FREE CHAPTER from ‘A Practical Guide to the Law of Medical Treatment Decisions – Second Edition’ by Ben Troke

09/09/2025

CHAPTER TWO – CHOICES AVAILABLE

  • Available options
  • Assisting suicide
  • DNACPR
  • Resource allocation
  • Discharge from hospital
  • Compulsion


Introduction

The core of this book is about how patients (and doctors) make choices about medical treatment (and, by extension, when and how decisions should be made in the “best interests” of patients who “lack capacity” to make their own decision – each of which are technical terms explained in Chapters 4 and 5).

But it is important to be upfront about one key issue – just because you want a particular treatment, it does not necessarily mean it will be available to you. Subject to the NHS (or private) funding being available, the clinician decides what options are available to you, and from those options, if you are able to make your own decision, you can consent / refuse or choose. Our autonomy in this context extends only to making decisions among the options which are actually available to us. We can choose option B or option C, if offered, even if the strong medical advice is in favour of option A, or we can refuse them all. But we cannot demand an option D if it is not being made available to us. Essentially, if we have capacity to do so, we can choose from the menu of options, however unwisely, or walk away, but we cannot set the menu ourselves. This works on several levels.


Clinical Indication

For an individual clinician, it is well established that a patient cannot force treatment to be offered which is not clinically indicated. To exaggerate to absurdity, I cannot walk into my local hospital, fit and well, and demand a knee replacement operation. If a clinician recommends this for me, however, I can choose whether to accept it or refuse, or (perhaps) choose between different types of operation being offered or a conservative option. (The same goes, by the way for the admission and accommodation in hospital that might come with any investigations or treatment being offered – see below).

The leading case is the tragic situation of Leslie Burke,[1] who was 45 years old when his case reached the Court of Appeal in 2005. Mr Burke suffered from a congenital degenerative brain condition called spino-cerebellar ataxia, which confined him to a wheelchair. In time, as his condition inevitably deteriorated, he would come to need clinically assisted nutrition and hydration (“CANH” – putting nutrition and fluids directly into his stomach by a tube) to survive.

He objected to guidance issued by the General Medical Council (GMC) in 2002,[2] which put the decision about giving or continuing CANH in the hands of the doctors, without regard for what the patient wants. Mr Burke wanted CANH to be given whenever he needed it, and though there was no suggestion that he would not be given CANH in due course, he said that the GMC guidance was unlawful because it gave that decision to the doctors rather than to the patient. Mr Burke won the declarations that he wanted in the first instance in the High Court, but the Court of Appeal reversed this and upheld the GMC guidance as lawful:

“Autonomy and the right of self-determination do not entitle the patient to insist on receiving a particular medical treatment regardless of the nature of the treatment. Insofar as a doctor has a legal obligation to provide treatment this cannot be founded simply upon the fact that the patient demands it”.[3]

Just so, at a more mundane level, my GP is under no obligation to prescribe antibiotics for me, just because I want them. They must be, in the view of the clinician, clinically indicated.

It is the clinical indication, a gangrenous foot for example, as well as the surgeon’s professional training and registration and the consent of the patient, that together make the amputation lawful which would otherwise be grievous bodily harm and likely to attract a lengthy prison sentence.

Consent of the “victim” in itself is no defence to such an assault, given the public interest in protecting people generally.[4] In a “healthcare” context, body modification carried out beyond the scope of proper clinical practice cannot be justified by a patient’s consent alone. In May 2024, the self-styled “Eunuch-maker”, Marius Gustavson, was imprisoned for life, with a minimum of 22 years, for a series of offences of grievous bodily harm with intent, for his practice of performing mutilating procedures, notwithstanding the consent of his victims; and equally three other men were sentenced for committing GBH on him with his own consent[5].

Subject to reform that may be imminent, at the time of writing, we have seen a similar point in the vexed law surrounding assisting suicide. Though suicide itself was decriminalised by the Suicide Act 1961, it remained a criminal offence to intentionally encourage or assist the suicide[6]. Time and again, the courts were asked by individuals to overturn what was said to be the discriminatory effect of this on those who may be unable to end their own lives without someone else’s help, for example due to a disability, from Diane Pretty in 2001[7] through Debbie Purdy in 2008[8] and Tony Nicklinson in 2014.[9]

Though the courts always refused to change the law, insisting that this must be a matter for parliament, there was a victory of sorts after the Purdy case when it was held unlawful that the Director of Public Prosecutions had given no guidance at all on the circumstances in which the offence of assisting suicide would be prosecuted, for example in taking a loved one at their insistence to Dignitas in Switzerland. In response, the DPP (Sir Keir Starmer QC by then, as it happens) published guidance in 2010, which has since been updated,[10] setting out the factors which might tend towards or against prosecution.[11]

For these purposes, the point is that wanting something does not necessarily mean that it must be made available to you, or even that giving you it would be lawful.


Resource allocation is not a best interests decision

Chapters 4 and 5 will explain capacity and best interests decisions more fully, but it is really important to get one thing straight here.

