England and Wales Court of Appeal (Civil Division) Decisions

IN THE SUPREME COURT OF JUDICATURE
COURT OF APPEAL (CIVIL DIVISION)
ON APPEAL FROM THE HIGH COURT (FAMILY DIVISION)
The Hon Mrs Justice Bracewell

		Royal Courts of Justice Strand, London, WC2A 2LL
		14/05/2004

Dame Elizabeth Butler-Sloss, P:

1. This is the judgment of the Court to which each member of the Court has contributed.
2. These two cases were listed sequentially since in each application for permission to appeal two general issues were raised as to the standard of proof to be applied to applications in care proceedings based upon the allegation that the parent had caused serious harm to the child and to the impact of the decision of the Court of Appeal (Criminal Division) in R v. Cannings [2004] EWCA Crim.1 upon such cases.
3. In the case of LU which we heard first in time, the child, LU, the subject of the proceedings was born on the 11^th July 2001 and is now two and a half. The mother sought to appeal the findings by Bracewell J, in a judgment handed down on the 7th November 2002 at the first (section 31 threshold) hearing, that the mother had been responsible for three incidents of deliberately imposed airway obstruction suffered by the baby in August 2001 and a further incident in October 2001. The second (disposal) hearing was fixed before Kirkwood J for the 17^thMarch 2004. The applications for permission to appeal, for extension of time of some 14 months and to adduce further evidence were adjourned by Thorpe LJ to the full Court. Mr McFarlane QC and Mr Gibbons represented the mother, Miss Vencatachellum for the father, Miss Rowe QC and Miss Stone for the local authority and Miss Hodgson for the guardian. We granted permission to adduce additional evidence. We refused permission to appeal and extension of time. We now give our reasons for our decision.
4. In the case of LB, the child, LB, was born on the 28^th October 1999. The mother sought to appeal the finding by Bracewell J on the 10^th January 2003 at the first (threshold ) hearing that she had been responsible for 11 incidents of rigor suffered by the child between the 25^th September and the 1st October 2001. The decision on the second (disposal) hearing was given by Holman J on the 12^th September 2003. Holman J made a care order and approved a care plan by which LB would continue to live with her paternal grandparents. The applications for permission to appeal and for extension of time were adjourned by Thorpe LJ to the full Court. Mr Cobb QC appeared for the mother and Mr Howard QC and Ms O'Connor for the local authority. The guardian was not represented on the appeal. We reserved our decision on appeal.
5. We invited the Department for Education and Skills to intervene in both cases and it was represented by Miss Eleanor Grey and Miss Waddicor.

The standard of proof in civil family cases

6. Counsel for the mother in each application submitted that the principles set out by the House of Lords in re H (Minors)(Sexual Abuse:Standard of Proof) ) [1996] AC 563, had been refined by subsequent dicta in three recent cases. Lord Nicholls in his speech in re H said at page 586et seq.

"Where the matters in issue are facts the standard of proof required in non-criminal proceedings is the preponderance of probability, usually referred to as the balance of probability. This is the established general principle. There are exceptions such as contempt of court applications, but I can see no reason for thinking that family proceedings are, or should be, an exception. By family proceedings I mean proceedings so described in the Act of 1989, sections 105 and 8(3). Despite their special features, family proceedings remain essentially a form of civil proceedings. Family proceedings often raise very serious issues, but so do other forms of civil proceedings.

The balance of probability standard means that a court is satisfied an event occurred if the court considers that, on the evidence, the occurrence of the event was more likely than not. When assessing the probabilities the court will have in mind as a factor, to whatever extent is appropriate in the particular case, that the more serious the allegation the less likely it is that the event occurred and, hence, the stronger should be the evidence before the court concludes that the allegation is established on the balance of probability. Fraud is usually less likely than negligence. Deliberate physical injury is usually less likely than accidental physical injury. A stepfather is usually less likely to have repeatedly raped and had non-consensual oral sex with his under age stepdaughter than on some occasion to have lost his temper and slapped her. Built into the preponderance of probability standard is a generous degree of flexibility in respect of the seriousness of the allegation.

Although the result is much the same, this does not mean that where a serious allegation is in issue the standard of proof required is higher. It means only that the inherent probability or improbability of an event is itself a matter to be taken into account when weighing the probabilities and deciding whether, on balance, the event occurred. The more improbable the event, the stronger must be the evidence that it did occur before, on the balance of probability, its occurrence will be established. Ungoed-Thomas J. expressed this neatly in In re Dellow's Will Trusts (1964) 1 W.L.R. 451, 455: "The more serious the allegation the more cogent is the evidence required to overcome the unlikelihood of what is alleged and thus to prove it."

