WALL J: On 24 June 1994, at the conclusion of a contested 5-day hearing, I found as a fact that a male child of 22 months (whom for the purposes of this judgment I will call AB) had suffered multiple non-accidental injuries whilst he was in the care of his parents. I have adjourned into open court for the purposes of this judgment because the case raises important issues both about the function of the court in the resolution of medical issues and about the role of medical experts in cases of alleged child abuse. For reasons which will rapidly become apparent, I propose to identify three of the experts who gave evidence in the case before me. I am, however, more than usually anxious that nothing in this judgment should be capable of leading to the identification of the child or his family. I do not, therefore, propose to identify by name the hospital to which he was admitted, the consultant paediatrician under whose care he was treated or the local authority which became involved in the case as a result of the admission. Furthermore, the initials by which I have identified the child are not his own.
The medical experts in the case
I heard evidence from four medical experts, all of whom directed their evidence specifically to the question of non-accidental injury. They were, in the order I heard their evidence:
(1) Dr H, the consultant paediatrician who treated AB;
(2) Dr Christine Hall, a consultant paediatric radiologist at the Hospital for Sick Children at Great Ormond Street and the Queen Elizabeth Hospital for children in Hackney in London;
(3) Dr Helen Carty, a consultant radiologist at Alder Hay Children's Hospital in Liverpool; and
(4) Dr Colin Paterson, senior lecturer and honorary consultant at the department of biochemical medicine in the University of Dundee.
I also had the benefit of seeing all AB's medical records, including all the records relating to his admission and treatment in hospital. Included in those records was a letter from Dr Tim Jaspan, consultant neuroradiologist of the Department of Neuroradiology at the Queen's Medical Centre in Nottingham and a leading authority on the interpretation of examinations of the brain using ultrasound techniques.
Dr Hall and Dr Carty are both radiologists of the highest distinction. Each has enormous experience in the field of paediatric radiology. Dr Carty told me that for the past 19 years she has worked at the same hospital, which has 120,000 children a year visiting its accident and emergency department, of whom about 80,000 a year are X-rayed. Dr Hall has also been a consultant for 19 years and, within paediatric radiology, she specialises in skeletal disorders. It would be difficult to imagine two experts of higher distinction in the field. Dr Hall was called by the local authority; Dr Carty was called by the guardian ad litem.
Dr Colin Paterson was called by the parents. As I shall have occasion in this judgment to be critical of Dr Paterson, it is only fair that I should record at this point that both Dr Hall and Dr Carty paid tribute to the work which Dr Paterson has done in the field of bone pathology. I should also record my own assessment of Dr Paterson as a highly intelligent man whose manner is sympathetic and whose evidence was given persuasively with both enthusiasm and charm.
Dr Paterson is not, of course, a radiologist, although I accept his evidence that he sees innumerable radiographs in the course of his work. He is, more importantly, not a paediatrician. However, in the instant case there was no material dispute between Drs Hall and Carty on the one hand and Dr Paterson on the other as to the existence, dating or extent of AB's injuries. The critical difference between the experts lay in the assessment of their causation.
The discovery of the injuries
AB was born on 29 August 1992. On 8 November 1992, when he was 10 weeks old, he was taken to the local hospital following what his parents described as an apnoeic attack. According to the note made by the houseman who first interviewed the parents:
'At 12.00 midday father was playing with him (looking out windows at cars) when he suddenly took four or five deep inspirations without apparently expiring and then went pale and floppy. He rushed downstairs by which time [AB] was cyanosed. They smacked his bum and blew at his face and he came round in about 1-2 minutes.'
I accept that as a broadly accurate description of what had occurred. The matter is of some importance because, when she came to give evidence to me, the mother told me that in addition to smacking AB on the bottom or back she had also shaken him in what she described as a desperate attempt to revive him.
The parents telephoned their doctor and a locum general practitioner, Dr R, attended. He examined AB and noted (inter alia) that he:
'. . . seems slightly unable to move the whole of both arms and shoulder joints appeared held back and rigid. There was a slight bruise approx size of a thumb print over right acromion process (the tip of the shoulder) and he seemed to be in some distress on the movement of right or left arm.'
After the main body of the note, which is in typescript, Dr R has added, in manuscript, '?NAI'.
Dr R gave the parents a letter to take to the hospital. That letter asked the hospital to investigate the apnoeic attack and to look at AB's upper limbs. In this latter aspect Dr R's note reads, 'slight hypotonic upper limbs (? normal gleno-humeral joints)'. In other words, the doctor was querying the normality of the appearance of the point in the child's body where the upper arm fits into the shoulder. This point is of importance, as later passages of this judgment will make clear.
Dr H, the consultant paediatrician at the local hospital, takes up the story:
'At the time of his admission he was pale; no bruises were apparent. The right shoulder was swollen and tender to touch. The left upper arm was tender and there were no spontaneous movements of his upper limbs. There were no retinal haemorrhages and physical examination was otherwise entirely normal.
Because of the apnoeic episode and because he was not moving his arms at the time of admission his chest was X-rayed along with both upper arms. This shows fractures. Full skeletal survey at that point was therefore performed.'
The injuries
The full extent of the injuries to AB was as follows:
The right arm and shoulder
[1] A fracture separation through the epiphyseal plate of the upper end of the humerus near the shoulder joint.
[2] Small metaphyseal fractures of the upper end of the humerus and the lower end near the elbow joint.
The left arm
[3] A fracture of the mid-shaft of the humerus.
The chest
[4] Fractures of the posterior ends of the right fourth to tenth ribs adjacent to the spine.
