How do you read a private hospital surgery bill? Four rules for the 50th and 90th percentiles, and two mechanisms that make the numbers incomparable, 2026

TL;DR Thirteen private hospitals publish actual bill statistics for 30 common operations on a government platform. The commonest misunderstanding is that the law requires this — it does not. It comes from a voluntary pilot scheme of 2016, and the price transparency provisions of the Private Healthcare Facilities Ordinance have still not commenced. And these figures cover standard wards only, and single-procedure admissions only: if you book a private room, or have two things done in one admission, you are not in the dataset at all.

---

#These figures are not required by law; the hospitals volunteer them

This is a voluntary scheme, not a statute. The Government and the Hong Kong Private Hospitals Association launched the Pilot Programme on Enhancing Price Transparency of Private Hospitals on 1 October 2016, and the Department of Health set up a dedicated platform to publish the data.

The Government's press release of 29 September 2016 states that all members of the Association would implement three measures on a voluntary basis: introducing budget estimates for services; publishing on hospital websites the price lists of major items; and publishing on hospital websites the actual historical bill size statistics of common operations and procedures [Note 1].

All three measures are voluntary, not statutory duties. The title of the platform's own page is worth noticing too: "List of Operations/Procedures Recommended for the Publication of Historical Bill Sizes Statistics" — recommended for publication, not required.

⚠️ The Ordinance commencing and its price transparency provisions commencing are two different things. The Private Healthcare Facilities Ordinance (Cap. 633) is itself in force — hospital licensing from 1 January 2021, clinic licence applications accepted from 13 October 2025 — but sections 61 to 63 of it, which deal with price transparency, had still not commenced as at 1 August 2026, with the draft subsidiary legislation expected to be tabled in the Legislative Council in 2026 [Note 2]. The Consumer Council's press release of 6 March 2025 puts it plainly: the provisions of the Ordinance relating to price transparency have not yet come into operation.

So the legal standing of the data you are looking at today is "the hospitals supply it voluntarily".

Participation: the scheme began with 11 private hospitals. According to the Health Bureau's paper to the Legislative Council of May 2025, all 14 private hospitals now take part on a voluntary basis (as at the end of 2024 Hong Kong had 14 licensed private hospitals with 5 194 beds between them). On the platform's hospital list, however, one of them carries a note the platform itself has added: "Historical Bill Sizes Statistics are not applicable". The effect is that fourteen take part and thirteen have bill data.

The format of the data was set out in a written reply to the Legislative Council on 8 May 2019: annual number of discharges, average length of stay, and the actual billing data for the 50th and 90th percentile of each specified treatment or procedure [Note 3].

⚠️ If you assumed the Government regulates private hospital prices, note that under the present system it does not. The Government's own position is stated clearly: private medical service is by its nature no different from other business transactions between consenting parties, where prices are determined by market forces, and the Government will not regulate the price level or price structure of services provided by private healthcare facilities.

But clause 12.3 of the Medical Council's Code of Professional Conduct provides that a doctor must not overcharge or accept excessive remuneration, and sets out three tests. So there is a route for "is this too expensive", not only for "was I told in advance" — though it runs through the Medical Council's professional misconduct machinery, not through government price control.

The scheme did not appear from nowhere. It starts with the Director of Audit's Report No. 59 of October 2012, which suggested that the Government explore ways to enhance the transparency of fees levied by private hospitals — for example by requiring hospitals to inform every patient, prior to or upon admission, of the estimated total fees likely to be incurred, and by enacting laws on transparency in medical fees [Note 4]. Everything after that is on the official record: a public consultation in December 2014; the voluntary pilot in October 2016; the Bill introduced into the Legislative Council in June 2017; passage on 15 November 2018 and gazettal on 30 November.

---

#Four rules for using this data

The platform sets out all four in the notes at the foot of every page — but they are below the table, and many people copy the numbers and leave.

Rule one: standard wards only, single-procedure cases only. The platform states that the information is based on data for patients accommodated in standard wards undergoing a single operation or procedure, and does not include patients undergoing multiple operations or procedures; and because the items covered may not be entirely the same at each hospital, you should ask the hospital directly what each operation or procedure covers [Note 5].

⚠️ So if you have booked a twin room or a private room, or you are having two things done in one admission, you are not in this dataset at all. And a good deal of medical insurance is written at the twin-room or private-room level.

Rule two: the percentiles are ranked on total charges, not on doctor's fees and hospital charges separately. In the platform's words: the percentile figures above are ranked by the amount of the "total charges", not by the "doctor's fees" and "hospital charges" within the total.

In one sentence: the "fiftieth doctor's fee" on a row is not "the median doctor's fee for this operation at this hospital"; it is "the doctor's fee in the case that ranks 50th by total charge".

⚠️ You can see the difference in the data directly: in the in-patient hernia repair table, the 90th percentile hospital charge on Canossa Hospital's laparoscopic row ($47,891) is lower than its own 50th percentile ($55,319). If the two columns were ranked independently, that could never happen. So: do not build hospital A's doctor's fee onto hospital B's hospital charges, and do not read a row's 90th percentile doctor's fee as "the most expensive doctor".

The platform also notes that all charge figures are shown in whole Hong Kong dollars, so that individual figures may not add to the total because of rounding.

Rule three: the data may be out of date, and out of date by different amounts at different hospitals. The Health Bureau's Legislative Council paper of May 2025 states that some private hospitals may only submit the previous year's data by the third quarter of the following year, so some information on the website may not be up to date [Note 6]. The Consumer Council's review of March 2025 is more specific: at 4 private hospitals, the historical charge statistics on the website were at least a year and a half old.

⚠️ The reporting period shown on the platform at the time of retrieval was 1 January to 31 December 2025, but the platform does not publish each hospital's own submission date. So a cross-hospital comparison may be comparing different years, and you cannot tell which row is which year.

Rule four: the 90th percentile is a distribution statistic — but do not overstate it. The platform's definition is that the 90th percentile is the amount below or equal to which 90 per cent of patients paid, with the remaining 10% paying more [Note 7]. At the same time, the first sentence of the same notes says that actual length of stay and total charges may vary with the patient's condition and the complexity of the disease.

So: a high 90th percentile reflects the upper end of the bill distribution for that procedure at that hospital. The platform does not say that upper end is purely "more items included", and does not say it is purely "more complex cases" — it has separated neither factor. Saying that much is right; pushing further is adding your own explanation.