This substantially accords with the approach adopted in authorities such as the well known judgment of Morris L.J. in Hornal v. Neuberger Products Ltd. (1957) 1 Q.B. 247, 266: This approach also provides a means by which the balance of probability standard can accommodate one's instinctive feeling that even in civil proceedings a court should be more sure before finding serious allegations proved than when deciding less serious or trivial matters."

7. In the first of the cases relied upon by Counsel, the Divisional Court heard an appeal, B v Chief Constable of Avon and Somerset Constabulary [2001] 1 WLR 340, against the making of a sex offender order under section 2 of the Crime and Disorder Act 1998(the 1998 Act). The justices had found that the defendant was a sex offender within the meaning of section 2(1)(a) of the Act and that he had acted on a number of occasions in a way which brought him within section 2(1)(b). Section 2(1) states that, if it appears to a chief officer of police that the following conditions are fulfilled …. in subsection (1)(a)

"that the person is a sex offender;" and

in subsection (1)(b)

"that the person has acted, since the relevant date, in such a way as to give reasonable cause to believe that an order under this section is necessary to protect the public from serious harm from him,"

then the chief officer may apply for an order under the section.

8. Lord Bingham of Cornhill, CJ in his judgment considered the standard of proof to be applied in that appeal and said at paragraph 31

"In a serious case such as the present the difference between the two standards is, in truth, largely illusory. I have no doubt that, in deciding whether the condition in section 2(1)(a) is fulfilled, a magistrates' court should apply a civil standard of proof which will for all practical purposes be indistinguishable from the criminal standard. In deciding whether the condition in section 2(1)(b) is fulfilled the magistrates' court should apply the civil standard with the strictness appropriate to the seriousness of the matters to be proved and the implications of proving them."

9. We understand from that passage that Lord Bingham was drawing a distinction between the standard of proof necessary to establish each of the two subsections and that the standard of proof set out in the speech of Lord Nicholls in re H (above) was appropriate to proving the facts required for section 2(1)(b).
10. In R (McCann and Others) v Crown Court at Manchester; Clingham v Kensington Royal London Borough [2002] UKHL 39;[2002] 3 WLR 1313, the House of Lords in two appeals was concerned with section 1 of the 1998 Act in relation to the antisocial behaviour of three teenage defendants in the first case and one defendant in the second case. The House of Lords held that applications under section 1 were initiated by the civil process of complaint and were not criminal proceedings. At paragraph 37 Lord Steyn considered the standard of proof to be applied in such cases. He referred to re H (above) and said

"Having concluded that the relevant proceedings are civil, in principle it follows that the standard of proof ordinarily applicable in civil proceedings, namely the balance of probabilities, should apply. However, I agree that, given the seriousness of matters involved, at least some reference to the heightened civil standard would usually be necessary: In re H (Minors)(Sexual Abuse: Standard of Proof) [1996] AC 563, 586D-H per Lord Nicholls of Birkenhead. For essentially practical reasons, the Recorder of Manchester decided to apply the criminal standard. The Court of Appeal said that would usually be the right course to adopt. Lord Bingham of Cornhill has observed that the heightened civil standard and the criminal standard are virtually indistinguishable. I do not disagree with any of these views. But, in my view pragmatism dictates that the task of magistrates should be made more straightforward by ruling that they must in all cases under section 1 apply the criminal standard."

11. Lord Steyn accepted the principles set out in Lord Nicholls' speech above; referred to, presumably, the passage in Lord Bingham's speech in the Chief Constable of Avon and Somerset Constabulary, but gave pragmatic advice to magistrates as how to apply the standard of proof in applications for antisocial behaviour orders.
12. In re ET(Serious Injuries: Standard of Proof) [2003] 2 FLR 1205, Bodey J heard a care application in which the baby had sustained skull, brain and other injuries alleged to be at the hands of her parents. The judge said at paragraph 2 of his judgment that the standard of proof was the civil standard of the balance of probabilities and directed himself according to the principles in re H. He then referred to Chief Constable of Avon and Somerset Constabulary and to McCann (above) and said at paragraphs 5 and 6, citing the passage in re H at page 586 adding his emphasis

"…Although the result is much the same, this [the cogency requirement] does not mean that where a serious allegation is in issue the standard of proof required is higher. It means only that the inherent probability or improbability of an event is itself a matter to be taken into account when weighing the probabilities and deciding whether, on balance, the event occurred."

So it may very well be that, in looking at these more recent dicta, one is (as Miss Ball put it) somewhat 'dancing on the head of a pin'; and no counsel has gone so far as to submit to me that, in a serious case such as this, it is now the criminal standard which should in terms be directly applied.

