The Code of Practice 2025 and the medicolegal definition of death

Abstract

In January 2025 the Academy of Medical Royal Colleges published an updated Code of Practice for the Diagnosis and Confirmation of Death, the first major update for 17 years. It represents the authoritative medical consensus on the diagnosis and confirmation of death in all contexts (and in all age groups) in the UK. This paper reviews the new Code, highlighting the major updates from the 2008 guidance and their relevance for intensive care clinicians. It goes on to explore the wider legal history and context, in particular the role that previous versions of the Academy’s guidelines have played during the emergence of a common law definition of death. It examines the important legal cases in which the Code of Practice was endorsed and ultimately adopted by the courts as the definitive medical and legal definition of death in the UK.

Introduction

Death represents a moment of enormous medical, social, legal, and philosophical importance. The Academy of Medical Royal Colleges 2025 Update to the Code of Practice for the Diagnosis and Confirmation of Death (the Code) represents the latest iteration in a series of guidelines that have been published over the past 50 years. Initially the focus was on providing the neurological criteria for so-called ‘brainstem death’, but since 2008 this was expanded to include circulatory criteria. The Code has been accepted and adopted by the courts in England, Wales, and Northern Ireland to create an increasingly sophisticated medicolegal understanding of death. This paper begins by reviewing the 2025 Code, highlighting important changes from the previous Code and their relevance for intensive care clinicians. In the second section, we explore how, during a succession of landmark legal cases, these clinical guidelines were gradually woven into the common law definition of death. We conclude by highlighting a currently unresolved legal issue regarding consent for DNC testing.

The Code of Practice 2025

The updated Code came into effect on 1 January 2025, ¹ replacing the previous version from 2008. It represents the first major update for 17 years. The Code provides a contemporary consensus endorsed by the Academy of Medical Royal Colleges (the Academy), the membership body for the UK’s medical royal colleges and faculties. ² It sets out an authoritative statement of practice for the diagnosis and confirmation of death in all circumstances and all ages groups. This ensures that ‘all deaths are diagnosed and confirmed in an accurate, standardised, and timely manner’. The Academy emphasises that the definition of death itself has not changed from 2008—indeed, the medical definition (predicated on the loss of brainstem function) has been accepted in UK clinical practice since 1979. ³ The updated Code instead reflects advances in medical practices during the intervening years, acknowledges the emerging international consensus around diagnosing death, and incorporates important lessons from recent cases in the UK and abroad. The international guidelines considered or incorporated into the new Code include those from the World Brain Death Project, ⁴ the Australian and New Zealand Intensive Care Society, ⁵ plus several recent publications from a special issue of the Canadian Journal of Anaesthesia. ⁶

Overview and layout

The 2025 Code is noticeably longer and more comprehensive that its predecessor, intended to provide greater conceptual clarity and clearer practical guidance. In broad terms, the 2025 Code reinforces the UK’s commitment to the so-called ‘unifying concept of death’. This has been the UK’s position since 1979 and re-iterated in all subsequent editions of the Code. In recent decades it has now also become a more globally accepted position in response to technological advances and philosophical deliberations. ⁷ The 2025 Code asserts that although there is only one underlying definition of death, it can be identified using three different sets of diagnostic criteria depending on the clinical context: somatic, circulatory, or neurological.

Section 1 emphasises the medical, social, and legal importance of the confirmation of death, whilst restating that the 2025 Code is not a redefinition of death—this ensures it remains supportive of the long-held medical consensus in the UK. Section 2 articulates the consensus definition of death (permanent loss of the capacity for consciousness combined with permanent loss of the capacity to breathe) and defines the point at which it can be confidently diagnosed (the permanent cessation of brainstem function). Section 3 introduces the three different diagnostic criteria which can be used to determine that the underlying definition of death has been satisfied. It also provides guidance on who can use these criteria and comments upon the level of competency required. Section 4 is entirely new for the 2025 Code, describing in detail the somatic criteria for the identification and confirmation of death (this had been alluded to only briefly in the 2008 version). Section 5 describes the circulatory criteria for diagnosing death, which remain very similar to the 2008 Code. Section 6 describes the neurological criteria for diagnosing death; it features the greatest number of clinical updates for intensive care clinicians compared to the 2008 guidance (see below). Section 7 is a new chapter intended to support communication with family and friends about the diagnosis and confirmation of death. Appendix 1 of the updated Code provides a summary of all the updates and modifications from the 2008 Code. Finally, Appendix 2 incorporates the Royal College of Paediatrics and Child Heath ‘2025 update on the diagnosis of death using neurological criteria in infants, children, and adolescents’.