Lack of capacity, briefly, means that a patient is unable to make a decision for themselves, when it is their decision to make, obliging us to make a best interests decision for them from the options which are available. The decision what options to make available is not the patient’s own decision to make, but a prior decision made by the clinicians or NHS. That remains true regardless of whether the patient has capacity to choose among them or if, instead, a best interests decision will be made.

Lack of capacity for any particular decision does not in itself entitle that patient to preferential access to resources. The Mental Capacity Act 2005 (the MCA) operates to prevent a patient who lacks capacity for any decision being disadvantaged as a result. It is not intended to give them a trump card or allow them to queue jump.

“Best interests”, defined by the MCA, is the legal test for how to pick among the available options for a patient who cannot choose for themselves. It is not appropriate, and obviously confusing, to talk about which options should be offered (as opposed to which of the available options should be selected) in terms of “best interests”. For the clinician’s decision what options to offer, the patient’s “capacity” and “best interests” are totally irrelevant.

So, a case of a patient with a learning disability who cannot make decisions about his ongoing renal dialysis may raise the challenging legal question of whether it is in his best interests for dialysis to be forced on him, against his wishes, when he is non-compliant, because the dialysis is actually available. But it does not follow that he must immediately be offered a kidney for transplant because it would be in his abstract “best interests”. There is a scarce resource and a waiting list limiting access to that. A best interests decision can only be made among options which are actually available now, just as the choice of a patient with capacity is limited in the same way.

The leading case is MN v A CCG.[12] MN was a profoundly disabled young man, in his early twenties at the time of the Supreme Court judgment in 2017. He was in a residential setting to meet his extensive care needs, but a dispute arose between his parents and the CCG funding his care. In essence, his parents wanted to have greater involvement in provision of his personal care and for there to be more frequent visits for MN back to the family home, arguing that these things were in MN’s best interests. But these things required either more staff from the placement or more funding from the CCG, or both, which were not on offer.

Lady Hale gave the leading judgment in the Supreme Court, noting that the Mental Capacity Act only permits the Court of Protection to make a decision that the patient could have made for himself:

“[The court] has no greater power to oblige others to do what is best than P would have himself. This must mean that, just like P, the court can only choose between the “available options”.[13]

The point is perhaps even clearer in a clinical context in the case of AVS[14]. A 57 year-old man had become progressively unwell with Creutzfeldt Jakob’s Disease, causing a serious brain injury and a very poor prognosis. He was thought to be in a persistent vegetative state. His brother – a solicitor as it happens – was insistent that an experimental treatment be given: a pump implanted under the skin in his chest delivering a drug – Pentosan Polysulphate) into AVS’s brain, and he found a doctor willing to provide that treatment, at least for the first three months until the pump failed and was due to be replaced. There had been no benefit evident at all, and no clinician was willing to replace it to continue that treatment. AVS’s brother brought the case to court arguing that the treatment should continue, in his brother’s best interests.

The court rejected this argument strongly, a decision that was upheld in the Court of Appeal. Essentially, the court said, it is nonsense to try to frame this argument in terms of best interests unless of until it is actually an option that is available:

“It is trite that the court will not order medical treatment to be carried out if the treating physician/surgeon is unwilling to offer that treatment for clinical reasons conscientiously held by that medical practitioner….”

“…[i]f there are clinicians out there prepared to treat the patient then the patient will be discharged into their care and there would be no need for court intervention. If there is no-one available to undertake the necessary operation the question of whether or not it would be in the patient’s best interests for that to happen is wholly academic and the process should be called to a halt here and now.”

Do not treat “best interests” as if a blank piece of paper or a blank cheque. It is a matter of choosing among the options that would have been available to the patient if he had capacity to make the decision for himself (see Chapters 4 & 5).

And as the philosopher Jagger noted, you can’t always get what you want.


CPR / DNACPR

One area where this can all get especially blurry is the provision of CPR (cardio-pulmonary resuscitation).

A DNACPR[15] form is a document recording a decision that in the event of cardio-pulmonary collapse no attempt at CPR should be made. But whose decision is this, and why would it be made?

There are actually (at least) three different routes to a DNACPR:

  • A patient can decide that they don’t want CPR
  • A doctor can decide that they wouldn’t give CPR, because it wouldn’t work; or
  • A doctor can decide that CPR could be done, but it is, on balance, best not to do it.

It really makes a difference which of these lines of thought lies behind any DNACPR notice, and it is unhelpful if this is not clear.

Let’s start with you, as the patient. If your heart stops, do you want every effort to be made to restart it? You may not have given this much thought, or it is possible you have a fixed and well-informed view already. But it is reasonably likely that any member of the public has a very distorted idea of the prospects of success, and what is really involved.

The likelihood of short-term survival from CPR for a cardiac arrest is said to be no better than 20%, even if it happens in hospital, even then with a likely slow decline and lots of collateral damage. The odds of longer-term survival neurologically intact are “abysmal”.[16] For children the prospects are woeful, with just a 5% survival rate from CPR after cardiac arrest and a flat trace, and only 1% of those would avoid grave brain damage.