I therefore propose, in applying the civil standard and the re H (Minors)(Sexual Abuse: Standard of Proof) [1996] AC 563…cogency test here, to have well in mind the dicta in the latter two cases just cited. So, whenever in this judgment I 'find' something occurred, or expressed myself 'satisfied' or 'persuaded' of some fact or other, it is in the light of the authorities which I have just been discussing and on the basis that, in this very serious case, the difference between the civil and the criminal standards of proof is 'largely illusory'."

Conclusions on the standard of proof

13. We understand that in many applications for care orders counsel are now submitting that the correct approach to the standard of proof is to treat the distinction between criminal and civil standards as 'largely illusory'. In our judgment this approach is mistaken. The standard of proof to be applied in Children Act cases is the balance of probabilities and the approach to these difficult cases was laid down by Lord Nicholls in his speech in re H. That test has not been varied nor adjusted by the dicta of Lord Bingham or Lord Steyn who were considering applications made under a different statute. There would appear to be no good reason to leap across a division, on the one hand, between crime and preventative measures taken to restrain defendants for the benefit of the community and, on the other hand, wholly different considerations of child protection and child welfare nor to apply the reasoning in McCann to public, or indeed to private, law cases concerning children. The strict rules of evidence applicable in a criminal trial which is adversarial in nature is to be contrasted with the partly inquisitorial approach of the court dealing with children cases in which the rules of evidence are considerably relaxed. In our judgment therefore Bodey J applied too high a standard of proof in the case of re ET and the principles set out by Lord Nicholls should continue to be followed by the judiciary trying family cases and by magistrates sitting in the Family Proceedings Courts.

The Impact of the Judgment in R v Cannings on Children Act Proceedings

14. In a criminal trial for murder or manslaughter arising out of a series of sudden infant deaths a jury is not entitled to convict and a guilty verdict cannot be deemed safe unless a natural cause of death, whether explained or unexplained, can be excluded as a reasonable (and not fanciful) possibility. In R v Cannings three of the defendant's four children had died in early infancy, although the murder charges related only to two. They had simply stopped breathing. The two other children had experienced apparent life threatening events taking a similar form. She was charged with, tried and convicted of murder. The Court of Appeal (Criminal Division) quashed the convictions. In that case there was no evidence other than the repeated incidents of breathing having ceased. There was serious disagreement between experts as to the cause of death. There was fresh evidence as to hereditary factors pointing to a possible genetic cause. In those circumstances it could not be said that a natural cause could be excluded as a reasonably possible explanation. The court was at pains to point out that it was solely concerned with a criminal prosecution. This is how it was put at paragraph 12 of the judgment:

"…If after full investigation, the deaths or ALTEs, continued to be unexplained, and there was nothing to demonstrate that one or other incident had resulted from the deliberate infliction of harm, so far as the criminal process was concerned, the deaths continued properly to be regarded as SIDS, or more accurately, could not properly be treated as resulting from unlawful violence."

15. Both Mr McFarlane and Mr Cobb founded their respective applications on the effect of the judgment in R v Cannings on family proceedings. Mr McFarlane stressed the following considerations:-

i) The importance of extraneous evidence, particularly where the expert medical evidence is not unanimous.

ii) The danger of interpreting the mother's behaviour as corroborative of guilt, when it may equally be construed as consistent with anxiety and distress.

iii) Nothing is proved by the adage that 'lightning does not strike thrice in the same place'. The absence of an acceptable alternative explanation does not establish an unnatural cause.

iv) The frontiers of medical science are always expanding. A natural cause at present unrecognised may emerge in the foreseeable future. The exclusion of causes known to medicine does not therefore justify a finding of abuse.

16. The paragraphs in the judgment of the Court in R v Cannings that were particularly stressed by Counsel for the applicants are 10, 11, 13, 14, 17, 18, 22, 29, 177 and 178. Mr Cobb even contended that, in cases where there is a serious disagreement between reputable experts leading to one opinion that natural causes could not be excluded as a reasonable possibility, then, absent clear extraneous evidence, Part IV proceedings should not be issued.
17. Mr Howard for the respondent local authority in the LB case identified three points in R v Cannings which might have an impact on care proceedings which are dependent, in whole or part, on expert evidence. They are as follows:-

a) The 'lightning does not strike three times in the same place approach', coupled with the building of a case on statistics.

b) The absence of a satisfactory alternative explanation does not prove deliberate infliction.

c) Where experts testify at the frontiers of scientific knowledge, and particularly where they disagree in such areas, absent cogent extraneous evidence, causation must remain unknown.