Major updates

This paper cannot provide a comprehensive review of all the changes and modifications in the updated Code; readers are strongly advised to consult the Code of Practice 2025 and its appendices for further details.

Inclusion of somatic criteria

The 2025 Code includes detailed guidance on the diagnosis and confirmation of death using somatic criteria; these were mentioned only briefly in the 2008 Code. These include signs of overwhelming physical trauma (e.g. decapitation, massive cranial destruction, incineration, etc.) or time-based findings (e.g. rigor mortis, livor mortis, decomposition, etc.). The Code notes that the use of somatic criteria to diagnose death may be particularly relevant in forensic, pre-hospital, midwifery and community settings, or may be used by non-healthcare professionals.

Circulatory criteria

The Code makes only minor modifications to the process of diagnosing death using circulatory criteria. It clarifies who can perform this and the level of competency required; that the whole period of examination requires 5 minutes of continuous observation in the physical presence of the body; and that the period of observation commences with mechanical (not electrical) asystole.

Neurological criteria

The 2025 Code makes a number of important modifications to the process of diagnosing death using neurologic criteria (DNC) which are relevant to adult and paediatric intensive care.

• ● Age categories:

• ○ <37 weeks corrected gestation, DNC diagnosis cannot confidently be made.

• ○ 37 weeks corrected gestation to 2 years, DNC diagnosis can confidently be made as per adults, except that:

• ○ 24-h wait period required from observed loss of brainstem reflexes to performing clinical testing.

• ○ 24-h interval required between clinical tests.

• ○ ancillary investigations cannot be used.

• ○ Over 2 years, DNC diagnosis can confidently be made using adult criteria.

The transition to adult criteria was previously accepted in the UK as occurring at 2 months of age (corrected gestation); the change to 2 years of age closer aligns the UK with international practice, in particular the USA. ⁸

• ● Apnoea test:

• ○ Starting PaCO₂ ⩾5.3 kPa (40 mmHg)

• ○ Minimum duration 5 min

• ○ Rise by PaCO₂ ⩾2.7 kPa (20 mmHg)

• ○ End PaCO₂ ⩾8.0 kPa (60 mmHg) and pH <7.3 These targets align the UK with international apnoea tests; they also allow the same start and end points to be used in all age groups.

• ● Time of death: When the doctors involved in clinical testing are satisfied the neurological criteria have been met: this is now after the second set of tests (or after the ancillary tests if these are performed after clinical testing). This is more intuitive for families, many of whom now witness the second set of tests.

• ● Minimum core temperature of 36 °C at the time of testing; patients with a core temperature <36 °C (either therapeutic or accidental) require a minimum of 24 h observation after correction to at least 36 °C prior to testing. Following rewarming, ‘transient and temporary’ reductions in temperature <36 °C do not require a further 24-h observation period. (Whilst not specified in the Code, ‘transient and temporary’ is generally accepted to mean <6 h in total over the proceeding 24 h.)

• ● Both eyes and both ears must be examinable; if they are not then an ancillary investigation will be required.

• ● Miscellaneous updates:

• ○ Preconditions are now more specific and highlight areas for ‘diagnostic caution’ (incorporating what were previously called ‘red flag’ situations).

• ○ Clarification on who can perform clinical testing.

• ○ Expanded discussions on cervical cord pathology, posterior fossa pathology, decompressive craniectomy, steroids, and patients requiring extracorporeal membrane oxygenation (ECMO).

• ○ Expanded discussion and guidance on the role, indications and limitations of ancillary investigations.

Inclusion of communication section

The 2025 Code includes an entirely new section providing guidance for communication with family and friends regarding the diagnosis and confirmation of death. It provides clear support and recommendations on where, when, and how these often challenging discussions should occur. The Code reminds healthcare professionals that as well as diagnosing death in an accurate, standardised, and timely manner, this must be communicated clearly and respectfully.

Inclusion of paediatric and neonatal guidance

Appendix 2 updates (and incorporates into one document) the 2015 Royal College of Paediatrics and Child Health ‘The diagnosis of death by neurological criteria in infants <2 months old’ ⁹ and the 1991 Report of a Working Party of the British Paediatric Association on ‘The diagnosis of brain-stem death in infants and children’. ¹⁰ Their assimilation into the 2025 Code brings welcome uniformity and consistency, ensuring that a single source now provides the guidance for the diagnosis of deaths in all contexts and in all age groups.

