When does low back pain actually need an MRI scan? Recognising red flags, where the 80% lifetime figure comes from, and the new referral-free physiotherapy rules

TL;DR "80%" is not a global figure. It is one point inside a range. The World Health Organization's low back pain fact sheet (19 June 2023) publishes no lifetime prevalence percentage at all, saying only that most people experience it at least once in their life [Note 1]. In the medical literature, Walker's systematic review of 2000 covered 1966–1998 and 30 methodologically acceptable studies, with lifetime prevalence ranging from 11% to 84%; Hong Kong's own Reference Framework for Common Musculoskeletal Problems in Primary Care Settings (Health Bureau, 2022) writes 58–84%. Hong Kong's own figures are far lower, and they measure different things: Lau et al. found 39% in 1995 (95% confidence interval 34%–44%; a sample of 652 people in two housing estates, 31 years ago); a Department of Health household survey (as quoted in the 2022 Reference Framework) put low back pain in the preceding 30 days among people aged 15 and above at 21.5%; and the Population Health Survey 2020-22 measured doctor-diagnosed chronic low back pain at just 1.8%. The sentence on the Hospital Authority's Smart Patient site (page updated 29 July 2024) that around 80% of adults experience low back pain at least once in their lives cites "medical reports" without naming one. The biggest myth: bad pain means an MRI straight away. The United Kingdom's NICE NG59 (published 2016, last updated 29 July 2026) states that imaging should not be offered routinely in a non-specialist setting; Hong Kong's Reference Framework states that imaging should not be offered unless the result is likely to change management. There are figures behind that: Brinjikji's review of 2015 covered 33 studies and 3,110 people with no symptoms at all, of whom 37% already had disc degeneration and 30% disc bulges in the 20-year-old group, rising to 96% and 84% respectively by 80. There is no exhaustive list of red flag symptoms — and this article will not pretend to have one. NICE's own wording is illustrative ("for example, cancer, infection…"); the table in Hong Kong's Reference Framework prints a complete six rows, but the body text of Chapter 5 of the same document adds several more; and the table 3.1 of the Primary Healthcare Commission's Clinical Guideline for First-contact Physiotherapists on Assessment and Cross-disciplinary Management Approach for Common Musculoskeletal Problems in Primary Care Setting, published in December 2025 (all 55 pages of which contain not one Chinese character, and which has no official Chinese title), is headed "Common Red Flag Conditions in Primary Care Setting that Require Urgent or Early Medical Care" — common is not the same as complete. So: not being on a list does not mean it does not matter. One practical change that took effect in July 2025: section 6 of the Physiotherapists (Registration and Disciplinary Procedure) Regulation (Chapter 359J) (consolidated version dated 25/07/2025, as amended by section 162 of Ordinance No. 33 of 2025) now has four referral-free routes. The clinical guideline route currently covers only two conditions in Hong Kong — osteoarthritis of the knee and low back pain — with a ceiling of 10 sessions or 30 days, whichever comes first, and is confined to first-contact physiotherapists who have completed recognised training. This article sets out only sourced facts: it will not tell anyone which exercise to do, which drug to take, or which clinic to go to. Individual diagnosis, medication, imaging and surgical decisions are for a doctor.

#"80% of people get back pain once in their life" — where does that come from, and how far can it be trusted?

The "80%" is the top half of a range that has been taken for a point estimate.

What is being measured comes first. Prevalence of low back pain has several entirely different formulations, and mixing them causes trouble:

• Lifetime prevalence: whether you have ever had it once in your life.
• Period prevalence: whether you have had it in, say, the past 30 days or the past 12 months.
• Point prevalence: whether you have it at the moment of asking.
• Doctor-diagnosed chronic low back pain: not "have you ever had pain" but "have you been diagnosed with chronic low back pain".

Those four measure four different things, and the values can differ many times over. Below is what each source writes for itself.

The WHO fact sheet publishes no lifetime prevalence percentage. The WHO's Low back pain fact sheet (page's own dateModified 19 June 2023) says only that low back pain can be experienced at any age, and that most people experience it at least once in their life [Note 1]. The WHO has its own Chinese text, but only in simplified characters with no traditional version; the simplified original is reproduced at [Note 1].

The same fact sheet gives case numbers rather than lifetime risk: in 2020, low back pain affected 619 million people globally, and it is estimated that the number of cases will rise to 843 million by 2050, driven largely by population expansion and ageing [Note 1].

"80%" is in fact a range of 11%–84%. Walker's systematic review of 2000 (J Spinal Disord 2000;13(3):205-17) covered the literature from 1966 to 1998: point prevalence ranged from 12% to 33%, one-year prevalence from 22% to 65%, and lifetime prevalence from 11% to 84% [Note 2]. That range appears across 30 "methodologically acceptable" studies, and Walker states that the studies differed significantly in design, in the ages of respondents and in how the data were collected. A spread running from 11% to 84% is not something "about 80%" can represent.

The figure the Hong Kong Government itself publishes is 58–84%. Section 1.2 of the core document of the Hong Kong Reference Framework for Common Musculoskeletal Problems in Primary Care Settings (Health Bureau, document self-dated 2022) states that the lifetime incidence of low back pain is 58-84%, and that 11% of men and 16% of women have chronic low back pain [Note 3]. The same passage also carries Hong Kong figures: the Department of Health reported that 21.5% of respondents aged 15 or above had low back pain during the 30 days preceding a household survey; 5.1% in the 15–24 age group, rising with age to 52.4% in the group aged 85 or above [Note 3].

The source of the "80%" in Hong Kong is a Hospital Authority patient education page, and it does not say which study it draws on. The Smart Patient site's low back pain page (page updated 29 July 2024, a date that exists only in an HTML comment and is invisible to the reader) states that low back pain is very common in Hong Kong and that, according to medical reports, around 80% of adults experience low back pain at least once in their lives [Note 4].

In one sentence: that figure is patient education text whose source is given as "medical reports" without naming one. This article will therefore not treat "80%" as a measured value for the world or for Hong Kong.

And the same page carries two more "80%" figures meaning entirely different things — all three are set out here, because they are easily confused. The section of the Smart Patient page on treatments for low back pain states that after reducing vigorous activity and resting for a few days, combined with appropriate exercise and strengthening of the back muscles, the condition can be relieved, and more than 80% of patients recover within a month; the very next sentence states that a slipped disc can cause low back pain, and that 80% of those patients are between the ages of 20 and 40 [Note 4]. Note one difference between the Chinese and the English: the Chinese writes 「八成以上」, more than 80%, while the English writes "80%" with no "more than". Both versions are the Hospital Authority's own text, and this article reproduces each without merging them.

There is no arithmetical relationship between those three figures. Seeing "80%" and taking it for the same thing is the easiest misreading on that page.

⚠️ If a headline saying 80% of people get it makes your own pain feel ordinary and not worth attention: how common something is and how serious it is are two different things — the same body of sources also states that fewer than 1% of low back pain seen in primary care is serious spinal pathology (see the next section), and that 1% does not disappear because the total is large.

Also worth knowing: a rising number of cases does not mean rising risk. The GBD 2021 Low Back Pain Collaborators (Lancet Rheumatol 2023;5(6):e316-e329) state that between 1990 and 2020, age-standardised rates of prevalence and years lived with disability decreased by 10.4% and 10.5% respectively [Note 5]. In other words, the rise in cases from 619 million to a projected 843 million is driven mainly by population growth and ageing, not by rising risk per person.

#What does "non-specific low back pain" mean, and why is there no test that points to where the pain is?

In about 90% of cases of low back pain there is no confirmable structural source of the pain — and that is not "we could not find it", it is that the tests available in primary care cannot localise it.

The two-page summary of the Hong Kong Reference Framework for Common Musculoskeletal Problems in Primary Care Settings (footer self-dated 2022) divides the low back pain seen in primary care into three classes with their proportions [Note 6]: specific spinal pathology, less than 1% in primary care, covering vertebral fracture, malignancy, spinal infection, axial spondyloarthritis and cauda equina syndrome; radicular syndromes, about 5–10%, covering radicular pain, radiculopathy and spinal stenosis; and non-specific low back pain, about 90–95%, presumed to be of lumbar musculoskeletal origin.

The footnote immediately following in the same document states that no tests are available in primary care to reliably specify the pathoanatomical source of low back pain [Note 6]. Which is to say: "non-specific" is a category defined by the limits of testing, not a way of saying the doctor could not find your problem.

The WHO fact sheet puts the same thing differently: non-specific means that the experience of pain cannot be confidently accounted for by another diagnosis such as an underlying disease, pathology or tissue damage, and it is non-specific in about 90% of cases [Note 1].

One branch that is easily missed: the pain is in the back, but the source need not be. The same Hong Kong summary has a separate column listing non-spinal causes [Note 6]: referred visceral pain (pancreatic or urological pathology, for example pancreatitis, pancreatic cancer, prostatitis, pyelonephritis); vascular causes (aortic aneurysm or dissection, femoral artery occlusion, peripheral vascular disease); hip pathology; and viral syndrome. The original introduces its items with "e.g.", so they are examples.

#Acute, subacute, chronic — two sets of definitions that cannot be mixed

There is a real divergence here, and it affects how the guidance you read applies:

⚠️ If you are reading a NICE recommendation such as "Do not offer opioids for managing chronic low back pain": that "chronic" uses NG59's own 3-month definition, not the WHO's 12 weeks. The same word, two different rulers.

#How the official sources distinguish low back pain from sciatica

The Hospital Authority's Smart Patient site (29 July 2024) states that low back pain usually means pain arising from the tissues of the lower back, such as muscles, tendons, intervertebral discs and vertebral joints, and can extend from below the ribs of the lower back to the buttocks and the back of the thighs; that low back pain differs from sciatica, in that the pain does not extend to the calf or ankle as sciatic pain does; and that back pain in middle-aged and younger people is mostly caused by muscular overuse and poor posture, while in older people it is mostly caused by degeneration of the skeletal system [Note 7].

The corresponding sentence in the English version of the same page is narrower: the English goes only as far as "foot", where the Chinese writes calf or ankle. Both are the Hospital Authority's own text, and this article reproduces each; the analysis below follows the Chinese original.

Those three passages are definitional distinctions, not a method of telling them apart yourself. That is, they describe how the term "low back pain" is bounded on the Hospital Authority's page, not that a person can classify themselves by where the pain reaches. Two things go with them: the Hong Kong Reference Framework puts radicular syndromes at 5–10% of primary care; and in the warning-sign list reproduced in the next section, weakness in the leg, loss of sensation in a dermatomal distribution, and urinary or faecal retention or incontinence fall in Priority category 1, the group to be seen as soon as possible. Pain or numbness reaching the foot is not a situation to be classified for oneself as "just sciatica" and watched.

#Red flags: why this article will not give you an exhaustive list

No authority has published a complete, closed list of red flags for low back pain — every source obtained for this article is written open-endedly, or needs to be read together with other passages to be complete.

This section is not here so that a reader can check a list and be reassured by not being on it. Precisely the opposite: because there is no exhaustive list, "not on the list" cannot be turned into "does not matter".

What each source writes for itself:

• NICE NG59 recommendation 1.1.1 is illustrative. It says: think about alternative diagnoses when examining or reviewing people with low back pain, particularly if they develop new or changed symptoms; exclude specific causes of low back pain, for example cancer, infection, trauma or inflammatory disease such as spondyloarthritis; and if serious underlying pathology is suspected, refer to the relevant NICE guidelines [Note 8]. NG59 uses "for example" itself and points to four other guidelines — that is, it does not set out to be complete here.
• The table in Hong Kong's Reference Framework prints a complete six rows, but the body text of Chapter 5 of the same document adds several more (age over 65, night-time pain, unresponsiveness to previous therapies, recent epidural or spinal instrumentation or procedures). So the table is not everything the document says.
• The Primary Healthcare Commission's guideline of December 2025 heads its table 3.1 "Common Red Flag Conditions in Primary Care Setting that Require Urgent or Early Medical Care" — "common" is by itself not "complete"; and that table covers all musculoskeletal problems rather than low back pain alone. Paragraph 3.1.3 of the same guideline states that table 3.1 lists common but important red flag symptoms and signs encountered in the primary care setting [Note 9].
• The profession itself states that the evidence base for red flags is thin. Finucane et al. 2020 (the IFOMPT international framework, J Orthop Sports Phys Ther 2020;50(7):350-372) state that red flags have historically been used by clinicians to identify serious spinal pathology, and that there is currently an absence of high-quality evidence for the diagnostic accuracy of most red flags [Note 10].

In one sentence: a red flag is a prompt to raise suspicion, not a test you can use on yourself to exclude serious disease. What follows reproduces each source's own list and adds nothing that is not in those sources.