18. However whilst accepting the relevance of those aspects of the judgment in R v Cannings, Mr Howard was quick to emphasise the distinctions between criminal and care proceedings. First, the latter are quasi-inquisitorial and the Court's function is child protection and not adult punishment. Second, different standards of proof operate in the two jurisdictions. Third, in criminal proceedings there is a heightened danger of an unjust verdict which cannot be subsequently analysed. By contrast the judge trying a family case makes findings of fact in the course of a reasoned judgment which is open to challenge on appeal.
19. In a helpful final submission Miss Grey for the Department of Education and Skills directly considered the application of R v Cannings to family proceedings. She, too, emphasised the different standard of proof in the jurisdictions, and added that in family proceedings the extraneous evidence relating to the family context would almost always be far wider.
20. Miss Grey then identified the following aspects in which R v Cannings may resonate in family proceedings:-

a. The need to recognise the limits of medical certainty.

b. The dangers of an over-dogmatic approach.

c. The recurrence of medically inexplicable events does not, on its own, entitle the Court to assume a sinister cause.

d. Particular care is required in cases involving a serious disagreement between medical experts. If a body of expert opinion allows that natural cause cannot be excluded as a reasonable possibility, particular caution is required.

e. In care proceedings surrounding evidence of families' circumstances is likely to be as abundant and as probative as medical opinion. The requirements for cogent evidence may be satisfied by a combination of the medical evidence and the extraneous evidence.

21. Miss Grey explained the concerns and indeed the confusion widely felt by local authorities with the responsibilities laid on them by Parts III and IV of the Children Act 1989. They would welcome some guidance from the Court of Appeal as to the applicability of the decision in R v Cannings to their statutory duties. In almost all cases where child protection issues are raised the local authority of the area in which the child is living has to investigate.

Conclusions on effect of R v Cannings on family proceedings

22. In family proceedings the procedures and the rules of evidence are different from criminal trials. In the first place the material available to the court is likely to be much more extensive than would be admitted in a criminal trial. In the second place the standard of proof to be applied before reaching a conclusion adverse to the parent or carer is, as we have set out above, also different. Given a similar background to that in R v Cannings a judge would be required to ask himself which of two possible explanations, human agency or unascertained natural cause, is the more probable. If persuaded by clear and cogent evidence that it was more likely to be the former the court is entitled to reach a conclusion adverse to the parent or carer.
23. In the brief summary of the submissions set out above there is a broad measure of agreement as to some of the considerations emphasised by the judgment in R v Cannings that are of direct application in care proceedings. We adopt the following:-

i) The cause of an injury or an episode that cannot be explained scientifically remains equivocal.

ii) Recurrence is not in itself probative.

iii) Particular caution is necessary in any case where the medical experts disagree, one opinion declining to exclude a reasonable possibility of natural cause.

iv) The Court must always be on guard against the over-dogmatic expert, the expert whose reputation or amour propre is at stake, or the expert who has developed a scientific prejudice.

v) The judge in care proceedings must never forget that today's medical certainty may be discarded by the next generation of experts or that scientific research will throw light into corners that are at present dark.

24. There may have been a tendency in some quarters to over-estimate the impact of the judgment in R v Cannings in family proceedings. The function of the Court in R v Cannings was to evaluate significant fresh evidence. The Court then had to ask the question : might that fresh evidence, if available to the jury at the trial, have resulted in a different verdict? If yes, the Court was bound to declare the conviction unsafe and set it aside.
25. Contrast the role of the Judge conducting the trial of a preliminary issue in care proceedings. The trial is necessary not to establish adult guilt, nor to provide an adult with the opportunity to clear his name. The trial of a preliminary issue is the first, but essential, stage in a complex process of child protection through the medium of judicial proceedings. The State, in the form of the local authority, in order to establish a foundation for intervention in the life of the family, must satisfy the Court –

"(a) That the child concerned is suffering, or is likely to suffer, significant harm; and

(b) That the harm, or likelihood of harm, is attributable to –

the care given to the child or likely to be given to him if the order were not made, not being what it would be reasonable to expect a parent to give to him;" see Section 31 (2) of the Children Act 1989.