Discussion

The 2025 Code is an update and evolution from previous guidance and specifically does not provide a redefinition of death. Reiteration of the UK’s previous definition of death, as found in the 1998 and 2008 Codes (although some words have been altered to provide greater clarity and reflect modern terminology), provides important continuity and reassurance that historical diagnoses of death remain valid within the updated guidance. By clearly articulating the definition of death and the three diagnostic criteria which can be used to satisfy that definition, the 2025 Code provides a clear framework with which to understand the underlying ‘unifying concept’ of death. The 2025 Code is accompanied by new national, professionally-endorsed diagnostic forms to document DNC. ¹¹ A single form for adults and children over 2 years replaces the ‘short’ and ‘long’ forms used previously; a similar revision has been made to the infant DNC form. (It had been erroneously suggested in a recent legal case that the selection of either the ‘short’ or ‘long’ form had a bearing on the performance or reliability of the DNC assessment. ¹² This potential area of confusion has therefore been removed.)

Despite the reaffirmation that DNC is a clinical diagnosis and that ancillary investigations are not routinely required, it seems likely that the use of ancillary investigations will increase with the 2025 Code. This is in part due to the expanded number of indications, preconditions, or cautions described; it might also be driven by persisting public unfamiliarity or unease with the concept of DNC. The Official Solicitor has ominously suggested that legal challenges to DNC diagnoses are becoming more commonplace. ¹²

The 2025 Code utilises a ‘permanence’ definition, although does briefly allude to the academic debate over ‘permanent’ and ‘irreversible’ terminology. ‘Permanent cessation’ describes function which has stopped and will not return (i.e. will not resume spontaneously and will not be restored through intervention), whereas ‘irreversible cessation’ describes function which cannot return. ¹³ Theoretically, ‘permanence’ has greatest philosophical validity when describing circulatory (as opposed to neurological) criteria, because circulatory arrest is typically assessed in a contemporaneous, prospective manner. ¹⁴ For clarity, simplicity and greater consistency with a growing number of international guidelines, the Academy has chosen to use ‘permanent’ throughout. The need to exclude reversibility in DNC is clearly articulated and strengthened compared to the 2008 Code. This is most evident in the DNC preconditions (e.g. an assessment period sufficient to exclude the potential for recovery, and exclusion of potentially reversible factors materially contributing to the coma or apnoea).

The Code and the legal definition of death

It is perhaps surprising that there is no statutory definition of death in the UK, despite numerous pieces of legislation regulating various matters surrounding death (including certification, ¹⁵ coronial investigations, ¹⁶ Medical Examiner processes, ¹⁷ inheritance, and organ donation ¹⁸ ). In this way the UK is similar to Canada (excluding Nova Scotia) in having a common law definition of death. There is a long history in the UK of the Codes being accepted by the courts as providing the authoritative criteria for diagnosing death using neurological criteria. ¹⁹ Because of this, some scholars have commented that ‘. . .the legal definition of death is merely the judicial application of the current medical definition of death’. ²⁰ It is worth noting that the traditional conception of death (i.e. when diagnosed using circulatory criteria) has rarely, if ever, been challenged in court. It has never been argued that a refrigerated corpse in a mortuary is not dead. Death in that context seems intuitive and self-evident, with bereaved relatives readily accepting the criteria used to establish death has occurred. Yet following a diagnosis of death using neurological criteria, a patient on ICU is just as dead as a corpse in the mortuary because their brainstem function has also permanently ceased. The diagnostic criteria may be different, but the underlying definition of death remains the same. To date, the courts’ role in DNC cases has therefore been to confirm that the appropriate clinical criteria outlined in the Code have been applied and satisfied—but not to question what those criteria are. It is interesting to examine how this legal situation came about, including the role that previous Codes of Practice have played in several key cases during the development of the legal definition of death.

Early legal cases

Soon after the early guidance from the Conference of Medical Royal Colleges, ³ a number of legal cases began to appear before the UK courts in which the new concept of death was considered. In R v Malcherek; R v Steel (1981), ²¹ a criminal case concerning two convicted murderers, the Court of Appeal acknowledged the shifting medical landscape: ‘There is, it seems, a body of opinion in the medical profession that there is only one true test of death and, that is, irreversible death of the brainstem. . . When that occurs it is said the body has died’. The judgement contrasted the differing professional responsibilities in this emerging area, commenting that, ‘It is no part the task of this court to inquire whether the criteria, the royal medical colleges’ confirmatory tests, are a satisfactory code of practice’.