Outside hospital, of course, we fare even worse. Of around 30,000 cardiac arrests outside hospital in the UK each year where emergency medical services attempt resuscitation, only 1 in 11 patients in England (1 in 12 in Scotland) survive until discharge from hospital.[17]

But the public’s impression of our chances is skewed by what we see. A study in 2018[18] reviewed 30 consecutive episodes of each of Grey’s Anatomy, Holby City and Casualty broadcast between October 2012 and November 2013 and found that these contained 36 attempts at CPR, with a wildly implausible success rate (i.e. achieving spontaneous return of circulation) of 62%.

Still, even a small chance might feel well worth a go, if trying CPR does no harm. But again the reality is not always well known, or fully portrayed in the media.

As described by Rachel Clarke:[19]

“A crash team descends like a tempest, pumping the chest, shocking the heart, slugging in the adrenaline, doing whatever it takes to try to claw back a life interrupted. Resuscitation is violent, bone-crunching work. During CPR a dead body is essentially assaulted by a team of doctors who hope to achieve its resurrection. But if that hope is vain from the outset – perhaps because the patient is too elderly, frail and overwhelmed with illness ever to stand a hope of having their heartbeat restored – then the ending it inflicts is invariably ugly, brutal and devoid of dignity”.

Sometimes,

“…. our heart stops because it is time for us to go. CPR in these cases is at best pointless, at worst a grotesque indignity”.[20]

And, she observes, there is more teaching on how to do CPR than on whether to do it, or how such decisions should be made.

So there are perfectly good reasons why a patient might well decide that they do not want CPR attempted in certain circumstances[21] and this, of course, must be respected. Normally, this would be recorded on a DNACPR form which would, you’d hope, be suitably prominent in your records to pre-empt any default to CPR, or else it’d be pretty worthless.[22]

However, as we’ve already explored, this doesn’t work the other way – a patient cannot demand that CPR be given in future.

It is also open to the clinicians to take a view that any attempt at CPR would be futile (i.e. there would be little or no real prospect of success) and so they would be unwilling to do it. In that case a DNACPR form may be appropriate but it would be at the doctors’ initiative rather than the patient’s.

However, though this is squarely in line with the principle embodied in Burke above that a doctor cannot be forced to provide treatment against their judgement, a decision of this gravity cannot properly be made entirely unilaterally and without consultation with the patient.

Janet Tracey was 63 when she suffered a broken neck in a car crash, having recently been diagnosed with terminal cancer, and was admitted to Addenbrookes Hospital in Cambridge. She sadly died little more than two weeks later. In that time, a DNACPR notice was put into her records, withdrawn, and then reinstated, and when the family later sued the hospital trust there had to be a seven-day hearing with a number of witnesses to sift through the conflicting evidence about the extent of discussions with Mrs Tracey or her family about this in that period. The bottom line, the Court of Appeal later held,[23] was that the hospital had breached her rights[24] by “failing to involve her in the process” by which a DNACPR was instated. Though the court reiterated the point in Burke that it is not open to a patient to compel a doctor to provide treatment which is not felt to be clinically indicated, the patient should be “consulted” about a decision such as this, “unless … the patient will be distressed by being consulted and that distress might cause the patient harm”.

Such “consultation” should not be misunderstood. It is more informing than asking. It does not empower a patient (or their family) to overrule the clinical decision, of course, but it does enable them to question it and, if appropriate, to seek a second opinion. This protects “the autonomy, integrity, dignity and quality of life of the patient”.

A subsequent case has confirmed that the same duties of consultation with the patient or, where appropriate their family, apply where the patient happens to lack the mental capacity to make their own decisions about the treatment.[25]

So it is fairly clear where you stand if the doctor[26] won’t do CPR, or if the patient doesn’t want it. But this can gloss over a more difficult middle ground where the doctor might be prepared to try CPR, but thinks it is better not to.

Of course, the Tracey consultation requirement applies just as much here, and would hopefully flush out and challenge any assumptions being made by the clinician, for example about the patient’s own values or views of their own quality of life. The capable patient might tell the doctor he’s prepared to take the gamble and – as long as the doctor is prepared to spin the wheel – it is the patient’s decision whether to do so.

Even more care may be needed, though, where it is a patient who lacks capacity to make their own decisions about the treatment and where the clinician is making a decision for the patient that CPR is not in their best interests, weighing up the balance of the benefits and burdens. If this is what is being done, it must be done with a proper best interests decision-making process, compliant with the Mental Capacity Act 2005, and referred to the Court of Protection in the event of a dispute (see Chapters 5 & 7).

This is a very different place to be than a decision that the doctor is unwilling to do CPR due to futility, and simply has an obligation to “consult” with (i.e. inform) the patient. And it makes a big difference: No one can force a doctor to do something he thinks futile. But if doctor just thinks something is not in a patient’s best interests, others can legitimately disagree, and the court can say he’s wrong and that a DNACPR order based on best interests is inappropriate.