#The six-row table the Hong Kong Government publishes (reproduced; not exhaustive)

Note that the last row writes "Older age", with no number. The body text of Chapter 5 of the same Reference Framework carries a number: features that may suggest underlying systemic disease include a history of cancer, age >65 years, unexplained weight loss, chronic pain of more than 1 month, persistent pain, night-time pain and unresponsiveness to previous therapies; and documented fever, intravenous drug use, recent infection (particularly bacteraemia), or recent epidural or spinal instrumentation or procedures increase the suspicion of spinal infection [Note 11]. That passage uses "include", so it too is open-ended.

This article will not print an age threshold for a first episode. In the Hong Kong official sources obtained here, the table says "Older age" with no number and the body text says "age >65 years"; the commonly circulated threshold of under 20 or over 50 appears in none of the sources cited here (the Hong Kong Reference Framework, the Primary Healthcare Commission guideline, NICE NG59, or the National Health Service). This article does not print a threshold that has no source.

#Which need to be seen at once and which within two weeks — Hong Kong's own published triage tiers

"See a doctor" is not something you can act on. What follows is the only Hong Kong official document found here that states, for each class of warning sign, how quickly it is to be handled.

Paragraph 3.1.4 of the same guideline explains why certain conditions sit in the most urgent tier. That paragraph has one general statement and three parallel examples, quoted in full at [Note 12]: in principle, any patient with a symptom or sign suggestive of a medical emergency must be referred to the emergency department without delay (that is, urgent, Priority category 1); examples of medical emergency include (a) potentially life-threatening conditions, for example heart attack, shock, status asthmaticus; (b) significant trauma with fracture or severe bleeding; and (c) conditions at high risk of rapid deterioration resulting in potentially irreversible complications or permanent disabilities unless promptly intervened, for example stroke and cauda equina syndrome. Note the word "include" — those three are examples, not a complete list.

Paragraphs 3.1.5 and 3.1.6 that follow deal with the other two tiers the same way: for Priority category 2 (24 to 48 hours) the examples given are raised blood pressure requiring a work-up for target organ damage, and infection requiring early antibiotics; for "Early" (within 2 weeks) the examples are suspected malignancy, cardiac or peripheral vascular disease, and depression. All three paragraphs open with "Examples … include", and all three are non-exhaustive.

⚠️ If you have low back pain and have recently developed urinary or faecal retention or incontinence, a change of sensation in the saddle area (perineum, ischium, inner thigh), or weakness in a leg, take note. One thing has to be stated clearly: it is not "both sides required". The Hong Kong guideline's cell reads "especially if unilateral" — "especially if unilateral", meaning one leg counts, not that both are needed [Note 13]. The National Health Service's 999 list writes "in both legs"; the two documents differ on that point, and this article lists both without taking the narrower. In the Hong Kong guideline this group is Priority category 1, emergency service, as soon as possible — not "watch it for a few days", and not "try physiotherapy first".

#The National Health Service's public version of the tiers — a more cautious reading

For the same symptoms, the version the United Kingdom's National Health Service publishes for the public has three tiers, and includes one thing the Hong Kong table does not (a change of sensation in sexual function). This is a more cautious reading, and this article lists both without choosing.

The Hospital Authority's Smart Patient site also has several Chinese warning sentences (29 July 2024). They are scattered across four different passages of the page and do not form a list, and are quoted in full in page order at [Note 14]: if symptoms persist without improvement, or severe nerve compression appears (causing muscle weakness, persistent numbness, faecal incontinence, urinary retention), or there is spinal instability or deformity, the patient will need surgery; a slipped disc can compress the spinal nerves, and if the patient develops muscle weakness, faecal incontinence, urinary retention, or persistent pain and numbness after physiotherapy, surgical treatment will be needed; patients with spinal stenosis, in the later stage as nerve compression continues, develop muscle weakness, numbness, faecal incontinence and urinary retention and need early surgical treatment; and a patient with unexplained fever, marked weight loss, poor general health or problems in other organs should seek medical attention early.

In other words: the same combination of symptoms (muscle weakness, faecal incontinence, urinary retention) appears three times on the same page, placed under two different sections, treatment and complications. A reader working down the page will easily see only one of the three.

Immediately below that 999 list, the same NHS page has an information box on how to get to the emergency department: do not drive yourself to A&E; ask someone to drive you, or call 999 for an ambulance; and bring any medicines you take with you [Note 15]. That is the British instruction; the emergency number in Hong Kong is likewise 999, and under the Hong Kong guideline above, Priority category 1 is likewise a referral to emergency services.

Not being on any of these tables does not mean nothing is wrong. The three sources list different items: the Hong Kong table has nothing on sexual function, the NHS does; the Hong Kong table says "Older age" while the body text of the same document says ">65"; and the Primary Healthcare Commission table covers all musculoskeletal problems rather than low back pain alone. And Finucane et al. 2020 state that there is an absence of high-quality evidence for the diagnostic accuracy of most red flags. Those differences are themselves the proof that no exhaustive list exists.

#Why does the doctor not send you straight for an MRI?

Two guidelines, one from Hong Kong and one from the United Kingdom, point the same way: imaging should not be done routinely. But the exception "unless the result is likely to change management" is not worded identically on both sides — the Hong Kong one does not distinguish by setting, while NICE writes it that plainly only for specialist settings (1.1.6). The reason is not cost. It is that "abnormalities" are just as common in people who have no pain.

Chapter 5 of the Hong Kong Reference Framework states that routine imaging and laboratory investigations are not recommended unless there is suspicion of cancer, spinal infection, signs of cauda equina syndrome, severe or progressive neurological deficits, or a risk of vertebral compression fracture; and its formal recommendation is — do not routinely offer imaging for individuals with low back pain unless it is likely to change management [Note 16].

The same document sets out its supporting evidence [Note 16]: many imaging findings identified in people with low back pain are also common in people without such pain, and their importance in diagnosis is unknown; studies did not show that imaging improved patient outcomes; and in patients with non-specific low back pain, X-ray and magnetic resonance imaging findings do not correlate with clinical symptoms or work capacity.

NICE NG59 (published 2016, last updated 29 July 2026) varies in the strength of its wording, and each is worth reading separately [Note 17]: 1.1.4 — do not routinely offer imaging in a non-specialist setting for people with low back pain, with or without sciatica; 1.1.5 — explain to people with low back pain, with or without sciatica, that if they are being referred for a specialist opinion they may not need imaging; 1.1.6 — consider imaging in specialist settings of care (for example a musculoskeletal interface clinic or hospital) only if the result is likely to change management.

In other words, it is not "you may not have one", it is "you will not have one automatically". If there is a red flag, or if the imaging result would genuinely change what happens next, both guidelines leave a door open — but the difference matters: the Hong Kong document does not distinguish by setting, while NICE's door opens only in the specialist setting (1.1.6); for the non-specialist setting where most readers are (1.1.4), NICE's original text carries no such exception.

#Do the arithmetic: how common "abnormal" imaging is in people with no pain

Brinjikji et al. 2015 (AJNR Am J Neuroradiol 2015;36(4):811-6, published about 11 years ago) systematically reviewed 33 papers covering 3,110 people with no symptoms at all: the prevalence of disc degeneration in asymptomatic individuals rose from 37% in the 20-year-old group to 96% in the 80-year-old group; disc bulge from 30% to 84%; disc protrusion from 29% to 43%; and annular fissure from 19% to 29%; and the authors' conclusion is that many imaging-based degenerative features are likely part of normal ageing and unassociated with pain [Note 18].

Do the arithmetic (converting the figures quoted above directly; the denominator is people with no symptoms):

• If 100 people aged 20 with no low back pain at all are scanned, at 37% and 30%: about 37 would be expected to show disc degeneration and about 30 disc bulge.
• Among 100 people aged 80 with no low back pain at all: at 96% and 84%, about 96 would be expected to show disc degeneration and about 84 disc bulge.
• By contrast: disc protrusion rises from 29% at 20 to 43% at 80, a far smaller rise than bulge (30% to 84%). That is, "something is wrong with the disc" describes very different degrees of commonness under different terms.

In one sentence: a report saying "disc degeneration" is close to the norm in the 80-year-old group (about 96%), and by itself it does not tell you whether the pain comes from it. That is why the Hong Kong Reference Framework, and NICE NG59 for specialist settings (1.1.6), both say imaging should not be offered routinely unless the result is likely to change management; and NICE's recommendation for the non-specialist setting (1.1.4, where most readers are) is blunter still — its original carries no such exception.

⚠️ If you already have an MRI report and the words "degeneration", "bulge" or "protrusion" have frightened you: the rates at which those words appear in people of the same age with no pain are above. How the report relates to you personally is for a doctor to decide.

#What imaging costs in the public system

The Hospital Authority's fee table (the page states an effective date of 1 January 2026), for eligible persons: non-urgent radiology service — basic items free, intermediate items $250 per item, advanced items $500 per item [Note 19]. Note that the fee table does not itemise which scan counts as basic, intermediate or advanced, so this article makes no statement about which tier an X-ray or an MRI falls into.

#What do the guidelines say to do? — the same long list of things not to do

In low back pain guidelines, the "do not" part is longer than the "do" part.

The treatment ladder in the two-page summary of the Hong Kong Reference Framework. What that ladder governs comes first. The column headings of that figure read "Non-specific LBP in primary care" and "Reassurance and self-management" — that is, it applies only to non-specific low back pain in primary care, not to all low back pain.

The ladder's content is [Note 20]: acute low back pain — non-pharmacological treatment, first line advice to remain active, second line exercise; pharmacological treatment, first line simple analgesics (NSAIDs, or paracetamol if there are no safe alternatives), second line multiple analgesics of different classes. Chronic low back pain — non-pharmacological treatment, first line advice to remain active and exercise, second line physiotherapy, nutrition intervention, psychological intervention and multidisciplinary rehabilitation; pharmacological treatment as above.

Two further lines sit immediately below the ladder in the same figure, and they are its exit conditions [Note 20]: consider referring to a specialist if there are red flag symptoms (see the table) or no clinical improvement; and review progress at pre-determined intervals and reassess. In one sentence: this ladder is not a road to be walked to its end — it states two conditions for leaving it, a red flag or no improvement.

Chapter 5 of the same Reference Framework has the corresponding formal recommendation: reassure individuals with non-specific low back pain and advise them to remain active and self-manage as first-line treatment. The body text that follows adds a qualification [Note 21]: for patients with acute low back pain, advice and education, reassurance, and encouragement to avoid bed rest, stay active and continue with usual activities including work, are sufficient as initial management; for patients with chronic low back pain, advice on staying active alone is not sufficient, and advice on appropriate exercises and promotion of self-management are also required.

The WHO fact sheet states that medicines can be used to reduce the symptoms of low back pain and should ideally be combined with other treatments, and that painkillers should not be the first-line treatment for low back pain [Note 1].

#Two official divergences this article will not smooth over

Divergence one: paracetamol. Hong Kong's Reference Framework (2022) permits "paracetamol if no safe alternatives" as first line; NICE NG59 (last updated 29 July 2026) recommendation 1.2.26 states that paracetamol alone should not be offered for managing low back pain. Both are current official documents, and this article reproduces both.

But that divergence is narrower than it looks, and the reason is in the next paragraph of the Hong Kong document itself. Section 5.3.2(b) of Chapter 5 of the Hong Kong Reference Framework explains why it nonetheless keeps paracetamol [Note 22]: according to a systematic review of paracetamol for low back pain, there was no difference between paracetamol and placebo for acute low back pain patients at 1 week, 2 weeks, 4 weeks and 12 weeks of follow-up; however, more than 90% of the participants analysed in that review came from a single large trial, and no trial evaluating paracetamol for chronic low back pain patients was found; so in selected patients for whom there are no safe alternatives, it is reasonable to consider a trial of paracetamol as initial therapy.

In one sentence: the two documents actually read the evidence on how effective paracetamol is the same way — both say there is no difference from placebo; the divergence is over whether it may be tried when there is no other safe option. The National Health Service's public page uses a third formulation: paracetamol on its own is not recommended for back pain, but it may be used with another painkiller [Note 23]. None of the three says paracetamol is the first choice for low back pain.

Divergence two: what physiotherapy includes. The Hospital Authority's Smart Patient site (29 July 2024) states that physiotherapy includes electrotherapy, heat therapy, lumbar traction, manual therapy, acupuncture, postural correction and exercise therapy [Note 7]. NICE NG59 states that several of those should not be offered (traction 1.2.6, acupuncture 1.2.8; on electrotherapy it separately lists transcutaneous electrical nerve stimulation 1.2.11 and interferential therapy 1.2.12).

But one item on that list NICE does support, and it is the only recommendation in the current version of NG59 marked as amended in 2026 — manual therapy. Recommendation 1.2.7 states: consider manual therapy (spinal manipulation, mobilisation, or soft tissue techniques such as massage) for managing low back pain, with or without sciatica, but only as part of a treatment package that includes exercise [Note 24]. "Only as part of a treatment package that includes exercise" is a condition, not a footnote — manual therapy on its own is outside what NICE supports.