This deference to medical expertise and the guidance was later reinforced in a subsequent civil case which directly addressed the issue of DNC. In Re A (1992), ²² Johnson J held an emergency hearing concerning a 1-year-old infant who had presented to hospital in cardiac arrest. Having been resuscitated and stabilised on ICU, clinicians suspected DNC and so tested him in accordance with the current medical guidance. Clinical testing confirmed the infant was dead according to neurological criteria, leading the parents to lodge an urgent application with the court to prevent the withdrawal of life-support. Johnson found the medical evidence compelling, including the results of the carefully conducted clinical tests, a second opinion, and the widely accepted medical criteria as described in the current guidance. He concluded, ‘I have no hesitation at all in holding that A has been dead since Tuesday of last week. . . A is now dead for all legal, as well as medical, purposes’. Re A therefore cemented into common law the notion that death can be diagnosed using brainstem criteria alone, although strictly speaking did not assert that all death could be defined in terms of brainstem function. The ruling also confirmed it would be lawful for clinicians to withdraw life supporting therapies, although did not actually direct them to do so. In this way, Johnson J had merely made determinations of fact as to the diagnosis of death and the legal status of the patient. DNC was therefore formally endorsed by the courts for the first time. This ruling again demonstrates the autonomy afforded to clinicians by the courts, with recognition that subsequent clinical decisions are best left to medical professionals. This sentiment has been echoed over subsequent decades, with one U.S. court pithily commenting that, ‘The judiciary is called upon to serve in black robes, not white coats. And it must be vigilant to stay in its lane and remember its role’. ²³

‘‘Bland’ and ‘Re A (A Child)’’

The legal position would later receive significant support from the House of Lords in Airedale Trust v Bland (1993). ²⁴ In considering the ongoing management of a patient in a vegetative state, the Law Lords had to address the preliminary issue of whether or not Anthony Bland was already dead: They held,

In the eyes of the medical world and of the law a person is not clinically dead so long as the brainstem retains its function. . . it has come to be accepted that death occurs when the brain, and in particular the brainstem, has been destroyed; . . .Anthony’s brainstem is still alive and functioning and it follows that. . . he is still alive and should be so regarded as a matter of law. ²⁵

This highly influential ruling left little doubt that clinical and legal death are synonymous. More significantly, it cemented a legal principle that all death could be defined in terms of brainstem function. Despite principally being a case concerned with the withdrawal of life-sustaining treatments from a living patient, Bland created a common law precedent for defining death that would be adopted in all subsequent court cases.

In Re A (A Child) (2015) ²⁶ this common law definition was conspicuously tied to the specific wording of the Code of Practice for the first time. A 19-month-old child had choked on a piece of fruit, leading to a cardiac arrest and severe hypoxic brain damage. DNC testing was conducted in accordance with the Code of Practice 2008, confirming that the child was dead. His parents however refused to accept that brainstem death ‘was synonymous with clinical/legal death’ and so refused the proposed withdrawal of life-support. Hayden quoted directly from the 2008 Code at length, emphasising its definition of death as one based upon absent brainstem function. He wrote, ‘Applying all of this, to A’s tragic circumstances, I conclude that. . . the criteria for death had been established’. Re A (A Child) therefore confirmed that DNC—established in accordance with the Code of Practice—constituted medical and legal death. This further reinforced the legal position from Bland, directly linking the diagnosis of death to the specific medical definition and criteria found within the Code. Several subsequent judgements would similarly quote directly from the 2008 Code. ²⁷

‘Re M’, ‘Battersbee’ and recent cases

From 2015 onwards the medicolegal situation surrounding the declaration of death appeared to be settled, yet that still did not prevent multiple challenges making their way to court. In Re M (Declaration of Death of Child) (2020), ²⁸ the Court of Appeal was again faced with the parents of a baby who would not accept a brainstem-centred definition of death. At first instance a High Court judge had ruled that, based on previous case law, the diagnosis of death was a factual issue for clinicians determined by applying the DNC criteria from the Code. The Court of Appeal upheld this reasoning, stating that,

. . .as a matter of law, it is the case that brainstem death is established as the legal criteria in the United Kingdom by the House of Lords decision in Bland. It is not, therefore, open to this court to contemplate a different test.

