This site is intended for health professionals only

At the heart of general practice since 1960

Read the latest issue online

CAMHS won't see you now

Should we put more GPs in A&E?

Dr Steven Laitner and Dr Craig Seymour debate

Dr Steven Laitner

Dr Steven Laitner


Simon Stevens has clearly stated that every hospital needs to have a comprehensive from door streaming service by next Christmas and that this will require a GP in every A&E in the country

He has acknowledged that this is only part of the solution, but he is correct that it is indeed an important part of the solution.

An A&E consultant I trained with once said to me: ‘There is no such thing as an inappropriate attendance at A&E, just an inappropriate service for that person’. And in relation to A&E we do rather tend to blame the patient, which is terrible. So let’s think about the person who attends A&E. They will be concerned, frightened, anxious and have a sense of urgency to have their concern dealt with as quickly as possible. They may not be registered with a GP, they may not have trust in their GP, they may have tried to get an appointment with their and be told ‘call back tomorrow or go to A&E’ (in my view a breach of our contract with the government and with society). So whilst it may not make sense to us that the person has attended A&E it makes perfect sense to the individual at that time.

So rather than blame them, tell them off, send them away, why don’t we just listen to their concern and manage their need, there and then, with the right professional with the right skills and experience?

And that’s where we, GPs, are fabulous!

That’s our sweet spot, listening to patients and their carers and managing ‘undifferentiated demand’, quickly, holistically and often without the need for expensive tests or specialist opinion. We get somatisation, we get mental health, we get families and communities, we get frailty, long term conditions, social isolation, we get patients' Ideas, Concerns and Expectations.

So let’s celebrate the fact that the chief executive of the NHS gets that too and understands that GP have so much to offer, not only in general practice, commissioning, 111 and urgent care centres but also, yes in A&E!

And it works. For example the Luton & Dunstable (L&D) model is based upon several years of collaboration between providers and commissioners. The end result of which is that now, during the hours of opening, the urgent GP service sees around 40% of patients who walk into the department. Many other A&Es in the country have had similar success with ‘GP front ends’ at A&E.

A GP friend of mine, attending A&E with his child with a head injury, related his frustration at a junior doctor behind the curtain next door taking so long to assess a patient with a simple sore throat, he wanted to go through and assess the patient himself in a few minutes, no hours!

When I was a 'fly on the wall' at a paediatric A&E recently I was amazed at the number of well looking children there who seemed to have fairly minor upper respiratory tract infections that I would routinely manage in my GP surgery. I overheard at least one mum say that she was there because her own GP practice receptionist told her they had no appointments left that day and she either had to call back tomorrow or go to A&E – a real failure of primary care delivery in my opinion. Yes, we must ensure GP practices deliver superb 'same day access' in their own practices, and that needs support too.

But let’s not deny people the benefit of an experienced GP in A&E, often it is all they need and let’s welcome the recognition and the funds that must follow into primary care (in all its wonderful forms).

Dr Steven Laitner is a GP in St Albans, Hertfordshire

Dr Craig Seymour

Dr Craig Seymour


There are already too many urgent care options for our patients leading to great confusion amongst the public. It is tempting to think GPs working in A&E will simplify and solve this issue; however, what is the problem we are trying to solve? Is it improving care, or improving A&E target performance? GPs working with A&E departments is not a new idea and the different approaches (walk in centres, urgent care centres, triage, streaming, colocation) have not been shown to lead to a meaningful or cost-effective improvement in patient outcomes, according to a Cochrane review. They may contribute to encouraging patients to pitch up at A&E, as attendances in 2016 were at their highest level since records began. 

In some areas GPs have been able to reduce ED attendance, but in the context of whole system change and led by primary care. GPs can’t just be parachuted in.

What can reduce A&E attendances and admissions to hospital is longitudinal, integrated, holistic continuity of care that can only be offered by general practice. This is particularly true for those with long term conditions or complex needs. This cannot be offered by a model of GPs offering ad hoc care at the front door of A&E.

How about an innovative solution that offers this care and makes the most of resources? The Islington iHUB, a GP Access Fund Pilot, has developed a model of redirecting patients triaged by A&E with primary care appropriate conditions, and registered with an Islington GP practice, to appointments with GPs working in a local practice. These GPs have access to that patient’s full clinical record and local community services, same as their home practice. For patients this makes these encounters valuable, as their problems can be put into context of their existing care and their comorbidities. Between November 2016 and March 2017, 50% of those eligible for the service with were seen in the hub, resulting in reduction of A&E traffic. With more fine tuning, it could be 100%. Patients were satisfied. It was a limited pilot, but it worked.

This kind of scheme keeps GPs working in general practice where they belong, rather than diluting an increasingly rare resource. The reality is GP numbers have, in real terms, gone down

There is an opportunity for patient behaviour change, encouraging accessing extended hours and in-hours primary care, and breaking the cycle of attending A&E by default. The act of creating a hub close to, but crucially not actually in, A&E would seem important in turning the tide on an unsustainable situation in A&E. It falls to commissioners and the government to recognise the value of such schemes, rather than investing scarce resource where the evidence doesn’t support it.

I would argue that limited governmental resource needs to be directed at boosting a struggling general practice, the foundations of our health service. Now is not the time to gamble taxpayers cash on evidence-light policy.

Dr Craig Seymour is a GP in Islington, North London and director of Islington GP Federation. With input from Dr John McGrath, a GP in Islington.

Rate this article  (4.4 average user rating)

Click to rate

  • 1 star out of 5
  • 2 stars out of 5
  • 3 stars out of 5
  • 4 stars out of 5
  • 5 stars out of 5

0 out of 5 stars

Readers' comments (23)

  • Having actually worked at luton and dunstable hospital there are issues with even this model. Steve interesting comments. How many clinical GP sessions do you do per week at your practice? I am full time at mine. I often find that people who work part time do not really grasp full time general practice. If they did they wouldn't make such patronizing statements

    - anonymous salaried!

