Cataracts and Dementia: Does Restoring Vision Matter More Than We Think?

Despite a longstanding career as an Operating Department Practitioner (ODP) in the NHS, my experience across the spectrum of surgical specialties never extended to ophthalmic surgery. Even during my training, our paths simply never crossed.

During a recent conversation with a consultant anaesthetist I was working alongside, who has extensive experience in ophthalmic anaesthesia, we got talking about our aging parents and dementia. The conversation moved on to cataract surgery and the potential benefits it may have for people with cataracts, particularly those living with dementia. It left me wanting to look a little more closely at the subject.

I had always thought of cataract surgery as being fairly straightforward. It is a common operation, usually relatively quick and often performed under local or topical anaesthesia. For most people, it can make a very noticeable difference to their vision.

What I hadn’t really considered was what losing that vision might mean for someone who already has dementia.

There is some interesting evidence around cataract surgery and dementia itself. A study published in JAMA Internal Medicine followed more than 3,000 adults aged 65 and over and found that those who underwent cataract surgery had around a 29% lower risk of subsequently developing dementia than those who did not have surgery. A similar association was found with Alzheimer’s disease.

That sounds quite significant, but it is worth being careful about what the study actually tells us. It was an observational study, so it shows an association rather than proving that cataract surgery prevents dementia. The researchers also acknowledged that further research would be needed to establish whether there is a causal relationship.

For me, though, the more interesting question is what happens when someone already has dementia and is also developing cataracts.

Dementia can make it more difficult to understand and navigate the environment around you. Poor vision potentially adds another layer to that. A familiar room may become less familiar. Objects can be harder to identify. Moving around safely can become more difficult, and recognising people may not be as easy.

The Royal College of Ophthalmologists recognises this in its guidance on cataract surgery for people living with dementia and recommends that surgery should not unnecessarily be delayed in this group. It also identifies the potential effects of poor vision on mobility, self-care, communication and behaviour.

That seems fairly logical when you think about it.

If someone is already struggling to make sense of their surroundings because of cognitive impairment, why would we want to add poor vision into the equation if there is something relatively straightforward that can be done about it?

There is some evidence that we may not always be doing this particularly well. The Royal College’s guidance refers to the PrOVIDe study, which suggested that at least a quarter of people living with dementia could benefit from cataract surgery. Yet the same work suggested that around three quarters of those who might benefit had not received surgery.

That makes me wonder whether we are sometimes looking at the cataract in isolation.

Are we simply asking whether someone’s eyesight has deteriorated enough to warrant surgery, or are we considering what that deterioration is actually doing to their everyday life?

NICE also recognises that visual acuity alone doesn’t necessarily capture the impact of cataracts on someone’s quality of life. This seems particularly relevant when assessing someone with cognitive impairment, who may find it more difficult to explain how their vision is affecting them.

There is also the fairly obvious issue of falls. Poor vision can increase the risk of falling, and NICE recommends referral for cataract surgery where visual impairment from cataracts is contributing to an increased risk of falls.

Then there is the anaesthetic side.

The Royal College of Anaesthetists also recognises the potential benefit of earlier cataract surgery in people living with dementia. It makes the fairly practical point that the pathway should be kept as simple as possible, because repeated hospital visits can be particularly difficult for these patients. The type of anaesthesia may also need to be adapted to the individual patient.

This brings me back to that original conversation.

As an ODP, I have spent a large part of my career thinking about patients in relation to anaesthesia, surgery and recovery. Yet I had never really considered cataract surgery in this particular context.

Maybe the question isn’t simply whether someone with dementia has a cataract that is bad enough to operate on.

Maybe we should be asking what that cataract is doing to them.

Is it contributing to falls? Is it making them less independent? Is it making an already confusing environment even more difficult to understand? Is it affecting their ability to recognise family members or communicate with those caring for them?

If restoring their vision can help with some of those things, then perhaps cataract surgery should be considered as part of the wider care of someone living with dementia, rather than simply as an operation to improve their eyesight.

I don’t think the evidence currently allows us to say that cataract surgery prevents dementia. That would be stretching it too far.

But there seems to be enough evidence to suggest that we should take the effect of poor vision in someone living with dementia seriously.

And that, for me, is what started with a fairly ordinary conversation about our aging parents and ended with me looking at cataract surgery rather differently.

If AI Is Here to Augment Healthcare, What Happens When Augmentation Becomes Substitution?

Professional expertise, workforce transformation and the future NHS clinician

Meta Description: Artificial intelligence promises to augment healthcare professionals, but could it reshape clinical roles, expertise and workforce design over time?

Whenever artificial intelligence is discussed in healthcare, we are offered a familiar reassurance.

AI isn’t going to replace doctors.

It is going to augment them.

The distinction between augmentation and substitution matters.

Augmentation occurs when technology helps professionals perform their role more effectively while preserving their responsibility, judgement and expertise. The clinician remains the decision-maker, using technology as an additional source of support.

Substitution is different. It occurs when technology begins to replace elements of professional judgement, expertise or decision-making. The clinician’s role shifts from actively thinking and deciding to monitoring and accepting recommendations generated elsewhere.

At first glance the difference may appear subtle. In reality, it represents a fundamental change in the relationship between healthcare professionals and technology. Augmentation strengthens human capability. Substitution risks allowing that capability to gradually diminish.

It will reduce administration, support clinical decision-making, improve access to information and give clinicians more time to spend with patients.

On the face of it, that sounds entirely reasonable. Healthcare contains an enormous amount of work that could be done differently. Clinicians spend considerable time documenting, searching for information, processing data and completing administrative tasks. If technology can remove some of that burden, why wouldn’t we want it to?

The trouble is that healthcare has heard similar promises before.

The introduction of physician associates, advanced nurse practitioners and anaesthesia associates was also presented as a way of increasing capacity and making better use of the workforce. The argument was not that doctors were no longer needed. Rather, it was that appropriately trained professionals could undertake work that did not always require a doctor while working alongside the existing medical workforce.

That sounds remarkably similar to the case now being made for AI.

I’m not suggesting that people and algorithms are equivalent. They clearly aren’t. But there is something useful in examining what happens when something introduced to support a workforce begins to reshape how that workforce is organised.

When Augmentation Changes the Role

The distinction between support and replacement sounds straightforward.

A new professional role can add capacity to an existing team.

Technology can help a clinician by providing information, automating documentation or supporting a decision.

The situation becomes more complicated when we stop focusing on individual tasks and start looking at how entire roles evolve.

If someone else can safely undertake a task previously performed by a doctor, that makes sense. We should not require the most highly trained professional in the system to carry out every task simply because they have traditionally done it.

The same principle applies to technology.

If AI can reliably undertake administrative work, clinicians should not need to spend valuable time doing it. If it can identify a possible abnormality on an image or assemble information from a patient’s record in seconds, there is an obvious benefit.

Once we’ve safely delegated one task, it is only natural to ask what else could be delegated or automated. Over time, the distribution of work changes. Professions do not suddenly disappear, but the work that defines them can gradually shift.

That possibility may be far more important than the familiar question of whether AI will “replace” clinicians.

The Associate-Profession Comparison Is Worth Considering

Debate around physician associates, advanced nurse practitioners and anaesthesia associates often becomes polarised.

I’m not interested in revisiting those arguments here.

There is a perfectly reasonable principle behind a multiprofessional workforce. A complex healthcare system does not need every task to be performed by a doctor, just as it does not need every task to be performed by a nurse, ODP or any other professional.

Good workforce design matches the expertise required to the work that needs to be done.

The real issue emerges after a new role has become established.

As new roles develop and demonstrate the ability to undertake additional responsibilities, role boundaries naturally expand. That is a normal part of workforce development.

It also creates a difficult question. If someone else can safely perform the work, why must it remain the responsibility of the original professional?

That is not the same as setting out to replace a profession. Yet over time it may still alter the shape of that profession.

The lesson is that the stated purpose of an innovation does not necessarily tell us what its eventual impact will be.

What Happens to the Time AI Gives Us?

This is perhaps the part of the AI discussion that interests me most.

We are repeatedly told that AI will save clinicians time. However, what happens next matters far more.

Imagine AI saves a doctor an hour a day by reducing documentation and administrative work. That time could be used to deepen patient interactions, support teaching and supervision, manage more complex clinical work, or simply see more patients.

All of those outcomes could be described as making better use of technology.

Yet they are not the same thing.

One is potentially an improvement in care. Another is an increase in capacity.

In a healthcare system under immense pressure, it is not difficult to imagine which option will appeal to organisations facing workforce shortages, financial constraints and growing waiting lists.

Technology can create capacity, but organisations decide what that capacity is used for. Whether it improves care, increases productivity or simply raises expectations of clinicians is ultimately a human choice, not a technological one.

AI Could Change Professional Roles Without Replacing Professionals

For that reason, I do not think the most useful question is whether AI will replace doctors.

It probably will not, at least not in the simplistic way that question is often framed.

The more realistic possibility is that AI progressively takes over parts of what doctors currently do. The same applies to nurses, ODPs and many other healthcare professionals.

