Allen, R (on the application of) v HM Coroner for Inner North London

This is an appeal against the refusal of Blake J to give permission to apply for judicial review of decisions made by the coroner Dr Andrew Scott Reid at an inquest into the death of Sandra Jane Allen. Mrs Allen died on 2 October 2006. She was 61 years of age. In July 2006, she had been detained by the Camden and Islington Social Care NHS Care Trust (“the Care Trust”) under section 3 of the Mental Health Act 1983 because she was suffering from a pre-existing schizophrenic disorder. She was detained at Highgate Mental Health Centre.

It will be necessary to consider in some detail the circumstances surrounding her death. At this stage, the barest outline account will suffice. At 0.15 hrs on 2 October, Mrs Allen was found by nurse Deborah Chambers collapsed on her bathroom floor. Dr Ruchit Patel, the senior house officer (SHO) was called and cardio-pulmonary resuscitation (CPR) was administered. The London Ambulance Service (LAS) was called and Mrs Allen was taken to Whittington Hospital. She was pronounced dead at 0.55 hrs. A post mortem report was made by Dr Freddie Patel on 5 October. He said that the cause of death was coronary artery disease leading to cardiac arrest. An internal inquiry was undertaken by the Care Trust. It found some evidence that staff generally were not familiar with the procedures for operating small oxygen canisters, including whole emergency kits, and that better comprehensive training was required.

The inquest was opened by Dr Reid on 19 October. It was then adjourned. Originally, the claimant (Mrs Allen’s widower) sought to adjourn the hearing by making an application for permission to apply for judicial review of various decisions of the coroner in relation to the evidence that was to be adduced, whether he should sit with a jury and other matters. Underhill J refused the application on 4 June 2007. The substantive inquest hearing was held before the coroner, sitting without a jury, on 5 June 2007. The claimant’s son, who represented the claimant on that occasion, again sought an adjournment. This was refused by the coroner. The claimant’s son then informed the coroner that the claimant did not wish to take part in the inquest because he considered that it would not be sufficiently full or effective.

The coroner called a number of witnesses and at the close of the evidence he adjourned to consider his verdict. On 12 July, he confirmed that the disease which caused the death was coronary artery disease. He described the “time, place and circumstances at or in which injury was sustained” in an appendix to the inquisition to which we will refer later in this judgment. His conclusion as to the death was that Mrs Allen “died from natural causes when she suffered an abnormal fatal heart rhythm”.

The claimant was dissatisfied with the inquest and he started these judicial proceedings seeking a fresh inquest. It will be necessary to examine his complaints in some detail. In summary, they are that in a number of respects the scope of the inquest was insufficient to satisfy the requirements of article 2 of the European Convention on Human Rights.

The application was refused on the papers by Underhill J on the grounds of delay and because he could see nothing wrong in law with the coroner’s approach.

The renewed application was refused by Blake J on 23 September 2008. He did not deal with the issue of delay, but he refused the application on the merits. At [8], he summarised the five particular points made on behalf of the claimant in the following terms:

“(1)

Dr Patel says he saw no nursing staff giving resuscitation assistance when he arrived upon the scene but the expert assumed that such assistance was beginning to be given in her overall assessment as to whether this was a cardiac arrest that might have been survivable by the deceased.

(2)

Dr Patel said that he had cleared the airways before providing the tubes into the deceased's throat, whereas the paramedic who arrived later at the scene suggested there was still food in the throat that had to be cleared for a clear airway to be obtained.

(3)

There was no investigation into precisely how long the paramedics had been held up because of the actions of the security guard who was asleep.

(4)

Given the very narrow window (possibly three to four minutes) in which in this kind of case, where there is some food in the airways as well as a cardiac incident, to try to get breathing and oxygen supply to the brain restored, any one or more of these failures could have been said to have materially contributed to the death.

(5)

He also pointed out that the coroner did not explore in the inquest or with the expert the conclusions of the health authority's internal inquiry that some of the staff were unfamiliar with the equipment and that may have caused delay contributing to death.”

