Griffiths v The Secretary of State for Health

The claimant, Mr Griffiths, is tetraplegic. This issue in this case is whether or not the defendant is, at least in part, responsible for his condition.

Mr Griffiths is a retired vet. His practice was on the Isle of Wight where he has lived for many years. He had alcohol problems. Every so often he would go on binges. They could last for days. His wife, Carol, is an Alcohol Specialist Nurse.

On the morning of 21 July 2010, Mr Griffiths was carrying out some DIY work at home which involved respraying decorative tools. This required the application of several coats of paint. He filled in the time while waiting for the paint to dry by working on his computer upstairs and drinking vodka and orange. Mrs Griffiths was, at the time, at the dentist’s.

At about noon, or shortly thereafter, Mr Griffiths was coming downstairs, empty tumbler in hand, when he lost his footing and fell. The consequences were disastrous. He broke his neck. He ended up sprawled out at the foot of the stairs with his legs resting on the lower steps.

When Mrs Griffiths arrived home she seriously underestimated the seriousness of her husband’s condition. She found him still lying at the bottom of the stairs and saw the abrasion where he had banged his head. She also saw the empty glass which smelt of alcohol. Mr Griffiths was, by this time, conscious and, not realising how serious his injuries actually were, protested that he would get up unaided in due course.

For her part, Mrs Griffiths was annoyed that he had been drinking again. She did what she had done on previous occasions when she had found him incapacitated. She took a photograph of him. Her intention was to confront him with the photograph later and thereby administer a dose of therapeutic discomfiture.

About twenty minutes later, Mr Griffiths was still lying at the foot of the stairs when his wife took another photograph of him. In the period between the two photographs she had removed the empty glass and her husband’s spectacles which had fallen off in his fall.

It was not until about 1.45pm that Mrs Griffiths began to suspect that her husband’s condition was potentially serious. She examined him and found that he had no grip strength. She shone a torch into his eyes which reacted unequally. She thought, mistakenly, that he might have had a stroke. She called the GP surgery and was advised to call an ambulance, which she did. Her call was received at 2.15pm. The ambulance arrived about ten minutes later. By this time, Mr Griffiths had been lying at the foot of the stairs for over two hours.

The ambulance crew, for whose actions the defendant is legally responsible, comprised a paramedic, Mr Hamilton, an ambulance care assistant, Mr Edisbury, and a student nurse who was simply there to observe. Of these, only Mr Edisbury was called to give evidence.

Almost from the outset, Mr Hamilton made a serious error. He assumed, as had Mrs Griffiths, that Mr Griffiths had suffered a stroke and he acted accordingly. There is no dispute that his assessment was negligent. The upshot was that no steps were taken to immobilise Mr Griffiths’ neck. He was lifted into a sitting position, carried into the lounge and sat on a sofa. Thereafter, he was taken to the ambulance in a carry chair. It was only after he arrived at hospital that the correct diagnosis was made.

The central issue to be determined is whether the admitted failure to immobilise Mr Griffiths’ neck was responsible for at least part of his residual disability.

Before moving on to the substantive issues, I must make a passing observation on the sheer scale of the documentation which has been accumulated and offloaded onto the court in this case. In an otherwise meticulously prepared claim, the approach to the preparation of the trial bundle has been egregiously over-inclusive. No fewer than seventeen lever arch files were deposited with the court. They contained well over five thousand pages of documentary material. About a dozen of these lever arch files remained unopened throughout the trial and reference to the contents of three others was confined to just a few pages. It would be no exaggeration to say that about 90% of the documentation was entirely redundant.

With disarming frankness, the claimant’s written opening provides at paragraph 10:

“There are seventeen trial bundles and a core bundle and application bundle. Files four to seventeen contain largely irrelevant documents…”

Sedley’s Laws of Documents (Footnote: ¹) provides, in so far as is material:

“Eighth Law: At least 80 per cent of the documents shall be irrelevant. Counsel shall refer in Court to no more than ten per cent of the documents...”

In marked contrast, Practice Direction 39A of the CPR 3.2 identifies ten limited categories of documents to be included in the trial bundle together with “any other necessary documents”.

In this case, the trial bundles entirely fulfilled the expectations of parody but signally failed to comply with the Practice Direction. Of course, the court should not be too ready to criticise solicitors who may understandably err a little on the side of generosity when deciding which documents to include but this pragmatic indulgence does not excuse wholesale profligacy. CPR 1.3 imposes a duty upon the parties to help the court to further the overriding objective. This duty is not fulfilled by documentary carpet bombing. Those responsible for putting trial bundles together should bear in mind that if they are in breach of the Practice Direction then, in appropriate cases, adverse costs consequences may flow.