Note that the next two sentences in the same passage of the Smart Patient site point the same way as NICE [Note 7]: these treatments can relax the muscles and reduce local pain; and in the long run, active exercise is what best helps you restore the normal biomechanics of the lower back and avoid abnormal irritation of the nerves.

That is: the divergence is not about whether exercise works, and not about manual therapy; it is about whether the passive treatments (electrotherapy, traction, acupuncture) should be offered. The Smart Patient site does not itself say those passive treatments can replace exercise — what it writes is that they relax the muscles and reduce local pain.

#The full "Do not offer" table in the current version of NICE NG59

#The "when you can" recommendation sitting beside each "do not"

The table above is NG59's "do not" side. NICE itself, in the same document and the same numbering sequence, placed recommendations that open the door again immediately beside several of them; this section reproduces those.

The most important: 1.2.27 points at 1.2.25 itself. The original of 1.2.27 in the table reads "Do not routinely offer opioids for managing acute low back pain (see recommendation 1.2.25)" — the bracket is NICE's own, and the 1.2.25 it points to states: consider weak opioids (with or without paracetamol) for managing acute low back pain only if an NSAID is contraindicated, not tolerated or has been ineffective [Note 25].

"Weak opioids" is a term NG59 defines in its own "Terms used in this guideline" section, and what that section writes is to see the information on weak opioids in the analgesics section of the British National Formulary — that is, NG59 does not list which drugs count as "weak" in this document, and points to another.

In one sentence: 1.2.27 says "do not offer routinely", not "must not be used"; and 1.2.25 states three circumstances in which it may be considered. Reading 1.2.27 without 1.2.25 leaves a patient with acute low back pain who cannot take NSAIDs believing the guideline has left no room for them. Note that both concern acute pain only (defined by NG59 as less than 3 months); on the chronic side, 1.2.28 carries no exception.

⚠️ If you cannot take NSAIDs such as ibuprofen because of a stomach ulcer, gastrointestinal bleeding, kidney problems, cardiovascular risk or allergy: the "do not" table above is not a conclusion written for you; NICE has 1.2.25 for your situation. What you personally can take and for how long is for a doctor — this article gives no advice on medication.

Five more in the same sequence are likewise qualifications on the entries in that table, quoted in full at [Note 26]: 1.2.18 — if a person is already taking opioids, gabapentinoids or benzodiazepines for sciatica, explain the risks of continuing them; 1.2.19 — as part of shared decision making about stopping them, discuss the problems associated with withdrawal; 1.2.20 — be aware of the risk of harms and the limited evidence of benefit from NSAIDs in sciatica; 1.2.21 — if prescribing NSAIDs for sciatica, take into account differences in gastrointestinal, liver and cardio-renal toxicity and the person's risk factors including age, think about appropriate clinical assessment, ongoing monitoring of risk factors and the use of gastroprotective treatment, and use the lowest effective dose for the shortest possible period; and 1.2.23 — when prescribing oral NSAIDs for low back pain, think about appropriate clinical assessment, ongoing monitoring of risk factors, and the use of gastroprotective treatment.

Those five settle three things. The two "do not offer" recommendations 1.2.16 and 1.2.17 are about not starting, not about telling people already taking these drugs to stop on their own — NG59 has 1.2.18 and 1.2.19 for people already taking them, and what those say is to explain the risks and discuss withdrawal, not to stop abruptly. 1.2.20 warns that the evidence of benefit from NSAIDs in sciatica is limited; but 1.2.21, immediately following, is not a prohibition but a set of prescribing conditions — 1.2.21's words are "If prescribing NSAIDs for sciatica", so NG59 does leave a place for NSAIDs in sciatica. Reading 1.2.20 without 1.2.21 leaves the impression that NG59 says nothing about how to use NSAIDs in sciatica; and equally, the "Consider oral NSAIDs" of 1.2.22, written for low back pain, cannot be taken directly as a recommendation for sciatica. 1.2.23 is the one sitting between 1.2.22 and 1.2.24, and says to think about gastroprotective treatment. None of the five is in the "do not" table.

The Hong Kong document's wording on opioids is worth setting beside this. Section 5.3.2(c) of Chapter 5 of the Hong Kong Reference Framework states that the efficacy of opioid analgesics in acute low back pain is unknown [Note 27]. The two documents say different things: NICE sets out a conditional place for them in acute low back pain (1.2.25), while the Hong Kong Reference Framework writes "unknown" on the same question. This article reproduces both without choosing.

NG59 has other positive recommendations too, quoted in full at [Note 28]: 1.2.1 — provide advice and information tailored to the person's needs and capabilities to help self-management; 1.2.15 — promote and facilitate return to work or normal activities of daily living; 1.2.22 — consider oral NSAIDs, taking into account differences in toxicity and the person's risk factors; 1.2.24 — prescribe at the lowest effective dose for the shortest possible period. On the invasive side there are two that open rather than close a door: 1.3.5 — consider epidural injections of local anaesthetic and steroid in people with acute and severe sciatica; and 1.3.8 — consider spinal decompression for people with sciatica when non-surgical treatment has not improved pain or function and their radiological findings are consistent with sciatic symptoms. Those two sit immediately before 1.3.6 and 1.3.9 in the table — 1.3.6 prohibits epidural injection for neurogenic claudication from central spinal canal stenosis, not all epidural injections; and 1.3.9 and 1.3.10 prohibit fusion and disc replacement for low back pain, not decompression for sciatica.

A note on currency: in the current version of NG59, recommendations 1.2.13 and 1.2.14 both read "This recommendation has been deleted." The original does not say at which update they were removed, and this article does not infer it. But the subheadings above the two deleted items are still printed on the page, "Psychological therapy" and "Combined physical and psychological programmes", so a reader can still see which subject the deleted recommendations came under, only not what they said. What matters is that any second-hand article quoting NG59 recommendation numbers should be checked against the current original — because numbers can be deleted while the page address stays the same, and old articles do not update themselves.

#Back belts: a conclusion with figures behind it

Section 5.1 of Chapter 5 of the Hong Kong Reference Framework has a recommendation first and then its supporting evidence. Both are reproduced here — because reading only the figures leaves out which sentence they support. The recommendation is: advise individuals to maintain an optimal level of physical activity for the prevention of low back pain [Note 29].

The supporting evidence in the same section is [Note 29]: a 2016 systematic review concluded that exercise combined with education, and exercise alone, reduced the risk of an episode of low back pain with relative risks of 0.55 (95% confidence interval 0.41 to 0.74) and 0.65 (0.50 to 0.86) respectively; while education alone at a relative risk of 1.03 (0.83 to 1.27), back belts at 1.01 (0.71 to 1.44) and shoe insoles at 1.01 (0.74 to 1.40) had no protective effect on low back pain.

Both sides together: in the same review against the same controls, the exercise groups' risk of an episode was 0.55 to 0.65 times that of the control group; the back belt group was 1.01 times, that is no different from control, with a confidence interval crossing 1. NICE NG59 recommendation 1.2.3 likewise states that belts and corsets should not be offered for managing low back pain. That review published only relative risks and did not publish the absolute episode rates of the two groups in this passage, so this article does not convert them into "so many fewer per 100 people".

As for which exercise and how much, that passage of the Hong Kong Reference Framework does not say; NG59 1.2.2 says to consider a group exercise programme (biomechanical, aerobic, mind–body or a combination of approaches), taking a person's specific needs, preferences and capabilities into account when choosing the type of exercise [Note 28]. This article specifies no movement and no number of repetitions; an exercise prescription depends on the individual and is for a doctor or physiotherapist.

⚠️ If you are about to buy a back belt or shoe insoles to prevent back pain: those two relative risks of 1.01 are the answer currently quoted in Hong Kong government documents.

#After July 2025, do you still need a doctor's letter to see a physiotherapist?

The default is still yes — but the referrer has widened from "registered medical practitioner" to "registered medical practitioner or registered Chinese medicine practitioner", and the law now has four referral-free routes; one of which, today, covers only two conditions in the whole of Hong Kong, and low back pain is one of them.

Section 6 of the Physiotherapists (Registration and Disciplinary Procedure) Regulation (Chapter 359J) (consolidated version dated 25/07/2025; amendment note "section 162 of Ordinance No. 33 of 2025") is quoted in full at [Note 30]. The four routes are: (a) a diagnosis certificate issued within 12 months (under subsection (2)(a)(ii), services may be provided only for the diagnosed condition); (b) a condition recognised by a clinical guideline; (c) an interdisciplinary collaboration arrangement of the Primary Healthcare Commission; and (d) circumstances specified by the code of practice as urgent or involving social services.

Note that the training requirement is not tied to one route alone — the Physiotherapists Board page quoted in the next section states that the training requirement must be met before any of sections 6(2)(a), 6(2)(b) and/or 6(2)(c), which means that even the route of taking a diagnosis certificate issued within 12 months to a physiotherapist is subject to the same personnel threshold.

But the section does not end at subsection (2). Subsection (3), immediately following, is what turns the ceilings in the guideline below into a statutory duty [Note 30]: a physiotherapist who provides services to a person in the circumstances described in subsection (2)(b) must comply with the requirements set out in the clinical guideline; subsection (4) does the same for subsection (2)(c); and subsection (5) provides that a physiotherapist whose name is in Part II must practise under the supervision of a physiotherapist whose name is in Part Ia.

And the words in that section are not a reference to any guideline whatever — subsection (6) of the same section defines them [Note 30]: "referencing authority" means the Department of Health, the Hospital Authority, the Primary Healthcare Commission or the Chinese Medicine Hospital of Hong Kong; and "clinical guideline" means a clinical guideline that is published by a referencing authority on its website, intranet or a similar electronic network and states that the guideline is published for the purposes of subsection (2)(b).

In one sentence: route (b) is not "any clinical guideline that has written about it". The guideline has to be published by one of those four named authorities and has to state for itself that it is published for the purposes of section 6(2)(b). And under subsection (3), when a physiotherapist provides services by this route, they must comply with the requirements set out in that guideline — so the ceilings in the next section are not soft advice but requirements under section 6.

The Chinese medicine referral route is itself new. The government press release on the arrangement for registered Chinese medicine practitioners to refer patients for allied health services (11 December 2025) states that the Supplementary Medical Professions (Amendment) Ordinance 2025 took effect on 25 July 2025, renaming the Supplementary Medical Professions Ordinance (Chapter 359) as the Allied Health Professions Ordinance; and that the Supplementary Medical Professions Council had passed codes of practice for radiographers, medical laboratory technologists, physiotherapists and occupational therapists allowing them to provide allied health services to patients referred by registered Chinese medicine practitioners, with those amendments taking effect on 11 December 2025 [Note 31].

#Route (b): Hong Kong currently covers only two conditions, and low back pain is one

Paragraph 3.2.9 of the Primary Healthcare Commission's guideline (document self-dated "First published: Dec 2025"; English only, with no official Chinese title, so it is referred to here throughout by its English name) states that at present the Hong Kong Reference Framework for Common Musculoskeletal Problems in Primary Care Setting covers two highly prevalent conditions, namely osteoarthritis of the knees and low back pain, which together constituted over 50% of primary care consultations for musculoskeletal problems in Hong Kong [Note 32].

The same guideline is hard-edged about the boundary [Note 32]: under this specific circumstance within the direct access arrangement, if the patient's presenting condition is not consistent with osteoarthritis of the knee, low back pain, or any other condition stipulated by a clinical guideline published by a referencing authority, the first-contact physiotherapist must refrain from providing direct access physiotherapy service, even if the condition can be effectively and safely managed by physiotherapy.

#How many layers of limit? — the eight written into that same guideline

The Chinese text of that Physiotherapists Board page gives the course specification and the enrolment requirements:

Hours: 8 hours (100% attendance) / Assessment: the examination pass mark is 75% (within six months of completing the course, a student may sit two further examinations, not counting the first sitting) / A physiotherapist in Part Ib needs to obtain from an employer proof of completing 2 000 hours of clinical experience, and to make, before a person specified in section 12 of the Oaths and Declarations Ordinance (Chapter 11), a statutory declaration of having completed 2 000 hours of clinical experience (that is, a Justice of the Peace, notary public, commissioner for oaths or other person authorised by law to administer oaths, any of whom may in the manner provided by section 14 take and receive a declaration made before them by any person). / (Note: the Department of Health and the Home Affairs Department provide free oath-taking services) (our translation from the Chinese original)

Chinese original:

「時數：8 小時（100% 出席率）」「評核：考試的合格率為 75%（學員於完成課程後六個月內，可額外重考兩次（不包括首次考試））」「第Ib部的物理治療師需從雇主獲得完成2 000小時臨床經驗的證明，並在《宣誓及聲明條例》（第11章）第12條所指明的人士面前作出完成2 000小時臨床經驗的法定聲明（即太平紳士、公證人、監誓員或其他獲法律授權監誓的人，均可按第14條訂定的方式監理和接受任何人在其面前作出的聲明）。」「（註：衛生署及民政事務署提供免費的宣誓服務）」

(The space inside 「2 000」 is the page's own, reproduced as found.)

