75 Per Cent Is the Number UK Regulators Quietly Share

Every English requirement in this series counts something abstract: a band, a sub-score, a grade, a number of years.

Two UK regulators count something else, in almost identical words. Three quarters of your conversations with patients.

SpeakShark is free for exactly the kind of conversation that threshold is about, three AI conversation sessions a day with no card.

In this guide: the number · the same sentence · who counts · three uses · why 75 · what it is not · using it · limits · method · FAQ

Key takeaways

75 per cent appears at the GDC and the GPhC, in nearly the same words.
Both name the same people: patients, carers, their families and other healthcare professionals.
The GDC puts it on a referee's signature; the GPhC on a candidate's evidence.
It is the only quantified behavioural threshold we have found in this series.
Neither regulator publishes why three quarters.

The same number in two places

The General Dental Council's structured English language reference form states that by completing it, the referee is confirming that at least 75 per cent of the applicant's day to day interaction with patients, carers, their families and other healthcare professionals in their employment or under their supervision has been in English.

The General Pharmaceutical Council's guidance, in the provision that lets it accept a test score more than two years old, requires that you have since worked as a pharmacy professional in a mainly English-speaking country and that at least 75 per cent of your day-to-day interaction with patients, carers, their families, and other healthcare professionals has been in English.

Two regulators. Two documents. One number, and very nearly one sentence.

Almost the same sentence

Set the two side by side and the overlap is not thematic, it is lexical.

	GDC	GPhC
Threshold	at least 75 per cent	at least 75 per cent
Of what	day to day interaction	day-to-day interaction
With whom	patients, carers, their families and other healthcare professionals	patients, carers, their families, and other healthcare professionals
Condition	has been in English	has been in English
Who asserts it	the referee	the applicant, with evidence

The interlocutor list is identical, in the same order, down to the same four categories.

Two bodies arriving independently at the same percentage would be a coincidence. Two bodies arriving at the same percentage attached to the same four-item list in the same order is drafting with a shared ancestor.

We have not found that ancestor and we are not going to name one. What we can say is that this is not two regulators improvising, and that a candidate meeting it at one is very likely describing the same working life the other has in mind.

Who the threshold names

Read the four categories carefully, because what is missing matters as much as what is there.

Patients. The core of it.

Carers and their families. Which is a harder conversation than the patient's, and often the one where explanation, reassurance and bad news actually happen.

Other healthcare professionals. Handover, referral, escalation.

What is not named: colleagues in general, administrative staff, or documentation. The GPhC's criteria mention reading and writing elsewhere, but the 75 per cent threshold itself is about spoken clinical interaction.

That is a narrower and more demanding claim than working in an English-speaking country. Somebody can spend years in an English-speaking workplace with most of their patient conversations in another language, and this threshold is written to catch precisely that case.

One number doing three jobs

The same figure turns up doing three different jobs across the two regulators.

Use	Body	What it does
Reference form	GDC	lets a signed reference stand in place of a test
Maintained proficiency	GPhC	lets a test score over two years old still count
In-service training	GPhC	qualifies a degree route, applied to clinical placement

The third is the least obvious. The GPhC's qualification route requires the entire course to have been taught and examined in English and at least 75 per cent of any in-service training, including clinical interaction and contact with patients, carers and other healthcare professionals, to have been conducted in English.

That closes an obvious gap. A degree can be taught entirely in English in a country where the hospital placements attached to it are conducted in the local language, and without the second clause such a degree would clear a requirement it has not actually met.

The GDC's use is the most consequential for an applicant, because it is the one that replaces the exam outright. We set out the whole form, including the requirement for an example of all four skills, in the dental reference route.

Why three quarters

Neither regulator publishes a reason, so what follows is our reading rather than anybody's policy.

A threshold of 100 per cent would be unusable. Real clinical workplaces are not monolingual. A nurse switching to a patient's first language for two minutes to explain a medication is doing good medicine, and a rule that punished it would fail exactly the clinicians it should reward.

A threshold of most or the majority would be unenforceable. A referee cannot sign a number that is not there, and a regulator cannot review a claim that means whatever the writer wanted.

Three quarters sits in the useful place. It is high enough that English is unambiguously the working language rather than one of two, and low enough to survive a genuinely bilingual setting. It is also a fraction people can estimate honestly, which matters when the person signing is a busy supervisor rather than an assessor.

Compare that with how the same regulators handle tests. There the numbers are precise because an instrument produced them: 7.0 in each area at the GPhC, 6.5 in each at the GDC, 8.0 for speech and language therapists at the HCPC. A behavioural threshold cannot be precise in that way, and 75 per cent is what a usable approximation looks like.

What the threshold does not cover

Three limits are worth being explicit about, because the route is narrower than it first appears.

It needs a witness. At the GDC the assertion is made by an employer or tutor who gives their name, occupation, place of employment, regulator registration number and the dates they supervised you. Nobody can self-certify it.

It needs relevant practice. Interaction with patients, carers and other professionals is not something available to somebody outside clinical work.

It does not stand alone. At the GPhC it is one half of a condition, the other half being work in a mainly English-speaking country, and those countries are defined by a list that differs between regulators, covered in three regulators, three lists.

Who can actually use this

Somebody in supervised clinical practice where the conversations are in English.
Somebody whose certificate has lapsed while continuing to work, which pairs with the reasoning in why English certificates last two years.
Somebody whose degree included clinical placement in English, for the GPhC qualification route.

It is not available to somebody applying from a non-clinical background, from a workplace where the patient-facing language is not English, or without a supervisor willing to be named and contacted.