⚠️ One technical warning as well. The platform cautions that bill data for procedures with fewer than 30 cases reported by a hospital should be interpreted with care (the English version says "less than 30 cases" and the Chinese says "not more than 30", which are not quite the same threshold). And discharge numbers are published only in bands, of which there are four: <30, 30 - 100, 101 - 200 and >200.

<30 is itself a published band meaning "fewer than 30 cases", not "not published". This article's approach: any row whose discharge band is <30 is treated as small-count data and is not used to build a range.

---

#Before you look at the numbers: four pairs not to confuse

• A price list is not a bill. A price list is the sticker price of single items; a bill is what is actually settled. And by the hospitals' own notes a price list excludes doctor's fees, so any estimate built by adding up a price list will understate.
• The median is not the typical case. The platform defines the 50th percentile as the amount below or equal to which 50 per cent of patients paid. It is a position in a distribution, not your bill.
• The doctor's fee is not the total. Within this same dataset there is no rule about which is larger — in the in-patient haemorrhoidectomy table, Matilda's 50th percentile hospital charge ($36,610) is higher than its doctor's fee ($22,849), while Hong Kong Adventist Hospital – Stubbs Road is the other way round (doctor's fee $45,850, hospital charges $34,137). "The doctor's fee is always most of it" is wrong.
• Day surgery is not in-patient, and a standard ward is not another ward. The platform splits day and in-patient into two tables, day meaning procedures discharged the same day (including those done as an out-patient service); and both tables are standard-ward data only.

---

#How a bill is built: two big classes of charge, issued by different people

A bill is not one "price for the operation" but at least six classes of charge stacked up — and two of the big ones, doctor's fees and hospital charges, are issued by different people.

The platform defines what each covers, word for word [Note 8]: doctor's fees include fees for the anaesthetist, the operating surgeon and ward rounds; hospital charges include the admission fee, accommodation, the operating theatre and associated materials, nursing procedures, investigation and examination fees, medication, meals and beverages, sundries and so on.

⚠️ So "doctor's fees" is not only the operating surgeon — the anaesthetist and every daily ward round are inside it. Many people get this wrong.

One level further down. St. Teresa's Hospital's price list of common items (the document is dated 1 July 2026) says the same thing in three separate places: every figure on the price list excludes doctor's fees [Note 9]. So adding up a hospital price list will not equal a bill.

---

#The first mechanism that makes headline figures incomparable: ward class reprices the whole bill

On the same price list, the basic operating theatre charge for one operation has four prices, one for each ward class:

Working one out (ward class): for the same open hernia repair, counting only the basic operating theatre charge, a standard room is $2,020 and a private room $3,130 — $1,110 more, or +55.0% (1,110 ÷ 2,020 = 0.5495). The operation is no different; what differs is which room you are in.

On the same price list, daily room charges in the special treatment unit: standard room $3,300, twin room $4,300, private room $5,300, deluxe private room $6,300, negative pressure room $7,000; day beds at the haemodialysis centre: standard room $400, single room $450. (These are the special treatment unit's and haemodialysis centre's own rates, not general surgical ward rates — general wards have their own, far lower, room charges on the same price list. They are used here only as an example of room charges following ward class.)

⚠️ And the hospital notes on the same page that medical and associated charges are commensurate with the bed type of ward [Note 10]. In one sentence: which room you choose affects not just the room charge but the whole bill.

The Consumer Council's study of March 2025 makes the same point and finds the rationale unclear: doctor's fees and hospital charges are often adjusted according to ward class, but the rationale behind that linkage is not clear; and some consumers regard it as unfair that the same treatment should cost different amounts merely because of a difference in accommodation.

⚠️ This mechanism collides directly with the data below: the percentiles the platform publishes are standard ward only.

---

#The second mechanism: one operation name can be two different techniques

The platform publishes 7 procedures split across two rows by surgical technique: cholecystectomy, colectomy, hernia repair, hysterectomy, open reduction and internal fixation of various fractures, ovarian cystectomy, and thyroidectomy. Laparoscopic and open are two rows, two bills and two lengths of stay.

In-patient cholecystectomy is the clearest example: average length of stay runs from 1.5 days (Matilda, laparoscopic) to 9.7 days (Hong Kong Baptist Hospital, open). Merging the two into a single "cholecystectomy range" treats a 1.5-day and a 9.7-day admission as the same thing.

Of the 18 rows of in-patient cholecystectomy data, 13 are laparoscopic and 5 open. Counting only the 10 laparoscopic rows whose discharge band is not <30, the 50th percentile total charge runs from $87,800 to $153,794 — that is the range for "laparoscopic cholecystectomy", not for "cholecystectomy", and a great many citations drop the label.

As for open, all 5 rows are in the <30 band and are small-count data, so this article publishes no range for them. One of them, Hong Kong Sanatorium & Hospital Limited, has identical 50th and 90th percentiles ($130,498／$130,498), which is the shape that appears when case numbers are very small.

⚠️ If you already have a quotation and want to shop it around: confirm three things before comparing — which ward, which technique, and whether doctor's fees are included. Change any one of the three and the two numbers are not comparable.

---

#In-patient haemorrhoidectomy: 13 rows, read row by row

The three tables below are the complete data the platform provided on 1 August 2026 for the reporting period 1 January to 31 December 2025, in the platform's own hospital order. This article does not compare which hospital is better or cheaper — every row is listed so that you can see where a "range" comes from.

How to read this table: across the 13 rows, the 50th percentile total charge runs from a low of $38,497 to a high of $79,987. But the highest row is in the <30 discharge band — fewer than 30 cases — which by the platform's own warning is to be interpreted with care.

Counting only the 10 rows whose discharge band is not <30, the range narrows to $38,497 to $61,892. Both ranges are true; the difference is whether you are willing to treat small-count data as a market range.

Working one out (spread between hospitals): $79,987 ÷ $38,497 = 2.078, so the highest row is 107.8% above the lowest.

The Consumer Council's study of March 2025, using the same arithmetic but on data from a different reporting period, found that the 50th percentile charge for in-patient haemorrhoidectomy ranged from $33,881 to $85,387 across private hospitals, a difference of 152.0% [Note 11].

⚠️ The two sets of figures cannot be run together. The Council's come from the data it saw when it did its study; this article's come from the reporting period January to December 2025. What they have in common is this: for the same operation, the gap between hospital medians is a multiple, and that gap is a persistent feature.

---

#Day haemorrhoidectomy: a table that teaches you when not to quote a number

⚠️ The most important cell in this table is "annual discharges" — none of the eight rows publishes a band. By the platform's own warning, all of them are small-count data.