At [9], the judge referred to the general submission by Mr Thomas that an adequate inquiry might have raised questions as to whether Mrs Allen was in the right place and whether she was receiving the right treatment.

At [10], the judge said:

“However, viewing the matter in the round, I have come to the conclusion that his two other submissions, namely that there was evidence of, or reason to believe that there was, a systematic failure that caused the death in a material way such as to engage the state's obligations with respect to mental patients who are undergoing treatment in a mental institution, is not a good one. The law is by no means crystal clear but those cases in which the courts have been minded to conclude that the Article 2 obligation existed were cases where there really was fundamental failures that caused the condition itself that caused the death. In Touche there was gross incompetence in the administration of anaesthetics to a healthy woman giving birth and in Takoushis there was an inexplicable failure to keep someone who was mentally disturbed and suicidal in the environment in which there might have been at least a possibility of protecting him. Instead he left the hospital and took his own life by jumping off a bridge into the Thames. In both those cases therefore, the concern was into either causing or failure to prevent the very thing which caused the death. In this case undoubtedly what caused the death was the natural condition of a heart attack in this lady, who obviously had her fair share of other health problems of one sort or another. The inquiry into whether the state's obligation through the health authority existed, would have been in respect of whether medical assistance promptly rendered might have prevented the natural cause of death taking effect. I consider that that is a material distinction in the present context when seeking to define the scope of Article 2 and also to some extent the question as to whether there was a systemic problem that should be addressed by a particularly broad ranging inquest or indeed some avoidable procedure or error or harm that would require or make it appropriate for the coroner to sit with a jury. I conclude that, notwithstanding the perfectly understandable concerns of the family in their letters, the facts of the case did not require examination of a broader inquest or an inquest with a jury.”

At [11], he returned to the five particular points and explained why a judicial review based on them would not succeed. He said:

“That leaves the simple first proposition as to whether there was an adequacy of inquiry on the points made. I have looked with care at the transcripts and there is always a danger of taking particular answers out of context and developing the case simply upon the basis of one answer. It is reasonably plain to me that Dr Patel had arrived on the scene when the nursing staff had done something to give assistance to the deceased and had got out the relevant equipment and so it was not simply a question of ignoring a dying lady who was in urgent need of assistance and there was discussion by the expert of the difficulties of getting a clear airway where one of the problems is, and was known to be, that the deceased had consumed a quantity of food. Indeed, that was the very factor that staff had noticed about her and why they wanted to monitor her because she seemed to be eating a lot of food rather quickly. Quite whether things might have been done better is not the test for whether this court should now entertain this application for judicial review. Certainly it is hardly inspiring to find that the ambulance crew could not get in in the vital minutes that are required in cardiac resuscitation because of the sleeping security guard, although the court understands that came to the attention of the health authority and this individual was dismissed. The court has to focus eventually upon the prospects of the claimant obtaining relief and the relief that the claimant would have to obtain in this case is the setting aside of this inquest on the basis that there was a fundamental failure of inquiry of the sort that would justify reopening the whole inquest de novo. Of course, it is not Mr Thomas' case that it is sufficient to obtain such relief, that one might have better explored with advantage certain questions in cross-examination.”

At [12], the judge said that he did not consider that there was a line of inquiry that was ignored that might have shed fresh light on the circumstances in which Mrs Allen met her death.

Thus it will be seen that the judge refused the application for three discrete reasons. First, a broader inquest of the kind required by article 2 was not necessary because article 2 was not engaged. Secondly, even if article 2 was engaged, the inquiry was not deficient in all or any of the specific respects alleged. Thirdly, there were no further lines of inquiry which the coroner should have undertaken.

On 26 February 2009, Arden LJ gave permission to appeal.