[5] Fractures of the posterior aspect of the right ninth, tenth and eleventh ribs away from the spine.
[6] Corresponding fractures of the posterior ends of the left ribs.
The left clavicle
[7] This was fractured.
The brain
[8] Shearing contusional injuries.
Although not directly relevant to the point which I had to decide, it is only fair to the parents to point out that on further investigation in hospital it was discovered that AB was suffering from a severe gastro-oesophageal reflux. This, Dr H told me, is known to be associated with life-threatening events, and thus may have been the cause of the apnoeic attack suffered by AB on 8 November 1992.
The dating of the injuries
There was no material dispute between any of the medical experts in the case that the injuries had occurred, although Dr Carty was more positive about the corresponding fractures of the posterior ends of the left ribs than Drs Hall and Paterson. Moreover, whilst there were minor differences of opinion about when the injuries had occurred, the broad consensus of opinion was that the fractures I have listed as [1] to [3] (those involving the arms and right shoulder) were recent and sustained within 7 days of 8 November 1992; that the rib injuries (numbered by me as [4] to [6]) were sustained within 2 to 4 weeks of 8 November 1992; and that the fracture to the left clavicle (number [7]) was sustained within 10 to 14 days of 8 November 1992.
As far as the injury to the brain is concerned, the evidence was that it had occurred some days before 8 November 1992. Dr Carty told me that it was, in her opinion, likely to be associated with the rib fractures.
Although the differences between the doctors were not substantial, I record, in order to assist the understanding of this judgment, that on the radiographic evidence I was able to find on the balance of probabilities:
(1) that the fractures of AB's right arm and shoulder and (probably) the left arm were occasioned on 6 November 1992;
(2) that all the rib injuries and the brain damage occurred at the same time on a date between 11 and 25 October 1992;
(3) that the injury to the clavicle was suffered on a separate occasion between 25 and 29 October 1992;
(4) that the injuries were either inflicted or occurred on a minimum of two and, more probably, on three occasions.
The dispute as to causation
Dr Hall and Dr Carty (and, from his different perspective, Dr H) were clear in their conclusion that AB's injuries were non-accidental. For ease of reference I quote from Dr Carty's report to the court:
'This child's bone density is normal. The constellation of injuries as seen in this child with healing rib fractures which are clearly older than both arm fractures, the arm fractures involving as they do, the metaphyses which are fresher are the typical radiological appearances of non-accidental injury. The presence of the slipped upper humeral
epiphysis on the right is further strongly suggestive evidence that this child has been yanked by the arm for this fracture to have occurred. This injury could not possibly have occurred throughout the normal handling of a child.
The rib fractures occur when the chest is squeezed, and the ultrasound appearances would further corroborate this. What happens is that as the child's chest is squeezed it is often vigorously shaken, and brain injuries occur during the shake. The arm fractures must have occurred from roughly snatching the child by the arms.
The fractures are of two different ages. This is further corroborative evidence of non-accidental injury. In my opinion this constellation of features add up to no other diagnosis. The child has no radiological evidence of osteogenesis imperfecta.'
In relation to the brain scan, Dr Carty's initial report to the court was hesitant. However, between the date of her report (2 December 1993) and the date of the trial, Dr Carty's knowledge and experience of brain scans had substantially increased and improved. She was therefore in agreement with Dr Jaspan that the shearing injuries to the brain were almost certainly caused by shaking: moreover, she was in agreement with the radiologist who had originally taken the scans that the shearing injury pre-dated the admission to hospital on 8 November 1992; I need not record the technical reasons for the dating of the damage to the brain. Furthermore, in Dr Carty's view (as the extract from her report makes clear) it was associated with the rib injuries and was caused at the same time.
For the purposes of this judgment, it is both fair to Dr Paterson and necessary for an understanding of my criticisms of him that I recite the whole of that section of his report to the court which he entitles 'Comments'.
'If the radiographs are considered in isolation, it would be entirely reasonable to include non-accidental injury in the list of possible causes. However, the appearances are also consistent with any of the forms of brittle bone disease.
This child was seen on many occasions by observers, other than the parents, both within and outwith the family. Had these fractures been sustained as a result of a series of deliberate injuries inflicted on a child with normal bones, it would be almost inconceivable that evidence of such injuries would not be obvious. In general, in genuine non-accidental injury, bruises are much more common than fractures. In this case, as I understand the evidence, the opposite is true.
The discrepancy between the radiological and the physical evidence of injury is emphasised by the fractures of the humeri. It is clear from the radiographs that the fractures were almost certainly recent, or very recent, when first seen radiologically. The child was reliably examined both by the general practitioner and by a number of hospital doctors and, apart from the deformity, there was little evidence of injury. In my view the major fractures of both the right and left humeri would have represented substantial trauma had the bones been normal. That no other evidence of trauma was recorded when the fractures were fresh
points to an abnormality of the bone.
Similar considerations apply to the rib fractures. A child's ribs are very flexible and substantial trauma is required to cause even one fracture. When rib fractures occur as a result of know accidental injury, the injury is usually very substantial (such as a road traffic accident) and in cases where more than about four rib fractures are seen, trauma is usually severe enough to cause other problems such as damage to the underlying vital organs. As I understand the evidence, there was no evidence that the child had bruises on the chest to suggest injury.
The discrepancy between the physical and the radiological signs of injury is the hallmark of all forms of brittle bone diseases. These are a large group of defects of the protein collagen, the fibrous material responsible for the strength of the bone and also for the structural integrity of other structures including the joints. Most forms of brittle bone disease are due to inborn defects in collagen related to defects in the genes responsible for its formation. The disorders vary enormously from very severe to very mild and the apparent severity of the conditions also varies from time to time.