#Red flags are a statutory duty within this arrangement

The government press release of 19 March 2025 states that physiotherapists and occupational therapists must at all times follow the recognised clinical guidelines, including referring a case promptly to a doctor for diagnosis and management on noticing that a patient has particular warning symptoms [Note 33]. The English version of the same press release explains what the 12-month proof can be: patients may seek direct physiotherapy or occupational therapy services for health conditions diagnosed by a registered doctor or Chinese medicine practitioner within the past 12 months without obtaining a new referral letter each time; patients must provide proof of the diagnosis, and apart from a referral letter, the proof can also be in the form of outpatient records, follow-up consultation records or discharge summaries [Note 33].

And the duty to refer onwards is not confined to seeing a red flag, nor to referral-free cases. The Physiotherapists Board's "First-contact Physiotherapist" page states [Note 34] that a first-contact physiotherapist–patient relationship is established when a patient, without a referral letter, receives assessment and subsequent treatment from a physiotherapist under sections 6(2)(a), (b) and/or (c) having given informed consent; and that whether or not the patient holds a referral letter, where the patient's needs go beyond the physiotherapist's professional scope, the physiotherapist should inform the patient and assist in finding a suitable and qualified person to provide the service needed.

The phrase "whether or not the patient holds a referral letter" is the key: this duty to inform and to help arrange onward referral does not exist only under the referral-free arrangement — someone attending physiotherapy with a doctor's letter is covered by the same sentence.

A comparison, not a calculation: direct access in Hong Kong and self-referral in the United Kingdom are not the same thing. The National Health Service's public page (page reviewed 5 March 2026) states that in many areas you may be able to get help, such as physiotherapy, from NHS community musculoskeletal services without needing a referral from a general practitioner [Note 23] — that is, open by area, with no statutory list of conditions. Hong Kong's version, on the table above, has at least eight layers of limit: the condition limit, the multiple condition limit, the ceiling on sessions and days, a red flag appearing mid-course, compulsory referral where there is no improvement after the course, review by a doctor even where it helped, the records limit and the personnel training limit. That "eight" is counted here from the passages of that guideline quoted above, not a number the guideline publishes — the guideline does not number or total these limits.

The Hong Kong Physiotherapy Association's submission to the Legislative Council of 19 March 2026 (LC Paper No. CB(3)214/2026(01)) notes that the guideline currently covers only low back pain and osteoarthritis of the knee, so that patients with other musculoskeletal conditions may be unable to obtain physiotherapy directly and in good time. (Only the English version of that document was obtained for this article, so it is restated here rather than quoted.)

As for which physiotherapists have met the training requirement — the Physiotherapists Board's "First-contact Physiotherapist" page carries no public register, but the same page states immediately afterwards where that proof should be lodged [Note 34]: physiotherapists should be responsible for keeping proof of completing the training body's course and using it where necessary, for example by submitting it directly to the Primary Healthcare Commission to update the Primary Care Directory, reflecting their qualification as a first-contact physiotherapist.

Following that up: the Primary Care Directory's physiotherapist search has a filter for first-contact physiotherapy service; searching all eighteen districts of Hong Kong on 3 August 2026 returned 58 entries.

That number has to be read with three limits: 58 is the number of search results and is not necessarily 58 people (a physiotherapist with several practice addresses is listed separately for each, and no de-duplication was done here); the entries are submitted by the service providers themselves, and the Directory is not the Board's registration record and does not verify whether training has been completed; and the figure is a snapshot of that day.

And the Primary Care Directory's own page still prints the rule as it stood before the 2025 amendment. That page states that under the current legislation and the codes of practice of occupational therapists and physiotherapists, patients generally have to obtain a doctor's referral before receiving the services of occupational therapists and physiotherapists [Note 35]. Its footer reads 「最後更新日期: 二零二二年七月一日」, earlier than the amendment to section 6 of Chapter 359J that took effect on 25 July 2025. Under section 6(1) as quoted above, a registered Chinese medicine practitioner is equally a qualified referrer; and that sentence makes no mention of the four referral-free routes either. Both are government pages, and this article reproduces both with their respective dates.

⚠️ If you have low back pain, no red flag symptoms, and would rather not queue for a public general outpatient appointment first: that route exists in law today, but it has a ceiling of 10 sessions or 30 days, is confined to particular trained physiotherapists, and requires referral to a doctor if there is no improvement after the course.

#Public and private: how long the wait actually is, and what it costs

For stable new orthopaedic cases in the public system, the median across the Hospital Authority's seven clusters is 23 to 45 weeks; the 90th percentile of the same data is 53 to 81 weeks.

The Hospital Authority's Waiting Time for Stable New Case Booking at Specialist Out-patient Clinics, for the period 1 July 2025 to 30 June 2026 (the document states a next update date of 30 October 2026). For orthopaedics across all clusters the number of new case bookings is 108,274, with triage proportions of urgent 14%, semi-urgent 16% and stable 69%.

Both sides together: within the same data set, the median for urgent new cases is 1 week or less, and for stable new cases 23 to 45 weeks. Triage decides which column you fall into — and the priority tiers in the earlier section are the clinical version of that ruler.

#Public charges

Note the structure of the fee table: it has no separate line for physiotherapy. On the table above, physiotherapy falls under the charges for "specialist clinic (including integrated clinics and allied health clinics)" or "community allied health service".

On the safety net taking effect the same day, the Hospital Authority's Chinese page states that from 1 January 2026, outside the medical fee waiver mechanism, the Hospital Authority has established a second layer of safety net, adding an annual cap of ten thousand dollars on public healthcare service charges with no means test — the "annual charge cap" [Note 19].

Do the arithmetic (adding the charges quoted in the table above directly; this is nobody's actual bill): a low back pain episode going once through the public system without admission — one family medicine clinic attendance ($150) plus two drug items (at $5 each, $10) plus one intermediate non-urgent radiology item ($250) plus three community allied health visits (at $100 each, $300) comes to $710. The same set of items with a specialist clinic attendance instead ($250) plus two drug items (at $20 each, $40) plus the same imaging and three community allied health visits comes to $840. Both are far below the annual charge cap of $10,000 — and what that cap does not govern is the waiting time, only the money.

And if your case is not one trip through — admission, repeated imaging, long-term follow-up, genuinely more than ten thousand dollars in a year — you need to know that the cap does not apply automatically. The Hospital Authority's page on the annual charge cap for eligible persons sets out four eligibility criteria [Note 36]: the patient must be an eligible person; the patient must have paid a cumulative total of ten thousand Hong Kong dollars in eligible medical fees and charges within the year; at the time of application the patient must have no outstanding eligible medical fees and charges anywhere in the Hospital Authority; and the hospital services the patient received must not have been determined by the Hospital Authority as having no clinical need.

Once all four are met, the patient still has to submit an application. The same page states that the annual application period begins on 1 January each year and closes on 31 March of the following year, with late applications not accepted; that eligible medical fees cover only bills issued between 1 January and 31 December of each year and paid in full at the time of application; that a fresh application is needed for each year; and that self-financed drugs and medical devices are excluded [Note 36]. The English version separately gives the route: eligible patients may submit their applications via HA Go or at any hospital's shroff office once their cumulative valid annual spending has reached $10,000.

Which is to say: ten thousand dollars is not an automatic ceiling but a second layer you reach only by paying the ten thousand first, then applying yourself, and applying again each year.

#Published bills for private spine surgery

To be clear: the figures below are the distribution of bills for patients who have already had surgery, not what low back pain costs. From the Hong Kong Reference Framework quoted above (fewer than 1% is specific spinal pathology) and NICE (which expressly prohibits certain operations), it follows that the great majority of low back pain never reaches this point; and NG59 1.3.9 states that spinal fusion should not be offered for people with low back pain except as part of a randomised controlled trial.

But "the great majority will not" is not "nobody needs it", and both guidelines say who does — that part has nothing to do with price and comes first. Section 5.3.3 of Chapter 5 of the Hong Kong Reference Framework states [Note 37] that selected groups of patients with low back pain should undergo evaluation for surgery — patients with suspected cauda equina lesions and worsening neurological deficits require immediate surgical investigation; and patients with intractable pain resistant to conservative treatment.

The same is true on the NICE side: 1.3.9 and 1.3.10 in the table above prohibit fusion and disc replacement for low back pain, while 1.3.8, sitting immediately before 1.3.9, says that spinal decompression may be considered for sciatica. They are not the same condition and cannot be set against each other.

In one sentence: surgery in low back pain is a matter for a minority, but the group with suspected cauda equina lesions and worsening neurological deficits is not "a minority, so it can wait" — the Hong Kong document says immediate. That group corresponds to the Priority category 1, emergency service cell earlier in this article.

The Department of Health's Office for Regulation of Private Healthcare Facilities operates the platform for the Pilot Programme on Enhancing Price Transparency of Private Hospitals, for the reporting period 1 January to 31 December 2025. The platform's own notes are quoted in full at [Note 38], of which four have to be read first: the data are based on patients accommodated in standard wards undergoing a single operation or procedure, and exclude those who had multiple operations; the percentile data are ordered by the amount of the total charge rather than by the doctor's fee or the hospital charge within it; the doctor's fee includes the fees of the anaesthetist, the surgeon and ward rounds; and billing data reported by any hospital for no more than 30 procedures should be interpreted with caution, because it is based on a small number of cases and may not reflect the general level of charges.

Counting only the hospitals with a published case number band (the platform's own note tells readers to treat data based on few cases with caution): the fiftieth percentile of total charges for laminectomy ranges from $142,963 to $203,421 (three hospitals), and the 90th percentile from $193,636 to $368,150; for spinal fusion the fiftieth percentile ranges from $320,684 to $324,261 (two hospitals) and the 90th percentile from $453,516 to $476,400. Taking all reporting hospitals together, including those without a published case number band, the spread of fiftieth percentiles is $121,031 to $343,775 for laminectomy and $232,252 to $571,065 for spinal fusion.

These figures cannot be used to compare hospitals. The platform states itself that the items covered may not be the same at each hospital and tells readers to enquire with the hospital directly; and the same percentile corresponds to a ranking of the total charge, not to the median of any single component.

⚠️ If you hold private medical insurance and are wondering about the worst case: the above is the published distribution of bills; how much your policy pays and whether it has itemised sub-limits are in the terms of your own policy, on which this article makes no statement.

#What to do next

This article will not judge what your back pain is, but it can narrow down which road you take next into a few things.

First, check for the group of symptoms that cannot wait. Urinary or faecal retention or incontinence, a change of sensation in the saddle area (perineum, ischium, inner thigh), weakness in the leg or loss of sensation in a dermatomal distribution — the Hong Kong guideline places that group in Priority category 1, emergency service, as soon as possible. And it is not "both legs": the original writes "especially if unilateral". The National Health Service's public version adds one more: a change of sensation during sex, being unable to get an erection or to reach orgasm. Neither list is exhaustive, so "not on the list" cannot be turned into "does not matter" — they are prompts to raise suspicion, not tests for ruling yourself out.

Second, if you do not have that group of symptoms, do not treat an MRI as the first step. The two guidelines point the same way: imaging should not be done routinely, because in people with no pain at all, disc degeneration is present in about 96% of the 80-year-old group and bulge in about 84%. A report saying "degeneration" does not by itself tell you whether the pain comes from it. Whether you need a scan is for a doctor to decide on your circumstances.

Third, if you want to go straight to a physiotherapist, establish three things first. The law still requires a referral by default (the referrers now include registered Chinese medicine practitioners); the referral-free clinical guideline route covers only two conditions today, osteoarthritis of the knee and low back pain, with a ceiling of 10 sessions or 30 days, whichever comes first; and it is confined to first-contact physiotherapists with recognised training. To find one, the Primary Care Directory's physiotherapist search has a filter for first-contact physiotherapy service — but that is not an official register, and the entries are submitted by the providers themselves. One more thing worth knowing in advance: after the course ends, even if it helped and you want to continue, the guideline still requires going back through a doctor for review.

Fourth, on what to buy. In the 2016 systematic review quoted in the Hong Kong government document, the relative risk for back belts is 1.01 and for shoe insoles 1.01 — that is, no different from control; in the same review, exercise combined with education is 0.55 and exercise alone 0.65. NICE likewise states that belts and corsets should not be offered. Which exercise and how much is for a doctor or physiotherapist to set for your circumstances; this article writes no prescription.

Fifth, on money. One trip through the public system, on the two sums added above from the published charges, is of the order of $710 or $840, and the annual charge cap of ten thousand dollars has to be applied for by you, each year. The published median bills for private spine surgery run, in the Chinese original's own rounding, from 十二萬 to 三十四萬 dollars for laminectomy and from 二十三萬 to 五十七萬 for spinal fusion — that is, from roughly a tenth of a million to about a third of a million, and from about a quarter of a million to a bit over half a million — but those are the bills of people who have already reached surgery, not the price of low back pain.