In our judgment, Blake J was wrong to hold that article 2 was not engaged in this case. His reasoning seems to have been that article 2 is only engaged in cases “where there really w[ere] fundamental failures that caused the condition itself that caused the death”. I can find no support for this approach in the authorities. In R (on the application of Smith) v Secretary of State for Defence [2009] EWCA Civ 441, this court reviewed some of the relevant authorities. The paradigm example of the case where article 2 requires the state to conduct its own investigation, for example, by an inquest is where the death has or may have been caused by an agent of the state or where the deceased is in prison or otherwise in the custody of the state at the time of death: see [83]. The vulnerability of detained persons, whether prisoners or detained mental patients, was emphasised in Savage v South Essex NHS Trust [2008] UKHL 74, [2009] 2 WLR 115 at [49]-[50] and [97] as the reason why the substantive article 2 obligation is owed to them by the authorities which detain them. In R(L) v Home Secretary [2008] UKHL 68, [2008] 3 WLR 1325 (a case involving the near-suicide of a prisoner), it was argued on behalf of the Secretary of State that article 2 did not require an independent investigation “unless there was some positive reason to believe that the authorities had indeed been in breach of their obligation to protect the prisoner”: see [58].

As to this, Lord Rodger of Earlsferry said at [59]:

“That argument is mistaken. Whenever a prisoner kills himself, it is at least possible that the prison authorities, who are responsible for the prisoner, have failed, either in their obligation to take general measures to diminish the opportunities for prisoners to harm themselves, or in their operational obligation to try to prevent the particular prisoner from committing suicide. Given the closed nature of the prison world, without an independent investigation you might never know. So there must be an investigation of that kind to find out whether something did indeed go wrong. In this respect a suicide is like any other violent death in custody. In affirming the need for an effective form of investigation in a case involving the suicide of a man in police custody, the European Court held that such an investigation should be held “when a resort to force has resulted in a person’s death": Akdogdu v Turkey, para 52.”

In our judgment, that reasoning applies to the present case. Mrs Allen was detained by the state as a mental patient in a hospital. It is possible that the medical authorities failed in their obligation to take general measures to save her from dying. That was sufficient to trigger an article 2 investigation, since without it, as Lord Rodger said, one might never know. A further reason for an article 2 investigation might have been the possibility that the death was caused by a breach of the operational obligation to take steps to save Mrs Allen from death.

We have felt it necessary to make these comments in the light of the reasons given by Blake J for holding that article 2 did not apply in this case. But it was in fact common ground before the coroner that the inquest had to satisfy the requirements of article 2. At the outset of the hearing on 5 June, the claimant’s son made detailed submissions to the coroner in support of his application that the coroner should sit with a jury. He identified a number of systemic failures which justified a jury. He referred the coroner to a number of authorities including R (Middleton) v West Somerset Coroner [2004]UKHL 10, [2004] 2 AC 182which remains the leading authority on the requirements of article 2 in relation to inquests. Miss Wolferstan, who represented the Care Trust, said to the coroner that the provisions of article 2 would be respected if the inquest took place that day.

The coroner then said this:

“Following the decision of the House of Lords in the Middleton Case, and also following the decision of the Court of Appeal in the Takoushis Case, it is now clear that the way in which the States obligation under Article 2 of the European Convention on Human Rights is discharged, is by a combination of the availability of other effective investigations including civil and criminal proceedings and the inquest in providing the legal system in England and Wales, the State discharges the obligation under Article 2 of the European Convention on Human Rights.

In a case such as this, in addition to the possibility or potential for remedy in the Criminal and Civil Courts, there is in any event going to be an inquest over which I will be presiding. Further in addition to those fora there are the NHS Complaints Procedures and also there is the role of the Mental Health Act Commission in looking at matters that are more germane to issues of clinical governance, health care rationing, and matters that are ones of health and government policy and not matters about the cause and circumstances of the deceased’s death.