So this article publishes no range for day haemorrhoidectomy: the eight rows are real numbers, but assembling them into a "day surgery price" would suggest they represent the market, and the data does not support that.

Notice the CUHK Medical Centre row: the 50th and 90th percentiles are identical ($54,409／$54,409) — that is what very small case numbers look like.

While we are on the dataset's coverage limits: of the 30 procedures, 10 have no day surgery table at all (caesarean section, cholecystectomy, colectomy, hysterectomy, laminectomy, ovarian cystectomy, spine fusion, thyroidectomy, tonsillectomy, vaginal delivery); LASIK is the opposite, with only a day table; herniotomy has 6 in-patient rows and 3 day rows, none of which publishes a discharge band.

---

#In-patient hernia repair: 24 rows, split laparoscopic and open

How to read this table: of the 13 hospitals, 11 have data for both techniques and 2 publish laparoscopic only (11 × 2 + 2 = 24 rows).

Counting only rows whose discharge band is not <30: laparoscopic, 9 rows, 50th percentile total charge $65,660–$95,715; open, 10 rows, $42,778–$71,718.

⚠️ Merge all 24 rows into one "hernia repair range" and you get $42,778–$119,217 — a figure spanning two surgical techniques and mixing in small-count rows, which looks authoritative and answers nobody's question.

If your doctor has already told you whether it will be laparoscopic or open, you should be looking only at the corresponding half. If you have not decided, look at both, because the data shows this choice can affect the bill more than the choice of hospital.

---

#How should one row be read? Five steps

Take St. Teresa's Hospital's in-patient haemorrhoidectomy row (its discharge band is >200, the largest band in the table, so it is not affected by the small-count warning).

That row: average stay 1.5 days; doctor's fee 50th percentile $25,000 and 90th $37,400; hospital charges 50th $17,417 and 90th $19,361; total charges 50th $42,417 and 90th $56,761.

Step one — check the structure. $25,000 + $17,417 = $42,417, exactly the 50th percentile total. The 90th is the same: $37,400 + $19,361 = $56,761.

That is not because the two columns are independent percentiles — quite the opposite: because the ranking is done on total charges, the three figures on a row are the breakdown of the same case, which is why they add back. The Canossa laparoscopic hernia row mentioned above, where the 90th percentile hospital charge is below the 50th, is the counter-proof. The practical conclusion: a breakdown within one row can be read as it stands, but you cannot build figures across rows, across hospitals or across percentiles.

Step two — measure the width of the distribution. $56,761 − $42,417 = $14,344; $14,344 ÷ $42,417 = 33.8%.

⚠️ Be clear what that number is: the platform's own definition is that the remaining 10% of patients paid more than this amount — $56,761 is the floor of the most expensive 10% of cases, not their representative figure, and the platform does not publish what that 10% actually spent. So 33.8% is the width of the stretch from the median to the 90th percentile. It is not the answer to "how much more does the top 10% cost"; how expensive the real top end gets has no ceiling here, because the platform does not publish it. That 33.8% can serve as a starting point for a budget, but it is not a cap.

Step three — remember what this figure does not include. Standard ward (not a twin room, not a private room); single-procedure cases (have more than one thing done in the same admission and it is out); and the platform says expressly that the items covered may not be the same at each hospital. (The 1.5 days is the average length of stay the platform publishes, not a boundary of coverage — staying longer does not mean your case is outside this data, only that your stay was above average.)

Step four — want to know what a private room would do? That figure is not published, by this or any official platform. All that can be done is to look at the mechanism: the same hospital's own price list shows the basic operating theatre charge for haemorrhoidectomy rising from $2,120 in a standard room to $3,290 in a private room, and the hospital's own note says medical and associated charges are commensurate with the bed type of ward.

⚠️ The price list and the bill data are from different periods (price list July 2026, bill data 2025), so you cannot subtract one from the other to produce a "private room bill" — only this: the direction is upwards, and the size is not published.

Step five — want to know what a different surgical technique would do? That one can be worked out. The same hospital's two in-patient hernia repair rows: laparoscopic 50th percentile total $82,226 (101–200 cases, 1.5 days), open $50,283 (>200 cases, 1.8 days), a difference of $31,943.

Broken down: doctor's fee $49,250 against $36,600, a difference of $12,650; hospital charges $32,976 against $13,683, a difference of $19,293. The two add up: $12,650 + $19,293 = $31,943, which reconciles with the total. So about 60% of the gap between the two techniques (19,293 ÷ 31,943 = 60.4%) sits in hospital charges and about 40% in doctor's fees.

As a reference for the mechanism: the same hospital's price list shows the standard-room basic operating theatre charge for laparoscopic hernia repair at $9,710 against $2,020 for open, a difference of about 4.8 times — the theatre charge is only one item within hospital charges, but it explains why the two techniques cannot be averaged together.

If you are budgeting from a platform figure: the median is the starting point, the 90th percentile is the floor of the most expensive 10% (not their representative figure — the real ceiling is not published), and ward class and surgical technique are two switches that reprice the whole bill. Look at all four together and an estimate means something; but remember there is an unpublished long tail beyond the 90th percentile.

---

#How long is the tail? Look at the Government's own data

The percentile tables cannot answer one question: how expensive is the most expensive group? The platform does not publish the highest case.

The only official data on how dispersed the distribution is comes from Annex II to the Health Bureau's Legislative Council paper of May 2025. Note that its scope is the same as the platform's: single procedures only, standard beds only.

The Government's own reading of that table: analysis reveals rather large standard deviations in the fees charged for many healthcare services, implying that the fee distribution is highly dispersed; this may show that healthcare facilities or healthcare professionals have charged similar patients at widely different prices [Note 12].

And the passage below is the one most worth remembering in this article — the Government's own demonstration using one of those services: oesophagogastroduodenoscopy is a minor procedure by nature, and among people covered by VHIS the average fee for the relevant hospital admission cases was approximately $30,000; but actual claims records show the fee in certain cases could be substantially higher, with about 10% of cases exceeding $50,000 and about 1% exceeding $100,000.

⚠️ In one sentence: a "minor" endoscopic procedure averages about thirty thousand, but about 10% of cases exceed fifty thousand and about 1% exceeds a hundred thousand. That tail is what an insurance benefit limit actually has to catch — buy a limit off the mean or the median and it will not catch it.