AI already assists with documentation and information retrieval. Its role in decision support, image interpretation, risk prediction and patient monitoring will continue to grow.

That does not eliminate the professional. It changes where professional time and attention are directed.

We can already see how this might develop. A radiologist who once reviewed every image may increasingly focus on those flagged by an AI system. A doctor who once drafted clinic letters may find themselves reviewing and editing text generated automatically. In both cases the professional remains involved, but the nature of their work has changed. In surgery, similar changes could occur in preoperative assessment, imaging, intraoperative monitoring and postoperative surveillance.

In many respects, that is exactly what we should want. If technology removes repetitive work, clinicians can spend more time exercising judgement, communicating with patients and dealing with situations that do not fit neatly into established pathways.

However, we should be careful about what we mean by human oversight.

There is a significant difference between a clinician using AI as a source of information and a clinician merely approving a recommendation generated elsewhere. If the system performs most of the work and the human intervenes only when something unusual occurs, the professional role has changed substantially.

The clinician may still be in the loop, but that does not automatically mean they retain the same level of capability.

That matters because professional capability is developed and maintained through practice. If clinicians increasingly supervise decisions rather than actively making them, we need to consider what happens to expertise over time.

Competence Isn’t the Same as Capability

A system can become extraordinarily good at specific tasks.

It can recognise patterns in images.

It can summarise information.

It can identify relationships in large datasets.

It can generate differential diagnoses.

It can predict risk.

Healthcare, however, is not simply a collection of isolated tasks.

Clinical practice happens in context.

Patients do not always follow the textbook. Information is often incomplete. Two reasonable options may conflict. A patient’s preferences may not align with the statistically most likely course of action.

Sometimes the most important skill is recognising that something simply does not feel right.

That is much harder to reduce to a defined competency.

The distinction between competence and capability therefore becomes important.

Competence is often demonstrated by performing a task to an expected standard.

Capability is broader. It involves combining knowledge, skills, experience and judgement, then applying them appropriately when circumstances change.

AI may become extremely competent at many things.

The challenge is ensuring that, as routine work becomes increasingly automated, we do not inadvertently erode the professional judgement that remains.

The issue is not simply today’s workforce. Future clinicians develop expertise through repeated exposure to complexity, uncertainty and decision-making. If technology removes too much of that experience, we may discover that adaptability is considerably harder than improving efficiency.

That is not an argument against AI.

It is an argument for being deliberate about the role we expect people to play within an increasingly technology-enabled system.

Defining the Human Role

The more useful question is not whether AI replaces clinicians, but where technology genuinely adds value. Some tasks are likely to be better performed by machines, others require human judgement, and much of healthcare will continue to depend on a combination of both.

The goal should not be to preserve human involvement for its own sake, nor to automate simply because we can. It should be to understand where expertise, adaptability and judgement contribute most and design systems around those strengths.

That sounds straightforward.

It becomes less straightforward when the technology creates significant workforce, operational and financial opportunities.

The Real Test Is What We Do with the Capacity

I do not think healthcare genuinely wants AI to replace doctors.

It wants greater capacity, quicker access for patients, better decisions, fewer errors and relief from the growing administrative burden that surrounds clinical work. AI potentially offers all of those things.

But there is an important difference between using technology to remove unnecessary work and using it to reduce the number of people required to deliver care. Both may happen, and the distinction between them may become increasingly difficult to see.

That is perhaps one of the more important lessons from workforce redesign over recent decades. A role can be introduced to support an existing workforce without any intention of replacing another profession. Yet as work is redistributed, the workforce itself inevitably evolves.

AI is likely to do something similar.

The key issue, therefore, is not what we say we want AI to do when it is introduced. It is what we choose to do with the capacity it creates.

If that capacity gives clinicians more time with patients, greater opportunity to exercise judgement and more space to manage complexity, then AI could make healthcare more human rather than less.

If instead it is used primarily to allow one clinician to manage more patients, make more decisions and undertake more activity, we may arrive at a very different form of productivity.

Perhaps the real measure of AI in healthcare should not be how many professionals it allows us to replace, but how much unnecessary work it removes while preserving and strengthening the capability of the people who remain.

Technology should make it easier for healthcare professionals to do the things that matter. It should not simply make it possible for fewer people to do more.

History suggests that technologies and workforce reforms rarely end up exactly where their architects intended. Once something creates significant new capacity, the question of how that capacity is used becomes an organisational and economic one as much as a technological one.

AI will not decide whether it augments healthcare or substitutes for parts of it. We will. And perhaps the most important decision is not what AI is capable of doing, but what we decide human beings should continue to be capable of doing.

The greatest risk may not be that AI replaces clinicians. It may be that clinicians gradually lose the expertise required to supervise AI safely.

Have We Confused Academic Achievement with Clinical Readiness? Rethinking Competence, Capability and Practice in Nursing and ODP Education

Edited by the Author 23/08/2026

Meta description:
Have we confused academic achievement with clinical readiness in nursing and ODP education? Exploring competence, capability and clinical practice.

There is little doubt that nurses and Operating Department Practitioners (ODPs) are better educated academically than they have ever been.

Degree-level education has brought significant benefits. It has given practitioners a stronger understanding of science, evidence-based practice, research, ethics and professional accountability. Those are important things, particularly in a healthcare system that is becoming increasingly complex.

So this isn’t an argument against university education.

But I do wonder whether, in making healthcare education increasingly academic, we have placed too much emphasis on what can be taught, measured and assessed in an educational environment, and not enough on what practitioners actually need to be able to do when they are faced with a real patient.

Perhaps the problem isn’t that healthcare education has become too academic. Perhaps academic achievement has become too easily confused with clinical readiness.

Because knowing something and being able to use that knowledge effectively are not necessarily the same thing.

Clinical competence is more than knowledge. It is more than being able to perform individual skills.

It is about bringing knowledge, technical ability, communication and judgement together when circumstances are uncertain, information is incomplete and things don’t go according to plan.

And that raises an uncomfortable question.

If our assessment systems cannot reliably tell us whether a student can bring all of those things together when it matters, can we really say that we have assessed clinical competence?

What did university education give us?

We shouldn’t romanticise the past.

Moving nursing and ODP education into higher education was not necessarily a mistake, and I don’t think anyone would seriously argue that healthcare professionals should return to an education system based largely on learning tasks and following established routines.

Healthcare has changed.

The science is more complex. Patients are often older and have more comorbidities. Treatments are more sophisticated. Professionals are expected to understand the evidence behind what they do, question established practice and recognise when something isn’t right.

Nurses and ODPs therefore need much more than a list of technical skills.

They need to understand why they are doing something, not simply how to do it.

For ODPs in particular, the move into higher education helped establish the profession as an autonomous, registered healthcare profession. The HCPC expects ODPs to practise safely and effectively, recognise the limits of their practice and exercise appropriate professional judgement.

That is a long way from simply being trained to complete a series of tasks.

And that is why I don’t think this debate can sensibly be reduced to academic versus vocational education.

Both have value.

The real question is whether we have got the balance right.

Competency isn’t the same as competence

I think there is an important distinction here that is easily missed.

We have become very good at defining competencies.

A competency can usually be described, taught, observed and assessed. It might be the ability to perform a particular procedure, demonstrate a particular skill or show that a particular learning outcome has been achieved.

That’s useful.

But competence is something bigger.

It is the ability to bring those individual competencies together and use them appropriately in the circumstances in which they are required. But there is another distinction worth making. Competence is often demonstrated against a defined standard. Clinical practice, however, doesn’t always happen in standard situations. The practitioner may have to adapt what they know to circumstances that are unfamiliar, changing or unpredictable.

That is where capability becomes important. It isn’t simply knowing what to do. It is being able to use what you know, exercise judgement and adapt when circumstances change.

And that is much harder to assess.

Imagine an anaesthetic emergency.

The patient isn’t responding as expected. The airway is becoming difficult. The equipment you need isn’t immediately where you expect it to be. Someone else in the team is asking you a question. The anaesthetist changes the plan. There is pressure building in the room.

The difficulty is that these things are not necessarily separate competencies. They interact. The practitioner has to recognise what is happening, decide what matters most, communicate effectively and adapt as the situation develops.

That is much closer to capability in practice than simply demonstrating a series of individual competencies.

There isn’t necessarily a single correct sequence of actions that can simply be recalled.

That is judgement, and it is considerably harder to assess than knowledge.

We have become very good at breaking competence down into manageable pieces: learning outcomes, competencies, skills, assessments and sign-offs.

All of these have a purpose.

But clinical practice doesn’t arrive in neatly separated compartments.

The patient doesn’t know that the student has successfully completed the airway competency, passed the physiology examination and demonstrated effective communication in three previous assessments.

What matters is whether those things can come together when they are needed.

That is the difference between demonstrating competency and demonstrating competence.

Are we measuring achievement or readiness?

Of course, modern healthcare education isn’t simply about university essays and written examinations.

That is a good thing.

