2025. Retrieved 3 August 2026.
• Department of Health Drug Office, "Oral Sleeping Medicines" (prescription status, insomnia symptoms, the table of medicines that may cause insomnia, the duration limits and contraindications by class, tolerance and dependence, rebound insomnia, complex sleep behaviours, driving waiting times, the six general advice items, and the page's own disclaimer): English version https://www.drugoffice.gov.hk/eps/do/en/consumer/news_informations/dm_07.html; Chinese version https://www.drugoffice.gov.hk/eps/do/tc/consumer/news_informations/dm_07.html. The page dates itself January 2026. Retrieved 3 August 2026. The two language editions are published independently and correspond broadly rather than sentence for sentence; the two divergences this article marks are at Notes 33 and 34.
• Centre for Health Protection, "Men's mental health — insomnia" (the one-month prompt, the definition of insomnia, the seven sleep hygiene items, the sleep knowledge quiz, and the "more than 40%" figure for men): English version https://www.chp.gov.hk/en/static/80083.html; Chinese version https://www.chp.gov.hk/tc/static/80083.html. The page carries no revision date. Retrieved 3 August 2026. This page sits in a men's health section and its statistic is specific to men.
• Centre for Health Protection, Non-Communicable Diseases Watch, March 2018 (48.0% and 6.8/7.3/6.0, all twelve values in Figure 3, the 6%–48% international gradient, help-seeking proportions, the alcohol misconception, the family doctor instruction, and co-existence with other sleep disorders): https://www.chp.gov.hk/files/pdf/ncd_watch_march_2018.pdf. Publication date March 2018. Retrieved 3 August 2026 (Figure 3 values retrieved 9 August 2026).
• Jockey Club Sleep Well Programme (the nature of the programme, who it serves, tiered support, the waiting period, research questionnaire obligations, the two sentences on charges, and "about 30% of Hong Kong people have insomnia symptoms"): https://jcsleepwell.cuhk.edu.hk/ and the eighteen paths reachable from its navigation. The site carries no revision date (the footer copyright year is 2023). Retrieved 9 August 2026. The site publishes in Chinese only; renderings of its wording in this article are ours and are marked at each quotation. On the readability question, see Note 40.
• Hospital Authority fee schedule (the service charges, footnote N7, the fee waiver mechanism and the HK$10,000 cap; no CBT-I item appears in the table): https://www.ha.org.hk/visitor/ha_view_content.asp?Content_ID=10045&Lang=ENG&Dimension=100&Ver=HTML. Effective 1 January 2026. Retrieved 3 August 2026.
• Hospital Authority psychiatric specialist out-patient waiting times and triage distribution (the seven cluster medians and longest figures, and the case numbers and shares of the three categories): https://www.ha.org.hk/opendata/sop/sop-waiting-time-en.json. Data period 1 July 2025 to 30 June 2026; the dataset states its next update date as 30 October 2026. Retrieved 3 August 2026. That dataset's fields are date_id_from, date_id_to, cluster, specialty, Category, Description and Value, and it has no service name field.
• NICE Medical Technologies Guidance MTG70, Sleepio to treat insomnia and insomnia symptoms (limited UK availability, the pathway attributed to the BAP consensus statement, medical assessment before referral, recommendations 1.1 to 1.3, £45 and 9 practices): https://www.nice.org.uk/guidance/mtg70/chapter/1-Recommendations. Guidance published 20 May 2022; the wording quoted follows that page's version of 10 September 2025. Retrieved 3 August 2026.
• Related subject: the article on this site about sleep apnoea testing — polysomnography, home sleep tests and screening devices, and what they cost.

This article was written from the sources listed above. It is about thresholds, evidence and what is actually available; what you should do about your own sleep, it does not answer — that is for a doctor, starting with your family doctor.

---

#Further reading

• On this platform: Sleep apnoea testing: PSG, home sleep tests and screening devices, and what they cost
• American Academy of Sleep Medicine patient information: sleepeducation.org
• Department of Health Drug Office, "Oral Sleeping Medicines": drugoffice.gov.hk