And the last thing, which is what this article most wants to leave you with: every list set out here is stated by its own source to be non-exhaustive; and the profession itself states that there is no high-quality evidence for the diagnostic accuracy of most red flags. So the use of these lists is to raise your suspicion, not to draw your own conclusion.

#Frequently asked questions

Is "80% of people get back pain in their lifetime" false?

Not false, but its source has been compressed. The Smart Patient site does say it, but attributes it to "medical reports" without naming one; what the literature measured is a range (Walker 2000: 11%–84%; the Hong Kong Reference Framework 2022: 58–84%), and the only local Hong Kong lifetime prevalence figure is 39% from 1995 (95% confidence interval 34%–44%). See the first section of this article.

Does back pain always need an MRI?

The Hong Kong Reference Framework (2022) states that imaging should not be offered routinely unless the result is likely to change management, and it does not distinguish by setting; NICE NG59 (last updated 29 July 2026) writes it separately: in a non-specialist setting (1.1.4, where most readers are) it should not be offered routinely, with no exception attached in the original, and the "unless it is likely to change management" exception appears only in the recommendation for specialist settings (1.1.6). Whether you personally need one is for a doctor.

My MRI report says "disc degeneration" or "bulge" — is that serious?

Brinjikji et al. 2015 measured, in 3,110 people with no symptoms at all, disc degeneration rising from 37% in the 20-year-old group to 96% in the 80-year-old group (the detailed figures and their conversion are in the "do the arithmetic" passage). How the report relates to your situation is for a doctor to interpret.

My symptoms are not on the red flag list — does that mean nothing is wrong?

None of the three lists set out here is exhaustive, and their contents differ — a change of sensation in sexual function, for example, appears only on the National Health Service's; and Finucane et al. 2020 state that there is an absence of high-quality evidence for the diagnostic accuracy of most red flags. Not being on a list does not mean it does not matter.

Does wearing a back belt prevent back pain?

The 2016 systematic review quoted in Chapter 5 of the Hong Kong Reference Framework: back belts at a relative risk of 1.01 (95% confidence interval 0.71 to 1.44), that is no protective effect against control; in the same review, exercise combined with education and exercise alone reduced the risk of an episode at relative risks of 0.55 (95% confidence interval 0.41 to 0.74) and 0.65 (0.50 to 0.86) respectively. NICE NG59 recommendation 1.2.3 likewise states that belts and corsets should not be offered for managing low back pain.

In 2026, do I still need a doctor's letter to see a physiotherapist?

The default under section 6 of Chapter 359J (consolidated version 25/07/2025) is still a referral, by a registered medical practitioner or a registered Chinese medicine practitioner; one of the four exceptions is a condition recognised by a clinical guideline, and the Primary Healthcare Commission's guideline of December 2025 states that this route currently covers only osteoarthritis of the knee and low back pain, with a ceiling of 10 sessions or 30 days, whichever comes first, and is confined to first-contact physiotherapists with recognised training (that training requirement applies equally to the routes under sections 6(2)(a) and 6(2)(c); a physiotherapist in Part Ib must additionally submit proof of 2 000 hours of clinical experience and a statutory declaration before enrolling on the course). The Physiotherapists Board itself keeps no public register, but its page points to the Primary Healthcare Commission's Primary Care Directory, which has a filter for first-contact physiotherapy service.

I cannot take anti-inflammatory painkillers (NSAIDs) — does that leave me with nothing?

This article gives no advice on individual medication, but both guidelines write about this situation. The original of NICE NG59 recommendation 1.2.27 is "Do not routinely offer opioids for managing acute low back pain (see recommendation 1.2.25)." — the word that matters is routinely; the bracket is NICE's own, and it points to recommendation 1.2.25, which states that weak opioids (with or without paracetamol) may be considered only if an NSAID is contraindicated, not tolerated or has been ineffective, and only for acute pain (defined by NG59 as less than 3 months). On the chronic side, 1.2.28 carries no exception. Chapter 5 of the Hong Kong Reference Framework says of acute low back pain that the efficacy of opioid analgesics is unknown. Which drug you take is for a doctor.

How long is the wait for public orthopaedics?

For the period 1 July 2025 to 30 June 2026, the median for stable new orthopaedic cases by cluster is 23 to 45 weeks, and the "longest" (90th percentile) is 53 to 81 weeks; the median for urgent new cases is 1 week or less. Over the same period there were 108,274 new orthopaedic case bookings, with triage proportions of urgent 14%, semi-urgent 16% and stable 69%.

#Notes: the original texts

[Note 1] World Health Organization, Low back pain fact sheet (page dateModified 19 June 2023): "LBP can be experienced at any age, and most people experience LBP at least once in their life." The corresponding sentence in the simplified Chinese version: 「腰痛可以在任何年龄出现，大多数人一生中至少经历一次。」 "In 2020, low back pain (LBP) affected 619 million people globally and it is estimated that the number of cases will increase to 843 million cases by 2050, driven largely by population expansion and ageing" "Non-specific means that the experience of pain cannot be confidently accounted for by another diagnosis such as an underlying disease, pathology or tissue damage. It is non-specific in about 90% of cases." "Medicines can be used to reduce the symptoms of LBP and should ideally be combined with other treatments. Painkillers should not be the first-line treatment for LBP."

[Note 2] Walker BF, J Spinal Disord 2000;13(3):205-17: "Point prevalence ranged from 12% to 33%, 1-year prevalence ranged from 22% to 65%, and lifetime prevalence ranged from 11% to 84%."

[Note 3] Hong Kong Reference Framework for Common Musculoskeletal Problems in Primary Care Settings, core document (Health Bureau, 2022), section 1.2: "The lifetime incidence of LBP is 58-84%, and 11% of men and 16% of women have chronic LBP." and "In Hong Kong, the Department of Health reported that 21.5% of respondents, who are aged 15 or above, had LBP during the 30 days preceding the household survey. … 5.1% of respondents in the age group of 15-24 reported LBP in the 30 days preceding the survey and the prevalence rises gradually in older age groups, with up to 52.4% of respondents in the age group of 85 or above reported LBP."

[Note 4] Hospital Authority Smart Patient site, low back pain page (page updated 29 July 2024, visible only in an HTML comment):

Low back pain is very common in Hong Kong. According to medical reports, around 80% of adults experience low back pain at least once in their lives. / After reducing vigorous activity and resting for a few days, combined with appropriate exercise and strengthening of the back muscles, the condition can be relieved, and more than 80% of patients recover within a month. / A slipped disc can cause low back pain, and 80% of those patients are between the ages of 20 and 40. (our translation from the Chinese original)

Chinese original:

「腰背痛(Low Back Pain)在香港十分普遍，醫學報告顯示，約有八成的成年人一生中會經歷最少一次的腰背痛。」「病人在減少劇烈運動和休息數天後，配合適量運動及鍛鍊腰背肌肉可使病情舒緩，八成以上病人會在一個月內康復。」「椎間盤移位可以引起腰痛，80% 病人的年齡介乎20至40歲之間。」

The corresponding sentences in the English version of the same page: "80% of people who suffer from it can recover within one month." and "80% of the people with this condition are between the ages of 20 and 40."

[Note 5] GBD 2021 Low Back Pain Collaborators, Lancet Rheumatol 2023;5(6):e316-e329: "Between 1990 and 2020, age-standardised rates of prevalence and YLDs decreased by 10·4% (10·9-10·0) and 10·5% (11·1-10·0), respectively."

[Note 6] The Hong Kong Reference Framework two-page summary Assessment of Low Back Pain (LBP) (footer self-dated 2022): "Specific spinal pathology (< 1% in primary care) - Vertebral fracture - Malignancy - Spinal infection - Axial spondyloarthritis - Cauda equina syndrome" "Radicular Syndrome(s) (~ 5-10% in primary care) - Radicular pain - Radiculopathy - Spinal stenosis" "Non-specific LBP* (~ 90-95 % in primary care) - Presumed lumbar musculoskeletal origin of LBP" The footnote: "No tests available in primary to reliably specify the pathoanatomical source of LBP" ("in primary" is the document's own wording, reproduced as found). The non-spinal causes column: "Non-spinal causes? - Referred visceral pain –pancreatic / urological pathologies (e.g. pancreatitis, pancreatic cancer, prostatitis, pyelonephritis) - Vascular causes-aortic aneurysm or dissection, femoral artery occlusion, peripheral vascular disease - Hip pathology - Viral syndrome"

[Note 7] Hospital Authority Smart Patient site, low back pain page:

Low back pain usually means pain arising from the tissues of the lower back, such as the muscles, tendons, intervertebral discs and vertebral joints, and can extend from below the ribs of the lower back to the buttocks and the back of the thighs. / Low back pain differs from sciatica; the pain does not extend to the calf or ankle as sciatic pain does. / Back pain in middle-aged and younger people is mostly caused by muscular overuse and poor posture, while back pain in older people is mostly caused by degeneration of the skeletal system, such as disc degeneration and ageing of the vertebral joints. / Physiotherapy includes electrotherapy, heat therapy, lumbar traction, manual therapy, acupuncture, postural correction and exercise therapy. / These treatments can relax the muscles and reduce local pain. / In the long run, active exercise is what best helps you restore the normal biomechanics of the lower back and avoid abnormal irritation of the nerves. (our translation from the Chinese original)

Chinese original:

「腰背痛通常是指由腰背位置的組織，如肌肉、肌腱、椎間盤、椎骨關節所引發之痛楚，範圍可由腰背肋骨對下伸延至臀部及大腿後端。」「腰背痛有別於坐骨神經痛，痛楚範圍不會如坐骨神經痛伸延至小腿或足踝。」「中年及年輕人背痛多是由於肌肉過勞及不良姿勢導致，而老年人背痛則多是由於骨骼系統退化引起，例如椎間盤退化及椎骨關節老化。」「物理治療包括：電療、熱療、腰椎牽引、手法治療、針灸、姿勢矯正和運動療法。」「這些治療可放鬆肌肉，減輕局部疼痛。」「長遠來說主動性的運動最能幫助你回復腰背正常的生物力學特性，並且避免神經受到不正常的刺激。」

The English version of the same page: "Low back pain differs from sciatica. It does not extend down to the foot as sciatic pain does."

[Note 8] NICE NG59 recommendation 1.1.1: "Think about alternative diagnoses when examining or reviewing people with low back pain, particularly if they develop new or changed symptoms. Exclude specific causes of low back pain, for example, cancer, infection, trauma or inflammatory disease such as spondyloarthritis. If serious underlying pathology is suspected, refer to relevant NICE guidelines on: metastatic spinal cord compression in adults / spinal injury / spondyloarthritis in over 16s / suspected cancer."

[Note 9] Primary Healthcare Commission, Clinical Guideline for First-contact Physiotherapists…, paragraph 3.1.3: "Table 3.1 lists common but important red flag symptoms and signs encountered in the primary care setting" The heading of table 3.1: "Common Red Flag Conditions in Primary Care Setting that Require Urgent or Early Medical Care"

[Note 10] Finucane LM et al., J Orthop Sports Phys Ther 2020;50(7):350-372: "Red flags (signs and symptoms that might raise suspicion of serious spinal pathology) have historically been used by clinicians to identify serious spinal pathology. Currently, there is an absence of high-quality evidence for the diagnostic accuracy of most red flags."

[Note 11] The Hong Kong Reference Framework, Chapter 5: "Features that may suggest underlying systemic disease include history of cancer, age >65 years, unexplained weight loss, chronic pain more than 1 month, persistent pain, night time pain, and unresponsiveness to previous therapies. Documented fever, intravenous drug use, recent infection (particularly bacteremia), or recent epidural or spinal instrumentation/ procedures increase the suspicion of spinal infection."

[Note 12] The same guideline as at [Note 9], paragraph 3.1.4: "In principles, patients with any symptom and/or sign suggestive of medical emergency must be referred to the Emergency Department without delay (i.e. Urgent, priority category 1) for stabilisation, prompt work up, interventions +/- further inpatient care. Examples of medical emergency include: (a) Potentially life-threatening conditions, e.g. heart attack, shock, status asthmaticus (i.e. acute severe asthma); (b) Significant trauma with fracture +/- severe bleeding; (c) Conditions at high risk of rapid deterioration resulting in potentially irreversible complication(s)/permanent disabilities, unless promptly intervened / requiring further inpatient care, e.g. stroke, cauda equina syndrome."