Furthermore, they added that it was impossible for them to assess whether a different test (such as the whole-brain criteria used in the USA) should replace the criteria in the 2008 Code. In this way, Court of Appeal confirmed that the ruling in Bland was binding precedent. They also deliberately declined the opportunity to scrutinise the robustness or coherence of the UK’s clinical criteria, ²⁹ once again signalling the diagnosis of death is a clinical matter for medical professionals with which the courts will seldom interfere. All that is required of the courts is to determine factual questions about whether the appropriate medical standards have been applied correctly. This legal deference bestowed significant authority and legitimacy on the Code of Practice.

The importance of this deference later became clear in the highly publicised saga of Archie Battersbee. During an early hearing, a High Court judge unfortunately caused confusion by declaring that, on the balance of probabilities, Archie was dead—despite not having been tested according to the Code’s criteria. ³⁰ The Court of Appeal promptly overruled this judgement, noting that none of the doctors had proposed declaring death in the absence of the DNC tests, no clinician had actually diagnosed death, and that the Code itself was clear that the DNC tests were required to confirm absent brainstem function ³¹ : They added,

We. . . strongly caution judges in future cases of this kind from being drawn into attempting to declare death on a basis outside the Code where none of the medical witness has themselves made a diagnosis of death.

The common-law precedent was clear: in order to declare death or DNC, the Code’s criteria and tests must be satisfied.

In St. George’s University Hospitals NHS Foundation Trust v Casey (2023), ¹² a declaration was sought that a brain-injured patient had died following an assault. In the High Court, MacDonald reiterated that his task in law was to confirm that the Code had been properly applied; declare the date and time of death; confirm that a patient’s best interests were no longer relevant; and declare the withdrawal of life support would be lawful. For the avoidance of doubt, he concluded ‘It is for doctors to diagnose and confirm death by brain stem testing carried out in accordance with the 2008 Code of Practice’. This case also went to the Court of Appeal, ³² which once again upheld the lower court’s ruling. As to the family’s suggestion that the prior case-law was not binding, the Court clearly stated,

The fact is that the approach in [Bland and Re M] reflects a widely accepted consensus in this country for almost 50 years that brain stem death, correctly diagnosed, is the proper indicator of death in the legal sense. The contrary is not reasonably arguable at any level of court.

The medicolegal position regarding the definition and diagnosis of death therefore appears entirely settled. As of November 2025, rulings from the Court of Protection ³³ and the High Court have now formally endorsed the updated version of the Code. In Barts Health NHS Trust v MC and ML (2025), ³⁴ Hayden J set out a concise summary of the preceding case law. He then quoted from the 2025 Code at length, confirming in law once again the Academy’s definition and confirmation of death, whilst also upholding newer elements such as the time of death occurring after the second set of DNC tests. He also commended the ‘accessible’ and ‘sensitive’ new lay summary provided by the Academy, ³⁵ commenting that family members should be made aware of it and directed to it by clinical teams.

Consent for testing

One area of UK law that remains unclear is whether consent for DNC testing is required. The updated Code makes no reference to the requirement (or not) to obtain consent for testing from family or friends. Many other international guidelines state there is no need to seek consent for testing.^4,36 There exists a tension between conducting DNC tests as part of a collaborative best interests decision-making process (in which the courts would be the rightful arbitrator of an intractable disagreement), versus the view as stated in Canada: ‘. . .clinicians probably do not and should not have a legal duty to obtain informed consent or any other consent before conducting DNC tests. There is no “decision” or “choice” for families to make’. ³⁶ In the UK this argument has not strictly been tested in the courts, so the legal issue remains unresolved for now—the decision ultimately appears to remain one for clinicians. Clearly it would be best practice to discuss and agree proposed testing with the patient’s family and friends, but in the case of intractable disagreement an application to the Court of Protection would likely be needed.^32,33 The issue is even less clear for children. Parental consent for testing has been considered tangentially during recent cases, ²⁸ but again not fully resolved. As with adults, in the event of significant disagreement an application to the Courts will almost certainly be required.

Conclusion

The 2025 Code represents the latest iteration in a gradual evolution of our medical understanding of death and how it is diagnosed. It provides welcome conceptual clarity and practical guidance, not only when it comes to establishing death using neurological criteria in intensive care units, but for all deaths in all contexts. In this way, clinicians, patients, their families and the wider public can be reassured that the diagnosis and confirmation of all death is accurate, reliable, standardised and timely. As of November 2025, the updated Code has now been formally endorsed by the courts in England and Wales. It seems clear from the common law precedent that the Code will continue to be accepted as the authoritative medicolegal consensus in all future cases.