But there is still an important distinction between demonstrating that somebody has achieved a learning outcome and demonstrating that they are ready to practise.

A student can produce excellent academic work, explain the relevant physiology, discuss the evidence, demonstrate the technical skill and complete the competency documentation.

Yet we may still not know how they will respond when several things happen at once.

That is not a criticism of students. It is a challenge for the way we educate and assess them.

We are generally very good at assessing things that are observable, repeatable and measurable. But some of the things that matter most in clinical practice are much harder to capture: judgement, prioritisation, adaptability and recognising when a situation is changing.

If we assess the individual components and then simply assume that the ability to integrate them will follow, are we actually assessing competence — or are we inferring it?

The question is not simply, “Can you demonstrate this?” but “Can you use it effectively when circumstances become unpredictable?”

There is a difference.

The NMC is asking some important questions too

This isn’t simply an argument being made by practitioners who remember the old days and think everything was better then.

The Nursing and Midwifery Council has been looking at practice learning and, in 2026, consulted on proposals intended to strengthen nursing and midwifery students’ practice-learning experiences.

The consultation followed work identifying variation in students’ experiences and supervision. That suggests the balance between university education and clinical experience is not necessarily a settled question, and perhaps that is no bad thing.

Professional education should never stand still. We should continually ask whether the way we educate people is actually preparing them for the reality they are going to encounter.

If there is a gap between academic achievement and clinical readiness, the answer is unlikely to be less education. The better response may be stronger integration between academic and clinical learning, longer periods of meaningful practice exposure, progressive responsibility under supervision, and assessment approaches that focus not only on what students know, but on how they apply that knowledge when circumstances are uncertain.

Simulation has a place. But it isn’t the real world.

Simulation is another area where this debate becomes interesting.

I am certainly not against simulation.

Used well, it allows students to practise emergencies they might rarely encounter, make mistakes safely, receive feedback and develop confidence.

But there is a difference between simulating a clinical situation and actually being in one.

Take an airway emergency.

A simulated scenario can test whether a student recognises the problem, knows what to do and can perform the necessary skills.

The real clinical environment adds something else.

There are interruptions, unexpected equipment issues, communication challenges, changing priorities and genuine consequences. That doesn’t mean simulation cannot reproduce some of these challenges. It can, and increasingly sophisticated simulation can provide very valuable learning. But simulation should enhance clinical experience rather than become a substitute for it.

The patient’s response may not follow the script, and neither does the team around them.

These factors are not peripheral to practice. They are part of practice.

The NMC itself recognises that simulated practice learning should enhance rather than simply replace experience in clinical practice.

That seems entirely sensible.

Perhaps the greatest value of simulation isn’t that it can reproduce clinical practice perfectly.

Perhaps its value is that it prepares students to make better use of the clinical experience that simulation cannot reproduce.

The ODP provides an interesting comparison

This is where I think the history of ODP education becomes particularly interesting.

Before ODP education moved firmly into higher education, the City & Guilds 752 Operating Department Assistant qualification provided a vocational route into operating department practice.

It wasn’t simply a classroom qualification.

It was closely tied to the clinical environment. Learners worked alongside experienced practitioners, spent prolonged periods in operating departments and progressively developed responsibility.

That matters.

The City & Guilds 752 qualification was more than a route into operating department practice. It helped provide a foundation on which the ODP profession was later built.

I’m not suggesting that the City & Guilds 752 system was perfect, or that everything that came afterwards was a mistake. Far from it. 

But the older model did place considerable emphasis on sustained exposure to clinical practice, progressive responsibility and learning alongside experienced practitioners. Those characteristics are worth considering when we think about what may have been lost as education became increasingly academic.

The move into higher education brought significant benefits. ODPs became graduates, autonomous registered professionals with a much stronger academic foundation and a clearer professional identity.

But something else changed.

The balance between learning the job and learning about the job changed.

That distinction is worth thinking about.

Students learned not only how to perform tasks, but how operating departments functioned, how teams behaved, how priorities changed and how practitioners responded when things did not go according to plan.

In essence, they were learning how to become practitioners.

The modern degree model provides something the old system could not provide to the same extent: a much stronger academic foundation.

Research literacy, critical appraisal, a stronger understanding of evidence, greater professional autonomy and a broader understanding of the science behind practice.

We shouldn’t throw any of that away.

But perhaps we should ask whether, during the transition from vocational education to higher education, we retained enough of the clinical immersion and progressive responsibility of the old model to complement everything the modern academic model gives us.

In other words:

What did we gain through academicisation, and what might we have lost?

And perhaps the more uncomfortable question is:

Have we retained enough of the clinical formation of the old model to make the modern academic model work as effectively as it should?

Graduates who can also be practitioners

I don’t think the answer is to produce less academic nurses or ODPs.

Quite the opposite.

We need professionals who can understand evidence, question practice and make decisions that can be justified.

But we also need professionals who can turn that knowledge into action.

A practitioner may understand the physiology, know the guideline and appreciate the evidence that underpins practice.

The real test comes when the patient does not behave like the scenario in the textbook, information is incomplete, priorities compete and someone has to make a decision.

That is where clinical competence lives.

Teaching, assessment and academic success remain essential. However, none of them automatically guarantees clinical readiness. The challenge is ensuring that knowledge, judgement and practical performance are developed and evaluated together, rather than assuming one is a reliable substitute for the others.

The challenge isn’t to choose between the two.

It is to make sure that the graduate walking into clinical practice is capable of being more than a graduate.

They need to be a practitioner.

The question is not whether healthcare education has become too academic. It is whether we have become too comfortable measuring what graduates know and can demonstrate individually, while being less rigorous in demonstrating what they can do when it matters most.

Ultimately, patients do not experience our curriculum, learning outcomes or competency documentation. They experience the practitioner standing in front of them. Perhaps we need to think less about whether someone has completed each individual competency and more about whether they can integrate those competencies when the situation demands it.

And when the situation becomes difficult, what matters is whether that practitioner can bring everything they know, everything they have learned and everything they have practised together — and exercise appropriate judgement when it matters.

That, surely, is what clinical competence is supposed to mean.

How Green Is My Anaesthetic Room?

Meta description:
Can reusable medical devices make anaesthesia more sustainable without compromising patient safety? Exploring decontamination, NIBP cuffs, laryngoscopes and the NHS.

Starting my career in the NHS, in the late 1980s, particularly in the anaesthetic room, the idea of reprocessing and reusing medical devices was simply part of everyday practice.

Some endobronchial tubes (red rubber, for those who remember them), laryngeal mask airways, Guedel airways (black rubber), anaesthetic face masks, laryngoscope blades and non-invasive blood pressure (NIBP) cuffs were all routinely decontaminated and returned to use. I can still remember the process of soaking, scrubbing and then placing the equipment into the steam steriliser, or “Little Sister”.

At the same time, things were beginning to change. The development of plastics and new materials for endotracheal tubes and their cuffs during the 1960s saw single-patient-use tubes gradually entering anaesthetic practice, while reusable rubber tubes continued to be used. The move towards single-use and the development of more formal reprocessing practices therefore happened alongside each other rather than one simply replacing the other. The emergence of HIV/AIDS in the early 1980s, and the growing awareness of the potential for blood-borne transmission and cross-infection, provided a further and significant impetus towards single-patient-use equipment. Now, more than two decades into the 21st century, we appear to be turning back the clock.

Environmental sustainability is increasingly challenging some of the assumptions that have underpinned anaesthetic practice for decades. We have already seen progress through reducing the use of nitrous oxide and eliminating desflurane, so inevitably other areas of practice are going to come under scrutiny.

And perhaps one of those areas is the single-use medical device.

Firstly, let me make a general assumption. If we are going to start reusing medical devices that have traditionally been single-use, the initial decontamination is probably going to fall to anaesthetic support staff.

Not that we don’t already have enough to do.

Whilst I absolutely support the drive towards environmental sustainability, I do have some concerns about how we are going to safely decontaminate things such as laryngoscope blades, laryngoscope handles and NIBP cuffs.

Take reusable NIBP cuffs.

The initial investment comes at a premium and, with a limited number of permitted uses — usually around forty — the economics only really begin to stack up if we get close to that maximum number of uses. On paper, they can therefore be more cost-effective than single-patient-use cuffs.

But then reality gets in the way.

How do we reliably record how many times each cuff has been used?

How do we make sure that every anaesthetic room has the correct range of cuff sizes available? We already know that using the wrong size cuff can affect the accuracy of a blood pressure reading.

Then there is the cuff itself.

The Velcro fastening creates areas where blood, skin debris and other contamination can potentially become trapped. How confident are we that every part of the cuff is being effectively cleaned every time?

And what happens when the cuff becomes visibly contaminated with blood or other body fluids? Presumably it comes out of service, whether it has reached its forty uses or not.

So the potential financial saving starts to look rather different.

Then there are laryngoscope blades.

Reusable blades, whether for direct or indirect laryngoscopy, require appropriate facilities and processes for their decontamination and, where necessary, sterilisation. This isn’t simply a case of giving the blade a wipe before putting it back in the drawer.