It is now permissible following the Middleton Judgement, for the Court to return either a traditional short form verdict, an extended narrative verdict, or a traditional short form verdict with a few other words added to the narrative that make non-judgmental findings of fact as to the cause and circumstances of the deceased’s death. But in doing so, in paragraph 36 of the Middleton Judgement, the House of Lords indicated that these should be central facts as the Court sees them in relation to the deceased’s death. ”

It is clear, therefore, that the coroner was not only aware of article 2 but that he intended to conduct the inquest in accordance with its requirements as explained in Middleton. Whether he did so depends on an analysis of what he said and did. But he plainly thought that article 2 was engaged.

It has been no part of the case of the Care Trust to say that the inquest did not have to satisfy the requirements of article 2. Indeed, although they have not been represented on this appeal, they served a fairly detailed summary defence to these proceedings in which they stated at para 24 “there has been a very full Article 2 compliant investigation into the circumstances of this death”.

The coroner heard evidence from Dr Freddie Patel, the pathologist, as to the cause of death. He said that he did not find any food material in the lung or the trachea or the main airways. He said that this showed that the collapse from coronary artery disease came first and the vomiting which occurred followed afterwards.

Staff nurse Deborah Chambers gave evidence that she was on duty at the Health Care Centre on the night of 1 October 2006. She said that Mrs Allen was eating sandwiches at 23.50 hrs. At 0.05 hrs, she saw Mrs Allen in the corridor, still eating and with food in her mouth. The next time she saw Mrs Allen was at 0.15 hrs when she found her on the floor in her bathroom. There was still some food in her mouth. Ms Chambers said that she could not feel a pulse and called her colleague Mr Madhewoo. They called for help and they started to attempt a resuscitation by press-ups and mouth to mouth resuscitation. They tried to open the airway. They removed the food from Mrs Allen’s mouth and tilted her head back. By this time, Dr Patel had arrived. He used a suction machine to try to clear the airway. Another member of staff called the ambulance.

Mr Madhewoo said that he and Ms Chambers removed food manually from Mrs Allen’s mouth and throat. He said that Dr Patel took over when he arrived and then they started the CPR. His role was to do the mouth to mouth resuscitation. He said that no airways were inserted by anyone.

Dr Ruchit Patel was the SHO for the entire Health Centre. He said that when he arrived at the ward, he did not see anyone performing CPR on Mrs Allen (p 87). He tried to clear her airway by a finger sweep. The emergency resuscitation bag was already present and the mask was used to begin ventilation and a G-airway was put in. Ventilation was taken over by Kris Madhewoo and Dr Patel continued with compressions. He asked one of the nurses to take over from him so that he could insert a cannula. He did this and administered 1 mg of adrenaline after checking that there was no cardiac rhythm. When the ambulance crew arrived, they took over the airway and inserted an ET tube. They also took over the compressions and ventilation. Dr Patel left them to it.

Staff nurse Eshwan Seedoyal was on the night shift with Miss Chambers on 1 October. His role included alerting the emergency services. He said that the first ambulance was delayed at the front reception gate because the security guard was asleep. He estimated that the ambulance was waiting for about a minute or so before it was admitted.

Emma Smith, an emergency medical technician employed by the LAS said that she arrived at the scene at 00.23 hrs. Her colleague Andrea Shields was already there carrying out basic life support assisted by staff on the ward. There was a doctor (who must have been Dr Patel) who was cannulating Mrs Allen. Ms Smith said that Andrea Shields cannulated Mrs Allen while she (Ms Smith) and her colleague Tony Wright tried to intubate her, but they were unable to do so until they had pulled out some food from the airway. Once the airway was clear, it was possible for intubation to take place. Meanwhile, life support continued and cardiac drugs were administered. She said that her colleague Ms Shields had told her that the staff had told her that effective life support was being carried out before she (Ms Shields) arrived. She also said that she applied monitors to see whether she could detect any rhythms in Mrs Allen’s heart. She said that she was asystolic ie no rhythms could be detected.