Our research over the last 6 years had indicated that in addition to these inborn disorders, a variety of "temporary brittle bone diseases" occur. These cause fractures particularly in the first 6 months of life. Fractures are generally spontaneous or with normal handling and many would not be identified but for X-rays done for another reason. One small pointer to this disorder is the fact that the anterior fontanelle was, even when I saw [AB], unusually large. This is a non-specific marker of several forms of brittle bone disease. In my experience metaphyseal fractures and other metaphyseal abnormalities occur not infrequently in this disorder. Apnoeic attacks occur both in affected children and in siblings; the nature of this association is not yet clear.
It is sometimes suggested that brittle bone diseases are less likely if bone density is normal. This view is incorrect. With ordinary radiographs normal appearances are the rule rather than the exception in most patients with known osteogenesis imperfecta in bones that have not sustained previous fractures. In part, this relates to the fact that a reduction in bone density cannot be assessed from ordinary films before about 50% of the bone is lost. In any event, when measurements of bone density are made with sophisticated densitometric methods, the majority of patients with known brittle bone disease have values within the normal range.
Conclusion
The large number of fractures in this case almost certainly reflects some form of brittle bone disease.'
The factual background
Before I turn to analyse Dr Paterson's assertions as to causation, it is necessary for me to summarise the relevant factual background of the case. Judicial findings of abuse can rarely if ever be made in isolation and on medical evidence alone: the factual substratum from which the allegations of abuse arise is usually of critical importance in an overall assessment of
the case. This, of course, is all the more so here, since a medical expert points to extraneous factors (absence of visible signs and the child's appearance as seen by others in the community) as supporting his thesis that non-accidental injury has not occurred.
The parents in this case were and remain married. The mother was 31 when AB was born. At the date of his birth she had a responsible job, which she told me she had enjoyed until she was promoted. Thereafter, she did not enjoy the responsibility which the new post entailed. The father was 32 when AB was born. At that point he was unemployed. That was clearly a fact which affected him deeply. He told me of his strenuous efforts to find work. Happily, he has recently been successful and is currently in full-time employment.
The wider family is highly supportive. The maternal grandfather is a retired medical man; the maternal grandmother regularly saw AB in the first 10 weeks of his life. The paternal grandmother also visited. All three gave evidence in support of the parents.
AB's conception came as a surprise to the parents, who were not planning to have children at that stage. The birth was highly traumatic for the mother. Labour lasted 18 hours: the mother had to have an epidural and an episiotomy and forceps had to be used. There is, however, no evidence that the use of forceps was in any way related to AB's subsequent injuries.
On 15 September 1992, when the health visitor made her first visit to the home, she found the mother pale and tired. She was suffering from mastitis in one breast and had an infected perineum from the episiotomy. By 1 October 1992 the mother was suffering from mastitis in both breasts and she had to abandon her attempts to sustain breast-feeding. I have no doubt that was a bitter disappointment to her. On 14 October 1992, when the health visitor visited, she reported the mother unwell and tired. She was expressing concern about AB's faeces, which she described as black. She was also under additional pressure because her employers were demanding that she write staff assessment reports against a specific deadline. The mother agreed with the description of herself which was put to her as a conscientious and meticulous person: she wrote the reports.
The impression which the mother and the father initially sought to give of these early weeks of AB's life was that all was well. I have no doubt that this was also the impression they sought (successfully) to give to the other members of the family. When AB was taken to the local clinic he was described by the mother as a content and placid baby who was sleeping well and rarely cried. She adhered to that description when she gave evidence before me. However, she accepted under cross-examination that she was in truth under pressure and that the ideal of motherhood which she had pictured prior to AB's birth had turned into a very different reality.
I had the advantage of hearing evidence from the health visitor, and of examining her records. From the records it is apparent that AB was seen on the following occasions:
(1) 17 September 1992 (a home visit) when the health visitor examined the baby;
(2) 23 September 1992 when the baby was examined by the clinical
medical officer at the local health clinic in relation to a bruise on the two small fingers of his left hand;
(3) 8 October 1992 when he was examined by GP on his 6-week check;
(4) 12 October 1992 when AB was seen in the clinic: the parents were concerned about black stools;
(5) 14 October 1992: a home visit and a discussion with the parents;
(6) 15 October 1992: an attendance at the clinic: the baby was weighed. He was also seen by the GP who told the mother that no blood had been found in the faeces (in her records the health visitor mistakenly dates this visit as 16 October 1992);
(7) 21 October 1992: AB was weighed at the clinic: no problems reported by the mother;
(8) 4 November 1992: AB again seen in the clinic;
(9) 5 November 1992: a home visit by the health visitor;
(10) 8 November 1992: AB was seen by Dr R prior to admission to hospital.
The events of 4 to 8 November 1992 were, inevitably, examined in some detail during the course of the hearing. On 4 November 1992 both parents attended the local clinic with AB. They reported he was a contented baby, that he rarely cried, that he fed well, and slept through the night. The parents expressed concern that he settled better with his mother and cried more when his father held him. He was reported to have a cold. He was weighed on this occasion.
On 5 November 1992 the health visitor visited AB at home. The parents again reported that he was a content baby who rarely cried, slept through the night and fed well. The mother stated that she did most of the caring for AB and admitted that she could be critical of the father's attempts at caring for him. As the mother told me, she wanted everything to be 'perfect' for AB: the father did things in a different way.