I will now return to the substantive issues in the case.

The victim of a serious spinal cord injury remains vulnerable to further trauma in the immediate aftermath of his injury. Those structures which protect both the cord and the integrity of the spine have been damaged and the cord itself is rendered more vulnerable to further deterioration through movement. As the British Orthopaedic Association (BOA) publication “The Initial Care and Transfer of Patients with Spinal Cord Injuries” 2006 records on page 16:

“Any spinal cord injury will have an area of cord tissue at critical levels of ischaemia which surrounds an area of necrosis. Any factor increasing this ischaemia has the potential to cause a deterioration of neurological function and, on occasion, this can be permanent.”

This is why, as the experts agree, any patient with a broken neck should be kept flat and his cervical spine immobilised. As the BOA publication goes on to state on page 16

“Mobilisation requires a graduated and carefully monitored approach. Simply allowing a patient with an acute cord injury to sit without such a programme is unacceptable.” [Emphasis not added].

In this case, it is agreed that Mr Griffiths had sustained a severe injury to his spinal cord as a result of his fall. His head had been thrust backwards by the impact and this had forced his neck into a degree of extension well beyond that which his spine was designed to accommodate. As a result, the third vertebra was pushed back and out of place with respect to the fourth vertebra (retrolisthesis). The space through which the cord ran was thus reduced and the neurons in the cord were damaged. An MRI scan performed on the day after the accident revealed extensive ligamentous disruption and an unstable injury.

Issues of fact arise as to the actual extent to which Mr Griffiths was manhandled by the ambulance crew. I will deal with these later in this judgment. Firstly, however, I must deal with the scientific literature relevant to the issue of causation.

Mr Gardner and Professor Fairbank, both highly regarded spinal injury specialists, reported and gave evidence on behalf of Mr Griffiths and the defendant respectively.

Both experts relied in part upon the scientific literature but, for reasons upon which I will expand, I am satisfied that, in one important respect, the literature failed to support the conclusions which each of them sought to draw from it.

Of particular significance is the extent to which the literature assists in determining the likelihood that, in any given case, inappropriate handling will lead to preventable deterioration.

Both experts relied upon a study of Poonnoose and others published in 2002 in the Journal of Trauma. The study comprised a retrospective review of the records of 569 patients. A prompt diagnosis of spinal cord injury had not been made in 52 of these cases. It was concluded that in 26 out of these 52 cases mismanagement had resulted in neurologic deterioration. Professor Fairbank contended, inter alia, that 26 is only 50% and so does not meet the legal test of balance of probabilities. Mr Gardner countered with the suggestion that a number of the 52 had suffered complete injuries from the outset and deterioration cannot always be demonstrated to occur in complete cases. Accordingly, at least some of these complete injury patients could safely be assumed to have suffered camouflaged deterioration and that this additional cohort would, therefore, tip the balance.

Having read and analysed the Poonnoose paper in preparation for this trial, I was left with serious reservations about the extent to which it assisted me. I raised these reservations during the course of Mr Bishop’s opening on behalf of Mr Griffiths. I noted, in particular, that:

Against this background, I formed the strictly preliminary view that any attempt to deploy this study to support a generic conclusion that, all other things being equal, manhandling would, on a balance of probabilities, cause neurologic deterioration in a spinal cord injured patient could well be spurious.

Having heard the experts, my view remains unchanged. The experts accepted that the value of the study was limited and that it fell far short of the first level of evidence for clinical studies. This is not a criticism of Poonnoose. A high quality prospective randomised trial, for example, would for obvious reasons have been wholly unethical. Accordingly, whilst acknowledging the elegance and ingenuity with which he presented his interpretation of the significance of this study, I have formed the view that, in his written evidence at least, Mr Gardner placed too much reliance upon it.

My conclusion is that this study alone does not entitle this court to conclude, without more, that, generically speaking, evidence of manhandling of a spinally injured patient must lead to the conclusion that it is more likely that not that such manhandling caused deterioration.

At first blush, my interpretation of the Poonnoose study might be thought to put the claimant in serious difficulties. After all, in his closing submissions, counsel for the claimant, realistically in my view, resisted the temptation to present this case as one in which an increase in material risk would be sufficient to establish legal causation. Thus, the conventional “but for” test is to be applied.