Dr Anne McGuinness is an A & E consultant. She gave expert evidence. She was asked by the coroner what are the chances of a person with asystole surviving a cardio-respiratory arrest outside an acute hospital, a coronary care unit or an acute psychiatric ward. She replied:

“If you have a cardiac arrest outside a hospital the chances are very low that you will survive. If you have it in a health care institution, or even a shopping mall where there is a defibrillator and if you are in a rhythm that can be defibrillated, then you have a better chance, but it is still slim. The vast majority of cases who are brought in from the street or from other institutions, by the London Ambulance Service, most of them, although we try very hard, they do not make it out of the resusc room.

If somebody has a cardiac arrest and they go into ventricular fibrillation or another shockable rhythm and the shock is brought in in a few minutes and that shocks the heart back into a normal rhythm and isn’t permanently damaged, then they’ve a reasonably good chance of survival, neurologically intact. But if the cause of the cardiac arrest meant that the heart is already diseased and can’t be restarted, then the chances are almost nil, if you have a diseased heart and it stops.

If it not in a shockable rhythm, which is asystole or PEA, (pulseless electrical activity) then there isn’t an option to restart it with an electric shock, it is, you know, hopeless and it can be brought into hospital with lots of good effective cardio-pulmonary resuscitation on going, and you cannot start a heart that is diseased and has stopped on it’s own accord.”

She was then asked about the possible effect of the delayed arrival of the LAS. She said:

“So if someone is doing chest compressions and someone else, or the same person is doing airways support, the fact that another team was delayed won’t influence the ultimate outcome.”

She added that if the heart and respiration have stopped, you only have a few minutes before you start to have permanent brain damage and the situation will then be irretrievable. She also said:

“It could have been up to 10 minutes ischaemia and the lack of respiratory effort makes me think that it must have been more than a few minutes, because people who collapse usually breathing goes, and the heart stops first, the person usually makes respiratory effort for a little while afterwards, sometimes for several minutes afterwards. So the lack of any respiratory effort and the fact that the, I believe the first defibrillator was attached showed no rhythm at all indicates to me that it was longer than a few minutes, more likely to be over 3 to 5 minutes. And if the heart is not in a shockable rhythm no matter who’s doing the resuscitation, you cannot get the person back.”

Dr McGuinness agreed with the opinion of Dr Freddie Patel that the cause of death was coronary artery disease and acute heart attack causing loss of output which leads to loss of consciousness and loss of breathing in a matter of minutes. This was not a choking incident, since there was no evidence of damage or food debris in the upper airways.

Finally, she was asked to comment on any concerns she had about the basic life support provided and any delays in delivering that basic life support. She said:

“I didn’t hear from the London Ambulance Service, but I heard that there was a delay of them having access, because of a security guard. In this case because there was proficient SHO who had been trained in A & E giving life support, I don’t think it would have affected the outcome. But I think if he had not been there and you were reliant on the LAS to do advanced life support, then that would have been significant and it would be in a future case you wouldn’t wish it to happen and the nurses seem to have recognised the situation quite rapidly and started life support, basic life support, and the SHO was there within a minute. So I don’t think it could really be improved on. I didn’t hear all of the evidence from the first nurse about exactly what she did, but I didn’t really hear all the evidence, but it seemed to me to be when they found the lady they recognised the scene, called for help, started basic life support and advanced life support arrived quite promptly. I might not have heard absolutely everything at the beginning.”

The final witness called by the coroner was Kate O’Regan, the assistant director at Highgate Mental Centre. She dealt with a number of issues including the question whether, as requested by the family, Mrs Allen should have been transferred to another ward which specialised in healthcare for the elderly. She explained that the decision not to move Mrs Allen was not based on considerations of her age but an on assessment of her needs.