26. It is for the purpose of satisfying that threshold that the local authority seeks to prove specific facts against the parent or parents. Only if it succeeds in that task can its application for a care or supervision order proceed. Thus the preliminary issue of fact constitutes the gateway to a judicial discretion as to what steps should be taken to protect the child and to promote his welfare. In those circumstances we must robustly reject Mr Cobb's submission that the local authority should refrain from proceedings or discontinue proceedings in any case where there is a substantial disagreement amongst the medical experts. For the Judge invariably surveys a wide canvas, including a detailed history of the parents' lives, their relationship and their inter-action with professionals. There will be many contributions to this context, family members, neighbours, health records, as well as the observation of professionals such as social workers, health visitors and children's guardian.
27. In the end the Judge must make clear findings on the issues of fact before the Court, resting on the evidence led by the parties and such additional evidence as the Judge may have required in the exercise of his quasi-inquisitorial function. All this is the prelude to a further and fuller investigation of a range of choices in search of the protection and welfare of the children. A positive finding against a parent or both parents does not in itself preclude the possibility of rehabilitation. All depends on the facts and circumstances of the individual case. In that context the consequences of a false positive finding in care proceedings may not be as dire as the consequence of the conviction of an innocent in criminal proceedings.
28. So it by no means follows that an acquittal on a criminal charge or a successful appeal would lead to the absolution of the parent or carer in family or civil proceedings. It is also worth remembering that the decision of the Court of Appeal (Criminal Division) in R v Cannings turned on the very particular facts of that case.
29. In summary the decision of the Court in R v Cannings has no doubt provided a useful warning to judges in care proceedings against ill-considered conclusions or conclusions resting on insufficient evidence. The extent of the retrospective effect remains to emerge. However practitioners should be slow to assume that past cases which have been carefully tried on a wide range of evidence will be readily reopened.
30. In our judgment the responsibilities of local authorities have not been changed by the decision of the Court of Appeal (Criminal Division). Theirs is the task to protect the child; to assess the issues within their competence and expertise; and to rely upon the legal team for the local authority to advise on the strength and credibility of the medical evidence. They will, with their legal advisers, continue to prepare applications for care orders in suitable cases based upon the civil standard of proof as explained by Lord Nicholls in re H. In our view the decision in R v. Cannings does not affect their responsibilities under the Children Act.
31. We turn now to the facts of the first case.

The background in the case of LU

32. The parents were born in England and are Muslim by religion. Their respective families came from the Indian sub-continent. The father was born in January 1981 and the mother was born in June 1983. The judge found that the mother had had an unhappy childhood. She came from a traditional Muslim Asian background and would have been expected to conform to an arranged marriage. She met the father in March 2000 and formed a relationship unknown to either family. She became pregnant at the age of 17 and concealed her pregnancy from her mother and other members of her family. At 8 months pregnant she was thrown out of her home and two days after giving birth she and LU went to live in the family of the father. She did not know any of them and despite the father's family taking her in, there was an atmosphere of disapproval by the members of the extended family. On the 20^th July 2001, the parents went through a religious ceremony of marriage at a mosque but a civil ceremony has not yet taken place.
33. On the 5^th, 14^th, 17^th August and the 13^th October 2001, LU suffered breathing problems whilst alone with the mother and on each occasion was admitted to hospital. On the first occasion on the 5^th August, other members of the father's family were in the house, the ambulance was called and the child was admitted to hospital for checks. On admission the child appeared to be well and was not suffering from breathing difficulties. Various tests were carried out and she did not appear to be suffering from any disease or illness. She was discharged home on the 6^th August. The treating consultant paediatrician Dr R considered it to be a one-off episode for which no explanation was found.
34. On the second occasion on the 14^th August the mother and LU were alone in the house. LU was admitted to the same children's hospital after the mother had called an ambulance. Tests were again performed but nothing was discovered. The child was discharged on the 16^th August. The next day, several members of the family were in the house and about 11.30 pm the mother alerted the father's brother to the condition of the child who was clearly unwell. The ambulance was called and the child again admitted to the same hospital. Again tests were carried out and there was nothing unusual. The child was discharged on the 20^th August.
35. In September LU had a development check and was found to be healthy and progressing well. On the 24^th September 2001, the mother was admitted to hospital having taken an overdose of drugs. On the 13^th October the mother and LU were in the house with another female relative. Again there was an incident and the female relative called the ambulance. The child appeared to the ambulance staff to be alert and responsive and there were no unusual signs. The child was again admitted to hospital but was seen to be well. On this occasion after discussions with social workers who had been alerted to these incidents, on discharge on the 26^th October, LU went to live with her maternal grandmother with whom she has remained pending the conclusion of the proceedings.
36. There was a further admission to hospital on the 11^th November 2001 when LU was living with her grandmother. On this occasion LU was febrile and was found to have an infection. The doctors concluded that she had suffered a genuine illness, diagnosed as retroviral enteritis.
37. In November and the following January the health visitor assessed the mother and considered her to be within the range of post-natal depression requiring additional support and monitoring.
38. The main issue before the judge was whether the evidence, medical and non-medical established to the satisfaction of the court on the balance of probabilities that the Children Act 1989 section 31 threshold had been crossed.

The Medical Findings

39. The judge summarised the alternative possibilities set out by the medical experts as follows: that LU probably either suffered induced upper respiratory obstruction or her symptoms were fabricated or they were described by an over-anxious mother. The judge's conclusions on the medical evidence were as follows:

"Drawing together the strands of the medical evidence, no natural explanation has been put forward as a likely explanation of [LU's] admission to hospital and the most likely explanation is that of suffocation.