The court cannot order that CPR must be given, and ultimately the decision whether or not to do it falls to the clinician on the spot at the time it needs done, though it helps everyone if the merits of doing it have been given a bit of thought beforehand. Remember that DNACPR is about planning for the future. It can be because a clinician says that it would be futile, or because a patient with capacity to do so decides that they do not want CPR in some or all circumstances. Or it can be a best interests decision, following the MCA appropriately.

In some cases, though, the issues can be muddled when a DNACPR is put in place, especially if the result is recorded but not the reasoning, and if the clinicians (if at their behest) are unclear whether it is a case of futility or a (covert) best interests decision.

The best way to tease out the clinician’s reasoning is to ask whether, if the patient had capacity to decide and wanted to take all the risks involved for the small chance of benefit, the doctor would be willing to provide the CPR. If they would, then it cannot be said to be “futile”, and any DNACPR being led by the clinicians rather than the patient’s own choice is really a best interests decision and ought to be seen as such, following the MCA appropriately.

As this is so often a cause of confusion, and at the risk of labouring these points, I’ll recap into 5 different scenarios for an elderly patient, Betty, and her doctor, Dr David:

  1. As a patient with capacity to make her own decision about this, Betty does not want the end of her life marked by attempts at CPR.
    • Fair enough – It’s her life and her body.
    • She just needs to make sure that the right people know to ensure that her wishes are respected.
    • A DNACPR form, after discussion with her treating clinician / GP should do this, but an Advance Decision to Refuse Treatment (see Chapter 5) in the correct legal form is binding.
    • If that’s what she wants, she should do an ADRT refusing CPR (in some or all circumstances) and make sure that her clinicians and family know about it.
  2. Dr David thinks it would be futile to attempt CPR for Betty. It just won’t work.
    • Fair enough – No-one (not Betty, not her family) can make him.
    • But it is best there are no surprises. Unless it is positively harmful to do so, David needs to consult with Betty (and/ or if appropriate with her family).
    • In this context that means making them aware of his view and his decision, and why. She or they can seek a second opinion.
    • That consultation might, on reflection, suggest that the thinking is more along the lines of scenario 3 or 4.
    • Dr David should document everything.
    • And remember to keep this under review or time limited as appropriate, if the assessment of futility is based on things that might change.
  3. Dr David thinks that CPR might be effective (in the sense of restoring circulation), but on balance it is not for the best, given the poor prospects and heavy burdens.
    • But Betty has capacity to make her own decision, so it is up to her.
    • Dr David should paint her an absolutely realistic picture.
    • But if Betty decides to take the heavy burdens for the small chance of limited benefit, well that’s her decision, as long as Dr David is willing to offer it (and if not, why not, given it might work, and Betty is willing to take the risk?).
    • Alternatively, Dr David might find that the discussion prompts Betty to decide she does not want this and she may decide to bindingly refuse CPR as above (in scenario 1).
  4. Or in the same scenario as 3, but this time Betty does not have capacity to make her own decision.
    • This then amounts to a best interests decision, and so Dr David must follow the MCA process.
    • This means involving Betty herself appropriately, and others such as family to help inform the best interests decision (see Chapter 5).
    • Any dispute about best interests can be referred to the Court of Protection (see Chapter 7).
  5. Finally, imagine a different doctor, Dr Smith, attends when Betty has collapsed and Dr Smith has to make a decision about CPR on the spot.
    • If there is a valid and applicable ADRT refusing CPR that is binding then Dr Smith must not give CPR (unless she has reason to think that Betty has changed her mind, or that the ADRT is not valid, in which case she is covered while providing treatment if need be in the interim – see Chapter 5).
    • If there is a DNACPR based on Betty’s capacitous wishes it should be treated like an ADRT.
    • If there is a DNACPR based on Dr David’s view of futility, Dr Smith may well rely on it, but she is entitled to take her own view. If she thinks CPR is not futile, Dr Smith then gets into making a best interests decision whether to do it or not, in circumstances when this is obviously difficult.
    • If there is already a DNACPR based on a proper best interests decision-making process this should not be second guessed without good reason.
    • But the absence of both an ADRT and DNACPR form does not oblige Dr Smith to provide CPR. She must take her own reasonable view at the time on futility and, if not futile, on best interests. (By definition, by the time that a patient needs CPR they will not be able to make their own decision about it, which is another good reason for planning ahead!)

You can see why it is helpful to be clear in any DNACPR about the reasoning as well as the decision, to minimise the uncertainty and the decision-making that must be done in a crisis when it could helpfully have been given some thought in advance.

I hope it should go without saying that the reasoning “DNACPR because learning disability” or “…old age” or “…dementia” is absolutely not ok.

And, for the reasons I have given, it would be hard to justify, either in legal or ethical terms, any sort of organisational policy that demanded that CPR must always be given to any patient unless a DNACPR is in place. Clearly, that decision on the spot should depend on the clinical situation and , as need be, an appropriate process of best interests decision making. It cannot be the case that CPR – if on offer – is automatically in someone’s best interests just because a DNACPR decision has not been made.