[Note 13] The same, table 3.1: "Marked weakness, especially if unilateral, paraplegic or corresponding to specific myotome(s)" and "Loss of sensation, especially if unilateral, dermatomal in distribution (e.g. saddle anaesthesia)"

[Note 14] Hospital Authority Smart Patient site, low back pain page, in page order:

If symptoms persist without improvement, or severe nerve compression appears (causing muscle weakness, persistent numbness, faecal incontinence, urinary retention), or there is spinal instability or deformity, the patient will need surgery. / The purpose of surgery is to relieve the nerve root or spinal canal, to stabilise and fuse unstable vertebrae, and to correct a deformed spine. / A slipped disc can compress the spinal nerves and damage the sciatic nerve. If the patient develops muscle weakness, faecal incontinence, urinary retention, or persistent pain and numbness after physiotherapy, surgical treatment to relieve the nerve root will be needed. / In the early stage the patient's only problem is an inability to stand or walk for long periods, but in the later stage, as nerve compression continues and the nerve is damaged, muscle weakness, numbness, faecal incontinence and urinary retention appear, and the patient needs early surgical treatment to relieve the nerve root and the spinal canal. / A patient with unexplained fever, marked weight loss, poor general health or problems in other organs should seek medical attention early to find the cause and exclude tumour, inflammation or visceral disease. (our translation from the Chinese original)

Chinese original:

「若症狀持續沒有改善、或出現嚴重神經壓迫(引致肌肉無力、持續麻痺、大便失禁、尿瀦留)、或有脊柱不穩或畸型，病人便需要施行手術。」「手術目的是放鬆神經根或椎管，固定及融合不穩定之椎骨，矯正變形之脊柱。」「椎間盤移位可以導致脊椎神經受壓，引致坐骨神經受到損害。若病人出現肌肉無力、大便失禁、尿瀦留、或接受物理冶療後仍持續痛楚、麻痺，便需要接受手術治療，放鬆神經根。」「病人早期的問題只是不能長期站立或步行，但後期由於神經持續受壓，引致神經受損，出現肌肉無力、麻痺、大便失禁、尿瀦留，病人便需要及早接受手術治療，放鬆神經根及椎管。」「病人如有不明原因的發燒、體重明顯減輕、出現不良健康狀態或其他器官問題，便要及早求醫，找出病因，排除患有腫瘤病、炎症或內臟病症。」

(「冶療」 in the fourth quotation is the page's own misprint, reproduced as found.)

[Note 15] NHS, Back pain (Page last reviewed: 05 March 2026): "Do not drive to A&E. Ask someone to drive you or call 999 and ask for an ambulance. Bring any medicines you take with you."

[Note 16] The Hong Kong Reference Framework, Chapter 5: "Routine imaging and laboratory investigations are not recommended unless there is suspicion for cancer, spinal infection, signs of the cauda equina syndrome, severe or progressive neurologic deficits or risk of vertebral compression fracture." "Recommendation — Do not routinely offer imaging for individuals with low back pain unless it is likely to change management. [A]" The supporting evidence: "Many imaging findings identified in people with low back pain are also common in people without such pain, and their importance in diagnosis is unknown. [1+] / Studies did not show imaging improved patient outcomes. [1+] / In patients with non-specific LBP, X-ray and magnetic resonance imaging findings do not correlate with clinical symptoms or work capacity. [2++]"

[Note 17] NICE NG59: "1.1.4 Do not routinely offer imaging in a non-specialist setting for people with low back pain with or without sciatica." "1.1.5 Explain to people with low back pain with or without sciatica that if they are being referred for specialist opinion, they may not need imaging." "1.1.6 Consider imaging in specialist settings of care (for example, a musculoskeletal interface clinic or hospital) for people with low back pain with or without sciatica only if the result is likely to change management."

[Note 18] Brinjikji W et al., AJNR Am J Neuroradiol 2015;36(4):811-6: "The prevalence of disk degeneration in asymptomatic individuals increased from 37% of 20-year-old individuals to 96% of 80-year-old individuals. Disk bulge prevalence increased from 30% of those 20 years of age to 84% of those 80 years of age. Disk protrusion prevalence increased from 29% of those 20 years of age to 43% of those 80 years of age. The prevalence of annular fissure increased from 19% of those 20 years of age to 29% of those 80 years of age." "Many imaging-based degenerative features are likely part of normal aging and unassociated with pain."

[Note 19] Hospital Authority fees and charges page (effective 1 January 2026):

Non-urgent radiology service — basic items: free / intermediate items: $250 per item / advanced items: $500 per item / From 1 January 2026, outside the medical fee waiver mechanism, the HA has established a second layer of safety net, adding an annual cap of ten thousand dollars on public healthcare service charges with no means test — the "annual charge cap". (our translation from the Chinese original)

Chinese original:

「非緊急放射科服務 — 基礎項目：免費／進階項目：每種服務250元／高端項目：每種服務500元」 「由2026年1月1日起，醫管局於醫療費用減免機制外建立第二層安全網，增設無需經濟審查的每年一萬元公營醫療服務費用上限–「全年收費上限」。」

The English version: "Non-urgent radiology service — Basic: Free · Intermediate: $250 per item · Advanced: $500 per item"

[Note 20] The Hong Kong Reference Framework two-page summary, treatment ladder: "Non-specific LBP in primary care — Reassurance and self-management" "Acute LBP — Non-pharmacological Treatment — First Line: Advice to remain active / Second Line: Exercise — Pharmacological Treatment — First Line: Simple analgesics (NSAIDs or paracetamol if no safe alternatives) / Second Line: Multiple analgesics of different classes" "Chronic LBP — Non-pharmacological Treatment — First Line: Advice to remain active; Exercise / Second Line: Physiotherapy; Nutrition intervention; Psychological intervention; Multidisciplinary rehabilitation — Pharmacological Treatment — First Line: Simple analgesics (NSAIDs or paracetamol if no safe alternatives) / Second Line: Multiple analgesics of different classes" The two exit condition lines: "Consider referring to specialist if there are red flag symptoms (Table) or no clinical improvement" "Review the progress at pre-determined intervals & reassess"

[Note 21] The Hong Kong Reference Framework, Chapter 5: "Reassure individuals with non-specific low back pain and advise them to remain active and self-manage their low back pain as first-line treatment. [A]" "Advice and education, reassurance and encouragement to avoid bed rest, stay active, and continue with usual activities, including work, are sufficient as initial management for patients with acute LBP. For patients with chronic LBP, advice on staying active alone is not sufficient. Advice on appropriate exercises and promotion of self-management is also required for patient with chronic LBP."

[Note 22] The same, section 5.3.2(b): "According to a systematic review on paracetamol for LBP, there was no difference between paracetamol and placebo for acute LBP patients at 1 week, 2 weeks, 4 weeks, and 12 weeks follow-ups… However, more than 90% of the participants analysed in the review were from one large trial. In addition, trial evaluating paracetamol for chronic LBP patients was not found. Thus, in selected patients for whom there are no safe alternatives, it is reasonable to consider a trial of paracetamol as initial therapy." (The ellipsis marks the values at each time point set out in that passage, which are not reproduced individually here.)

[Note 23] NHS, Back pain: "paracetamol on its own is not recommended for back pain but it may be used with another painkiller" and "In many areas you may be able to get help, such as physiotherapy, from NHS community musculoskeletal (MSK) services without needing a referral from a GP."

[Note 24] NICE NG59 recommendation 1.2.7: "Consider manual therapy (spinal manipulation, mobilisation or soft tissue techniques such as massage) for managing low back pain with or without sciatica, but only as part of a treatment package that includes exercise. [2016, amended 2026]"

[Note 25] NICE NG59 recommendation 1.2.25: "Consider weak opioids (with or without paracetamol) for managing acute low back pain only if an NSAID is contraindicated, not tolerated or has been ineffective. For guidance on safe prescribing of opioids and managing withdrawal, see NICE's guideline on medicines associated with dependence or withdrawal symptoms. [2016]" The "Terms used in this guideline" section: "Weak opioids — See the information on weak opioids in the analgesics section of the BNF."

[Note 26] NICE NG59: "1.2.18 If a person is already taking opioids, gabapentinoids or benzodiazepines for sciatica, explain the risks of continuing these medicines." "1.2.19 As part of shared decision making about whether to stop opioids, gabapentinoids or benzodiazepines for sciatica, discuss the problems associated with withdrawal with the person." "1.2.20 Be aware of the risk of harms and limited evidence of benefit from the use of non-steroidal anti-inflammatory drugs (NSAIDs) in sciatica." "1.2.21 If prescribing NSAIDs for sciatica: take into account potential differences in gastrointestinal, liver and cardio-renal toxicity, and the person's risk factors, including age / think about appropriate clinical assessment, ongoing monitoring of risk factors, and the use of gastroprotective treatment / use the lowest effective dose for the shortest possible period of time." "1.2.23 When prescribing oral NSAIDs for low back pain, think about appropriate clinical assessment, ongoing monitoring of risk factors, and the use of gastroprotective treatment."

[Note 27] The Hong Kong Reference Framework, Chapter 5, section 5.3.2(c): "The efficacy of opioid analgesics in acute LBP is unknown."

[Note 28] NICE NG59: 1.2.1 "Provide people with advice and information, tailored to their needs and capabilities, to help them self-manage… Include: information on the nature of low back pain and sciatica / encouragement to continue with normal activities" 1.2.2 "Consider a group exercise programme (biomechanical, aerobic, mind–body or a combination of approaches)… Take people's specific needs, preferences and capabilities into account when choosing the type of exercise" 1.2.15 "Promote and facilitate return to work or normal activities of daily living" 1.2.22 "Consider oral NSAIDs… taking into account potential differences in gastrointestinal, liver and cardio-renal toxicity, and the person's risk factors, including age" 1.2.24 "Prescribe oral NSAIDs for low back pain at the lowest effective dose for the shortest possible period of time" 1.3.5 "Consider epidural injections of local anaesthetic and steroid in people with acute and severe sciatica" 1.3.8 "Consider spinal decompression for people with sciatica when non-surgical treatment has not improved pain or function and their radiological findings are consistent with sciatic symptoms"

[Note 29] The Hong Kong Reference Framework, Chapter 5, section 5.1: "Advise individuals to maintain optimal level of physical activity for prevention of low back pain. [A]" The supporting evidence: "A 2016 systematic review concluded that exercise combined with education and exercise alone reduced the risk of an episode of LBP with relative risks of 0.55 (95% CI 0.41 to 0.74) and 0.65 (95% CI 0.50 to 0.86) respectively. Education alone with relative risk of 1.03 (95% CI 0.83 to 1.27), back belts with relative risk of 1.01 (95% CI 0.71 to 1.44), and shoe insoles with relative risk of 1.01 (95% CI 0.74 to 1.40) had no protective effect on LBP. [1+]"

[Note 30] Physiotherapists (Registration and Disciplinary Procedure) Regulation (Chapter 359J), section 6 (consolidated version dated 25/07/2025; amendment note "section 162 of Ordinance No. 33 of 2025"):

6. Restrictions on the practice of physiotherapists (1) A physiotherapist must not provide any service of the physiotherapist's profession to a person unless that person is referred by any of the following (each a qualified referrer) — (a) a registered medical practitioner; (b) a registered Chinese medicine practitioner. / (2) However, subsection (1) does not apply if — (a) the physiotherapist — (i) has obtained or been provided with a certificate (or other written document) that is — (A) issued by a qualified referrer not more than 12 months previously; and (B) states a diagnosis of the person's condition (the diagnosed condition); and (ii) provides services only for the diagnosed condition; (b) the person's condition is of a class recognised by any clinical guideline as one for which services of the physiotherapist's profession may be provided without a referral; (c) the person is registered in an interdisciplinary collaboration arrangement of the Primary Healthcare Commission, under which a physiotherapist may provide services of the physiotherapist's profession to the registered person without a referral; or (d) the circumstances of the case meet the circumstances specified by the code of practice as urgent or involving the provision of social services… / (3) A physiotherapist who provides any service of the physiotherapist's profession to a person in the circumstances described in subsection (2)(b) must comply with the requirements set out in the clinical guideline. / (4) A physiotherapist who provides any service of the physiotherapist's profession to a person in the circumstances described in subsection (2)(c) must comply with the requirements set out in the guideline published by the Primary Healthcare Commission for the interdisciplinary collaboration arrangement concerned. / (5) A physiotherapist whose name is in Part II must not practise except under the supervision of a physiotherapist whose name is in Part Ia. / (6) In this section — referencing authority means — (a) the Department of Health; (b) the Hospital Authority; (c) the Primary Healthcare Commission; or (d) the Chinese Medicine Hospital of Hong Kong; clinical guideline means a clinical guideline for the professional services engaging physiotherapists that is — (a) published by a referencing authority on its website, intranet or a similar electronic network; and (b) states that the clinical guideline is published for the purposes of subsection (2)(b). (our translation from the Chinese original)

Chinese original:

「6. 對物理治療師的執業限制 (1) 除非某人是由下列任何人士轉介(各人皆為合資格轉介人)，否則物理治療師不得向該人提供該物理治療師專業的任何服務 —— (a) 註冊醫生； (b) 註冊中醫。」「(2) 然而，如有以下情況，則第(1)款不適用 —— (a) 有關物理治療師 —— (i) 已取得或獲提供符合以下說明的證明書(或其他書面文件) —— (A) 由合資格轉介人在不超過12個月前發出；及 (B) 列明對有關的人的狀況的診斷(已診斷狀況)；及 (ii) 只為已診斷狀況提供服務； (b) 該人的狀況屬以下說明的種類：獲任何臨牀指引認可，可在沒有轉介的情況下，就該狀況提供物理治療師專業的服務； (c) 該人獲登記加入基層醫療署的跨專業協作安排，而於該安排下，物理治療師可在沒有轉介的情況下，向該登記的人提供物理治療師專業的服務；或 (d) 有關個案的情況符合執業守則視作緊急狀況或涉及提供社會服務的指明情況…」「(3) 物理治療師如向在第(2)(b)款所述的情況下的人，提供該物理治療師專業的任何服務，則須遵從臨牀指引所列明的規定。」「(4) 物理治療師如向在第(2)(c)款所述的情況下的人，提供該物理治療師專業的任何服務，則須遵從由基層醫療署為有關跨專業協作安排而發布的指引所列明的規定。」「(5) 名列第II部的物理治療師須在一位名列第Ia部的物理治療師的督導下執業，否則不得執業。」「(6) 在本條中 —— 參考機關 (referencing authority)指 —— (a) 衞生署； (b) 醫院管理局； (c) 基層醫療署；或 (d) 香港中醫醫院； 臨牀指引 (clinical guideline)指為聘用物理治療師的專業服務，而符合以下說明的臨牀指引 —— (a) 由參考機關發布在其網站、內聯網或相類似的電子網絡；及 (b) 述明該臨牀指引是就第(2)(b)款而發布的。」

[Note 31] Government press release on the arrangement for registered Chinese medicine practitioners to refer patients for allied health services (11 December 2025):

The Supplementary Medical Professions (Amendment) Ordinance 2025 came into effect on 25 July 2025, renaming the Supplementary Medical Professions Ordinance (Chapter 359 of the Laws of Hong Kong) as the Allied Health Professions Ordinance (the Ordinance)… / …the Chinese Medicine Practitioners Board… noted that the Supplementary Medical Professions Council has passed the codes of practice for radiographers, medical laboratory technologists, physiotherapists and occupational therapists, allowing them to provide allied health services to patients referred by registered Chinese medicine practitioners, with the relevant amendments also taking effect today (11 December). (our translation from the Chinese original)

Chinese original:

「《2025年輔助醫療業（修訂）條例》於二○二五年七月二十五日生效，該條例將《輔助醫療業條例》（香港法例第359章）重新命名為《專職醫療業條例》（《條例》）…」「…中醫組…知悉輔助醫療業管理局已通過放射技師、醫務化驗師、物理治療師及職業治療師的執業守則，容許他們可為註冊中醫師轉介的病人提供專職醫療服務，有關修訂同樣於今日（十二月十一日）生效。」

[Note 32] Primary Healthcare Commission, Clinical Guideline for First-contact Physiotherapists…, paragraph 3.2.9: "At present, the HKPRF for Common Musculoskeletal Problems in Primary Care Setting covers two highly prevalent conditions, namely, osteoarthritis of knees and low back pain, which together constituted over 50% of primary care consultations for musculoskeletal problems in Hong Kong." and: "Notably, under this specific circumstance within the direct access arrangement, in case if the patient's presenting condition is not consistent with osteoarthritis of knee, low back pain, or any other conditions stipulated by a clinical guideline published by a referencing authority, the first-contact physiotherapist must refrain from providing direct access physiotherapy service, even if the condition can be effectively and safely managed by physiotherapy."

[Note 33] Government press release (19 March 2025):

Physiotherapists or occupational therapists must at all times follow the recognised clinical guidelines, including referring a case in good time to a doctor for diagnosis and management on noticing that a patient has particular warning symptoms. (our translation from the Chinese original)

Chinese original:

「物理治療師或職業治療師必須時刻遵照獲認可的臨床指引，包括在察覺病人出現特定警示病情徵狀時及時將個案轉介醫生診斷和處理。」

The English version: "Patients may seek direct physiotherapy or occupational therapy services for health conditions diagnosed by a registered doctor or CMP within the past 12 months without obtaining a new referral letter each time. Patients must provide proof of the diagnosis. Apart from a referral letter, the proof can also be in the form of outpatient records, follow-up consultation records, or discharge summaries."

[Note 34] Physiotherapists Board, "First-contact Physiotherapist" page:

A first-contact physiotherapist–patient relationship is established when a patient, without a referral letter from a registered medical practitioner or registered Chinese medicine practitioner, receives assessment and subsequent treatment services from a physiotherapist under sections 6(2)(a), (b) and/or (c) of the Physiotherapists (Registration and Disciplinary Procedure) Regulation (Chapter 359J), having given informed consent. Whether or not the patient holds a referral letter, where the patient's needs go beyond the physiotherapist's professional scope, the physiotherapist should inform the patient and assist in finding a suitable and qualified person to provide the service needed. / Physiotherapists should be responsible for keeping proof of completing the training body's course and using that proof where necessary. For example, by submitting it directly to the Primary Healthcare Commission to update the Primary Care Directory, reflecting their qualification as a first-contact physiotherapist. (our translation from the Chinese original)

Chinese original:

「當病人在沒有註冊醫生或註冊中醫師的轉介信下，根據《物理治療師（註冊及紀律程序）規例》（第359J章）第6(2)(a)、(b)及／或(c)條，並已作出知情同意，接受物理治療師的評估及其後治療服務時，即建立了「首診物理治療師—病人關係」。無論病人是否持有轉介信，若病人的需要超出物理治療師的專業範疇，物理治療師均應通知病人，並協助尋找合適及具資格的人士提供所需服務。」「物理治療師應負責保留完成培訓機構課程的證明，並在必要時使用該證明。例如，直接提交給基層醫療署以作更新基層醫療指南，反映其作為首診物理治療師的資格。」

[Note 35] Primary Care Directory:

Under the current legislation and the codes of practice of occupational therapists and physiotherapists, patients generally have to obtain a doctor's referral before receiving the services of occupational therapists and physiotherapists. (our translation from the Chinese original)

Chinese original:

「根據現行法例及職業治療師和物理治療師的專業守則，一般而言，患者須先獲得醫生轉介方可接受職業治療師和物理治療師的服務。」

(That page's footer reads 「最後更新日期: 二零二二年七月一日」.)

[Note 36] Hospital Authority, annual charge cap page for eligible persons:

Must be an eligible person* / The patient has paid a cumulative total of ten thousand Hong Kong dollars in "eligible medical fees and charges" within the year / At the time of applying for the "annual charge cap", the patient has no outstanding "eligible medical fees and charges" anywhere in the HA / The hospital services the patient received must not have been determined by the HA as having "no clinical need" / The annual application period begins on 1 January each year and closes on 31 March of the following year. Late applications will not be accepted, and eligible medical fees cover only medical bills issued between 1 January and 31 December of each year and paid in full at the time of application. / A fresh application for the "annual charge cap" is required for each year / Self-financed drugs and medical devices are excluded. (our translation from the Chinese original)

Chinese original:

「必須為符合資格人士*」「病人於年度內累計繳付的「合資格醫療費用及收費」達到一萬港元」「在申請「全年收費上限」時，在醫管局轄下無任何拖欠的「合資格醫療費用及收費」」「病人接受之醫院服務不得被醫管局裁定為「無臨床需要」」「年度接受申請期由每年的1月1日開始，至次年的3月31日截止。遲交的申請將不獲接納，合資格醫療費用只包括在每年1月1日至12月31日所發出的醫療賬單，並在提交申請時已全數繳付。」「每一個年度需要重新申請「全年收費上限」」「自費藥物及醫療器械除外。」

The English version: "Patients who met the eligibility criteria may submit their applications via HA Go or at any hospitals' shroff offices once their cumulative valid annual spending reached $10,000."

[Note 37] The Hong Kong Reference Framework, Chapter 5, section 5.3.3: "Selected groups of patients with LBP should undergo evaluation for surgical evaluation: - patients with suspected cauda equina lesions and worsening neurologic deficits require immediate surgical investigation; and - patients intractable pain that is resistant to conservative treatment." ("patients intractable pain" is the document's own wording, reproduced as found.)

[Note 38] Notes on the Department of Health / Office for Regulation of Private Healthcare Facilities platform for the Pilot Programme on Enhancing Price Transparency of Private Hospitals:

The information provided by private hospitals 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. / The percentile data above are ordered by the amount of the "total charge", not by the "doctor's fee" or the "hospital charge" within the "total charge". / The doctor's fee includes the fees of the anaesthetist, the surgeon and ward rounds. / Readers should interpret with caution billing data reported by any hospital for no more than 30 procedures, because such data are derived from a small number of cases and may not reflect the general level of charges. (our translation from the Chinese original)

Chinese original:

「私家醫院提供的資料是根據病人住宿於標準病房及進行單一手術或程序的數據，並不包括進行多次手術或程序的病人的資料。」「以上百分位數據是按「總收費」金額順序排列，而非按「總收費」內的「醫生費」及「醫院費」排列。」「醫生費包括麻醉科醫生，外科醫生和巡房的費用。」「讀者須謹慎闡釋由每間醫院彙報不超過 30 宗程序的帳單數據，因此類數據只基於少數個案而得出，未必能反映一般的收費情況。」

#What this article does not state

• There is no current Hong Kong lifetime prevalence figure. The only Hong Kong lifetime prevalence survey found for this article is Lau et al. 1995 (39%, 95% confidence interval 34%–44%; a sample of 652 people in two housing estates), 31 years ago; the Population Health Survey 2020-22 measures doctor-diagnosed chronic low back pain (1.8%), a different construct, and the two are not interchangeable. This article therefore gives no figure for how many people in Hong Kong will have back pain in their lifetime.
• The threshold of "a first episode under 20 or over 50" appears in none of the sources cited here. The Hong Kong Reference Framework's table writes "Older age" with no number, and the body text of the same document writes ">65". This article does not print that threshold.
• This article gives no figure for what percentage of acute back pain becomes chronic, nor for what percentage of chronic back pain patients eventually need surgery, nor for what percentage of disc herniations improve with conservative treatment — none of the three has corresponding data in the sources cited here.
• How long the effects of epidural injection and radiofrequency ablation last — no statement is made. NICE NG59 has recommendations on referral for assessment (1.3.2, 1.3.3, 1.3.5) but publishes no duration of effect in months; and no Hong Kong data on this was obtained.
• This article gives no per-session private physiotherapy fee, no private MRI charge and no injection treatment charge. None of the sources cited here carries a first-hand private outpatient price list; the published private spine surgery bills cover only laminectomy and spinal fusion.
• This article gives no public physiotherapy waiting time. What the Hospital Authority publishes is specialist outpatient waiting times for new cases; no Hospital Authority series specific to physiotherapy was found.
• No Chinese version was found of the Hong Kong Reference Framework (core document, two-page summary or Chapter 5) or of the Primary Healthcare Commission's Clinical Guideline for First-contact Physiotherapists on Assessment and Cross-disciplinary Management Approach for Common Musculoskeletal Problems in Primary Care Setting (the Reference Framework's traditional Chinese PDF path returns 404; all 55 pages of that Primary Healthcare Commission guideline contain no Chinese character and it has no official Chinese title, so this article does not devise one for it). Where those documents are quoted here, what is inside quotation marks is the English original word for word, with any rendering placed outside the quotation marks.
• The Chinese version of the WHO fact sheet exists only in simplified characters; this article reproduces the simplified original and does not convert it and then treat the result as a quotation.
• This article relies only on the English report of the Population Health Survey 2020-22; whether a Chinese version exists is outside what this article relies on, so that report's figures are restated rather than quoted. The Hong Kong Physiotherapy Association's Legislative Council submission of March 2026 is likewise available to this article only in English, and is likewise restated.
• What NG59 recommendations 1.2.13 and 1.2.14 originally said — no statement is made. The current page prints only "This recommendation has been deleted."; the subheadings above the two (Psychological therapy; Combined physical and psychological programmes) remain, but the recommendations themselves are gone, and NICE's update note page was not obtained.
• For eight of the conditions in table 3.1 of that Primary Healthcare Commission guideline, this article reproduces only the name and not the full symptom column (vital signs, mental state, asthma, suicide risk, open wound, cellulitis, depressive symptoms, and abnormal vital signs in Priority category 2). Those eight bear less on low back pain, but they are on the same table; their full content is in the original table 3.1. No condition name has been dropped, so that no reader takes the table to have only nine entries.
• The red flag provisions of the United Kingdom's NICE Clinical Knowledge Summaries could not be consulted (that site restricts access to within the United Kingdom). The British public-facing list cited here comes from the National Health Service's web pages, not from the clinical wording of the Clinical Knowledge Summaries.
• The WHO's 2023 guideline on the management of chronic primary low back pain was not obtained. The WHO fact sheet cited here (page dated 19 June 2023) still describes that guideline as "under development", and this article does not treat that sentence as a statement of the current position.
• The American College of Physicians' guideline of 2017 was not obtained and is therefore not cited.
• The Hong Kong Reference Framework's sentence on a lifetime incidence of 58–84% carries that document's own reference number, which this article has not traced to its ultimate source; the Department of Health household survey quoted in the same passage (30-day prevalence 21.5%) is likewise available to this article only as the Reference Framework's restatement, the original survey report not having been obtained independently.
• The revision date of the Physiotherapists Board's "First-contact Physiotherapist" page could not be obtained (that field is blank in the raw HTML); nor is any public register of qualified first-contact physiotherapists visible on that page — but the page points to the Primary Care Directory, which has a filter for first-contact physiotherapy service (see above). This article will not treat the Directory's 58 search results as the number of qualified first-contact physiotherapists in Hong Kong: entries are submitted by the service providers, one person may have several practice address entries, and no de-duplication was done here.
• The Hospital Authority's fee table does not itemise which radiological examinations count as basic, intermediate or advanced, and has no separate line for physiotherapy; this article therefore makes no statement about which fee tier a particular scan falls into.
• The update date of the Hospital Authority Smart Patient page (29 July 2024) exists only in an HTML comment and is invisible to the reader; that date is about 24 months ago. The Hong Kong Reference Framework document is self-dated 2022, about 4 years ago.