I find that, on the medical evidence, [LU's] presentation is most likely to be as a result of smothering."

The Non-Medical Findings

40. The judge then turned to the non-medical evidence upon which the local authority relied. She said

"Although the medical evidence is of very great importance, it is not the only evidence in the case. Explanations given by carers and observers and the credibility of those involved with the child are of great significance. All the evidence, both medical and non-medical, has to be considered in assessing whether the pieces of the jigsaw form a clear and convincing picture of what happened and whether section 31 is satisfied."

41. The judge considered the mother's particular history and how it impinged upon her capacity to care for her child. She found a number of factors relevant to this mother. The mother was very young. She had not had a happy background and was in conflict with a physically abusive brother. She had had no emotional support from either parent and had made an attempt at suicide in 1998. She had rejected the traditional Asian Muslim approach to marriage and became pregnant outside marriage at 17 which was a very great shock. She had had to conceal the pregnancy from her mother and when it was revealed she was rejected by her family. In the father's household she lived in an atmosphere of disapproval and had no resource upon which to rely. She made a second attempt at suicide when the baby was 74 days old after an argument with the father. She told the doctor that her character had undergone a change after the birth of the child and she had become irritable, tearful and depressed. She was not allowed to visit her family nor to go out of the house. She was isolated and unhappy, troubled and frightened by her rejection by her own family and by the strong disapproval of the father's family. The judge found that she was torn apart emotionally by the stress and lack of communication between the two families. This made her vulnerable. The father was well motivated but did not appreciate the extent to which the mother needed his support and his absence was a time of great pressure for the mother. She was unsure of her parenting skills with a difficult and colicky baby. She was deprived of sleep and was tired and stressed and somewhat depressed.
42. The judge found that the father was a truthful witness and so were the members of his family in relation to the relevant facts. On each of the four occasions when these episodes of breathing difficulties occurred, the mother was alone with the child but on three occasions other members of the family were also in the house and gave evidence that the child was unwell prior to the ambulance being called. The judge said

"If reliable, the evidence of the four members of the family is extremely damaging to the mother's case because the evidence is of an observed illness for which there is no medical explanation."

43. Counsel for the mother submitted that the evidence of the members of the family was unreliable but the judge accepted the evidence of the family that the child was ill on each of the three occasions.
44. She also found that the mother's evidence was not credible where it was in conflict with other evidence. The judge found that the mother did not give a truthful account of numerous incidents in which her version was at odds with the professionals and she gave inconsistent accounts, in particular surrounding the admissions to hospital of LU and the mother's account of the baby's medical condition. On the 13^th October she gave two inconsistent versions as to whether the baby had or had not stopped breathing. On admission to hospital she did not give an account to the admitting doctor of the baby's previous health problems until pressed. The judge found her reaction to be extraordinary in her failure to keep appointments with professionals.
45. The judge concluded that

"There are many positives to be taken into account in relation to mother. She loves her daughter, she is attached to her, she is able to comfort her. No criticism has been made of the quality of care and there is no hint of any neglect of her daily needs. ….But it does appear to me that father, by reason of his inexperience of parenting, has failed to spot the signs of stress, and he has underestimated just how difficult and challenging was the mother's position. He failed to recognise her lack of ability and her lack of confidence as a parent.

In conclusion, I find that the combination of medical evidence, the lack of credibility of the mother and the stresses to which she was subject all compel me to a finding that, to the standard of proof envisaged by H and R, this mother has on four occasions between 5^th August and 13^th October deliberately obstructed the upper airway of [LU] thereby causing her harm which involved hospital admissions and invasive investigation."

The Appeal

46. Before this Court the submissions concentrated primarily upon an attack upon the medical evidence. Mr McFarlane did not criticise the reliance placed by the judge on that evidence in the light of the approach of courts prior to the decision in R v Cannings. He submitted however that, in the light of that decision, this Court should approach the medical evidence in a different way. The additional evidence he sought to adduce potentially threw a new light on the medical evidence given at the trial.
47. 
     