For everybody else the route is the test, and the part that decides most floors is speaking, because it is the skill with no obvious way to practise alone. Role play scenarios and a daily speaking partner are the practical routes, testing your speaking level free shows the gap, and CEFR levels for speaking explains the scale. IELTS for UKVI by CEFR level maps bands to levels, the GMC requirement for doctors covers medicine, and English certificates that expire covers validity. For how examiners mark spoken English, Cambridge speaking criteria by level works from real scored candidates.

What we could not verify

We did not find a common source for the wording. Two regulators using nearly identical language suggests a shared origin, and we have not identified one. Treat the shared ancestry as an observation about the text, not a documented fact.

Our explanation of why 75 per cent is inference. Neither body publishes a rationale and we have marked our reading as ours.

We did not check other regulators for the same threshold. It may appear elsewhere. Absence from the documents we read is not absence generally.

We do not know how the threshold is assessed. The referee or the applicant asserts it. What verification follows is not published.

We did not verify how often either route succeeds. Regulators publish processes, not outcomes.

This is not registration advice. It is a reading of published guidance and a published form, all linked below.

How we researched this guide

We were not looking for this. We read the GDC's reference form for a different post and noted 75 per cent as an unusual thing for a regulator to quantify, since everything else in the series counts bands and years.

It became a finding a document later. Reading the GPhC guidance for its score table, the same number appeared attached to the same list of four interlocutors, and at that point it stopped being a quirk of one form.

The check that made it solid was comparing the phrasing rather than the number. A shared percentage could be coincidence; a shared percentage plus patients, carers, their families and other healthcare professionals in the same order is not. That comparison is why the table above sets the two sentences against each other clause by clause instead of just noting that both say 75.

The third appearance, inside the GPhC's qualification route, came from reading past the part we needed. It is applied to in-service training rather than to employment, which is a different situation with the same test, and it closes a gap that the taught and examined in English condition leaves open on its own.

We stopped short of naming a source for the shared wording. Two documents agreeing is evidence that a third exists; it is not evidence about what the third one is.

Practise speaking, from SpeakShark

SpeakShark is an AI English speaking practice app, and this threshold describes the skill we exist for more precisely than any band score does.

It is not asking whether you can pass an exam. It is asking whether the conversations you have with patients, their families and your colleagues happen in English, most of the time, without falling back.

You talk, the AI answers what you actually said, and you get speaking feedback while the conversation is still running. The free tier gives basic feedback; the detailed pronunciation and grammar breakdown is on Premium.

Being straight about the limits: a free session runs five minutes with four turns, we issue no certificates, and practising here is not evidence to any regulator. Nobody can sign a reference on the strength of it.

Start free with three sessions a day and no card. Paid sessions run ten minutes with unlimited turns. Limits are on the pricing page, and how it works walks through a session. Sign up here if the work-based route is closed to you and the test is what is left.

We are a speaking improvement tool. We are not an exam preparation provider, and we are not affiliated with Cambridge English, the IELTS partners, Pearson or any exam board. For test format, booking and official practice material, go to the exam body directly. Use SpeakShark to make your spoken English stronger, and use official material to learn the test.

Site: speakshark.com
Support: speaksharksupport@gmail.com
Free tier: 3 AI conversation sessions per day, no card

Sources

GDC, structured reference to assess knowledge of English: the referee's confirmation that at least 75 per cent of the applicant's day to day interaction with patients, carers, their families and other healthcare professionals in their employment or under their supervision has been in English, and the requirement for examples of all four skills
GPhC, guidance on evidence of English language skills, June 2025: the maintained proficiency provision requiring work as a pharmacy professional in a mainly English-speaking country with at least 75 per cent of day-to-day interaction with patients, carers, their families and other healthcare professionals in English, and the qualification route requiring the entire course to be taught and examined in English with at least 75 per cent of in-service training conducted in English
GDC, guidance on English language controls: the provision for a structured reference completed by an employer or tutor, and the criteria that evidence be recent, objective, independent, robust and readily verifiable
HCPC, certificate of English language proficiency: the test-only design used for contrast, with no work-based route of this kind

FAQ

What is the 75 per cent English rule?

A threshold used by UK healthcare regulators for evidence based on work rather than on a test. It asks whether at least 75 per cent of your day-to-day interaction with patients, carers, their families and other healthcare professionals has been in English. It appears at the General Dental Council and at the General Pharmaceutical Council.

Where does the GDC use it?

On its structured English language reference form. By completing the form, the referee confirms that at least 75 per cent of the applicant's day to day interaction with patients, carers, their families and other healthcare professionals in their employment or under their supervision has been in English. The referee asserts it, not the applicant.

Where does the GPhC use it?

Twice. It appears in the provision allowing a test score more than two years old, where the candidate must have worked in a mainly English-speaking country with at least 75 per cent of day-to-day interaction in English. It appears again in the qualification route, applied to at least 75 per cent of any in-service training.

Why 75 per cent rather than all of it?

Neither regulator publishes a reason. Our reading is that a realistic clinical workplace is never entirely monolingual, so a threshold of everything would fail honest applicants and a threshold of most would be unenforceable. Three quarters is high enough to mean the working language and low enough to survive a bilingual ward.

Who counts as the people you interact with?

Both regulators name the same group: patients, carers, their families and other healthcare professionals. Colleagues in general are not named, and neither is paperwork. The threshold is about conversation with the people the profession exists to serve, which is a narrower thing than working in an English-speaking office.

Can I use this instead of a test?

Only if your regulator offers a work-based route and you have somebody able to attest to it. At the GDC the threshold sits on a reference form completed by an employer or tutor. At the GPhC it sits inside provisions for older certificates and for qualifications. Neither is available to somebody with no relevant supervised practice.