On 7 November 1992 the father contacted the general practitioner by telephone to report that AB was crying when his right arm was raised. The call was taken by the same locum general practitioner who called on the following day and whose note of that visit I have already in part recited. He was told on 7 November 1992 that AB was otherwise well and had only been in discomfort for the 'past day or so'. He did not therefore visit or see the child on 7 November 1992.
That night the parents went out together without AB for the first time. The maternal grandmother babysat for them for about 4 hours. She fed and changed AB. She did not think that he was in pain. I have already recited the events of 8 November 1992.
On this evidence it seems to me unlikely that the injuries to AB's arms and right shoulder occurred before the visit by the health visitor to the home on 5 November 1992. On 4 November 1992 the child was weighed at the clinic: had he been in pain from the fractures to the arms on that occasion, I think it would have been noticed. Equally, whilst the health visitor did not examine AB on 5 November 1992, I think it more likely than not that she would have noticed any discomfort in moving his arms when she saw the child on that day. On 7 November 1992 the parents reported symptoms which are plainly consistent with the injury: in my
judgment, therefore, it is most probable that these injuries occurred, as Dr Hall argued from the radiological evidence, on 6 November 1992.
It is plain, therefore, that on the facts AB was medically examined within 48 hours of the likely date of the injury to his arms and that the injury was immediately apparent on that medical examination.
Moreover, when one looks at the timings for the rib fractures given by Drs Hall and Carty (between 11 and 25 October 1992) one sees that AB was not seen by the doctor or the health visitor between 15 and 21 October 1992 nor between 21 October and 4 November 1992. Furthermore, the evidence from Drs Hall and Carty was that whilst the fractures to the ribs would be painful when they occurred and would also be painful when the baby was moved, such injuries would not be painful when the baby was lying still. Crying because he was hungry and crying because he was in pain could be difficult to distinguish, and crying because he was in pain could easily be masked by the fact that he had to be picked up in order to be fed. Similar considerations apply to the injury to the clavicle, which occurred on the radiological evidence, between 25 and 28 October 1992.
For the purposes of this judgment it is sufficient for me to say that on the evidence which I heard the following propositions were established.
(1) AB actually had the injuries listed. Therefore, the fact that they were not remarked upon by the wider family or the health visitor has to be examined in the factual context of the case.
(2) When the factual substratum of the case is carefully analysed, there is no inconsistency on the one hand between the fact that AB was seen on a number of occasions by doctors, the health visitor and regularly by members of the wider family and, on the other, the proposition that his injuries were non-accidental.
The duties of experts in children's cases
Before I turn to the detail of my findings on the medical evidence, I propose to summarise my understanding of the duties owed by medical experts in giving evidence in children's cases. It is against that background that I must judge their evidence.
The locus classicus for the exposition of the duties of experts in children's cases is the decision of Cazalet J in Re J (Child Abuse: Expert Evidence) [1991] FCR 193 at p 226, also reported as Note: Re R (A Minor) (Experts' Evidence) [1991] 1 FLR 291. The latter report contains only the judgment given in open court; the former gives both the chambers and the open court judgments. I respectfully endorse and adopt every word of the latter judgment as being of general application in all cases in which expert evidence is involved.
The judgment given in chambers at [1991] FCR 193 is also of interest because this was a case in which a non-ambulant child suffered both fractures and brain damage; furthermore, causation was the issue. Both Dr Paterson and Dr Jaspan gave evidence to Cazalet J. I must also declare a personal interest in the case as I appeared as leading counsel for the parents and myself called Dr Paterson as a witness on their behalf.
In my judgment it is of critical importance in discussing the role of the expert witness in children's cases to bear in mind throughout the respective functions of expert and judge. The expert forms an assessment
and expresses his opinion within the particular area of his expertise. The judge decides particular issues in individual cases. It is therefore not for the judge to become involved in medical controversy except in the extremely rare case where such a controversy is itself an issue in the case and a judicial assessment of it becomes necessary for the proper resolution of the proceedings. The reason for this is obvious. Whilst the judge has knowledge and experience from practice and previous cases, he or she rarely has more medical knowledge than the intelligent lay person: the judge, almost by definition, is not an expert in the field about which the witness is giving evidence.
The judge brings to the inquiry forensic and analytical skills and has the unique advantage over the parties and the witnesses in the case that he or she alone is in a position to weigh all its multifarious facets. This process, of course, involves an evaluation of the expert opinion in the context of a duty to make findings of fact and assessments of the credibility of witnesses. It follows that the dependence of the court on the skill, knowledge and, above all, the professional and intellectual integrity of the expert witness cannot, in my judgment, be over-emphasised. The judge's task is difficult enough as it is in sensitive child cases. To have, in addition, to resolve a subtle and complex medical disagreement or to make assessments of the reliability of expert witnesses not only adds immeasurably to the judge's task but, given his fallibility and lack of medical training, may help to lead him to a false conclusion.
It is partly for this reason that the current practice of the courts in children's cases is to require disclosure of all medical reports and to invite the experts to confer pre-trial. By these means the ambit of agreement and disagreement can be defined.