    

The Medical Evidence

48. Mr McFarlane's principal submission was that the three medical experts were guilty of paediatric dogma in arriving at their unanimous conclusion. He said that theirs was a systemic approach, elevating an unexplained episode to child abuse on the simple basis of repetition. That approach was condemned by paragraph 13 of the judgment of the Court in R v Cannings and that condemnation applies with equal force in family proceedings. Whilst Mr McFarlane did not attack any of the three experts individually he submitted that the lack of medical disagreement is of no consequence, given that all three adopted the wrong approach.
49. In seeking to make good that submission Mr McFarlane emphasised certain passages in the expert evidence selected either from their written reports or from the note of the discussion at their meeting with the guardian ad litem and her solicitor, or in the transcript of their oral evidence. We will consider the selected passages in relation to each expert in turn.
50. Dr R was the treating paediatrician in doctor / patient relationship with the baby. In his original report he said:-

"The original referral to Social Work was not made by myself although I concur with the reasoning. The basis for the referral is that although there could be a medical cause for these 4 presentations of not breathing, such a number of presentations without medical cause found makes it more likely that there may well be a non-medical cause for these and investigation for non-medical causes is as important, if not more important, than any medical investigation at this point."

51. In his second report Mr McFarlane relied upon the following paragraph:-

"In 4.1, Dr.[D] has stated that the medical evidence supports the diagnosis of induced illness in the form of imposed upper airways obstruction. It is my opinion that this explanation of L's problems is the most likely and it is hard to think of medical reasons for the reported problems, although, children with laryngomalacia and gastro-esophageal reflux, (the two frequently go together), can have similar events without their being interval signs of upper airway stridor or recognisable vomiting."

52. During the course of the discussion meeting there were two exchanges cited by Mr McFarlane. The first was as follows:-

"Q – How much notice would you have taken of the earlier episode which occurred on the 5^th August?

A – A lot, because as we said it is unusual to repeat these things. We are all used to seeing 'one-offs' and we could entertain two, but when we get a second one, in this situation, you're working the statistics out. Also if you get a different presentation, but still within the well-described area of parents causing problems, you take a lot more notice. So I think people did. . . . One's grappling with all of those things but the second episode on top of the first one is very important. It is the start of what is the cumulative effect."

53. Later Dr R made this observation:-

"It could be an innocent cause in the sense that we see children with this story the first time they come in, the only time they come in generally. It is not unusual and we do not know what causes it and they don't do it again. That's what I meant by innocent. We don't know but it's a one-off.

We come back to this thing of adding things up statistically. I think we are going to come back to this statistic of one plus one, plus one, plus one, plus one."

54. There are three passages in Dr. R's oral evidence criticised by Mr McFarlane. The first was in chief when counsel invited him to consider probabilities thus:-

"Q – Clearly, Dr. R, you can only speak from your experience, but can you give us any idea of how often you have seen, starting with two presentations to hospital, similar to [LU]?

A – Starting with a group that presents in that kind of way with a first episode, I would feel that it is certainly much less than one in twenty of those children that have anything similar subsequently.

Q – So 5% the second time. What about third or a fourth time?

A – Well, I said less than 5%.

Q – I am sorry. So that is less than 5%, and I go on to say what about a third and fourth time?

A – It is to some extent guesswork because we don't have the data. We don't have the data because, I think, it doesn't seem to happen. I suppose in that situation figures as guesses may be meaningless and it may mislead the Court if I were to guess in that kind of way."

55. His cross-examiner returned to this area in the following exchange:-

"Q – We each perhaps know from our own experience of life that coincidences do occur. Why could there not be a concentration of different explanations such as have been mooted in this case for [LU]'s admissions?

A – There could be. If we knew the specific cause of each presentation or type of presentation and you knew how frequently those things occur, one statistically can work it out by multiplying one by the other and coming out with a very large denominator of one in such a big number. So we could do that but we addressed ourselves to the balance of probabilities rather than the rare statistical chance."

56. We turn to the evidence of Dr. D. To his written report he annexed a chronology of LU's medical records with his comments. In relation to the first admission he commented

"The significance as part of an evolving pattern of events is obviously far more important as will be seen."

57. In relation to a later admission he commented

"Of course any child can have a single seizure without necessarily having corroborating features, but in isolation this episode is unusual and as part of a pattern of events it takes on great significance."

58. He said that circumstances surrounding another admission were a matter of concern "particularly as part of an evolving pattern."
59. Dr. D's contributions to the experts' meeting were particularly criticised by Mr McFarlane. We set out the five passages complained of in turn.

a) "I agree. The presence of the three other episodes changes the probabilities in relation to the one. We cannot look at them individually. It is like a child who has four linear bruises on their face. You cannot look for four different explanation for each linear bruise because four of them together add up to a hand mark."

b) "As part of an evolving pattern, I would say it was very worrying."

c) "It is possible that children can have episodes like this for no identifiable reason. It is entirely possible that you could take any episode individually. It is the accumulating pattern which makes the diagnosis. No one episode is diagnostic. . . . In probability terms, the whole history is more likely to be seen in fabricated rather than natural illness."

d) "You have to look at the collective probabilities. You have to look at the fact that each event alters the probability of all the others being due to abuse. It is the probability overall which is what has to be looked at."

e) "This is the reason that we have got to look at the whole thing as a sequence and as a pattern of events. The situation on a statistical argument would mean that any one piece of information could not be diagnostic. It is the sequence of events."