There has also been interest in whether ultraviolet-C (UV-C) light could be used in the decontamination of laryngoscope equipment.

Perhaps it can.

But I think we need to be careful here.

UV-C isn’t some magic box that we put a dirty laryngoscope into and take a sterile one out of. Its effectiveness depends on the light reaching the surfaces that need to be treated. Shadowing, the shape of the equipment and areas that the light cannot reach all potentially become important.

The same applies to manual wipe-based systems such as the Tristel Trio system.

These systems provide an alternative to conventional processing and are specifically designed for devices such as laryngoscopes. But they still rely on the process being followed correctly.

And that, for me, is the important bit.

The process has to work every time.

It isn’t enough for the product to be effective. The device has to be cleaned properly, the disinfectant has to reach all the relevant surfaces, the correct contact time has to be achieved and the person doing it has to know exactly what they are doing.

That also means having appropriate training, competency assessment, facilities and traceability.

In other words, we cannot simply replace a single-use device with a reusable one and assume that we have automatically made healthcare more environmentally friendly.

We are replacing a product with a process.

And that process needs to be examined very carefully.

There is a bit of a conundrum here.

We quite rightly want to do what is better for the planet, but we also have a responsibility to do what is safe for the patient. The two shouldn’t be in competition with each other.

There is also something slightly Dickensian about the whole situation.

At the moment, the NHS can sometimes feel a little like Oliver Twist standing in front of the workhouse master, bowl in hand, asking:

“Please, sir, I want some more.”

More productivity.

More efficiency.

More savings.

And now, quite rightly, more environmental responsibility.

The problem is that all of these things eventually find their way down to the people actually delivering the care.

Manufacturers have a much bigger role to play here. If a device is going to be reused, then perhaps it should be designed as a reusable device from the outset, with the manufacturer specifying and validating how it should be decontaminated, how many times it can safely be reused and how its performance can be assured.

There is a significant difference between a reusable device designed for repeated use and a single-use device that a healthcare organisation decides to reprocess. We shouldn’t blur that distinction simply in the name of sustainability.

If we want reusable devices, buy devices that are actually designed, validated and supported as reusable — don’t turn a single-use device into a reusable one by local ingenuity.

And, of course, there are plenty of things we can do ourselves. Turning lights and computers off when they aren’t being used, reducing unnecessary waste and simply becoming more aware of what we throw away are all relatively easy things to do.

But is there a clear answer?

I fear not.

I’m certainly not suggesting that we should go back to the anaesthetic room of the 1980s and start scrubbing everything with a brush again.

Nor am I suggesting that single-use devices are automatically the environmentally responsible option.

What I am suggesting is that we need to look at the whole picture.

If we are going to move from single-use to reusable devices, we need to know that the decontamination process is safe, reliable, achievable and sustainable in the real world.

Because ultimately, being green is important.

But so is getting the patient safely through the anaesthetic room.

From dependency to independence: why social care reform matters to the future of the NHS

Estimated reading time: 8 minutes

When I wrote about the difficulties of integrating health and social care a couple of years ago, I was trying to understand why something that seems so logical remains so difficult to achieve. Political interference, fragmented funding, competing priorities and organisational silos all contribute to a system where the needs of the person can easily become secondary to the needs of the organisations providing their care.

Looking at the issue again now, I think there is an even more important question. Even if we successfully integrated health and social care, what would we actually want that integrated system to achieve?

For me, the answer has become increasingly clear: social care reform should help people remain independent for as long as possible, rather than simply expanding our ability to respond once independence has already been lost.

That changes the conversation. Instead of asking how we can provide more care, we begin asking how we can reduce the need for intensive care in the first place. Instead of treating social care as a safety net beneath the NHS, we start treating it as one of the foundations of a healthy society.

Recent NHS England and Department of Health and Social Care policy documents increasingly recognise that health and care need to become more integrated, preventative and community-based. The problem is that the system still largely rewards activity and crisis response rather than independence and resilience.

After forty years working in frontline healthcare, what strikes me most is how often people enter hospital because the systems around them have failed long before the ambulance arrived. A fall, a chest infection, confusion, dehydration, medication problems or carer exhaustion rarely occur in isolation. They are often the final stage of a gradual loss of independence that could have been recognised earlier and managed differently.

I think social care reform needs to happen at three interconnected levels. Structural reform changes how the system works; community reform changes the conditions in which people live; and population health reform addresses the factors that create or accelerate dependency in the first place.

Structural reform: creating person-centred health and social care

The NHS and social care continue to operate with different funding arrangements, accountability structures, eligibility rules and organisational cultures. We talk endlessly about collaboration, yet fragmentation remains one of the defining characteristics of the system.

The consequence is familiar to anyone working in hospitals or community services. Patients move between organisations that often have different priorities, incompatible information systems and separate budgets. Delayed discharge is usually presented as a hospital problem, but it is more accurately a coordination problem. The hospital is often where the problem becomes visible, rather than where the problem originated.

An older person admitted after a fall may require rehabilitation, home adaptations, medication support, social care, community nursing and family assistance. None of these needs fits neatly within a single organisation. The patient experiences one journey; the system experiences multiple disconnected episodes.

Integrated neighbourhood services

One of the most important structural reforms is the development of genuinely integrated neighbourhood services. The aim is not organisational merger for its own sake. It is continuity of care.

People with frailty, multiple long-term conditions, dementia or disability benefit from consistent relationships and coordinated decision-making. When support is organised around institutions rather than people, gaps become inevitable.

Integration also needs to occur at the level of capability, not just governance. Community teams need the authority, skills and information to respond rapidly when someone begins to deteriorate. Social care staff should be able to trigger rehabilitation, clinical review or housing interventions before problems become crises.

Shared accountability

One of the persistent failures of reform is that organisations are often rewarded for protecting their own budgets rather than improving overall outcomes.

A local authority may invest in social care that reduces hospital admissions, while the financial benefit accrues largely to the NHS. Hospitals may discharge patients quickly while community services absorb the resulting pressure.

A reformed system needs shared accountability for outcomes such as maintaining independence, avoiding unnecessary admissions, reducing delayed discharges and improving quality of life.

Outcome-based funding

Much of the current funding architecture still makes activity easier to measure and reward than outcomes such as independence, confidence or wellbeing.

An outcome-based approach would incentivise services that maintain functional ability, reduce unplanned admissions, support people at home and improve quality of life.

The objective should be to create a system where the greatest reward comes from helping people need less intensive support, not from delivering more episodes of care.

Community reform: building independence through housing, carers and local support

Structural integration is essential, but organisations alone do not keep people independent. Independence is sustained within homes, families, neighbourhoods and communities.

This is where I think social care reform has often been too narrowly defined.

Housing as a health intervention

Housing is one of the most underappreciated health interventions available.

Poor housing contributes to falls, frailty, respiratory disease, social isolation, delayed discharge and loss of independence. Lack of adaptations can turn a manageable disability into a hospital admission.

Investment in home adaptations, accessible housing, supported living and technology-enabled care should be treated as health infrastructure, not simply social policy. Measures such as stairlifts, level-access showers, improved heating and appropriate housing support can often prevent much greater downstream costs associated with hospital admission or long-term care.

Supporting unpaid carers

No serious social care reform can ignore unpaid carers.

Millions of family members and friends provide substantial amounts of unpaid care across England. They support medication, nutrition, mobility, personal care, emotional wellbeing and coordination with services. They are, in many respects, the largest and least recognised workforce in the care system.

When carers become exhausted, isolated, financially stressed or physically unwell, the consequences are felt across the NHS. Emergency admissions, carer breakdown and residential care placements often follow.

Supporting carers is not a peripheral issue. It is a preventative health intervention.

Respite services, flexible support, financial recognition, training and proactive identification of carers within healthcare settings should be central components of reform.

Building community resilience

Prevention is frequently discussed as if it were simply about encouraging healthier lifestyles. In practice, community resilience is about practical support before crisis occurs.

That includes falls prevention, strength and balance programmes, dementia support, social prescribing, loneliness reduction and rapid-response home support.

People rarely become dependent overnight. Most experience a series of small losses: reduced mobility, bereavement, social isolation, difficulty shopping, medication confusion, declining confidence and reduced activity. Community-based support can interrupt that trajectory before it becomes a hospital admission.

Population health reform: preventing dependency before it begins

The deepest level of reform is population health.

Health and social care systems tend to intervene most intensively when people are already frail, disabled or seriously unwell. Yet the conditions that create dependency often develop over decades.

Prevention across the life course

Prevention should be understood as maintaining functional ability, not merely preventing disease.

That means supporting physical activity, nutrition, mental wellbeing, cognitive health, social connection and the early management of long-term conditions.

Reablement should become the default assumption. Every intervention should ask whether it can increase independence rather than simply manage dependency.

Tackling the wider determinants of health

Perhaps the most important lesson from population health research is that healthcare services account for only part of health outcomes.

Poverty, housing, education, employment, transport, social isolation and local environments all shape demand for health and care services. An older person living in poor housing, with limited income, no local transport and minimal social support is far more likely to experience deterioration that eventually reaches the NHS.