Mr Kennerley, for the local authority, referred me to an article in Archives of Diseases in Childhood, vol 68 at p 712 in which the author, discussing expert evidence in cases of child abuse, says:
'Experts obviously are involved in pursuing theories, in research, and in trying to persuade the rest of the medical profession of the validity of their conclusion. However, a court is not a suitable forum in which to advance untested hypotheses. The place for that kind of debate is in the medical literature and in conferences and meetings, where others, who are eminently well qualified to do so, may challenge any findings. Judges and magistrates are not in such a position. They must take much of the expert evidence presented to them on trust. As Cazalet J stated: "Outside the legal field the court itself has no expertise and for that reason frequently has to rely on the evidence of experts". Consequently, if an expert believes that current medical opinion on, for example, the occurrence of fractures in babies, is subject to challenge, he or she should present their arguments to medical peers for scrutiny. Only after there has been such scrutiny, followed by acceptance of the validity of the expert's argument, should evidence be presented in a court.'
Having referred to comments made in Report of the Inquiry into Child Abuse in Cleveland 1987, Cm 412 (1988) ('the Cleveland report'), the author then quotes a letter from the then President of the British Paediatric Association which is in these terms:
'The regulation of medical practice is achieved best when it is accomplished within the medical profession. New stances based on a new awareness of clinical signs, or new significance being attached to them, require first to be established within the profession. This takes some time and requires persuasion and scientific evidence of validity, based on the accepted method of communication to professional journals or scientific meetings. It is the dedicated research worker and the pioneering enthusiast who so often change medicine for the better and uncover deficiencies in medical practice and the understanding. In the end, however, any new development has to be fitted into the complex jigsaw which constitutes balanced medical practice. As well as the benefits, the possible adverse effects of any revision of accepted practices have to be taken into account.'
I respectfully agree with the broad thrust of what is being asserted in these two passages. My only caveat is that there are sometimes cases in which there is a genuine disagreement on a scientific or medical issue, or where it is necessary for a party to advance a particular hypothesis to explain a given set of facts. Where that occurs, the judge will have to resolve the issue which is raised. Two points must be made. In my view, the expert who advances such a hypothesis owes a very heavy duty to explain to the court that what he is advancing is a hypothesis, that it is controversial (if it is) and to place before the court all the material which contradicts the hypothesis. Secondly, he must make all his material available to the other experts in the case. It is the common experience of the courts that the better the experts the more limited their areas of disagreement, and in the forensic context of a contested case relating to children, the objective of the lawyers and the experts should always be to limit the ambit of disagreement on medical issues to the minimum.
In Note: Re R (A Minor) (Experts' Evidence) [1991] 1 FLR 291 at pp 292H-293B, Cazalet J said this:
'It should be borne in mind that a misleading opinion from an expert may well inhibit a proper assessment of a particular case by the non-medical professional advisers and may also lead parties, and in particular parents, to false views and hopes.
Furthermore, such misleading expert opinion is likely to increase costs by requiring competing evidence to be called at the hearing on issues which should in fact be non-contentious.
In wardship cases the duty to be objective and not to mislead is as vital as in any case because the child's welfare, which is a matter of extreme importance, is at stake, and his/her interests are paramount. An absence of objectivity may result in a child being wrongly placed and thereby unnecessarily put at risk.'
I would like to reiterate and re-emphasise this passage and in particular Cazalet J's reference to the instillation into parents of false hopes. Nobody wishes to prevent a parent who is accused of the non-accidental injury of a child from putting forward a tenable and innocent explanation for the child's injuries. As Dr Carty put it in her oral evidence to me, a false
accusation of abuse can have a devastating effect on family life. However, the corollary to the exercise of extreme caution in reaching a diagnosis of non-accidental injury is, in my view, the duty of the expert who puts forward an alternative, innocent explanation to do so with equal caution, care and objectivity.
Parental denial is a commonplace of child abuse: to overcome that denial and to accept responsibility for abuse which has in fact occurred is correspondingly painful and difficult for the parent. It is, however, a highly important factor in the consideration of the issue of child protection. Accordingly, if the medical evidence points overwhelmingly to non-accidental injury, an expert who advises the parents and the court that the heavy duty upon him to ensure that he has considered carefully all the available material and is, moreover, expressing an opinion which takes that material fully into account and which can be objectively justified.
If in the situation just posited the truth is that the parent has injured the child non-accidentally, the damage done by an opinion which exonerates the parent is severe. The process of acceptance and recognition is either set back or destroyed; the parent's conviction that he or she has not injured the child is reinforced; the question of rehabilitation of the child is rendered more complex and the risks to the child of a return to parental care become even more difficult to quantify. In short, both the parents and more importantly the child, whose interests are paramount, are ill-served.
In the instant case, both parents, and in particular the mother, told me that because of Dr Paterson's report, no finding of mine would make any difference to their view of the causation of AB's injuries. That assertion meant that the mother and the father, who are both intelligent people, had sat in court throughout the medical evidence called by the local authority and the guardian ad litem with entirely closed minds. I do not know how the parents will react to the judgment which I delivered on 24 June 1994. If, as I fear, their perception of themselves and their conduct does not change, it will render my task on the final hearing of this case the more difficult.
Nobody, of course, least of all a judge, would wish a parent falsely to confess to something which he or she had not done. That was the burden of the parents' complaint in the instant case. Yet, paradoxically, that is precisely what I have found to have happened in this case. The parents' denial, reinforced by their acceptance of Dr Paterson's report, led the mother to assert that she thought she had innocently caused the brain damage to AB by shaking him in the desperate attempt to revive him on 8 November 1992. For reasons which I gave in detail in my chambers' judgment and which I need not now repeat, I do not believe that the mother did shake AB on 8 November 1992, but even if she did, the medical evidence is unequivocal: AB's brain injuries were not caused on that day. Accordingly, in the instant case, the mother's acceptance of Dr Paterson's report has in my view played a significant part in leading her to believe and admit something which is factually impossible.