Finally, the focus here has been on DNACPR, but it is worth noting that these important discussions about care planning, reflecting people’s wishes, are often now dealt with in terms of the ReSPECT form[27], addressing a person’s care overall, rather than being limited only to the issue of CPR alone, which seems to be a better way to do it.


Rationing

It is not just the case that an individual clinician is a gatekeeper of access to any and all healthcare your heart desires. There is also – within the NHS at least – a system level restriction, which you might call “rationing”, but we are typically more comfortable putting in terms of resource allocation and prioritisation.

It has never been possible for the NHS to meet every healthcare need. As the demands on the service have grown over time, and the cost increased as a share of GDP accordingly, the tension between supply and demand has become more evident, and especially so in the last 15 years or so. On average, since 1948 the NHS has received an increase in funding of around 4% pa. Since 2010, in the era of austerity, that figure has been closer to nil, until the more recent boost around Covid. And while that represented a victory of sorts compared to the departments which faced swingeing cuts over that period, the very large reduction in funding of adult social care in that period has simply pushed further demand into the NHS.

This is not the place to get into the complexities of NHS structure, but it is significant that after the reforms in April 2013[28] the burden of determining what the NHS ought to provide was shifted. Previously the key bit of English legislation was the NHS Act 2006, s3 which provided that the Secretary of State must provide hospitals, doctors, nurses etc “to the extent that he considered necessary to meet all reasonable needs”. After 2013 alongside the newly established National Commissioning Board (usually now known as NHS England) taking on certain national responsibilities, the burden of commissioning was put onto the new Clinical Commissioning Groups (replacing the previous Primary Care Trusts), which were obliged to ensure the appropriate provision in their view (not the Secretary of State’s) or everything reasonably necessary to meet the needs of the people in their area (not nationwide). Pitched as enabling clinically led and locally focussed prioritisation (the more negative phrase for which is “the postcode lottery”), putting it largely in the hands of 211[29] or so relatively small, new organisations to decide what NHS services ought to be commissioned.

Since the first edition of this book, CCGs themselves have been replaced in turn by Integrated Care Boards (ICBs)[30] which have inherited pretty much all their functions, since July 2022, on a larger footprint with 42 ICBs across England.

In the time that I have been working on this second edition (in the Spring of 2025), it has been announced that NHS England is to be abolished (or “merged” into the Department of Health and Social Care, depending on the framing), with its functions presumably to be split between going “up” into the DHSC and “down” into ICBs. But ICBs themselves have been ordered to cut 50% of their running costs in the next few months, predicted to drive mergers among the smaller of the 42 ICBs to create more economies of scale, pushing their number down towards half that. At the same time, there is likely to be a far-reaching reconfiguration of the ICBs’ role and responsibilities, focussed on “strategic commissioning”, and with many of their other current functions divested off to NHS providers or elsewhere[31] (TBC).

Alongside the continual churn of what a client has called “NHS re-disorganisation”, one constant is that the strain between resources and demand has tightened, as we have all grown older, less healthy, and generally more demanding. At the same time, the public finances are tight, and there is real concern about the economy generally and the world around us, and especially the need to find more money to invest in defence.

In some ways the growing demand reflects the successes of the NHS (as well as wider social improvements). Maladies that used to inevitably kill us relatively quickly may instead now be beaten, or at least fought to a draw and leave us with many years of ongoing healthcare need, often living long enough to have several overlapping chronic conditions (co-morbidities) causing a new layer of complexity. On top of this, each ICB has an explicit statutory duty to carry out its functions “effectively, efficiently and economically”,[32] and a duty to stay within its budget.[33]

On any view, juggling limited resources against ever greater demand is a thankless task, and the courts have generally been pretty sympathetic to the NHS about this when the issues have, occasionally, been tested.

One high profile example of this – in the mid 1990s, when I was first getting interested in such things – was the case of Jaymee Bowen,[34] a 10 year-old girl with leukaemia. Various treatments had already failed, and the health authority took a view that her prospects of success from further treatment could not justify the very high cost. Her parents challenged this decision, and the judge in the first instance hearing said that in these circumstances, with the life of a young girl in the balance, the health authority must “do more than toll the bell of tight resources. They must explain the priorities that have led them to decline to fund the treatment”. But the Court of Appeal immediately overturned this:

‘I have no doubt that in a perfect world any treatment which a patient, or a patient’s family, sought would be provided if doctors were willing to give it, no matter how much it cost, particularly when a life was potentially at stake. It would however in my view, be shutting one’s eyes to the real world if the court were to proceed on the basis that we do live in such a world. It is common knowledge that health authorities of all kinds are constantly pressed to make ends meet. They cannot pay their nurses as much as they would like; they cannot provide all the treatments they would like; they cannot purchase all the extremely expensive medical equipment they would like; they cannot carry out all the research they would like; they cannot build all the hospitals and specialist units they would like. Difficult and agonising judgments have to be made as to how a limited budget is best allocated to the maximum advantage of the maximum number of patients. That is not a judgment which the court can make. In my judgment, it is not something that a health authority such as this Authority can be fairly criticised for not advancing before the court.’