Social care reform therefore cannot be separated from housing policy, local government, employment support and community development.

Reducing inequalities

Health inequalities inevitably become care inequalities.

Communities with higher levels of deprivation often experience greater disability, earlier frailty, higher levels of multiple long-term conditions and greater pressure on both health and social care services.

A sustainable reform programme must allocate resources according to need and risk, not simply historical funding patterns. Reducing inequalities is not only a matter of fairness; it is also a strategy for reducing long-term demand on the NHS.

From dependency to capability

The strongest argument for reform is not that social care saves money, although effective social care can improve efficiency. The stronger argument is that good social care creates capability: it helps people retain the ability to live, participate and make choices within their own communities.

A good care system enables people to remain active, connected, confident and independent for longer. It supports families, strengthens communities and prevents avoidable deterioration.

The NHS will always be needed for acute illness, trauma, surgery and specialist treatment. But many of the pressures on hospitals are rooted in the gradual erosion of independence that occurs outside hospital walls.

The future of the NHS will not be secured solely through hospital expansion, workforce growth or technological innovation. It will depend equally on our ability to help people remain independent for longer within their homes and communities.

Social care should not be viewed as a service that steps in once dependency occurs. It should be recognised as a system that actively creates and protects independence. If reform focuses on that goal, we may finally move beyond managing growing demand and begin reducing it.

AI in the NHS: trusted guide or taskmaster? Human factors, clinical judgement and patient safety

AI in the NHS is no longer a future possibility. It is becoming part of everyday clinical practice. From documentation and diagnostics to workforce planning and clinical decision support, artificial intelligence is increasingly shaping how healthcare is delivered.

The real question is no longer whether AI has arrived. It clearly has.

The more important question is whether AI is strengthening clinical judgement and patient safety, or quietly reshaping them in ways we do not yet fully understand.

When I first wrote Artificial Intelligence in the NHS: Taskmaster or Trusted Guide, I argued that the future of healthcare should be guided by both the human heart and the digital brain, working together rather than one replacing the other.

Almost twelve months later, I think that question has become even more relevant.

What I perhaps underestimated was not the speed of technological change, but the speed with which AI would become woven into everyday clinical and organisational practice. That matters because AI is no longer simply another piece of technology sitting alongside healthcare. It is becoming part of the environment in which healthcare is delivered.

And once that happens, it becomes a human factors issue.

AI in the NHS has entered everyday clinical practice

AI is increasingly being used in clinical decision support, diagnostics, administration, documentation, workforce planning and operational management. NHS organisations are actively exploring its potential, and growing numbers of NHS staff now have access to AI tools.

That is a significant shift.

We are moving from asking whether AI works to asking what AI is doing to the way clinicians think, make decisions, communicate and take responsibility for those decisions.

That is a much more important question than simply whether the technology is accurate.

Why AI is a human factors issue

One of the strongest parts of my original article was the argument that AI should be considered not simply as a technology issue, but as a human factors issue.

Twelve months on, I think that has become even more important.

AI is already helping clinicians with documentation, information summaries, draft correspondence, data analysis and workflow support. On the face of it, that has to be a good thing.

If AI can reduce some of the administrative burden, surely that gives clinicians more time to do what they do best: care for patients.

But perhaps it isn’t quite that simple.

The question is no longer just whether AI is accurate. We also need to ask what it does to the way we work.

Does it improve our attention and situational awareness, or could we gradually become less attentive because we expect the system to pick things up for us?

Does it support professional judgement, or does it encourage us to defer to the recommendation?

Does it improve communication and learning, or are we simply becoming more efficient at completing tasks?

These are not really questions about the technology itself. They are questions about the interaction between the technology and the people using it.

Cognitive ergonomics and clinical decision-making

Reading this alongside some of my more recent thinking around human factors, psychological safety and the Dreyfus model of skill acquisition, I think there is another layer to the argument that is becoming increasingly important: cognitive ergonomics.

Good clinical practice has never been about simply following prompts.

Whether it is a doctor assessing a deteriorating patient, a nurse recognising that something isn’t quite right, or an Operating Department Practitioner anticipating that a situation is beginning to change, clinical expertise involves much more than following a predetermined pathway.

It involves interpreting the situation, recognising patterns, noticing weak signals and making sense of uncertainty.

This is where Dreyfus becomes particularly interesting.

Expertise is not simply about knowing more. With experience, clinicians develop an ability to recognise situations in context. They begin to understand what matters, what does not, and when something does not quite fit.

AI can undoubtedly be very good at recognising patterns within the information it is given. But healthcare is rarely as neat or predictable as the information contained within a computer system.

Patients do not always follow the textbook.

Sometimes the most important piece of information is the thing that does not quite fit.

Automation bias and professional judgement

Perhaps the biggest risk of AI in the NHS is not that AI occasionally gets things wrong.

The bigger risk is that clinicians gradually become less likely to question AI recommendations because the system is right often enough to earn their trust.

This is the problem of automation bias.

The issue may not be that the AI is wrong. It may be that we become less inclined to challenge it.

That creates an interesting paradox.

The more reliable AI becomes, the harder it may become psychologically to question it when it gets something wrong.

This is where professional judgement becomes so important.

If an AI system suggests one course of action but an experienced clinician thinks something does not quite add up, what happens?

In theory, the clinician should simply exercise their professional judgement.

But real healthcare does not always work like that.

Psychological safety and the confidence to challenge AI

The organisational environment matters.

If the culture says that AI is objective, efficient and evidence-based, while questioning it is seen as resistance, inefficiency or failure to follow the process, then we have created a problem.

This is where psychological safety becomes central.

A clinician should be able to say:

“The system is recommending this, but I don’t think it is right for this patient.”

That should not automatically be seen as resistance to technology.

It may actually be evidence of good clinical practice.

Psychological safety is therefore not just about whether people feel comfortable speaking up in meetings. It is also about whether they feel able to challenge the systems and processes that increasingly influence their work.

If we genuinely want AI to support professional judgement, people need to feel safe enough to disagree with it.

Otherwise, we risk creating a system where the technology does not have to force compliance.

The culture does it for us.

Productivity versus performance in healthcare

This brings me to another concern that has become increasingly prominent in my thinking about the NHS more generally: the danger of confusing productivity with performance.

AI offers enormous opportunities to increase productivity. It can reduce administrative work, speed up processes, automate repetitive tasks and potentially give clinicians more time with patients.

But doing more in less time does not automatically mean that we are performing better.

We can become more efficient at completing tasks without necessarily improving the outcome.

AI could reduce the time taken to produce a clinical letter while increasing the risk that something important is missed.

It could make information processing faster while reducing the opportunity for reflection.

It could increase throughput while adding cognitive workload elsewhere in the system.

It could improve utilisation without improving outcomes.

That does not mean we should not pursue those efficiencies.

It means we need to be careful about what we call success.

Efficiency, utilisation, productivity and performance are related, but they are not the same thing.

If the success of AI is measured primarily by how many tasks it completes, how much time it saves or how much activity it generates, we risk creating another version of the productivity paradox that exists elsewhere in the NHS.

We may become very good at measuring activity while becoming less good at measuring value.

Perhaps the better question is not simply:

How much work can AI save us?

but:

What does AI enable healthcare professionals to do better?

That changes the conversation.

The real test: is the human-AI system safe?

Of course we should care about time, cost and productivity. The NHS cannot ignore those things.

But they should sit alongside questions about quality, safety, clinical judgement, professional autonomy, teamwork, patient experience and resilience.

Does AI improve decision-making?

Does it reduce cognitive burden, or simply move it somewhere else?

Does it help clinicians recognise what matters?

Does it create more time for meaningful patient interaction?

Does it support learning and professional development?

Does it allow people to challenge the system when something does not look right?

And ultimately, does it improve things for patients?

These questions take us beyond thinking about AI governance as simply a technical exercise.

The question is not only whether the algorithm is safe.

It is whether the human-AI system is safe.

That brings me back to the question I asked almost twelve months ago.

Is AI going to become a taskmaster or a trusted guide?

Perhaps the answer is not really determined by the technology.

It depends on how we choose to use it.

An AI system designed primarily around productivity, standardisation and throughput may well become a taskmaster.

An AI system designed around professional judgement, human factors, psychological safety and patient outcomes has a much better chance of becoming a trusted guide.

That, I think, is where my thinking has changed over the past twelve months.

In 2025, I was asking how we could make sure the human heart and digital brain worked together.

In 2026, I think the question has become more fundamental.

How do we ensure that the way we introduce AI strengthens rather than erodes human expertise, professional judgement and patient safety?

Human factors helps us understand how people interact with the system.

Professional judgement reminds us why humans still need autonomy.

Psychological safety allows people to challenge what the system tells them.

And the distinction between productivity and performance reminds us that doing more is not necessarily doing better.

AI may be able to process information faster than any of us. It may eventually perform tasks we once thought required uniquely human capability.

But healthcare is not simply an information-processing exercise.