However, there is a further and particular feature of this case which requires careful consideration. Mr Gardner relies upon studies which tend to show that neurological motor deterioration is quite uncommon. He also refers to a letter dated 27 August 1999 from the joint author of a National Acute Spinal Cord Injury Study (NASCIS 3) of 1997, which reveals that of 371 patients showing some motor deficit on admission only 3 were in a worse ASIA category at six months. An earlier study, Marshall and others (1987), had found a somewhat higher proportion of patients in which deterioration had occurred of just short of 5%. However, in most of these cases decline in function could be attributed to specific management procedures which were not in themselves indicative of suboptimal care but the deployment of which reflected the challenges of dealing with these types of injury.

Professor Fairbank properly points out that there are limitations upon the extent to which the NASCIS results can be relied upon. In particular, the trial itself was not designed to find out whether manhandling made any difference to the outcome and there is no evidence to suggest that the NASCIS patients were all perfectly handled before arriving at hospital. Nevertheless, he accepted in paragraph 18 of his report of 15 April 2015 that: “mishandling is the main cause of neurological deterioration when it is clearly established”.

A combination of the literature and the expert opinion in this case thus leads me to the conclusion that where there is evidence of material deterioration after significant manhandling then the court would normally conclude that, on a balance of probabilities, the deterioration would not have occurred but for the manhandling.

The resolution of the issue of causation in this case therefore rests upon the answers to two questions:

The starting point must be that any failure to immobilise a broken neck which is compromising the spinal cord has the potential to cause further injury. In this case there is some measure of agreement as to what took place:

There is an issue as to whether Mr Griffiths was sat on the edge of the sofa and then fell backwards. I am satisfied on a balance of probabilities that this is what happened. The defendant points out that neither Mrs Griffiths’ letter of complaint of 1 August 2010 nor the letter before action makes specific mention of this. This is true but I am not satisfied that the omission of this level of detail is a compelling reason for concluding that this is something that Mrs Griffiths and her husband imagined or invented after the event.

I note, in particular, that Mr Hamilton, the paramedic, was interviewed on 11 August 2010 by an investigating officer about the sequence of events. I have concluded that his responses were very candid. When asked on reflection what he would have done differently he said that “he would have done 101 things differently”. He said he would have immobilised the spine with a spinal board and put a collar on. Thus, I have formed the view that he was doing his best to give honest answers in his interview.

He said that Mr Griffiths sat upright in the sofa and that he got the impression that he was supporting himself. Mr Hamilton did not give oral evidence at trial but his colleague, Mr Edisbury did. He said that Mr Griffiths was placed on the sofa with his head supported by cushions. In his interview shortly after the events in question he said that Mr Hamilton’s assessment was carried out when Mr Griffiths was sitting back in the cushions of the sofa but he gave no description of how he got there.

Mr Griffiths’ evidence was that he recalled sitting up for a time because he remembers seeing his own feet directly beneath him. Mrs Griffiths supports his account. I was impressed by Mr Griffiths as a witness. For example, when examined in chief about the likelihood that he would use various items of equipment in respect of which claims had been made, his responses were measured and restrained. When he considered that some items might be of limited value or seldom used he was frank about it. It would have been very easy for him to increase the potential value of his claim by adopting a tone of confected enthusiasm. This he did not do and he thereby enhanced his general credibility.

The evidence is unclear and contradictory about whether attempts were made to put Mr Griffiths in a standing position. On balance I am not satisfied that he was ever raised to the vertical despite Mrs Griffiths’ recollection of hearing the crew saying: “Come on mate. Let’s get you to your feet.” Mr Hamilton gave a fairly detailed account of how Mr Griffiths was taken to the sofa in his interview and I conclude that he was giving an honest and accurate account which did not involve attempting to get Mr Griffiths on his feet. I recognise, of course, that Mr Hamilton did not give evidence and was not therefore subject to cross examination but this is a matter which goes to weight and not admissibility. On this issue, however, I am able to give his evidence sufficient weight to lead me to the conclusion that he was accurate in his recollection.

Accordingly, I conclude that in addition to the factors listed in paragraph 34 above Mr Griffiths was also allowed to slump from a more or less upright seated position on the sofa back onto the cushions behind him.