In the appendix to the inquisition, the coroner said:

“On 24^th July 2006, she was detained by the Camden and Islington and Social Care trust under Section 3 of the Mental Health Act because her pre-existing schizo-affective disorder of manic type had relapsed due to non-compliance with medication and non-engagement with the Mental Health and Social Care Services. The nature and degree of her illness included thought disorder, grandiose delusions, including a delusion that she was pregnant, and behaviour intrusive and provocative of other patients. Her medication included olanzapine and sodium valporate, an anti psychotic and mood stabiliser respectively. In August 2006, a clinical decision was made to prescribe and administer a trial of acuphase, a short acting intramuscular depot injection. On 17.8.06, a 75mg dose was given i.m. On 19.8.06, a 50mg dose was given i.m. After the second dose, there were a number of episodes when she was found on the floor by staff and it could not be determined whether she had tripped, slipped or fallen as a result of side effects of the medication or whether she was on the floor as a result of her behaviour when acting out her delusion that she was pregnant. The trial of the short acting depot-medication was, therefore, discontinued and the third dose was withheld. Also in August 2006, it was alleged that Mrs Allen had been assaulted by another patient. In September 2006, she was referred by the psychiatric Senior House Officer to the surgical out-patients department. She was treated for a presumptive diagnosis of pneumonia with antibiotics on 07.09.06. She tripped, slipped or fell on the 11.09.06 and a fractured hip was excluded by sending her to the A & E department for an X-ray. On 18.09.06 she developed bilateral ankle swelling but a diagnosis of deep vain thrombosis was excluded by Dopple ultrasound scanning. By the evening of the 01.10.06, she was again taking olanzapine and sodium valporate medication. She was seen on the ward by a number of staff during the evening shift. A member of staff advised her not to eat so quickly. She was seen by the same member of staff at 00.05 on 02.10.06. There was food in her mouth apparent from the distension of her cheeks. Mrs Allen spoke to the Staff Nurse to tell her she was going to the toilet. When the staff member returned to check Mrs Allen’s room at 00.15, she found Mrs Allen collapsed in the toilet, partially cleared her airway and called for assistance. Cardio-pulmonary resuscitation was carried on by other members of staff and the London Ambulance Service was called but she could not be resuscitated. The independent evidence obtained at autopsy by a Home Office Forensic Pathologist and independent evidence about basic and advanced life support, and the effect of olanzapine by a Consultant in Accident and Emergency medicine, indicated that the cause of collapse and sudden death was a fatal cardiac arrhythmia due to underlying coronary artery disease. The probable sequence of events was that the fatal cardiac rhythm was followed by regurgitation or vomiting of the food in the mouth and pharynx rather than a collapse due to choking or aspiration of the oral or gastric contents.”

We make the preliminary observation that it was not incumbent on the coroner to investigate, still less to state his conclusion in relation to, every issue raised by the claimant, however peripheral to the main questions to be determined. In Middleton at [36], Lord Bingham of Cornhill said in relation to an article 2 inquest: “it must be for the coroner, in the exercise of his discretion, to decide how best, in the particular case, to elicit the jury’s conclusion on the central issue or issues” (emphasis added). The coroner was, therefore, required to do no more than focus the investigation and the inquisition on the central issue or issues in the case.

We can now turn to the particular criticisms made by Mr Thomas. In his oral submissions, he formulated the respects in which the coroner failed to conduct a sufficiently comprehensive inquiry somewhat differently from the way in which they were summarised by Blake J. His first criticism is that the coroner failed properly to investigate whether choking and/or a blockage of the airway may have contributed to Mrs Allen’s death. He submits that the coroner should have informed Dr Freddie Patel and Dr McGuinness that there was evidence from Ms Smith that some food was found in the airway after the collapse. He argues that, if they had been informed of this fact, they might have concluded that the failure to clear the airway contributed to the death.

We cannot accept this criticism. First, Dr Freddie Patel was told by the coroner (p 22 of the transcript) that those who attended Mrs Allen in her collapsed state had found that her airway had been blocked. Secondly, knowing this fact Dr Freddie Patel was clearly of the opinion that the cause of death was coronary artery disease and that the food played no part in the death. Dr McGuinness was of the same view: see p 92 of the transcript.