The Government also states the limits of this data: because of the limitations of VHIS data it may be difficult to obtain the full picture; in Hong Kong's actual context, household out-of-pocket payment and non-VHIS products still account for a larger portion of health expenditure, and the latter is more likely to be used to settle more expensive cases, such as patients who opt for semi-private or private rooms [Note 13].

⚠️ So the expensive cases — particularly those in twin and private rooms — are quite likely not in this data at all, which means that if the tail above is biased, it is biased towards understatement.

If you are deciding a sum insured from an average: the Government has demonstrated with official data that the mean and the tail can differ by more than three times.

---

#Is the public system always cheaper? Price and waiting are two sides of one decision

Public charges are a policy price, not a market price; they are far lower, and the cost is in waiting. Both belong in the same view, from the same official figures.

Price first. Under the Hospital Authority's fee schedule effective 1 January 2026, for eligible persons: acute beds $300 a day (halved to $150 a day for children under twelve and for babies who cannot be discharged with their mothers); convalescent, rehabilitation, infirmary and psychiatric beds $200 a day (the same halving, $100 a day); specialist out-patient clinics $250 per attendance and $20 per drug item; day procedures and treatment at day facilities $250 per attendance; injections or dressings $50 per attendance [Note 14].

Plus the second safety net: from 1 January 2026 the Hospital Authority has, alongside the medical fee waiver mechanism, set an annual cap of HK$10,000 on public healthcare fees for eligible patients, without requiring financial assessment.

⚠️ But that cap has several practical gates, all on the Authority's own page: you must have cumulatively paid $10,000 before you can apply (it does not take effect automatically); you must have no outstanding balance anywhere in the Authority; the hospital services received must not be assessed as being without clinical need; the application period runs from 1 January each year to 31 March of the following year, and late applications are not accepted; only fully settled bills count, and until the result is announced the applicant must still pay every bill issued from the date of application; you must re-apply each year; assessment usually takes 14 days; once approved you must say so when registering for hospital services; and the Authority conducts periodic reviews of approved cases. Anything paid above $10,000 before approval goes towards eligible medical fees in later years, with no time limit. And self-financed drugs and medical devices are excluded.

Now waiting. Hospital Authority specialist out-patient waiting times for stable new cases (statistical period 1 July 2025 to 30 June 2026), surgery: medians across the seven clusters run from 27 weeks to 48 weeks; the "longest" in the same data runs from 62 weeks to 99 weeks. The Authority defines "longest" in its own footnote as the 90th percentile, meaning 90% of new cases are seen within the time shown.

There is a neat symmetry here: private bills and public waiting are both reported by official data at the 90th percentile, and on both sides it means "90% of cases are within this number". On the private side the 90th percentile measures what you might pay; on the public side it measures how long you might wait. The two numbers are two sides of one decision.

For scale, the Consumer Council in March 2025: Hong Kong's healthcare system runs on a dual track, with about half of Hong Kong's doctors employed by the private sector, providing about 68% of out-patient services but only about 10% of in-patient services, which reflects an imbalance between the public and private healthcare markets.

If your condition is "stable" and you can afford to pay: your decision is not "which side is cheaper" but "wait some tens of weeks, or spend money to buy time" — and the figures for both are above.

---

#Paying with insurance: what to watch

This section is only about how a bill gets settled; plan comparisons, premiums and tax deductions are in the VHIS article.

First, ward class. The platform's data is standard ward only, while a good deal of medical insurance is designed around ward class, semi-private or private.

⚠️ Your policy covering a private room does not mean the platform's figure applies to you — your bill is not in the dataset at all. Add the hospital's own note that medical and associated charges are commensurate with the bed type of ward, and choosing a higher ward class lifts the whole bill. This is the single most important line in this article for anyone insured.

Second, itemised limits, especially the surgeon's fee. VHIS has a Schedule of Surgical Procedures classifying procedures into four categories and regulating the minimum standard benefit limits for the surgeon's fees insurers must bear: Minor $5,000, Intermediate $12,500, Major $25,000 and Complex $50,000; insurers may offer higher benefits depending on product design [Note 15].

One comparison: cholecystectomy is classified Major under VHIS, so the minimum standard benefit for the surgeon's fee is $25,000; the same Annex II shows the whole bill for cholecystectomy averaging $96,200 (2021) to $100,700 (2023).

⚠️ Note these are not the same thing — $25,000 is the minimum benefit for one item, the surgeon's fee, not a ceiling on the whole bill, and an individual plan may cover far more. But the comparison explains why "I have VHIS" does not mean "I will not have to top up".

Third, get a written budget estimate first, then have the insurer calculate the claimable amount. The 2019 Legislative Council reply states that VHIS does not mandate private healthcare facilities to provide packaged charges; nonetheless, upon receipt of the estimated charges provided for non-emergency surgical procedures, insurers should provide consumers with claimable amount estimates for the procedures concerned [Note 16].

So the budget estimate is the tool in your hands — once you have it, you can ask the insurer to calculate a claimable amount against it.

Fourth, packages are not a cure-all. The Consumer Council's March 2025 study (a sample of 13 private hospitals providing those 30 treatments and procedures): apart from the commoner colonoscopy, gastroscopy and caesarean section, for which 10 private hospitals offered packages, the other treatments and procedures among the 30 rarely had them; and the number of packages offered varied widely, 1 hospital offering packages for 26 treatments and another 1 for only 2 [Note 17].

The same study also notes: medical packages often do not disclose the prices of their exclusions, and common exclusions such as medication, consultation fees and doctor's fees can be substantial, making it hard for consumers to compare packages fairly and like with like.

If you are insured and about to have non-emergency surgery: you should have two pieces of paper before admission — the hospital's budget estimate, and the insurer's claimable amount calculated against it.

---

#What are you entitled to ask?

The present system does not control prices, but it does control disclosure.

The four clauses of section 12 of the Medical Council of Hong Kong's Code of Professional Conduct for the Guidance of Registered Medical Practitioners [Note 18]:

The three tests under 12.3 are: the difficulty, cost and special circumstances of the service, and the time, skill and experience it requires; the fee customarily charged in Hong Kong for similar services; and the doctor's own experience and ability in providing the service.

In one sentence: this Code governs not only whether you are told, but whether the charge is excessive. Ask, and the doctor must tell you; do not ask, and there is no automatic duty to quote — but 12.3 stands on its own, and forbids overcharging.