There is also no doubt that unnecessary investigation of medical issues is very expensive and time-wasting. In the instant case, four specialists were called: they all came from different parts of the country. More than a day was spent hearing the medical evidence.
There is uniformity of view across the various Divisions of the High Court in relation to the duties of experts. In National Justice Compania Naviera SA v Prudential Assurance Co Ltd [1993] 2 Lloyd's Rep 68, Cresswell J sitting in the Commercial Court said that the duties and responsibilities of expert witnesses included the following:
'(1) expert evidence presented to the court should be and should be seen to be the independent product of the expert uninfluenced as to form or content by the exigencies of litigation;
(2) an expert witness should provide independent assistance to the court by way of objective unbiased opinion in relation to matters within his expertise . . . An expert witness in the High Court should never assume the role of advocate;
(3) an expert witness should state the facts or assumptions on which his opinion is based. He should not omit to consider material facts which detract from his concluded opinion. . .
(4) an expert witness should make it clear when a particular question or issue falls outside his expertise;
(5) if an expert's opinion is not properly researched because he considers that insufficient data is available then this must be stated with an indication that the opinion is no more than a provisional one. . .
(6) if after exchange of reports, an expert witness changes his view on a material matter, such change of view should be communicated . . . to the other side without delay and when appropriate to the court;
(7) where expert evidence refers to photographs, plans, calculations . . . survey reports or other similar documents these must be provided to the opposite party at the same time as the exchange of reports.'
Once again, I respectfully adopt every word of Cresswell J's judgment and it is against the background of Re R and National Justice Compania Naviera SA v Prudential Assurance Co Ltd that I propose to assess the expert evidence in the instant case.
My assessment of the underlying causation of the injuries
(1) The brain damage
In my judgment, the only finding which can properly be made on the medical evidence is that the brain damage was caused by the child being shaken on a date prior to 8 November 1992. In my judgment, no other conclusion is possible. The mother's evidence that she inadvertently caused the damage by shaking the child in the desperate attempt to revive him on 8 November 1992 must be rejected as scientifically impossible.
Dr Paterson is, of course, neither a radiologist nor a neuroradiologist. He could not therefore dispute the records of the scans and the evidence relating to it given by Dr Carty and supplemented by Dr Jaspan's letter. That matter does not, however, end there.
Dr Paterson's report is entirely silent on the question of the brain damage. His only reference to the baby's skull is to record at an earlier stage in his report that 'views of the skull were obtained on 8 November
1992 and 14 December 1992. Appearances were normal'. As the passage in his opinion which I have cited demonstrates, he nowhere seeks to fit the causation of the brain damage into the causation of the other injuries. Yet the two are obviously related, and when Dr Paterson was cross-examined he was constrained to accept that if the injuries recorded on the brain scan were older than 8 November 1992 then that fact drove a coach and horses through any suggestion that the brain damage was caused during the attempt at resuscitation. Moreover, Dr Paterson does not even mention, let along discuss, the mechanism of shaking, and the possibility that the rib injuries were occasioned during the shaking episode.
Thus in putting forward an opinion that 'the large number of fractures in this case almost certainly reflects some form of brittle bone disease' Dr Paterson has omitted to take into account a vital piece of information which is wholly inconsistent with his thesis. I regard that omission as a grave breach of Dr Paterson's duty to the court as an expert witness in putting before it an opinion upon which he invites the court to rely. It would be serious enough in any medical witness: it is compounded, in my judgment, in Dr Paterson's case by the fact that he has been criticised for his evidence in other cases and by the fact that he was one of the doctors who gave evidence in Re R.
(2) The absence of bruising
As the extract from Dr Paterson's report which I have cited makes clear, Dr Paterson places considerable emphasis on the absence of external signs of injury on AB. He says:
'Had these fractures been sustained as a result of a series of deliberate injuries inflicted on a child with normal bones, it would be almost inconceivable that evidence of such injuries would not be obvious.'
I must examine this statement in the light of the other expert evidence.
The evidence from Dr Hall and Dr Carty was that fractures in young children frequently occur without evidence of bruising. Dr Hall explained the reason. In elementary terms, a bruise is an injury to underlying tissue which is accompanied by bleeding. Bruising, therefore, depends on soft tissue damage. If, because of the site of the fracture or the nature of the mechanism which causes it there is no significant soft tissue damage, there will be no evidence of bruising. She told me it was common for bruising not to be seen in conjunction with fractures in young children, both accidental and non-accidental. Dr Carty said the same.
Drs Hall and Carty are supported by the literature. In an article entitled Child abuse: some myths and shibboleths published in 1991, the late Dr Leonard Taitz of Sheffield wrote this:
'Absence of bruising
The other major myth perpetuated by those unwilling to believe that parents assault their babies is the notion that the absence of bruising means that the injury occurred with minimal force, thus implying the presence of metabolic bone disease. This is an example of classical myth-making as the primary assumption is untrue (ie fictional), yet,
if it is accepted, it forces the clinician to search for an alternative explanation.'
Dr Carty, in an article published in the Archives of Disease in Childhood in 1988 wrote:
'Fractures both accidental and non-accidental can occur with or without bruising over a fracture site. The presence or absence of bruising over a fracture site gives no indication as to the cause of a fracture. The fact that patients with osteogenesis imperfecta may have a tendency to bruise easily is irrelevant in the context of fractures.'
When he came to be cross-examined, Dr Paterson told me that he had been asked to investigate the cases of approximately 100 children with unexplained fractures where non-accidental injury had been suspected. Of that sample, some 80% had multiple fractures and in the majority of those cases there was no evidence of external injury.