Sadly, though the wide press coverage of the case brought an anonymous donor forward to pay for the further treatment, Jaymee died anyway, just a few months later[35].

The NHS can still come unstuck in court if the decision making is flawed. For example, it cannot be an irrational decision, or be undermined by bias or prejudice, or by taking into account factors it should not have (or vice versa). Human rights issues ought to be appropriately considered. In some cases, promises may have been made by the NHS which it can reasonably be expected to keep.[36] And in some circumstances human rights arguments for particular individual or groups of patients can add another layer, which is beyond the scope of this chapter. But at the end of the day, demand often outstrips resources and it is usually perfectly ok for the NHS to say “no”, if it does so by a reasonable and proper process.

Typically, in our system, resource allocation tensions have been largely displaced into a mechanism of waiting lists, though there have also always been examples of particular treatments which are deemed to be non-core business for the NHS (most routine dentistry might be an example, or cosmetic surgery) driving patients for the most part towards private sector provision. In 1999, NICE[37] was established to provide national guidance, including explicitly considering the cost / benefit of new technologies or medication, and to bring some consistency in approach. Even so, you may still find that the scope of NHS provision for you individually may depend on the approach of the ICB in whose area you happen to be registered with a GP, or where you live.[38]


Discharge from Hospital

Just as we have no right to demand particular treatment, so we have no right to demand admission to hospital, and indeed no right to stay in hospital once deemed medically fit for discharge.

This is no small issue – according to the Nuffield Trust, in May 2024 over 12,000 people were in hospital in England who no longer needed to be there, down slightly from a peak of over 14,000 in January 2024. A great many of those are desperate to leave hospital, but are unable to do so while there are delays in finding the right social care or other support after discharge.

For a small number of cases, a patient who is medically optimised may simply refuse to leave. Here, the cliché that “possession is nine tenths of the law” has some basis – it is very much easier not to admit a patient than it is to make an inpatient leave against their wishes.

But the law is clear that there is no right to reside in a hospital bed, other than by the hospital offering it to you on the basis of clinical need, and there are various legal powers available to hospitals which need to take steps to secure that a patient leaves.

For instance, the Criminal Justice and Immigration Act 2008 s119 makes it a criminal offence to cause a nuisance or disturbance on the premises of an NHS hospital. There is an exception only if they are there to receive medical treatment, which would not apply to someone who has been declared to be no longer in clinical need of the bed. It has not yet been properly tested in the courts whether simply refusing to leave hospital in itself is sufficient to be considered a nuisance or disturbance, but I have certainly seen cases in which someone’s behaviour does amount to this. As well as s119 of that Act making this a criminal offence, s120 provides for a power to remove someone from the premises who is committing that offence, which can be exercised by a police officer or by someone authorised by the Hospital, and which allows for the use of reasonable force, if necessary, except only if doing so would endanger the person’s physical or mental health.

One problem is that this power includes no authority to take a patient to another particular place – say a care home, or back to their own property – but only to remove them from the NHS premises.

In the great majority of cases, though, it is sufficient to write a formal letter to a patient who is reluctant to leave setting out that they have no right to reside and are expected to leave in a reasonable timescale, referring to these powers, and – perhaps – to the fact that if court proceedings are needed then the trust may seek to recover its legal costs from the (former) patient.

Some cases are more complex, and it is open to the hospital trust to make an application to the court, which is typically framed in terms of possession proceedings as if this is a property dispute. Such court cases are rare, and there are only a handful of reported judgments – for example in University College London Hospitals NHS FT v MB, seeking possession of a hospital bed in the early days of the pandemic[39], and most recently in Northampton General Hospital v Mercer, which set out some broad principles for such cases[40]

In Mercer, the patient had been medically fit for discharge after only 11 days of the admission, and had then remained in hospital, refusing to leave, for more than 18 months. The judge considered the various arguments in public and private law that an individual might have to successfully oppose such an application by the hospital. The judge was clearly sympathetic to the hospital’s position, opening the judgment with “bed-blocking in NHS Hospitals is a huge problem”. Helpfully, he says that a hospital bed or room – even if occupied long-term – is probably not to be considered a “dwelling” requiring a court order for eviction under s3 Protection from Eviction Act 1977, but he appreciates why a hospital might prefer to seek a court order for possession or an injunction in cases where some force might be needed to remove someone from hospital against their will.

With respect to the learned judge in that case though, I would caution against one part of the judgment in the Mercer case, where he considers the question of her “capacity to consent to discharge from hospital”. Apart from the fact that this was academic here, as there was no doubt about the patient’s capacity, the more fundamental problem is that leaving hospital is not a matter of consent. It is not the patient’s decision whether to leave – they must leave once the bed is no longer being offered to them. The only question for the patient is where they go, not if they go; they cannot stay in hospital.

As such, the proper question is whether they have capacity to make a decision about their destination on discharge, and – if they cannot make that decision – then a best interests decision needs to be made among the available options. But staying in hospital is not an available option, and does not become so just because someone lacks capacity to (or refuses to) make the decision about where to go on discharge.