⚠️ The Chinese version of the Code carries its own note that the English text governs, and in English 12.1 says a fee "should be made known … on request" while 12.2 says a doctor is "strongly advised"; the wording of 12.3 is the same in both. So this article states only the consequence both versions agree on for the disclosure duty; the prohibition on overcharging in 12.3 is quoted directly and is unaffected by that divergence.

The statutory version the Government proposes would go further [Note 19]: the budget estimate would have to list separately the estimated hospital charges, doctor's fees, and charges for medication, implants and consumables; the components of the estimated doctor's fee would include attending doctor's visit fees, surgeon's fees, anaesthetist's fees and specialist's fees; but a patient requiring admission within 12 hours of booking could be exempted from a budget estimate. And it would require: where the final fee significantly exceeds the budget estimate (say, by 20% or more), the private hospital must give the patient or family a written explanation.

⚠️ Note the tense: these are proposals, not current requirements. The Legislative Council Secretariat's paper of March 2026 states that the draft subsidiary legislation is expected to be tabled in 2026.

If asking about price feels awkward: clause 12.4 requires the doctor to display a notice in the practice telling patients they may ask about charges before treatment. Asking is what the system expects you to do.

---

#What to do next

1. Confirm three things first, or two quotations are not comparable: which ward, which technique, and whether doctor's fees are included.
2. Look only at the half of the data that matches your situation. Laparoscopic and open are two sets of figures, and a range that merges them answers nobody's question.
3. Treat any row in the <30 discharge band as small-count data. The platform itself tells you to interpret it with care.
4. Do not build figures across rows, hospitals or percentiles. The percentiles are ranked on total charges, and a row's breakdown is the breakdown of one case.
5. If you have booked a private room, do not budget from these figures. The direction is upwards and the size is not published.
6. Ask for the budget estimate, then ask the insurer to calculate a claimable amount against it. Two pieces of paper before admission.
7. Do not buy a sum insured off an average. The Government's own demonstration: a "minor" procedure averaging about thirty thousand, with 10% of cases over fifty thousand and 1% over a hundred thousand.
8. If you think a charge is excessive, there is a route beyond complaining about disclosure — clause 12.3 of the Medical Council's Code.
9. Out-patient consultation fees and specialist clinic charges are not in this dataset; see the article on private doctors' fees, and the VHIS article for comparing the plans themselves.

---

#Frequently asked questions

Does the law require these charge statistics to be published?

No. They come from the Pilot Programme on Enhancing Price Transparency of Private Hospitals launched in October 2016, and the Government's press release says it is "on a voluntary basis". The Private Healthcare Facilities Ordinance (Cap. 633) is itself in force (hospital licensing from 1 January 2021), but sections 61 to 63 of it, which deal with price transparency, had still not commenced as at 1 August 2026, with the draft subsidiary legislation expected to be tabled in the Legislative Council in 2026.

Can I add the 50th percentile doctor's fee to the 50th percentile hospital charge and call it my budget?

That is not how to use it. The platform states that the percentile figures are ranked by the amount of the "total charges" and not by the "doctor's fees" and "hospital charges" within the total — a row's breakdown is the breakdown of one case, not two independent medians. For the total, read the "total charges" column directly; and do not build figures across hospitals.

I have booked a private room. Can I budget from these figures?

Not directly. The platform says expressly that the data is "based on data for patients accommodated in standard wards undergoing a single operation or procedure". One hospital notes on its own price list that medical and associated charges are commensurate with the bed type of ward, and there is no official publication of whole-bill figures for private rooms.

Why can the median for the same operation differ by a factor of two between hospitals?

There is no single official explanation. The Government's position is that private healthcare prices are determined by market forces and that it will not regulate the price level or structure; the Consumer Council points out that doctor's fees and hospital charges are often adjusted according to ward class and that the rationale is unclear. The data itself also has several features that make it incomparable: different surgical techniques, different items covered at each hospital, and figures that may not all be from the same year.

Does the 90th percentile mean unusually complex cases?

The platform says only that the 90th percentile is the amount below or equal to which 90 per cent of patients paid, and warns in the same notes that actual length of stay and total charges may vary with the patient's condition and the complexity of the disease. It has not separated the two factors, so it can be described as the upper end of the distribution but its cause cannot be asserted.

So how expensive is the most expensive case?

The platform does not publish the highest case, and the 90th percentile is the floor of the most expensive 10% rather than their representative figure. The only official data on the tail is the Health Bureau's analysis of VHIS claims: a "minor" endoscopic procedure by nature averaged about $30,000, but about 10% of cases exceeded $50,000 and about 1% exceeded $100,000. And the Government itself points out that the expensive cases, those opting for semi-private or private rooms, are more likely not to be in that data at all, so if the tail is biased it is biased towards understatement.

How long is the wait for the same operation in the public system?

Hospital Authority surgical specialist out-patient stable new cases (statistical period 1 July 2025 to 30 June 2026): medians of 27 to 48 weeks across the seven clusters, and a "longest" (the 90th percentile) of 62 to 99 weeks. That is the wait for a first specialist out-patient appointment, not the wait for the operation itself.

What can I do if I think a charge is excessive?

Under the present system the Government does not regulate price levels, but clause 12.3 of the Medical Council's Code of Professional Conduct provides that a doctor must not overcharge or accept excessive remuneration, and sets out three tests: the difficulty, cost and special circumstances of the service and the time, skill and experience required; the fee customarily charged in Hong Kong for similar services; and the doctor's own experience and ability. That runs through the Medical Council's professional misconduct machinery, not government price control. Clause 12.1 separately provides that refusing or failing to disclose charges on a patient's reasonable request may be regarded as professional misconduct.

---

#What this article does not state

• The text of sections 61 to 63 of the Private Healthcare Facilities Ordinance (Cap. 633) is not quoted here. Everything said about those three sections comes from explanatory documents of the Government or the Legislative Council Secretariat, and is cited as such; the provisions themselves are neither quoted nor paraphrased.
• Day procedure centres publish almost no charge statistics, and are not required to. Consumer Council, March 2025: of 20 day procedure centres sampled, 15 published no price list online for the specified treatments and 1 had no website; and neither the Ordinance nor the Code of Practice for Day Procedure Centres requires them to publish historical charge statistics. As at the end of 2024 Hong Kong had 260 licensed day procedure centres. All percentile data in this article covers private hospitals only, not day procedure centres.
• Of the fourteen private hospitals, this article has read the full price list of only one, St. Teresa's Hospital. So the item-by-item prices in the section on how a bill is built are one hospital's publication, not a market survey, and should not be taken as any other hospital's prices.
• The platform does not publish each hospital's data submission date, so a comparison between hospitals may be comparing different years; nothing in the public record resolves this.
• The platform publishes neither the lowest nor the highest actual case amount, nor the exact number of discharges (only bands). The statutory version the Government proposes would add actual ranges, but it has not commenced.
• There is no official source, anywhere, for whole-bill figures for private or twin rooms.
• Whether a private hospital or clinic requires a referral letter, and which will settle directly with an insurer, is not stated here for want of a first-hand source.
• Percentages from the Consumer Council's consumer questionnaire (such as the proportion given verbal estimates, or the proportion whose final bill exceeded the estimate) are not cited, because the press release does not state that survey's sample size or field dates. Only figures with an explicit basis of calculation are used from that study.
• This article does not compare or rank any private hospital, and offers no charge estimate or medical advice for any individual reader.