One finds in Dr Paterson's report no discussion of the proposition that fractures in children with normal bones can occur without signs of bruising. Furthermore, the underlying assumptions which Dr Paterson makes about the facts of the case are demonstrably unsound.
First, for these injuries to occur, it was not, on the medical evidence, necessary for this child to be the subject of systematic abuse or, as Dr Paterson puts it, to have suffered 'a series of deliberate injuries'. The evidence is consistent with a maximum of three occasions of injury, well separated in time and on each of which the person responsible suffered a momentary loss of control. Dr Paterson's suggestion that systematic abuse was required is not only plainly wrong, but has had the undoubted effect of reinforcing the parents' denial that anything untoward had occurred. It also reinforced the opinion of the maternal grandfather that because systematic violence would have caused bruising and there was no bruising the injuries must have an innocent explanation.
Secondly, Dr Paterson's assertions as to the events surrounding the injury to the arms and shoulder on 6 November 1992 are factually incorrect. For ease of reference I repeat what Dr Paterson says:
'The discrepancy between the radiological and the physical evidence of injury is emphasised by the fractures in the humeri. It is clear from the radiographs that the fractures were almost certainly recent, or very recent, when first seen radiologically. The child was reliably examined both by the general practitioner and by a number of hospital doctors and, apart from the deformity, there was little evidence of injury. In my view the major fractures of both the right and the left humeri would have represented substantial trauma had the bones been normal. That no other evidence of trauma was recorded when the fractures were fresh points to an abnormality of the bone.'
I confess I have some difficulty in following this paragraph. When the child was examined on 8 November 1992, 2 days after the arm fractures occurred, he was found to have broken ribs, brain damage and a broken collar bone. Moreover, if the injury occurred, as I find it did, when the
child was yanked by the arm on 6 November 1992, the symptoms reported by the parents on 7 November 1992, the examination by Dr R on 8 November 1992 and the subsequent X-rays in hospital are, in my judgment, all entirely consistent with such a non-accidental injury, and inconsistent with anything else.
I therefore have no hesitation in preferring the evidence of Drs Hall and Carty on this part of the case.
(3) 'Some form of brittle bone disease'
Prior to Dr Paterson giving his oral evidence, some time was taken up discussing the different forms of osteogenesis imperfecta and whether or not AB's injuries could be consistent with a rare variant of the condition. There were several learned articles in the court bundles which I read.
When he came to give his oral evidence, Dr Paterson accepted that AB had never suffered from osteogenesis imperfecta, an inherited condition characterised by extreme fragility of the skeleton. To be fair to Dr Paterson, I understand that part of his reason for making this concession was that AB had suffered no further fractures since his admission to hospital in November 1992. Since AB's skeleton was now entirely normal, Dr Paterson's thesis, as developed in evidence before me, was that AB had, during the first 6 months of his life, suffered from what Dr Paterson described as 'temporary brittle bone disease'. In an article published in the American Journal of Medical Genetics in 1993, Dr Paterson suggests that unexplained fractures in infants in the first year of life can be explained by a temporary deficiency of an enzyme, perhaps a metalloenzyme involved in the post-translational processing of collagen, the fibrous protein which is an important feature of bone strength.
It is right to record that the body of medical opinion, represented in the instant case by Drs Hall and Carty does not acknowledge the existence of temporary brittle bone disease in young infants. Accordingly, to examine whether or not there is a scientific basis for suggesting that AB suffered from the condition (assuming for this purpose that it exists) it is necessary to look at Dr Paterson's published work on the subject.
In the article to which I have referred, Dr Paterson produces a number of tables. One of these lists the possible risk factors in temporary brittle bone disease. The five factors listed are birth at less than 37 weeks, birth at less than 33 weeks, multiple pregnancy, artificial feeding and one parent with joint laxity which Dr Paterson described in the body of the article as 'striking' in the cases in which he says he found it. None of these factors applied to AB. The only one which comes near it is the fact that AB's mother had a minor degree of joint laxity measuring 2 on a scale of 1 (normal) to 9 (abnormal). In another table, Dr Paterson lists clinical and laboratory findings in infants with temporary brittle bone disease. The factors listed are vomiting, diarrhoea, oedema, apnoeic attacks, hepatomegaly, and haemoglobin and neutrophils lower than given levels. Dr Paterson accepted that all these symptoms are non-specific. AB did have an apnoeic attack: his haemoglobin was once marginally below the level posited as were his neutrophils. He also, according to Dr Paterson, had blue sclerae and an enlarged fontanelle. Dr Paterson did not give either factor any substantial weight. That, however, is the limit of the scientific evidence.
Having heard the evidence of Dr H, Dr Hall and Dr Carty, I am satisfied that there is no scientific evidence that this child ever suffered from temporary brittle bone disease (once again assuming for the purpose of this part of the case that the condition exists). On any view, however, by no stretch of the imagination could it properly be said under any circumstances that on the scientific evidence AB 'almost certainly' suffered from some form of brittle bone disease.
A further matter of considerable concern to myself was Dr Paterson's answers to me when I queried with him how the instant case would be treated in the context of his further research. Dr Paterson told me that even if I made a specific finding of non-accidental injury, he would still regard the case as being one in which AB suffered from temporary brittle bone disease and the case would be so logged in his research findings. I find this very worrying indeed.