#Sources

• Hospital Authority Smart Patient site, "Low back pain" (the three 八成 / 80% figures on the same page: at least once in a lifetime, recovery within a month, and the age distribution of slipped disc patients; the distinction between low back pain and sciatica and the Chinese–English difference in it; the content of physiotherapy and the two sentences that follow; the four Chinese warning sentences, spread across the treatment and complications sections): https://www.smartpatient.ha.org.hk/zh-hk/smart-patient-web/disease-management/disease-information/disease/LowBackPain and English version https://www.smartpatient.ha.org.hk/en/smart-patient-web/disease-management/disease-information/disease/LowBackPain (page updated 29 July 2024, visible only in an HTML comment; retrieved 3 August 2026)
• Hong Kong Reference Framework for Common Musculoskeletal Problems in Primary Care Settings — core document, section 1.2 (lifetime incidence 58–84%; 30-day prevalence 21.5% and the age gradient): https://www.healthbureau.gov.hk/phcc/rfs/src/pdfviewer/web/pdf/musculoskeletal/en/02_CoreDocument/13_en_musculoskeletal.pdf (document self-dated 2022; retrieved 2 August 2026)
• The same — two-page summary Assessment of Low Back Pain (LBP) (the three categories and their primary care proportions; the red flag warning sign table; non-spinal causes; the treatment ladder and its column headings "Non-specific LBP in primary care — Reassurance and self-management" and the two exit condition lines below it): https://www.healthbureau.gov.hk/phcc/rfs/src/pdfviewer/web/pdf/musculoskeletal/en/10_en_Twopagesummary-LBP.pdf (footer dated 2022; retrieved 3 August 2026)
• The same — core document Chapter 5 (section 5.1's prevention recommendation and the relative risks from the 2016 systematic review; section 5.2's red flag body text and the imaging recommendation with its supporting evidence; section 5.3.1(a)'s first-line treatment recommendation and the qualification for chronic low back pain; section 5.3.2(b)'s reasoning on paracetamol; section 5.3.2(c) on opioids; section 5.3.3 on who should be evaluated for surgery): https://www.healthbureau.gov.hk/phcc/rfs/src/pdfviewer/web/pdf/musculoskeletal/en/02_CoreDocument/08_en_musculoskeletal_chapter5.pdf (document self-dated 2022; retrieved 3 August 2026)
• Primary Healthcare Commission, Clinical Guideline for First-contact Physiotherapists on Assessment and Cross-disciplinary Management Approach for Common Musculoskeletal Problems in Primary Care Setting (the complete seventeen red flag conditions across the three tiers of table 3.1; paragraphs 3.1.3, 3.1.4, 3.1.5, 3.1.6, 3.1.18, 3.1.19, 3.1.20, 3.1.21, 3.2.9, 3.2.10 and 3.2.11; figure 3.1): https://www.healthbureau.gov.hk/phcc/rfs/assets/pdf/home/discipline-based_guidelines/clinical_guideline_on_first_contact_assessment_and_cross-disciplinary_management_approach_for_common_msk_problems.pdf (document self-dated First published: Dec 2025; all 55 pages contain no Chinese character and it has no official Chinese title; retrieved 3 August 2026)
• Physiotherapists (Registration and Disciplinary Procedure) Regulation (Chapter 359J), the whole of section 6 (subsections (1) to (6), including the duty under subsection (3) to comply with the clinical guideline, and the definitions of "clinical guideline" and "referencing authority" in subsection (6)), in the bilingual text on the Physiotherapists Board's website: https://www.ahp-council.org.hk/pt/file/pdf/Cap.359J-s.6.pdf (consolidated version dated 25/07/2025; amendment note: section 162 of Ordinance No. 33 of 2025; retrieved 3 August 2026)
• Physiotherapists Board, "First-contact Physiotherapist" (the training requirement; course hours and assessment; the enrolment requirements set out separately by register part, including the 2 000 hours of clinical experience and statutory declaration for Part Ib and the note on free oath-taking services; the definition of the first-contact physiotherapist–patient relationship and the duty to inform and assist "whether or not the patient holds a referral letter"; keeping proof of training and submitting it to the Primary Healthcare Commission to update the Primary Care Directory): https://www.ahp-council.org.hk/pt/tc/content.php?page=fcp and https://www.ahp-council.org.hk/pt/en/content.php?page=fcp (the page's "last updated" field is blank and could not be obtained; retrieved 3 August 2026)
• Physiotherapists Board, "Registration Qualifications" (the distinction between Part Ia and Part Ib turning on whether the applicant has not less than one year of post-qualification "recognized experience" at registration): https://www.ahp-council.org.hk/pt/en/content.php?page=reg_quareg (the page's "Last Update" field is blank and could not be obtained; retrieved 3 August 2026)
• Primary Healthcare Commission, Primary Care Directory physiotherapist search (which has a filter for first-contact physiotherapy service; a search of all eighteen districts of Hong Kong on 3 August 2026 returned 58 entries; the same site still prints the referral statement last updated 1 July 2022): https://apps.pcdirectory.gov.hk/Public/tc/ (footer 「最後更新日期: 二零二二年七月一日」; retrieved 3 August 2026)
• Government press release on the arrangement for registered Chinese medicine practitioners to refer patients for allied health services (the Ordinance taking effect on 25 July 2025; the code of practice amendments taking effect on 11 December 2025): https://www.info.gov.hk/gia/general/202512/11/P2025121100448.htm (11 December 2025; retrieved 2 August 2026)
• Government press releases relating to the Supplementary Medical Professions (Amendment) Bill (the duty to refer on warning symptoms; the 12-month proof of diagnosis being able to take the form of outpatient records, follow-up consultation records or discharge summaries): https://www.info.gov.hk/gia/general/202503/19/P2025031900427.htm and English version https://www.info.gov.hk/gia/general/202503/19/P2025031900434.htm (19 March 2025; retrieved 2 August 2026)
• Hong Kong Physiotherapy Association, Operational Concerns on the "Clinical Guideline for First-Contact Physiotherapists…", LC Paper No. CB(3)214/2026(01) (the profession's concern that the guideline covers only two conditions): https://www.legco.gov.hk/yr2026/english/panels/hs/papers/hs20260320cb3-214-1-e.pdf (19 March 2026; retrieved 2 August 2026)
• NICE guideline NG59, Low back pain and sciatica in over 16s: assessment and management, Recommendations (all current provisions 1.1.1, 1.1.4–1.1.6 and 1.2.1–1.3.10, including 1.2.7, 1.2.18, 1.2.19, 1.2.20, 1.2.21, 1.2.23, 1.2.25, 1.3.5 and 1.3.8; the three definitions of acute, chronic and "Weak opioids" in "Terms used in this guideline"; 1.2.13 and 1.2.14 having been deleted with their subheadings remaining): https://www.nice.org.uk/guidance/ng59/chapter/Recommendations (the page states Published 30 November 2016 · Last updated 29 July 2026; retrieved 3 August 2026)
• WHO, Low back pain fact sheet (at least once in a lifetime; 619 million and 843 million; about 90% non-specific; the definitions of acute, subacute and chronic; painkillers not to be first line): https://www.who.int/news-room/fact-sheets/detail/low-back-pain and the simplified Chinese version https://www.who.int/zh/news-room/fact-sheets/detail/low-back-pain (page dateModified 19 June 2023; retrieved 2 August 2026)
• NHS, Back pain (the three tiers of 999 or emergency department, urgent, and general practitioner; the "Do not drive to A&E" information box below the 999 list; the public-facing wording on paracetamol not being used alone; self-referral to community musculoskeletal services): https://www.nhs.uk/conditions/back-pain/ (the page states Page last reviewed 05 March 2026; retrieved 3 August 2026)
• GBD 2021 Low Back Pain Collaborators. Global, regional, and national burden of low back pain, 1990-2020… Lancet Rheumatol 2023;5(6):e316-e329 (age-standardised prevalence falling by 10.4%; the definition of low back pain) (DOI 10.1016/S2665-9913(23)00098-X; PMID 37273833)
• Walker BF. The prevalence of low back pain: a systematic review of the literature from 1966 to 1998. J Spinal Disord 2000;13(3):205-17 (lifetime prevalence 11%–84%) (PMID 10872758)
• Lau EM, et al. Low back pain in Hong Kong: prevalence and characteristics compared with Britain. J Epidemiol Community Health 1995;49(5):492-4 (Hong Kong lifetime prevalence 39%) (PMID 7499992)
• Brinjikji W, et al. Systematic literature review of imaging features of spinal degeneration in asymptomatic populations. AJNR Am J Neuroradiol 2015;36(4):811-6 (age-specific imaging findings in 3,110 asymptomatic people) (PMID 25430861; PMCID PMC4464797)
• Finucane LM, et al. International Framework for Red Flags for Potential Serious Spinal Pathologies. J Orthop Sports Phys Ther 2020;50(7):350-372 (an absence of high-quality evidence for the diagnostic accuracy of most red flags) (PMID 32438853)
• Centre for Health Protection / Department of Health, Report of Population Health Survey 2020-22 (Part I) §3.1.8 (doctor-diagnosed chronic low back pain 1.8% among those aged 15 and above; musculoskeletal disease 8.4%; the distribution of treatment modalities): https://www.chp.gov.hk/files/pdf/dh_phs_2020-22_part_1_report_eng_rectified.pdf (retrieved 2 August 2026)
• Hospital Authority, "Fees and Charges" (charges for eligible persons effective 1 January 2026; the annual charge cap of ten thousand dollars): https://www.ha.org.hk/visitor/ha_view_content.asp?Content_ID=10045&Lang=CHIB5&Dimension=100&Ver=HTML and the English version at the same address with Lang=ENG (the page states an effective date of 1 January 2026; retrieved 1 August 2026)
• Hospital Authority, "Annual Charge Cap" (for eligible persons) (the four eligibility criteria word for word; the annual application period and the refusal of late applications word for word; the need to reapply each year word for word; the exclusion of self-financed drugs and medical devices word for word; and the English version's sentence on applying through HA Go or a hospital shroff office): https://www.ha.org.hk/visitor/ha_view_content.asp?Content_ID=281820&Lang=CHIB5&Dimension=100&Ver=HTML and the English version at the same address with Lang=ENG (effective 1 January 2026; retrieved 1 August 2026)
• Hospital Authority, Waiting Time for Stable New Case Booking at Specialist Out-patient Clinics (medians and 90th percentiles for urgent, semi-urgent and stable new orthopaedic cases by cluster; the number of new case bookings and the triage proportions): https://www.ha.org.hk/haho/ho/sopc/dw_wait_ls.pdf (period 1 July 2025 to 30 June 2026; next update date 30 October 2026; retrieved 1 August 2026)
• Department of Health / Office for Regulation of Private Healthcare Facilities, Pilot Programme on Enhancing Price Transparency of Private Hospitals (the percentiles of bills for laminectomy and spinal fusion as inpatient surgery; the platform's own notes): https://apps.orphf.gov.hk/Public/tc/ and https://apps.orphf.gov.hk/Public/en/ (reporting period 1 January 2025 to 31 December 2025; retrieved 1 August 2026)

This article was written from the sources listed above; information date 2 August 2026. It is general information and does not constitute medical advice. Individual diagnosis, imaging, medication and surgical decisions are for a doctor.

---

#Further reading

• This site: Reading laboratory and imaging reports
• This site: Public waiting too long? Private and public-private partnership alternatives
• This site: Training injuries: rhabdomyolysis and overtraining
• Hospital Authority orthopaedic specialist services: ha.org.hk Orthopaedics
• NICE guideline — low back pain and sciatica in over 16s: nice.org.uk NG59
• This site: Tianjiu moxibustion