Sometimes, there may be a dispute about where a patient should be discharged to, or perhaps the proposed discharge in their best interests would be very much against their wishes, as might be the case with someone who only wants to go back to their own home for whom residential care is now thought to be necessary. In those cases, if the patient cannot make that decision for themselves, it may be necessary to make an application to the Court of Protection for a decision among the available options[41], and for this to be done with the urgency necessary to reflect the need to get that patient out of a hospital bed that they no longer need, while there are so many others who do.


Compulsion

The focus of this chapter has been on getting access to resources that a patient (or a family) want, and how this is managed.

But, for completeness, there is another side to this coin, which has also been highlighted by recent events – to what extent can treatment be forced on someone unwillingly?

In normal times, this is largely an issue for the Mental Health Act 1983, which is beyond the scope of this book but in broad terms provides for treatment without consent of a person’s mental disorder in certain circumstances, alongside a raft of procedural safeguards for that patient’s rights. It does not allow treatment of their physical health, unless it is directly linked to the mental disorder for which they are being sectioned[42], for example in some cases of treatment for the physical consequences of a person’s eating disorder.

Other than for mental health treatment, the Mental Capacity Act 2005 (see Chapters 4 & 5) can be used to make best interests decisions about physical medical treatment, including against a patient’s wishes, when they lack capacity to make the relevant decision for themselves.

Apart from that, for a patient with capacity but at risk of spreading an infectious disease as a result of non-compliance with treatment there are (rarely used) powers under the Public Health (Control of Disease) Act 1984 together with the Public Health (Infectious Disease) Regulations 1988.[43] Under sections 37 and 38 of the Act a local authority can ask a justice of the peace for an order to remove a person suffering from one of the specified diseases to hospital and to detain them there if, broadly, proper precautions cannot be taken or are not being taken, causing serious risk of infection to other persons. The Act contains no time limits on detention.

Even so, of course, this was nothing next to the swingeing, draconian powers included in the Coronavirus Act 2020, rushed through parliament in just a few days in response to the pandemic.

The Coronavirus Act provided for an enormous range of issues, from emergency registration of health and social care professionals, through to food supply, postponing elections, banning events and the use of video links in criminal trials.

Schedule 21 confered wide “powers relating to potentially infectious persons” exercisable by the relevant authorities in England, Wales, Scotland and Northern Ireland. Those could include a public health official compelling someone suspected of being infectious with coronavirus to go somewhere specified for up to 48 hours, to undergo screening and assessment, including providing biological samples and, if need be, to remain isolated for up to 14 days or impose restrictions for up to 14 days on their movement, activities or contact with others. The 14-day period could be extended in certain circumstances.

Those were not normal times.


Key points

  • Patient choice is limited by what is actually being offered. You can’t always get what you want.
  • Available options can be restricted by what is clinically indicated (or futile), by affordability (in the NHS, at least), or what the clinicians are willing to do.
  • Decisions about what options are available are for the NHS organisations or individual clinicians to make, and not for the patient, and so the patient’s “capacity” and “best interests” are not relevant (I’ll explain capacity and best interests more in Chapters 4 and 5). This means that any challenge to this decision-making is a matter of public law (i.e. judicial review), not for the Court of Protection.
  • Decisions about DNACPR in particular need clarity, especially about the different possible reasons for any such decision. It can be a patient deciding that they do not want it. Even if it is purely the clinician’s decision not to offer CPR as it would be futile, it brings a duty of consultation with the patient and/or family as appropriate. But sometimes a clinician’s DNACPR is really a best interests decision that ought to be recognised as such and the Mental Capacity Act followed accordingly (see Chapters 4 and 5).
  • You have no more right to remain in hospital once the bed is no longer being offered to you than you have a right to insist on being admitted in the first place.
  • Our decisions (and best interests decisions on our behalf) can only be made among the options that are being offered to us, on the basis of clinical judgement (and, in the NHS, funding and availability).

MORE INFORMATION / PURCHASE THE BOOK ONLINE

[1]     Burke, R (on the application of) v General Medical Council & Ors [2005] EWCA Civ 1003

[2]     “Withholding and Withdrawing Life-prolonging Treatment: Good Practice and Decision Making”, GMC, August 2002.

[3]     Paragraph 32.

[4]     The key case here, by the way, is one of the very few I remember from my law degree, and not only because it came out while I was at university. In R v Brown, House of Lords, [1994] 1 AC 212, a group of gay men who had been taking part in sado-masochistic activities – the phrase “genital torture” sticks in the mind – were exposed, as it were, by an unrelated police investigation. Though none of them had complained, they were successfully prosecuted for actual and grievous bodily harm under the Offences Against the Person Act 1861, their consent counting as no defence. https://www.bailii.org/uk/cases/UKHL/1993/19.html

[5]     https://www.bbc.co.uk/news/uk-england-london-68977469#:~:text=A%
20self-styled%20%22eunuch%20maker%22%20who%20mutilated%20
paying%20customers,counts%20of%20causing%20grievous%20bodily%20harm%20with%20intent

[6]     Suicide Act 1961 s2.