The second criticism is that the coroner did not ensure that Dr McGuinness heard all the relevant evidence and permitted her to assume that the nurses gave basic life support at the outset when in fact they did not do so. This criticism is founded on the fact that Dr Ruchit Patel said that he did not see anyone doing CPR when he arrived. That seems to be borne out by the evidence of Nurse Madhewoo. On the other hand, nurse Chambers said that she and her colleague started resuscitation before the arrival of Dr Patel. At p 88 of the transcript of the proceedings, in one of his questions to Dr McGuinness, the coroner said: “There’s some uncertainty in the evidence as to exactly what [the staff nurse and site practitioner] were doing, compared to what the SHO observed when he arrived, but ultimately when the SHO arrived he led the arrest call”. It is clear, therefore, that the coroner did alert Dr McGuinness to the fact that there was uncertainty as to precisely when the attempts to resuscitate started. He did not permit her to assume that the nurses gave basic life support at the outset. If Dr McGuinness expressed her opinion on a false view of the facts, that would not have misled the coroner. He had to decide when the resuscitation started, if he considered that this was a central issue for him to determine.

The third criticism made by Mr Thomas is that the coroner failed to inform Dr McGuinness that there was an issue on the evidence as to how long it was before the LAS arrived. But it is clear from the answer given by Dr McGuinness (see [30] above) that she did not base her conclusion that the death was unavoidable on the performance of the LAS. Her point was that the life support provided by the SHO was proficient, but even that could not have affected the outcome.

The fourth criticism is that Dr McGuinness was allowed to say that the life support provided was proficient without her being told that the airway was not completely cleared until the LAS arrived. But, as I have said, that point had already been disposed of by Dr Freddie Patel. The coroner was not obliged to canvas it again with Dr McGuinness.

The fifth criticism is that the coroner failed to explore the staff’s inability to use the oxygen cylinders. Mr Thomas draws attention to the fact that the internal inquiry undertaken by the Care Trust had said that the staff had found it difficult to operate the small oxygen canister which was provided with the emergency kit and that they were unfamiliar with the operating procedures for the canister. One of the recommendations of the report was that there should be more comprehensive training in the use of this equipment.

This report was before the coroner and it is true that he did not explore the oxygen canister issue with any of the witnesses. But in our judgment this was not a defect in the investigation. As I have said, the coroner was only obliged to investigate those issues which were, or at least appeared arguably to be, central to the cause of the death. I cannot accept the submission of Mr Thomas that causation is irrelevant in an article 2 investigation. He refers to [31] in R (Amin) v Home Secretary [2003] UKHL 51, [2004] 1 AC 653 where Lord Bingham identified the purposes of an article 2 investigation. It is true that there is no reference to causation in that passage, but it does not follow that Lord Bingham considered that causation was irrelevant. The investigation is directed to seeing whether there has been at least an arguable breach of article 2. It is implicit in such an investigation that what is being investigated caused or may have caused or contributed to the death. Otherwise the link between the investigation and article 2 is severed.

It might have been preferable if the coroner had investigated the oxygen canister issue, since it had been raised in the internal inquiry report. But it is clear that he would have concluded that it did not cause or contribute to Mrs Allen’s death and almost certain that he would have made no reference to it in his inquisition. The sad fact is that by the time Dr Ruchit Patel saw her, Mrs Allen was already asystolic and, as Dr McGuinness said, a person who is asystolic is almost certainly beyond recovery even in a specialist coronary care unit.

The final criticisms made by Mr Thomas are that (i) Mrs Allen was known to have a history of choking (she had no dentures) and yet there was no inquiry as to why she had no dentures; and (ii) she was frail and known to be susceptible to falling and yet there was no inquiry into the question why she had not been moved to a ward for elderly patients. As to (i), for the reasons already given, choking was not the cause of death. As to (ii), these issues were canvassed to some extent with Ms O’Regan. In my judgment, article 2 did not require them to be investigated at all, since on the evidence the fact that Mrs Allen had not been moved to another ward was not even arguably a possible cause of the death.

Sadly, Mrs Allen died of coronary artery disease and nothing could have been done to save her.

For these reasons, Blake J was right to refuse permission to apply for judicial review, although not entirely for the right reasons. The appeal is dismissed.