The importance of the issue as to whether Mr Griffiths’ condition deteriorated over the period from the time when the ambulance crew arrived to the time of his arrival at hospital has been properly recognised by both sides. The defendant advanced the following proposition in its written closing submissions:

“…the issue of causation devolves to this simple question: did the risk, inherent in the failure properly to immobilise the neck, of further neurological deterioration in an existing spinal cord injury, actually eventuate in this case?

There is, or appears to be, consensus that that question is, in turn, answered by the determination of whether or not there was any deterioration in neurological function between the time of the Claimant’s fall down the stairs and his arrival in the Emergency Department.”

I agree with this approach. Particularly in the light of my assessment of the limited value of the Poonnoose study, I would not be prepared to conclude, without more, that the level of manhandling which I have found to have occurred was causative of a poorer outcome than would otherwise have been the case. However, if I were to find in addition that there was evidence to show that on a balance of probability Mr Griffiths was in a worse state when he arrived at hospital than he had been before he had been manhandled then this would establish that Mr Hamilton’s negligence probably caused a poorer outcome for Mr Griffiths.

At the time Mr Griffiths was examined in hospital he was found to have no voluntary muscle power or conscious sensation below the level of his injury. The central question, therefore, is whether there is evidence that this marked a deterioration from his earlier state or whether his condition was no worse than it had been before the ambulance crew arrived.

Mr Gardner relies upon four indications that there had been such deterioration. Each indication depends upon what findings of fact I make. I propose to deal with each in turn.

The question arises as to whether before arriving at hospital Mr Griffiths was able to move his limbs and torso below the level of his injury.

Doubtless when Mrs Griffiths took the two photographs of her husband at the bottom of the stairs she had no conception of the forensic use to which they would subsequently be put. Indeed, until a few weeks ago no apparent difference between Mr Griffiths’ posture in the photographs had been spotted. However, a careful comparison appears to show that Mr Griffiths’ left arm extends further from his body in the second photograph than in the first. The photographs were taken from different angles and so some care should be exercised. However, Mrs Griffiths took measurements at her home based on the photographs and concluded that the movement was of some seven or eight inches. I accept her evidence on this issue and find it to be consistent with what is depicted in the photographs.

The fact that the movement is limited and undramatic explains why it had not been noticed earlier. The defendant denies that the photographs depict movement but argues, in the alternative, that the most likely explanation is that Mrs Griffiths moved the arm when she was testing her husband’s grip. However, in her witness statement she stated that she did not touch or move him. It was about 1.45pm when she tested his grip. At the time she signed her witness statement she had not appreciated the difference in the position of her husband’s left arm illustrated in the photographs and so had no motive, subconsciously or otherwise, to misrepresent the extent and timing of her intervention. The second photograph was taken at about 1.15pm which was about half an hour before she tested her husband’s grip. I am therefore satisfied that Mrs Griffiths did not move her husband’s arm during the period of twenty minutes between the photographs. He achieved this himself as a result of some level of residual function remaining below the level of his spinal injury.

In interview, Mr Hamilton described Mr Griffiths as being in a position which, on the face of it, is different from that depicted in either of the photographs. However, this evidence is inconsistent with all of the other evidence on the issue in this case and on this point I am not satisfied that his recollection, albeit honest, was an accurate one.

Mr Griffiths’ case is that for about thirty seconds he was supporting himself on the sofa before falling back. If this is correct then it is evidence that he had retained at least some level of function below the neck. I have already addressed this issue with reference to the degree of manhandling to which Mr Griffiths was exposed at the hands of the ambulance crew. I find that Mr Griffiths did indeed remain unsupported for a relatively short time but long enough at least to demonstrate some residual function.

Professor Fairbank made the point under cross examination that it would be unlikely that Mr Griffiths could have had truncal support without movement in his upper limbs. However, the value of this point is inevitably diminished by my finding that Mr Griffiths did, indeed, manage to move his left arm whilst he was lying at the bottom of the stairs.

Mr Gardner deployed a vivid anthropomorphic simile in his description of the presentation of a patient with complete flaccid tetraplegia as being like a “puppet whose strings have been cut”. He made the point that an experienced paramedic would not be likely to miss this. In his interview, Mr Hamilton recorded that his assessment of Mr Griffiths included not only the observation that he appeared to be supporting himself but that he had bilateral arm weakness and little to no grip strength. This description is not consistent with complete flaccidity. I am satisfied that an important (but not exculpatory) reason why Mr Hamilton did not appreciate that Mr Griffiths was spinally injured was that he did not initially present as being entirely flaccid.