---

#Notes: the official wording

Where a source is quoted below in English, the wording is the source's own.

[Note 1] Government press release, 29 September 2016. The Chinese version states that under the pilot programme all members of the Hong Kong Private Hospitals Association, being the eleven private hospitals in Hong Kong at the time, would implement three price transparency measures on a voluntary basis: introducing budget estimates, so that hospitals and doctors give patients a reference for the overall total charge when they undergo a non-emergency operation or procedure (such as thyroidectomy, colonoscopy, LASIK or knee arthroscopy); publishing on hospital websites the price lists of major items (such as operating theatre charges and the charges of common nursing procedures); and publishing on hospital websites the actual historical bill size statistics of common operations and procedures (such as vaginal delivery and caesarean section). The English version of the same paragraph uses the words "on a voluntary basis". The platform's page title reads "List of Operations/Procedures Recommended for the Publication of Historical Bill Sizes Statistics".

[Note 2] The status of the price transparency provisions. Consumer Council press release, 6 March 2025, states that the provisions of the Ordinance relating to price transparency have not yet come into operation.

Legislative Council Secretariat Research Office, Information Note ISE03/2026 (2 March 2026):

In May 2025, the Health Bureau briefed the Panel on Health Services of the Legislative Council ("LegCo") on legislative proposals to make existing voluntary price transparency measures statutory. With the draft regulation expected to be tabled in 2026, there are calls to draw on overseas experiences.

Note 7 of the same document is the clearest description this article found of sections 61 to 63 (it is the Secretariat's description of the provisions, not the provisions themselves):

The Private Healthcare Facilities Ordinance empowers the Secretary for Health to make regulations for the purposes of Sections 61 and 62 on price transparency … Section 63 prescribes hospitals to publish historical statistics on the fees and charges for the specified treatments and procedures in the way specified by the Director of Health.

The Government's position on prices themselves, from the 2019 Legislative Council reply:

as private medical service by its very nature is no different from other business transactions between consenting parties where prices are determined by market force, the Government will not regulate the price level or price structure of services provided by private healthcare facilities.

[Note 3] The data format, written reply of the Secretary for Food and Health to the Legislative Council, 8 May 2019:

Private hospitals should also publish historical bill sizes statistics of 30 common treatments/procedures in a standardised format, which covers annual number of discharges, average length of stay, as well as the actual billing data for the 50th percentile and 90th percentile of each specified treatment/procedure.

[Note 4] The recommendation of the Director of Audit's Report No. 59 of October 2012, as described in note 7 of the Health Bureau's Legislative Council paper of May 2025:

Issued on October 2012, the Director of Audit's Report No. 59 reviewed the issues of regulatory control of private hospitals and land grants for private hospital development at the time. The report suggested that the Government should explore ways to enhance the transparency of fees levied by private hospitals … such as requiring private hospitals to ensure that every patient is informed, prior to or upon his admission to the hospital, of the estimated total fees which are likely to be incurred, and enacting laws on the transparency in medical fees.

[Note 5] Platform note, rule one. The platform states that the information private hospitals provide is based on data for patients accommodated in standard wards undergoing a single operation or procedure, and does not include information on patients undergoing multiple operations or procedures; and that because the items covered may not be entirely the same at individual hospitals, patients should ask the hospital directly for details of what each operation or procedure covers.

Rule two: the platform states that the percentile figures above are ranked by the amount of the "total charges" and not by the "doctor's fees" and "hospital charges" within the "total charges".

[Note 6] Currency of the data, note 8 of the Health Bureau's Legislative Council paper of May 2025:

However, actual experience shows that some private hospitals may only submit data of the previous year by the third quarter of the following year, and thus some information on the website may not be up to date.

Consumer Council, March 2025: its review further found that at 4 private hospitals the historical charge statistics shown on the website were at least a year and a half old.

[Note 7] The percentile definitions. The platform states that the fiftieth percentile is the amount below or equal to which 50 per cent of patients paid, and that this figure may serve as a reference for estimating the median charge; and that the ninetieth percentile — the column the platform's English pages head "90th Percentile" — is the amount below or equal to which 90 per cent of patients paid, with the remaining 10 per cent paying more, and that this figure may serve as a reference for estimating the higher charges.

The first sentence of the same notes: actual length of stay and total charges may vary according to the patient's condition and the complexity of the disease. The small-count warning: the platform cautions that readers should interpret with care the bill data a hospital reports for procedures done in small numbers; the English version says "less than 30 cases" and the Chinese "not more than 30", which are not quite the same threshold.

[Note 8] The definitions of the two classes of charge, in the platform's English:

"Doctor's fees include fees for anaesthetist, operating surgeon and ward round. Hospital charges include admission fee, accommodation, operation and associated materials, nursing procedures, investigation and examination fees, medication, meals and beverage, sundries, etc."

[Note 9] Price lists exclude doctor's fees, in three places in St. Teresa's Hospital's price list of common items. Note 6 on the rooms and wards page:

"6. The above prices are exclusive of doctor's fee, which includes fees for operation, anaesthetic, doctor's visit, special consultation, etc. Please refer to your physician for details."

The two notes on the endoscopy centre page are numbered (2) and (3) in the original. At (2), the price list sets out only the basic charges of common items at the endoscopy centre for reference, and those charges exclude doctor's fees, monitored anaesthesia charges, pathology charges, drugs and materials, additional instruments and equipment, and the performance of any additional treatment or procedure such as polypectomy. At (3), the actual charge depends on the complexity of the case.

The note to the common nursing procedures table states that the price list sets out the charges of common nursing procedure items for reference, and that those charges exclude doctor's fees and drugs and related materials.