It is uncertain, complex and deeply human.

Perhaps, therefore, the real test of AI in the NHS is not whether the technology becomes more intelligent.

It is whether we remain intelligent enough in the way we choose to use it.

The ongoing challenges facing UK A&E departments: causes, consequences and practical solutions for NHS Trusts

Meta description: Why are UK A&E departments overcrowded? An evidence-based analysis of patient flow, bed capacity, staffing, corridor care and NHS emergency care reform.

Estimated reading time: 9–11 minutes

Executive summary

Accident and Emergency (A&E) departments are often described as being in crisis because of rising demand. The evidence suggests a different conclusion. Overcrowding, corridor care and prolonged waits are primarily symptoms of a wider failure in patient flow across hospitals and the health and social care system.

In March 2026, England recorded a record 2.43 million A&E attendances, yet 77.1% of patients were admitted, transferred or discharged within four hours — the best monthly performance since July 2021. Despite this improvement, performance remained well below the NHS constitutional standard of 95%. The central challenge is not simply the number of patients arriving at the front door, but the ability of hospitals to move patients safely through assessment, treatment, admission, discharge and community care.

This article argues that A&E performance should be viewed as a barometer of whole-system performance, not simply the performance of the emergency department itself.

---

A&E pressure in England: the key facts

Emergency care by the numbers

• 2.43 million A&E attendances in March 2026 (record monthly activity)
• 77.1% admitted, transferred or discharged within four hours
• NHS constitutional standard remains 95%
• Corridor care remains widespread across emergency departments
• Delayed discharge and limited inpatient capacity continue to drive overcrowding

Source: NHS England A&E statistics 2026.

---

Why are UK A&E departments overcrowded?

The most important question is not how many patients arrive in A&E.

It is what happens after a decision has been made to admit them.

When inpatient beds are unavailable, patients remain in A&E for prolonged periods. They occupy cubicles, trolley spaces and clinical resources that are needed for newly arriving patients. The consequences are visible across the entire emergency pathway: delayed ambulance handovers, crowded waiting rooms, corridor care and increasing clinical risk.

NHS England specifically monitors waits of more than four hours following a decision to admit, recognising that emergency department performance depends heavily on the availability of inpatient capacity.

The patient flow paradox

A&E departments can assess patients efficiently, but they cannot create inpatient beds.

As a result, overcrowding is often a hospital flow problem rather than an emergency department problem.

Key insight: An overcrowded A&E department is often the visible consequence of constrained inpatient capacity elsewhere in the hospital.

---

Demand is only part of the problem

Rising demand undoubtedly increases pressure on emergency departments. However, demand alone does not explain the scale of deterioration seen over recent years.

A hospital operating close to full occupancy has little ability to absorb unexpected emergency admissions. High utilisation may appear efficient on paper, but excessive occupancy reduces operational resilience.

This distinction is crucial.

The objective should not be to maximise bed occupancy. The objective should be to maintain sufficient capacity and flexibility to preserve patient flow when demand fluctuates.

---

Why bed occupancy matters

Hospitals frequently operate at occupancy levels that leave very little resilience.

When occupancy becomes too high:

• emergency admissions wait longer for beds,
• elective activity becomes more vulnerable,
• ambulance handovers are delayed,
• patients remain in A&E for extended periods,
• staff spend increasing amounts of time managing congestion rather than delivering planned care.

In operational terms, flow becomes the constraint.

---

Staffing shortages across the emergency pathway

Emergency care depends on a large multidisciplinary workforce.

Doctors, nurses, operating department practitioners, healthcare support workers, porters, radiographers, laboratory staff, pharmacists and bed management teams all contribute to the patient journey.

Staff shortages affect more than individual workload. They reduce the system’s ability to:

• respond to demand surges,
• complete investigations,
• transfer patients,
• discharge patients safely,
• maintain resilience during periods of exceptional pressure.

Persistent overcrowding also contributes to fatigue, burnout and retention problems, creating a self-reinforcing cycle in which pressure further reduces workforce capacity.

---

Patient flow: the problem beyond A&E

A&E departments cannot function efficiently if patients cannot move onwards.

The typical pathway is straightforward:

1. Arrival
2. Assessment
3. Investigation
4. Treatment
5. Decision to admit
6. Transfer to an inpatient bed

The bottleneck usually occurs at step six.

If an inpatient bed is unavailable, the patient remains in A&E. Congestion then spreads backwards through the system.

Evidence from RCEM

RCEM’s 2025 survey of 58 Type 1 emergency departments found:

• 19% of patients were being cared for on trolleys or chairs in corridors
• 34.5% of departments reported patients being cared for in ambulances outside the emergency department

This demonstrates that congestion rapidly affects every stage of the emergency care pathway.

---

The discharge bottleneck

Beds become available only when patients can leave hospital safely.

A patient may be medically fit for discharge but still require:

• rehabilitation,
• intermediate care,
• domiciliary care,
• residential care,
• housing support,
• community nursing.

NHS England now measures patients who no longer meet the criteria to reside, recognising that delayed discharge is usually a system failure rather than an individual patient problem.

When discharge is delayed, inpatient beds remain occupied longer than necessary. This limits admissions from A&E and increases emergency department overcrowding.

---

How social care affects A&E performance

The relationship between social care and emergency care is often underestimated.

A patient waiting for home care is also occupying a hospital bed.

That occupied bed may prevent the admission of an emergency patient from A&E.

RCEM’s January 2026 survey provides particularly strong evidence:

• 92.6% of emergency department clinical leads identified social and community care discharge problems as a major cause of overcrowding
• 79.0% identified organisational discharge delays
• 74.1% identified a lack of alternatives to admission

This is one of the clearest demonstrations that A&E overcrowding cannot be solved by A&E alone.

---

Corridor care is a symptom of system failure

Corridor care is the most visible manifestation of inadequate system capacity.

It creates risks relating to:

• privacy,
• dignity,
• infection prevention,
• clinical observation,
• communication,
• medication safety,
• patient experience.

RCEM’s research found that corridor care was also affecting staff wellbeing, with 44% of clinical leads reporting negative effects on staff mental health.

Temporary workarounds risk becoming normal practice.

That should concern every NHS organisation.

---

Why resilience matters

One of the least visible consequences of sustained overcrowding is the loss of resilience.

A hospital permanently operating close to maximum capacity has limited ability to respond to:

• winter pressures,
• infectious disease outbreaks,
• major incidents,
• multiple trauma,
• sudden increases in emergency admissions.

RCEM’s 2025 survey found that while 88.7% of departments said their major incident plans required the emergency department to be cleared, only 9.5% believed this would be achievable.

A system without resilience is not operating efficiently.

It is operating precariously.

---

What NHS policy is trying to achieve

NHS England increasingly recognises emergency care as a whole-system flow problem.

The Urgent and Emergency Care Plan 2025/26 emphasised:

• improving patient flow,
• same-day emergency care,
• urgent treatment centres,
• community capacity,
• discharge improvement,
• reduction of corridor care.

For 2026/27, NHS England set an ambition for 82% of patients to be admitted, transferred or discharged within four hours, supported by the rollout of the Model Emergency Department.

These initiatives acknowledge an important reality: A&E performance depends on how the entire system functions around the clock.

---

Five practical reforms that would improve A&E performance

1. Real-time hospital flow coordination

Dedicated operational teams should manage bed capacity, admissions and transfers across the whole hospital.

2. Seven-day discharge capability

Discharge planning, pharmacy, transport and community coordination should operate consistently throughout the week.

3. Strengthened intermediate care

Greater community and rehabilitation capacity would reduce delayed discharge and avoid unnecessary admissions.

4. Dedicated transfer coordinators

Specialist coordinators with real-time bed visibility could reduce transfer delays and improve patient movement.

5. Flow-focused performance metrics

Hospitals should measure decision-to-admit waits, discharge delays, transfer times and occupancy-related constraints alongside traditional A&E targets.

---

The real question

The debate about A&E often focuses on a single issue:

• demand,
• staffing,
• funding,
• bed numbers,
• social care,
• management.

The evidence suggests these factors are interdependent.

Pressure moves backwards through the system.

When discharge slows, beds become unavailable.

When beds become unavailable, admissions slow.

When admissions slow, A&E becomes overcrowded.

When A&E becomes overcrowded, ambulance handovers are delayed.

Eventually, the entire emergency care pathway is affected.

The question is therefore not simply:

How do we make A&E more efficient?

The better question is:

How do we make the whole patient pathway work better?

A&E departments are not isolated services. They are the front door to a much larger health and social care system.

Until patient flow improves across that entire pathway, overcrowding, corridor care and prolonged waits will remain symptoms of the same underlying problem.

The patients will continue to wait.

---

Frequently asked questions

Is A&E overcrowding caused mainly by inappropriate attendances?

The evidence suggests that constrained inpatient capacity and delayed admissions are often more important drivers of overcrowding than inappropriate attendance alone.

Why do patients wait in A&E after a decision to admit?

Because an appropriate inpatient bed is not immediately available.