The defendant draws attention to the fact that in her letter of complaint to St Mary’s Hospital in the Isle of Wight Mrs Griffiths described her husband’s arms as being flaccid. However, I bear in mind that, unlike Mr Hamilton, she was not trained or qualified to make a formal assessment and I find it likely that her examination was limited to a test of grip strength which is not determinative of whole limb flaccidity.

There are references to loss of limb function and inability to move in the ambulance crew notes and Emergency Department records but these fall short of describing complete flaccidity. The notes from Southampton to the effect that Mr Griffiths was found to be quadriplegic with flaccid paralysis are ambiguous and capable of being interpreted as being a reference to Mr Griffiths’ condition on admission and not when at home. References to “quadriplegic@scene” in the Daily Review Sheets take the matter no further. It is not disputed that Mr Griffiths was quadriplegic at the scene. This issue is as to whether his quadriplegia was complete.

There was a level of agreement between the experts that if a complete tetraplegic is suddenly raised he is likely to suffer a sharp fall in blood pressure which could well manifest itself in paleness, faintness or perhaps loss of consciousness. The risk is higher with movement to a standing position than to a sitting posture. There were no signs of this in Mr Griffiths’ case and he consistently scored highly on the Glasgow Coma Scale with respect to verbal response.

Of the four indicia relied upon by Mr Gardener, however, I find this to be the least strongly probative of residual function. The absence of symptoms of sudden loss of blood pressure are consistent with the suggestion that he was not suffering from a complete injury before he was manhandled but I would not have considered this factor, if standing alone, to have been a sufficiently robust indication to discharge the burden of proof.

I summarise my findings thus:

Professor Fairbank suggested that the anatomical nature and extent of the extension injury was such that, but for the application of a very considerable level of further force, there was little further damage to be done. The strength of this argument, however, is substantially undermined by the MRI results which reveal an unstable fracture. I do not, therefore, consider that this point is sufficiently strong to rescue the defendant from the conclusions I have drawn from the rest of the evidence in the case.

Professor Fairbank also relies on the undisputed fact that Mr Griffiths has made a remarkably good recovery. He contends that such a good outcome is inconsistent with a serious injury upon which further trauma caused by manhandling is superimposed. He further points out that there is no literature to the effect that manhandling will cause a less good eventual recovery as opposed to deterioration.

I am unpersuaded by these arguments and share the scepticism of Mr Gardner in this regard. Mr Griffiths’ good recovery does not prove that he suffered only one insult rather than two or more successive insults to the spinal cord. The starting point from which the good recovery proceeded was the complete tetraplegia recorded upon his admission to hospital. I fail to see how his later progress would provide any reliable evidence as to whether his condition upon admission had been caused by one severe trauma or the combination of a less serious trauma which had been made more severe by later manhandling.

I am equally unimpressed by the contention that there is no evidence that manhandling can lead to a less good recovery as opposed to a deterioration. The likely extent of recovery is, in part, determined by the severity of the original injury. The more severe the initial presentation the greater the chance that the outcome will be poorer. I find that the manhandling caused additional damage to Mr Griffiths’ spinal cord. I agree with Mr Gardner that the logical conclusion is that, good as it was, Mr Griffiths’ recovery would have been even better but for the manhandling.

I accept Mr Gardner’s description of the etiology which lies behind this conclusion. Any given neuron in the spinal cord will, if damaged, either remain permanently incapable of conducting electrical impulses or it will recover sufficiently in time to function effectively once more. The greater the extent of the damage to the neuron the lower the chance that it will recover. Accordingly, additional damage sustained by neurons caused by manhandling will lead to a lower proportion of neurons as a whole returning to full function. The consequence is a more limited recovery than would otherwise have been the case. In many respects Mr Griffiths’ recovery has brought him to the cusp of useful functionality and thus even if a modestly higher proportion of neurons had returned to full function this would have been likely to have brought significant benefits. The manhandling has deprived him of those benefits.

In his reports, Mr Gardner has attempted to assess what the extent of Mr Griffiths’ recovery of function would have been but for the manhandling. The defendant seeks to persuade me that this exercise, upon which Professor Fairbank has declined to embark, is so speculative that no useful conclusions are permissible and the claim must fail in any event. I disagree. Of course, the task which Mr Gardner has set himself has involved an inevitable element of speculation but his approach is both coherent and consistent with the reasoning which lies behind it. Accordingly, this claim must now be quantified, either by agreement or further adjudication by the court, with reference to the “but for” assessments of comparative function which are set out in Mr Gardner’s reports.