[Note 10] Charges commensurate with ward class, note 7 of the same price list — the note carries the number 7 in both language versions of that document:

"7. Medical and associated charges are commensurate with the bed type of ward. Please refer to Accounts Department or the respective departments / wards for the pricing information of other services provided."

Consumer Council, March 2025: doctor's fees and hospital charges are often adjusted according to ward class, but the rationale behind that linkage is not clear; and some consumers regard it as unfair that the same treatment should be charged differently merely because of a difference in accommodation.

[Note 11] The Consumer Council on the spread between hospitals in March 2025 (those figures are from a different reporting period), in its English release:

"the 50 th percentile price for in-patient haemorrhoidectomy ranged from HKD33,881 to HKD85,387 among different PHs, representing a 152.0% price difference."

The same study on the scale of the two sectors: Hong Kong's healthcare system runs on a dual track, with about half of Hong Kong's doctors employed by the private sector, providing about 68% of out-patient services but only about 10% of in-patient services, reflecting an imbalance between the public and private healthcare markets.

[Note 12] Paragraph 16 of the Health Bureau's Legislative Council paper of May 2025:

Analysis reveals rather large standard deviations in the fees charged for many healthcare services, implying that the fee distribution is highly dispersed. This may show that healthcare facilities or healthcare professionals have charged similar patients at widely different prices.

The demonstration at paragraph 17:

For example, oesophagogastroduodenoscopy (OGD), a minor procedure by nature, is quite commonplace among hospital admission cases. Annex II shows that the average fee for relevant hospital admission cases was approximately $30,000 among VHIS-insured persons. However, actual claims records show that the fee charged in certain cases could be substantially higher, with about 10% of cases exceeding $50,000 and about 1% exceeding $100,000.

The title of Annex II: "Fees of Representative Healthcare Services under the Voluntary Health Insurance Scheme (Cases where patients receive single procedure and use standard beds in private hospitals)". The 2023 hysterectomy cell in the original document is missing its dollar sign, and is reproduced here as printed.

[Note 13] The limitation stated at paragraph 21 of the same document:

Due to the limitations of VHIS data, it may be difficult to obtain the full picture from the above analysis. In the actual context of Hong Kong, household out-of-pocket payment and non-VHIS products … still account for a larger portion of health expenditure, and the latter is more likely to be used to settle more expensive cases (e.g. patients who opt for semi-private or private rooms etc.).

[Note 14] Footnote 4 to two cells of the Hospital Authority fee schedule provides that children under twelve, and babies who cannot be discharged at the same time as their mothers, pay half the hospital charge for the bed type occupied, with other charges the same as for adults.

The annual fee cap, from the Authority's own page: from 1 January 2026 the Hospital Authority has established a second safety net alongside the medical fee waiver mechanism, setting an annual cap of HK$10,000 on public healthcare fees and charges for eligible patients without requiring financial assessment — the "Annual Fee Cap". Once an application is approved, the patient need not pay any further "eligible medical fees and charges" within that year. The English version of that sentence uses the words "without requiring financial assessment".

The definition of "longest" in specialist out-patient waiting: the longest (90th percentile) waiting time means 90% of new cases can be seen within the waiting time shown.

[Note 15] The four VHIS minimum benefit levels for surgeon's fees, from the note beneath Annex II of the Health Bureau paper:

The VHIS has set up a Schedule of Surgical Procedures to classify various procedures into four categories, regulating the minimum standard benefit limits for the relevant surgeon's fees to be borne by the insurance companies, namely Minor ($5,000), Intermediate ($12,500), Major ($25,000) and Complex ($50,000). Insurance companies may offer higher benefits depending on the design of their products.

[Note 16] The 2019 Legislative Council reply on budget estimates and claimable amounts:

VHIS does not mandate private healthcare facilities to provide consumers of Certified Plans with healthcare services at packaged charges. Nonetheless, … upon receipt of the estimated charges provided for non-emergency surgical procedures, the insurers should provide consumers with claimable amount estimates of the procedures concerned for consumers' reference.

[Note 17] Packages, Consumer Council study of March 2025 (a sample of 13 private hospitals providing those 30 treatments and procedures): apart from the commoner colonoscopy, gastroscopy and caesarean section, for which 10 private hospitals offered packages, the remaining treatments and procedures among the 30 rarely had packages; and the number of packages offered by private hospitals varied considerably, 1 offering packages for 26 treatments and another 1 for only 2.

The same study: perhaps because it is difficult to set standard charges for exclusions, medical packages often do not disclose the relevant prices; and common exclusions such as medication, consultation fees and doctor's fees may all be substantial, making it difficult for consumers to compare medical packages fairly and on a like-for-like basis.

On day procedure centres, the same review found that at 4 private hospitals the historical charge statistics on the website were at least a year and a half old.

[Note 18] Section 12 of the Medical Council of Hong Kong's Code of Professional Conduct for the Guidance of Registered Medical Practitioners (revised October 2022). The Chinese version carries its own note that the content of the Code is governed by the English text, and this article follows the English; in English 12.1 reads "should be made known … on request" and 12.2 reads "strongly advised", while the wording of 12.3 is the same in both versions. Clause 12.1 provides that a doctor's fee should be made known to the patient on request, that during diagnosis and treatment a doctor should on request disclose, so far as he knows, all charges likely to be incurred before the service is provided, and that refusing or failing to disclose charges on a patient's reasonable request may be regarded as professional misconduct. Clause 12.2 provides that although a doctor is under no obligation to quote fees in advance, where the fee involved is substantial the doctor is strongly advised to inform the patient of the charges before providing the service, so as to avoid complaint and dissatisfaction. Clause 12.3 provides that a doctor must not overcharge or accept excessive remuneration, and that the Medical Council applies the following principles in deciding whether a charge is excessive: (a) the difficulty, cost and special circumstances of the service and the time, skill and experience required; (b) the fee customarily charged in Hong Kong for similar services; and (c) the doctor's own experience and ability in providing the service. Clause 12.4 provides that a doctor must display a notice in the practice informing patients of their right to enquire about the charges involved before receiving treatment.

[Note 19] The proposed statutory version, as quoted by the Legislative Council Secretariat in ISE03/2026:

The budget estimate must separately list out the estimated hospital charges, doctor's fees as well as charges for medication, implants and consumables. The components of estimated doctor's fee include attending doctor's visit fees, surgeon's fees, anaesthetist's fees and specialist's fees. However, if a patient requires admission within 12 hours of booking, the budget estimate can be exempted.