[7]     House of Lords [2001] UKHL 61 https://www.bailii.org/uk/cases/
UKHL/2001/61.html

[8]     House of Lords [2009] UKHL 45 https://www.bailii.org/uk/cases/
UKHL/2009/45.html

[9]     Supreme Court [2014] UKSC 38 https://www.bailii.org/uk/cases/
UKSC/2014/38.html

[10]   https://www.cps.gov.uk/legal-guidance/suicide-policy-prosecutors-respect-cases-encouraging-or-assisting-suicide

[11]   For example, one factor tending towards prosecution for assisting suicide is if the victim is in the care of the offender, who was acting in their capacity as a doctor, nurse or other healthcare professional.

[12]   [2017] UKSC 22 https://www.supremecourt.uk/cases/docs/uksc-2015-0238-judgment.pdf

[13]   At paragraph 35.

[14]    AVS v An NHS Foundation Trust [2011] EWCA Civ 7, https://www.
bailii.org/ew/cases/EWCA/Civ/2011/7.html – strictly this was a decision to refuse permission to appeal, rather than a substantive judgment, but the reasoning was given in detail with the intention of setting a clear precedent.

[15]   Do Not Attempt Cardio-Pulmonary Resuscitation.

[16]   “The Language of Kindness” by Christie Watson, Vintage, 2018, see p317. Oddly, she explains, if you’re going to have a cardiac arrest anywhere, you should choose a Las Vegas casino, where there is a staggering 75% recovery rate from CPR, which she speculates may be due to the combination of high levels of staff training and very high levels of supervision to prevent cheating, so any collapse is spotted immediately.

[17]   https://www.bhf.org.uk/what-we-do/influencing-change/cpr-in-schools

[18]   “Cardiopulmonary resuscitation on television: are we miseducating the public?” Colwill M, Somerville C, Lindberg E, et al. Postgrad Med J 2018;94:71–75.

[19]   “Dear Life”, Little, Brown, 2020, p96.

[20]   Ibid, p108.

[21]   You may, perhaps, want to make an exception if the cardiac arrest happens in a Las Vegas casino, see note 16 above.

[22]   There is a case of compensation being awarded for an attempt at CPR being made, against a patient’s wishes, where the doctors ought to have known about a DNACPR form.

[23]   [2014] EWCA Civ 822

[24]   Right to private and family life under Article 8 of the European Convention of Human Rights

[25]   Winspear v City Hospitals Sunderland NHS Foundation Trust [2015] EWHC 3250 (QB).

[26]   I am aware that resuscitation may often be decided or done by other clinicians, but with apologies to them I am trying to keep this simple.

[27]    https://www.resus.org.uk/respect

[28]   Under the Health and Social Care Act 2012 reforms led by then health secretary Andrew Lansley.

[29]   Ever since then there have been mergers, in many cases to move them back towards the size of their predecessor PCTs.

[30]    Health and Care Act 2022 – https://www.legislation.gov.uk/ukpga/2022/31/contents

[31]    https://www.digitalhealth.net/wp-content/uploads/2025/05/Model-Integrated-Care-Board-%E2%80%93-Blueprint-v1.0.pdf

[32]   NHS Act 2006 s14Z33, as amended by the Health and Care Act 2022.

[33]   NHS Act 2006 s223GC, as amended by the Health and Care Act 2022.

[34]   Reported as R(B) v Cambridge Health Authority [1995] EWCA Civ 43.

[35]    “Jaymee: The Story of Child B”, Sarah Barclay, Viking, 1996 – published, poignantly, at a time when the cover could say “so far the treatment has proved successful and Jaymee is in remission”.

[36]   For example the case of Coughlan (R v North and East Devon Health Authority (ex parte Coughlan)), [1999] EWCA Civ 1871, which is the leading case on the vexed question of where we draw the line between NHS funded care, which is free of charge, and social care, which is means tested, which is far too sticky to deal with here.

[37]   The National Institute of Clinical Excellence, in 2013 given an extended remit to include social care interventions and renamed as the National Institute for Health and Care Excellence, though still abbreviated as NICE.

[38]   Note that the rules about which ICB is actually responsible for commissioning your care are messy and complex in places, and again warrant more unpicking than can be done here – not least because all the reform mentioned above is likely to move these goalposts again shortly.

[39]   [2020] EWHC 882 (QB) https://www.bailii.org/ew/cases/EWHC/QB/
2020/882.html

[40]    [2024] EWHC 2525 (KB) https://www.bailii.org/ew/cases/EWHC/KB/
2024/2515.html

[41]    At times, there may be additional complication from the possibility that the situation after discharge might amount to a deprivation of that person’s liberty, which may need prior court approval – for more on this see https://www.lawbriefpublishing.com/product/deprivationofliberty/

[42]   Mental Health Act 1983, s63.

[43]   Applying to England, Wales and Northern Ireland and only to specified diseases including cholera, plague, smallpox, typhus, AIDS, anthrax, leprosy, rabies and tuberculosis.