What is corridor care?

Corridor care refers to patients receiving treatment in corridors, on trolleys or chairs, because clinical spaces within the emergency department are full.

Why is discharge so important?

Discharge creates inpatient bed capacity. Without timely discharge, emergency admissions cannot move from A&E into wards efficiently.

What is the single biggest improvement that could reduce A&E overcrowding?

Improving whole-system patient flow, particularly inpatient capacity, discharge processes and community care availability.

---

References

1. NHS England (2026). A&E Attendances and Emergency Admissions 2026–27.
2. NHS England (2026). Performance Report 2024/25.
3. Royal College of Emergency Medicine (2025). Research Reveals Devastating Reality of So-Called Corridor Care.
4. NHS England (2026). Patients Who No Longer Meet the Criteria to Reside.
5. Royal College of Emergency Medicine (2026). Clinical Lead Snap Survey – 12 January 2026.
6. Royal College of Emergency Medicine (2025). Emergency Departments Too Crowded to Cope with a Disaster.
7. NHS England (2025). Urgent and Emergency Care Plan 2025/26.
8. NHS England (2026). The Model Emergency Department: High Performing Urgent and Emergency Care Pathways.

---

What Does an Emergency Anaesthetic Practitioner Do? A Frontline NHS Perspective

My job title is Emergency Anaesthetic Practitioner (EAP). I am a member of a seven-strong team that provides 24-hour cover, 365 days a year. Although all of us come from an Operating Department Practitioner (ODP) background, the role is not exclusive to ODPs and is open to appropriately trained and experienced nursing colleagues. The EAP role has existed within the Trust for approximately 15 years, and most of the team have been part of it since the beginning.

The EAP role developed in response to the impact of the European Working Time Directive (EWTD) on anaesthetic training, particularly the former Senior House Officer (SHO) grade. SHOs had traditionally provided airway support as part of the cardiac arrest team, but changes to working patterns left a gap in out-of-hours cover between 8 pm and 8 am. The Department of Anaesthesia played a central role in developing the EAP position, enabling experienced ODPs to provide dedicated airway support across the hospital.

My professional background is as an Operating Department Practitioner (ODP), having originally qualified with the City and Guilds 752 Operating Department Assistant qualification in 1990. This predates both the National Vocational Qualification (NVQ) framework and the modern degree qualification for ODPs.

My role within the hospital has two distinct but closely connected elements. As an ODP, I provide skilled support as part of the anaesthesia team and deliver planned, individualised patient care within the operating theatre suite, working across a broad range of surgical specialities. As an EAP, I extend that support beyond the theatre environment, providing advanced airway assistance and patient care wherever it is required, both during routine hours and out of hours.

That may involve responding to emergencies on medical wards, in interventional radiology or cardiology, critical care areas, and other clinical environments across the hospital. I am also an integral member of the Trauma Team in A&E Resus and the Cardiac Arrest Team, where my role is to support the anaesthetist or, when appropriate, manage the airway independently.

To support this role, I hold Resuscitation Council UK Advanced Life Support (ALS) certification and have recognised advanced airway training, with competence in basic airway management, the use of supraglottic airway devices, and both direct and indirect laryngoscopy. I always work within my scope of practice and level of competence, with appropriate senior support available whenever required.

The Trust also recognises the EAP role as a non-medical referrer for diagnostic X-ray imaging, specifically to confirm the position of a central intravenous catheter and identify potential complications.

As both an EAP and an ODP, I am expected to maintain competence in intravenous cannulation and intravenous drug administration in accordance with Trust policy.

The multidisciplinary teams we support value both the EAP role and the experience that team members bring to challenging clinical situations. Doctors in training who rotate through the Trust often recognise the continuity and stability that the EAP team provides. A significant part of our role is helping to create and maintain a safe environment for the administration of anaesthesia and advanced airway management outside the normal theatre setting.

Whatever role I am performing, and wherever I am working, the patient remains at the centre of everything I do.

Completing the perioperative safety pathway: a case for an Emergence Time Out and post-extubation Sign-Out within the NatSSIPs 2 framework

9 minute read

The World Health Organization (WHO) Surgical Safety Checklist has undoubtedly made surgery safer. The structured process of Sign In, Time Out, and Sign Out has improved communication, teamwork, and situational awareness in operating theatres across the world. 

After many years working in anaesthesia, spanning a broad spectrum of surgical specialties, there is a question that, following the introduction of the WHO checklist, has increasingly occupied my mind. 

When do we consider the operation to be safely complete? 

For the surgeon, we know that that moment is often considered to be the completion of the procedure. For the anaesthetist, it is usually later. However, a patient whose surgery has finished but who has not yet safely emerged from anaesthesia has not completed the perioperative journey. 

That distinction matters. 

The extubation safety gap 

For many patients, extubation represents one of the most physiologically vulnerable phases of the perioperative pathway. The transition from controlled ventilation to spontaneous breathing can rapidly expose airway obstruction, inadequate reversal of neuromuscular blockade, aspiration, laryngospasm, bronchospasm, hypoventilation, and haemodynamic instability. 

These complications often develop within minutes of extubation, precisely when the theatre team may be completing documentation, preparing equipment, arranging patient transfer, or mentally transitioning to the next case. 

In most operating theatres, the WHO sign out is completed before extubation. As a result, the final formal multidisciplinary communication occurs before one of the highest-risk phases of anaesthesia has been completed. 

This is not simply a question of checklist timing. It is a human factors issue. 

Once the WHO sign out has concluded, the structured team focus that existed during the operation begins to dissipate. Surgeons turn to documentation, scrub staff begin instrument processing, circulating staff prepare for transfer, and attention gradually shifts away from the patient towards task completion. The shared situational awareness that the WHO checklist is designed to create is no longer actively reinforced across the multidisciplinary team. 

The consequence is that the patient may enter a predictable period of increased physiological risk without the same level of structured team alignment that exists during other critical phases of care. 

A NatSSIPs 2 opportunity, not another checklist 

This is where NatSSIPs 2 becomes particularly relevant. 

NatSSIPs 2 encourages organisations to develop Local Safety Standards for Invasive Procedures (LocSSIPs) that address specialty-specific risks and strengthen safety throughout the entire procedural pathway. 

That changes the conversation. 

I am not suggesting replacing the WHO checklist or introducing another national checklist. I am suggesting that an Emergence Time Out immediately before extubation, followed by a brief post-extubation Sign-Out, should be considered an anaesthesia-specific LocSSIP that extends the existing sequential safety process into the emergence phase of anaesthesia. 

This is not about creating more paperwork. It is about closing a recognised safety gap using the governance framework that NatSSIPs 2 already provides. 

The missing Time Out before emergence 

The WHO checklist recognises that high-risk phases of care benefit from a structured pause. The Time Out immediately before incision aligns the team, confirms the plan, identifies hazards, and establishes a shared mental model before a critical phase begins. 

What is striking is that there is no equivalent structured pause before emergence from anaesthesia. 

Extubation is not a single technical act. It is the beginning of a complex physiological transition from controlled ventilation to spontaneous breathing and from pharmacological support to independent airway protection. 

The period immediately before extubation is therefore analogous to the period immediately before incision. It is a predictable transition into a phase of increased risk. 

Yet in many operating theatres, emergence begins without a formal multidisciplinary pause. The anaesthetist may have a clear extubation strategy, but that strategy is not routinely shared with the wider theatre team. Equipment may be available but not explicitly confirmed. The recovery practitioner may not yet be fully integrated into the emergence plan. Potential airway concerns, aspiration risk, haemodynamic instability, or the possibility of re-intubation may not have been collectively acknowledged. 

An Emergence Time Out would provide that missing pause. 

Immediately before extubation, the anaesthetist would briefly confirm the extubation strategy, airway risk, oxygenation and ventilation status, reversal of neuromuscular blockade, haemodynamic concerns, required equipment, and the postoperative destination. 

More importantly, it would ensure that the entire theatre team shares the same mental model of the patient’s condition and is prepared for immediate intervention if extubation is unsuccessful. 

This is not fundamentally about the checklist itself. It is about creating a shared understanding of what is about to happen and what the team will do if it does not go to plan. 

The missing stop moment 

The Emergence Time Out addresses preparedness before extubation. A post-extubation Sign-Out addresses confirmation after extubation. 

This would be a brief structured discussion led by the anaesthetist immediately after successful extubation and before transfer to recovery. 

The discussion would simply confirm that: 

• the airway is patent; 

• ventilation and oxygenation are satisfactory; 

• neuromuscular blockade has been adequately reversed where appropriate; 

• haemodynamic stability is acceptable; 

• analgesia and antiemetic plans are in place; 

• any airway concerns have been communicated; 

• the recovery team understands the immediate priorities and escalation plan. 

In most cases, this would take less than a minute. 

The crucial point is not the wording of the checklist. The crucial point is the shared situational awareness it creates before responsibility transfers from the anaesthetic team to recovery. 

Recovery handover starts before the patient leaves theatre 

One of the recurring themes in patient safety is that harm often occurs at the interfaces between teams. 