While final fees may be adjusted due to changes in clinical conditions, private hospitals must provide a written explanation to the patient, the patient's family member(s) or authorized person(s) when final fees significantly exceed the budget estimate (say, by 20% or more)

---

#Sources and dates checked

• Launch of the pilot programme, the three measures, and "on a voluntary basis": Government press release "Government launches pilot programme to enhance price transparency of private hospitals", https://www.info.gov.hk/gia/general/201609/29/P2016092900791.htm (English); Chinese version https://www.info.gov.hk/gia/general/201609/29/P2016092900793.htm. Document dated 29 September 2016; retrieved 1 August 2026.
• The format of the statistics (discharges, average length of stay, 50th/90th percentiles), the 30 procedures, and the Government not regulating price levels: Legislative Council written reply "LCQ4: Prices for private healthcare services", https://www.info.gov.hk/gia/general/201905/08/P2019050800331.htm, 8 May 2019; retrieved 1 August 2026.
• Bill percentiles, discharge bands, average length of stay and platform notes by hospital and procedure (including the standard-ward/single-procedure scope, ranking on total charges, the definitions of doctor's fees and hospital charges, the percentile definitions and the small-count warning): Department of Health, Pilot Programme on Enhancing Price Transparency of Private Hospitals public enquiry platform, https://apps.orphf.gov.hk/Public/en/ (Chinese version https://apps.orphf.gov.hk/Public/tc/). Reporting period 1 January 2025 to 31 December 2025; retrieved 1 August 2026.
• The list of participating hospitals (14, 1 of them marked "Historical Bill Sizes Statistics are not applicable") and the list of 30 procedures "recommended" for publication: same platform, https://apps.orphf.gov.hk/Public/Info/Hospitals.aspx and https://apps.orphf.gov.hk/Public/Info/HBSS.aspx; retrieved 1 August 2026.
• Passage, gazettal and licensing commencement dates, and price transparency as a key element of the new regulatory regime: Office for Regulation of Private Healthcare Facilities, https://www.orphf.gov.hk/tc/regulatory_regime/new_licensing_scheme (English version /en/); the page is marked "Review date: 23 May 2025"; retrieved 1 August 2026.
• Sections 61 to 63 not yet commenced: Consumer Council press release, https://www.consumer.org.hk/tc/press-release/p-private-healthcare-services-study-report, 6 March 2025; retrieved 1 August 2026.
• Subsidiary legislation expected in 2026, the description of sections 61 to 63, the proposed content of budget estimates and the written explanation where the final fee exceeds the estimate by 20%: Legislative Council Secretariat Research Office, Information Note ISE03/2026 "Enhancing price transparency of private healthcare services", https://www.legco.gov.hk/research-publications/chinese/essentials-2026ise03-enhancing-private-healthcare-price-transparency.htm (English version /english/), 2 March 2026; retrieved 1 August 2026.
• The recommendation of the Director of Audit's Report No. 59 of 2012, the voluntary participation of 14 private hospitals, 14 licensed private hospitals / 5 194 beds / 260 licensed day procedure centres as at the end of 2024, data possibly not up to date (note 8), VHIS average claim fees and standard deviations (Annex II), OGD with 10% over $50,000 and 10% of those over $100,000, the four minimum benefit levels for surgeon's fees, and the limitations of VHIS data: Health Bureau paper to the Legislative Council Panel on Health Services, "Enhancing Price Transparency of Private Healthcare Services", LC Paper No. CB(3)639/2025(03), https://www.legco.gov.hk/yr2025/english/panels/hs/papers/hs20250509cb3-639-3-e.pdf (Chinese version /chinese/…-c.pdf), 9 May 2025; retrieved 1 August 2026.
• In-patient haemorrhoidectomy 50th percentile $33,881–$85,387 (a 152.0% difference), doctor's fees and hospital charges adjusted by ward class, day procedure centre disclosure, package availability, and the public/private split (about 68% of out-patient and about 10% of in-patient services): Consumer Council press release, https://www.consumer.org.hk/tc/press-release/p-private-healthcare-services-study-report, 6 March 2025; retrieved 1 August 2026.
• Basic operating theatre charges by ward class, room charges, price lists excluding doctor's fees, the exclusions on the endoscopy and nursing procedure items, and "Medical and associated charges are commensurate with the bed type of ward": St. Teresa's Hospital, price list of common items, https://www.sth.org.hk/download/FeeSchedule.pdf, the document dated 1 July 2026; retrieved 1 August 2026.
• The duty to disclose fees (clauses 12.1 to 12.4) and the note that the English text governs: Medical Council of Hong Kong, Code of Professional Conduct for the Guidance of Registered Medical Practitioners (revised October 2022), https://www.mchk.org.hk/english/code/files/Code_of_Professional_Conduct_(English_Version)_(Revised_in_October_2022).pdf (Chinese version https://www.mchk.org.hk/tc_chi/code/files/Code_of_Professional_Conduct_(Chinese_Version)_2022.pdf); retrieved 1 August 2026.
• Public healthcare charges effective 1 January 2026 (acute beds $300 a day, specialist clinics $250 per attendance, day procedures $250 per attendance, injections or dressings $50 per attendance and so on): Hospital Authority "Fees and Charges", https://www.ha.org.hk/visitor/ha_visitor_index.asp?Content_ID=10045&Lang=CHIB5, effective 1 January 2026; retrieved 1 August 2026.
• The $10,000 annual fee cap, no financial assessment, eligibility and the application window: Hospital Authority "Annual Fee Cap" (for eligible persons), https://www.ha.org.hk/visitor/ha_view_content.asp?Content_ID=281820&Lang=CHIB5&Dimension=100&Ver=HTML, implemented 1 January 2026; retrieved 1 August 2026.
• Surgical specialist out-patient stable new case median and 90th percentile waiting times, and the definition of "longest": Hospital Authority, Specialist Out-patient Stable New Case Waiting Time, https://www.ha.org.hk/haho/ho/sopc/dw_wait_ls.pdf, statistical period 1 July 2025 to 30 June 2026; retrieved 1 August 2026.

This article was compiled by the editorial team from official sources, with every charge and every institutional fact sourced where it appears; it is health and consumer information, not medical advice, and is not a recommendation of any healthcare institution.

This article was written from the sources listed above. It is about how to read this table; what your own bill will come to, it does not answer — that is determined by your doctor and the hospital, on your situation.