The matter was raised because in a critique of Dr Paterson's work by Dr Taitz (to whom reference has already been made) the latter raised a question relating to two cases which Dr Paterson had claimed were brittle bone disease. Dr Taitz's letter reads as follows:
'Dr Paterson seems to be in two minds in ascribing "unexplained" fractures in young babies to either osteogenesis imperfecta or copper deficiency. In the recent case of R v Lees (heard in the Court of Appeal, Criminal Division in February 1987) Dr Paterson first gave evidence at the trial for the defence diagnosing osteogenesis imperfecta. Later after further research he changed his diagnosis to copper deficiency. On appeal, when the accused sought a new trial, the Lord Chief Justice expressed regret that Dr Paterson had not done his research before giving evidence initially. Lord Lane said that no doubt the result of the trial was a disappointment to Dr Paterson and he was moved thereby to carry out extensive research into the literature of copper deficiency. The court stated that it was unimpressed by much of what Dr Paterson had said and rejected his evidence as a basis for a retrial.
This rejection was echoed in a recent wardship case, where the judge declined to accept Dr Paterson's evidence. On that occasion the court found that Dr Paterson was mistaken in diagnosing anaemia, having relied on erroneous American tables for haemoglobin concentrations, and rejected Dr Paterson's evidence that he had measured the anterior fontanelle of the child at 4.8 cms across. Dr Paterson's approach seems to be to find evidence of collagen defect such as hyperextensibility of the joints in a parent or relative on the basis of clinically controversial procedures and use this as a peg on which to hang the diagnosis of either copper deficiency or osteogenesis imperfecta. Such procedures are not entirely satisfactory. It seems that Dr Paterson now seems to believe that of his two suits osteogenesis imperfecta type IVA is probably the stronger. On the evidence he has produced so far, it is not strong enough.'
This passage was put to Dr Paterson by counsel. Whilst I have not seen a transcript of the judgment on the application for leave to appeal, it is apparent that in the case of R v Lees (unreported) the judicial investigation resulted in a criminal conviction beyond reasonable doubt that the child
had been assaulted and therefore had suffered non-accidental injury. In the wardship case it appears that there was similarly a finding of non-accidental injury. When I asked Dr Paterson how these two cases were treated in his research data, he replied that notwithstanding the judicial findings, both were treated as proved cases of brittle bone disease.
Whilst the courts of course accept that there may be cases where there is a divergence between judicial and clinical findings, I regard as worrying in the extreme Dr Paterson's failure to record in his research material of cases of proven brittle bone disease judicial findings to the contrary. In my judgment this is a factor which must cast the gravest doubts on his findings.
None of these matters was discussed in Dr Paterson's report. He gives no indication that his theory of causation is a matter of controversy and he ignores credible material which contradicts his views.
Conclusions
I set Dr Paterson's report and oral evidence against the guide-lines laid down by Cazalet J in Re R at [1991] 1 FLR 291. In my judgment and to put the matter at its lowest the following criticisms can be made of Dr Paterson:
(1) By failing to deal with the brain damage, he misled by omission to a very serious extent.
(2) By failing to disclose the controversial nature of his research, Dr Paterson lacked objectivity and omitted factors which did not support his opinion.
(3) His attitude to judicial findings in previous cases and his failure to record the fact of such findings casts considerable doubt over the validity of his research data.
(4) By concluding that the large number of fractures 'almost certainly' reflected some form of brittle bone disease he was promoting a particular case rather than reaching a wholly objective conclusion.
(5) His report has reinforced the parents in their false views and hopes that the outcome of the litigation would be to exonerate them.
(6) His report and the need for detailed evidence to be called undoubtedly increased the costs of the case.
If additional considerations were necessary, it is plain to me, looking at Cresswell J's overlapping checklist that Dr Paterson assumed the role of advocate to promote his belief in temporary brittle bone disease.
In Re R, Cazalet J said this of Dr Paterson:
'He accepted that he has been criticised in certain previous cases for developing particular theories as to causation. In the present case, I think he may have developed a theory of causation rather than a diagnosis.'
Mr Kennerley, for the local authority, invited me to make the same finding in the instant case. I regret to say that I think such a finding justified.
As I have already stated, no court wishes to prevent parents accused of non-accidental injury putting before the court an innocent explanation for the injuries their child has suffered. Again, as I hope I have made clear, judges decide individual cases. It is not for me to enter into medical controversy save insofar as it is necessary for the decision of the case before me. However, despite these views, I feel obliged to say that unless a report from Dr Paterson conforms in the future to the guide-lines laid down in Re R and National Justice Compania Naviera SA v Prudential Assurance Co Ltd, his evidence in any case involving the alleged non-accidental injury of infants in the first year of life should, in my judgment, be treated with the greatest caution and reserve.
In the High Court, the judges of the Division and the nominated judges who undertake the bulk of the public law work are experienced in hearing and sifting expert evidence. Moreover, in a case proceeding in the High Court there is likely to be both competing medical evidence of high quality and experienced counsel. My grave anxiety would be that before a less forensically experienced tribunal, notably I have to say the family proceedings court, there may be a real risk that Dr Paterson's evidence would not be subjected to the rigorous scrutiny given to it in the hearing before me and that it might, accordingly, be uncritically accepted. Whilst one would hope that such a case as the present would not be proceeding in a lower court, it is not difficult to imagine circumstances in which a non-accidentally abused child might be returned to abusing parents on the false premise that the child had not been abused. That in my judgment is just as much an injustice as a false finding that a parent has injured a child. In either case, the cause of the proper investigation of alleged child abuse is set back, to the inevitable detriment of the children concerned.
Solicitors: The names of instructing solicitors are omitted in the interest of preserving anonymity.
PATRICIA HARGROVE
Barrister