Theatre to recovery is one of those interfaces. 

Recovery practitioners frequently receive excellent handovers, but the quality of communication can vary depending on workload, interruptions, staffing pressures, and the complexity of the case. 

A formal post-extubation sign-out helps ensure that critical information about airway difficulty, emergence complications, respiratory risk, or anticipated deterioration is communicated consistently. 

It is important to distinguish this proposal from the recovery handover itself. 

Recovery handover is a transfer of information between the anaesthetic team and recovery practitioners. Post-extubation sign-out is a multidisciplinary confirmation that emergence has been achieved safely before that transfer occurs. 

In other words, it creates a structured point at which the whole theatre team acknowledges that the patient has not only completed surgery but has also safely emerged from anaesthesia and is ready for transfer with a shared understanding of ongoing risks. 

This is as much about human factors as airway management 

During emergence, the anaesthetist is managing multiple competing tasks: airway, ventilation, oxygenation, analgesia, haemodynamic stability, documentation, and preparation for the next case. 

A structured stop moment reduces the risk of omission. 

More importantly, it preserves the collective situational awareness that can be lost once the formal WHO sign out has concluded and attention begins to fragment across multiple tasks. 

It also gives every member of the theatre team permission to speak up before the patient leaves the theatre. 

That is entirely consistent with the human factors principles that underpin both the WHO Surgical Safety Checklist and NatSSIPs 2. 

The strength of the WHO checklist has never been the physical checklist itself. Its greatest contribution has been the creation of moments in which communication becomes deliberate, expectations are aligned, and hierarchy is temporarily reduced in the interests of patient safety. 

Emergence from anaesthesia deserves the same discipline. 

A practical quality improvement project 

The advantage of framing post-extubation sign-out as a LocSSIP is that it can be implemented through existing governance structures. 

It could be introduced using a simple Plan–Do–Study–Act (PDSA) cycle and evaluated through measures such as: 

• unplanned airway interventions; 

• postoperative hypoxia; 

• re-intubation rates; 

• recovery handover omissions; 

• staff perceptions of communication and situational awareness. 

The intervention requires no new national documentation, can be incorporated into existing LocSSIP processes, and is readily auditable through routine quality improvement methodology. 

Equally importantly, it is reversible. If a pilot demonstrates no measurable benefit, the process can be modified or discontinued. If it improves communication, team preparedness, or patient safety, it provides a scalable model that other organisations could adopt. 

That is precisely how many successful safety interventions have evolved: through local innovation, structured evaluation, and iterative improvement. 

Completing the perioperative safety pathway 

I do not believe the WHO checklist has failed. Quite the opposite. It has transformed perioperative safety by recognising that structured communication improves outcomes. 

The question is whether our structured safety process should end when the operation finishes, or when the patient has safely emerged from anaesthesia and been safely transferred to recovery. 

NatSSIPs 2 views safety as a sequential process. 

From that perspective, the absence of a structured emergence pause and a structured post-extubation confirmation is not a criticism of the WHO Checklist. It is an opportunity to extend its underlying principles into a phase of care that remains physiologically vulnerable and heavily dependent on team coordination. 

If the purpose of the WHO Surgical Safety Checklist is to protect patients through structured communication at critical moments, then emergence from anaesthesia deserves to be recognised as one of those moments. 

An Emergence Time Out and post-extubation Sign-Out may simply be the logical next step in completing the perioperative safety pathway. 

GRAPHIC: Safe Surgical Checklist: Extending the SAFETY PATHWAY through emergence and recovery graphic rendered using ChatGPT AI (generate a referenced surgical safety checklist graphic showing: Sign In → Time Out → Procedure → Emergence Time Out → Extubation → Post-extubation Sign-Out → Recovery) Authors conceptual image (05/08/2026)

Beyond Aviation: What Scuba Diving and Skydiving Can Teach Healthcare About Human Factors

5-minute read

There is little doubt that aviation has influenced patient safety, shaping many of the checklists, communication strategies and safety cultures we see across healthcare today.

The events surrounding the death of Elaine Bromiley are now well known, as is the remarkable work undertaken by her husband, Martin Bromiley OBE. Drawing on his extensive background in the aviation industry, he has dedicated himself to helping healthcare understand the importance of Human Factors, teamwork and systems thinking in the hope of preventing similar tragedies from occurring again.

Jordan Nicholls recently argued in a LinkedIn article that aviation may not be the most appropriate industry against which to compare healthcare because the two operate under fundamentally different conditions. His article prompted me to consider a different question altogether.

His argument is not that aviation has nothing to offer—far from it. Rather, aviation and healthcare operate in fundamentally different environments. Whilst aviation provides valuable lessons in Human Factors, teamwork and safety culture, meaningful improvement comes not from trying to replicate the aviation model, but from recognising healthcare’s unique complexity and adapting proven safety principles to fit that reality.

It raises an equally interesting question. Had Martin Bromiley’s background been in professional SCUBA diving or skydiving rather than aviation, might the NHS have adopted a different perspective on Human Factors?

Both activities demand technical competence, disciplined preparation, calm decision-making under pressure and an acceptance that human performance is influenced by stress, fatigue and the environment. Perhaps they also offer lessons that are just as relevant to healthcare—not because they mirror clinical practice, but because they acknowledge that high-risk professions all face the same challenge: minimising the consequences of inevitable human fallibility.

Rather than asking which industry provides the best comparison, perhaps we should ask what each can teach us.

Lessons from Scuba Diving

Scuba diving takes place in an environment where communication is restricted, visibility can change rapidly and there is little margin for error. Success depends not only on technical competence but on discipline, preparation and trust.

One important lesson is the danger of normalisation of deviance. A diver who repeatedly skips a safety check because “nothing happened last time” gradually begins to accept unsafe practice as normal. Healthcare is no different. Whether bypassing patient identification, omitting hand hygiene or taking shortcuts during equipment checks, repeated success can create a false sense of security until circumstances eventually align and harm occurs.

Divers are also taught a simple mantra:

Stop. Breathe. Think. Act.

Rather than reacting instinctively, they pause long enough to regain control before making decisions. It is a principle equally applicable during an airway emergency, major haemorrhage or rapidly deteriorating patient, where panic can quickly narrow thinking and impair judgement.

Buddy checks offer another valuable lesson. Before entering the water, divers systematically check one another’s equipment. The purpose is not to question competence but to recognise that everyone, regardless of experience, can overlook something. Healthcare already embraces this philosophy through surgical safety checklists and medication verification, yet there is perhaps more we can do to encourage peer checking without it being perceived as criticism.

Lessons from Skydiving

Skydiving offers a different perspective on Human Factors.

Every skydiver has a predetermined decision altitude. If a parachute malfunction has not been resolved by that point, there is no further debate—the reserve parachute is deployed. The decision has already been made before the emergency occurs.

Healthcare frequently faces similar moments.

In a Can’t Intubate, Can’t Oxygenate emergency, for example, the challenge is often not knowing what to do but recognising when to stop repeated attempts and move immediately to an emergency front-of-neck airway. Defining these decision points in advance may reduce hesitation when every second matters.

Skydivers also practise emergency procedures repeatedly until they become almost instinctive. They understand that stress degrades memory, attention and decision-making. The same principle underpins simulation training in healthcare. Repetition allows clinicians to rely on well-rehearsed actions when cognitive capacity is compromised.

Finally, experienced skydivers establish personal minimums before they ever leave the ground. They decide in advance the conditions under which they will not jump, whether because of fatigue, weather or equipment concerns.

Should healthcare encourage the same mindset?

Recognising when fatigue, workload or experience limits safe practice should not be viewed as weakness. Instead, it should be seen as an important patient safety intervention, allowing support to be sought before a situation escalates.

Looking Beyond Aviation

Perhaps the lesson is not that aviation is the wrong comparison. Rather, it is that aviation is only one comparison.

Healthcare shares characteristics with many high-risk professions, each of which has developed practical ways of managing uncertainty, complexity and human performance under pressure.

Aviation has given healthcare invaluable insights into teamwork, communication and safety systems. Scuba diving reminds us of the dangers of complacency and the value of structured cross-checking. Skydiving demonstrates the importance of predetermined decision points, repetitive practice and recognising personal limits.

Instead of searching for a single industry to emulate, perhaps healthcare should become better at learning from them all.

After all, every high-risk profession is ultimately trying to answer the same question:

How do we reduce the impact of inevitable human fallibility when the consequences of getting it wrong are so high?

That may be the most valuable Human Factors lesson of all.

Acknowledgements

This article was inspired by the pioneering work of Martin Bromiley OBE on Human Factors in healthcare and by Jordan Nicholls’ recent LinkedIn article questioning whether aviation remains the most appropriate comparator for healthcare. Any comparisons with scuba diving and skydiving, and the opinions expressed, are my own.

GRAPHIC: “BEYOND AVIATION: What Scuba Diving and Skydiving Can Teach Healthcare About Human Factors“: graphic image rendered using ChatGPT AI (create a suitable image to support the article) 15/07/2026