    Unsuitable or offensive? Report this comment

  • Thanks, I am sure issues with all models, I'd be interested to hear of your experience

    I do 2-3 session per week in general practice, rest working on clinical service redesign. I'm in awe of ppl who do full time GP

    Interest to hear what you feel i said is patronising

    All the best


    Unsuitable or offensive? Report this comment

  • I don't think Steve is being patronising. He says we manage Undifferentiated demand - we take all comers and we can handle that. The reason is we are good at general medicine in a way that hospital specialists aren't. We are so general we can accommodate all branches of medicine( gynae psychiatry etc). To put it bluntly there are no doctors working in hospitals who can do that. We deliver highly abstract diagnoses rather than specific ones as we work with simple tools, very quickly. We understand when the medical model doesn't apply- somatisation- and we specialise in that - as we see it all the time across the board.
    The problem is that when they use us in casualty we have an image problem. We are the triage, the junior taking the walk ins and the bouncers. Any junior doctor viewing how we are utilised will soon get the idea that we are the "sore throat" doctor. And that I'm afraid is why we should avoid working there.

    Unsuitable or offensive? Report this comment

  • very helpful Gary, thank you. I agree then model needs working through carefully - perhaps we are the "?diagnosis (but not immediately life threatening) doctor"
    perhaps the challenge is our wide range of skills?

    Unsuitable or offensive? Report this comment

  • Craig - I agree with a lot of what you say in your piece, especially about the need for whole urgent care system solutions.

    Unsuitable or offensive? Report this comment

  • I'm afraid I couldn't agree with Steve's comments.

    As you stared L&D model isn't simply putting a GP in AED and we know evidence of this is poor. My understanding of that model is not seeing the patient in AED but to redirect them to appropriate service.

    I also disagree with notion it is never the patient's fault. Whilst I agree the healthcare has become unneccesarily complex due to government's desire to constantly change/pilot/re-invent healthcare, patients must hold some responsibilty. After all, no other country would use tax to suppliment patient's failure to selfcare/navigate the health system their outcome are often better then UK.

    I'm sure you know from your own assertion of well children in paediatric AED. We see huge number of patients who do not need but want health care. Personally, decades of treating patients like children who are incapable of making decisions and hold no responsibility of action has led to the state we are in now.

    What I think we need is;
    1) patients to be educated on correct use of healthcare
    2) inform patients how much health care costs for individual episodes
    3) health care being protected from litigation when agreed and approved pathway of re-direction is followed
    4) shared patient record so each organizations can understand and manage episodes of care
    5) single point of access run by senior clinicians to triage appropriate service (sorry NHS111 doesn't cut it!)

    and yes, all of the above needs some form of collaborative work

    Unsuitable or offensive? Report this comment

  • Steve- doctor specialising in consultation skills perhaps. Though this is easily misunderstood as anyone can take a history. And your point about "not life threatening" is unfortunate.
    Take the simple sore throat. Common knowledge is that a monkey could do that. But it's all about the black swan event. Over the years I have recognised tonsillar cancers, epiglottitis in a 60 year old(intubated), addisonian crisis ( was hypothermic), a fatal lymphoma, as I recall. Every GP has their own war stories here. Add in the ubiquitous quincies, oral thrush and neutropenic patients, oh and the scarlet fever occasionally and you have a diagnostic slippery slope. Yet we get through these in a few minutes apiece.
    So - people will always see us diagnosing non life threatening diseases whereas what we do is diagnose everyone who has anything. More to the point we do this best in our own surgeries where we work with patients we have prior knowledge of. Bayesian!

    Unsuitable or offensive? Report this comment

  • Gary - I like it!

    By non life-threatening I mean like immediate from a stabbing, shooting, major RTA etc

    There will be a lot that is not that where our unique skills and experience will be highly valuable I am sure.

    Ideally we will see people in our surgeries because our access is superb, continuity (where needed) is excellent and hence everyone chooses to see us there.

    In the meantime.....

    Unsuitable or offensive? Report this comment

  • Good comments although I don't think Steve was patronising ...
    I already work in the ED as a GP. I am fully integrated so I advise on GP type presentations and will set up a mini surgery in minors when demand is high. I also work in resus (initially terrifying!) deal with minor injuries, see majors patients and sometimes review patients on our ED ward admitted by the night team. I do 4 sessions of this and 3 of Gp as a locum locally. I love the job! It has given me some great skills to take back to general practice and helped out the ED. The new model is quite prescriptive with how they want the new service to run with GPs being on site but totally separate from the ED. They have also stipulated that we should not have access to diagnostics. This follows the L and D model. The problem is one model does not necessarily suit all hospitals. The first thing that strikes me with what is proposed is that they will have difficulty with recruitment - why would GPS want to work in this model when they could earn a lot more in the ooh service down the road? I do the job because I enjoy being part of the team. I was fed up with the grind of seeing hundreds of patients isolated in a room on my own - I am in awe of anyone who can do that full time! You may think that this job has taken me away from GP - it has not . Nothing would persuade me to do 9/10 sessions salaried or locum Gp work and I am not really interested in running a business so partnership does t appeal either. I enjoy the fact that I have instant access to investigations and that I see a different cohort of patients - definitely sicker and more likely to have significant underlying pathology. So bring on ED GPS! But let's think carefully how it is done.

    Unsuitable or offensive? Report this comment

  • Love it Kirsty!

    Thank you and keep enjoying your job and doing good!


    Unsuitable or offensive? Report this comment

View results 10 results per page20 results per page50 results per page

Have your say