60. In Dr. D's oral evidence Mr McFarlane drew particular attention to the fact that at page 6 of the transcript, when asked if four presentations would be highly unusual unless caused by deliberate airway obstruction, Dr D answered

"That is correct. The probability tilts very much towards it being child abuse."

61. In answer to the Judge's question "A differential diagnosis?" he answered

"That is right. The history being a crucial part of that. But I agree with the inference to the question, there is no specific laboratory test or x-ray or anything like that one can do."

62. Finally we turn to the evidence of Dr. H. In his written report he considered possible medical explanations for LU's symptoms, noting that

"many infants presenting with an isolated, brief apnoea will be found to be healthy on clinical examination and no cause will be elicited." However, having considered the relevant features he stated "although potentially plausible medical explanations could be advanced for some individual episodes, there is no single explanation that unites all the features of each of the four separate presentations of this child . . . therefore, having considered the medical evidence in this case, I have formed the opinion that, on the balance of probabilities, imposed airway obstruction is the most likely explanation for the symptoms and presenting features described."

63. During the course of the experts' meetings Mr McFarlane relied on the following three contributions:-

a) "It is more likely in my view that there is a common thread that explains these than there is a separate medical explanation for every single one of these events. So if you take them all individually we get a picture that I think is the wrong way to look at it – that is my view."

b) "You could advance the case that reflux caused all the episodes and was silent in between . . . on the balance of probabilities it is fairly unusual for reflux to be present in this way. On four occasions with such severe apparent effects on the baby with absolutely nothing in between, when one has a fit, healthy and well-fed infant, this makes it less probable but not improbable."

c) "It goes to what Dr. R has said about when the probability shifts so that the future events do not alter it a great deal. The probability in my view becomes high after the second event and it stays high in this case."

64. In the course of Dr H's oral evidence Mr McFarlane drew attention to passages at pages 5, 9, 14, 16 and 17 of the transcript, the effect of which was to emphasise that for Dr. H the sequence of the four admissions was highly influential in reaching his conclusion of deliberate airway obstruction.
65. Miss Rowe QC for the local authority and Miss Hodgson for the child pointed to other passages in the expert evidence which they submitted, rightly in our judgment, demonstrate that, far from adopting an improper or dogmatic approach, the medical experts offered their opinion on the foundation of a fair and careful analysis of the relevant medical evidence.
66. Dr R, as the treating paediatrician, came to the diagnosis of deliberate airway obstruction more reluctantly than his colleagues. In his initial report his starting point was that it was not possible to say whether a non-medical cause as explanation for the episodes was more likely than a medical cause. However in his second report, having read Dr. D's opinion, he conceded that imposed upper airways obstruction was the most likely explanation for LU's problems. However he continued:-

"I am not convinced from Dr. D's report and my re-reading of the medical records that [LU]'s episodes were due to imposed upper airway obstruction. Although it is the most likely explanation, I think there is enough doubt as to make us proceed with caution in determining the outcome care for this child and family, as well as other consequences that might follow."

67. At the experts' meeting, although agreeing in the general conclusion, he demonstrated the fairness of his approach when he said:-

"I agree with that. I am not sure if the question implies that if there is a medical cause mark it down and if there isn't a medical cause then don't. There is a third way – i.e. that as doctors we can find no explanation in our experience or learning background."

68. In his oral evidence he was asked the question:-

"Q – During the course of the experts' meeting there were times – correct me if I'm wrong – where you appeared to display some underlying unease that the nature of the overall opinion being expressed of upper airway obstruction being the explanation for these four admissions. First of all is that right? Is there in fact a sense of unease that you feel in relation to this diagnosis or opinion and, secondly, if that is the case, why?

A – No, I don't think there's any unease with regards my position with respect to the summary opinion / diagnosis, no unease at all. Whenever you get a group of individuals – doctors are no different than anybody else – in a room, even if they hold the same views that are based on good information they will still have a difference in emphasis. I think that was just what was displayed but it didn't alter the summary of opinion in any way."

69. Dr D has a great deal of forensic experience. In his written report he expressed his conclusion thus:-

"I have come to this conclusion with full regard to High Court rulings that took the view that allegations of more serious abuse, being inherently less probable, should be supported by more compelling evidence than usual. I am satisfied in this case that the probability of induced illness is much greater than the probability of some as yet unrecognised natural illness or innocent environmental factor, both of which I think can be discounted in any practical sense."

70. Furthermore in his oral evidence he demonstrated his awareness of the dangers of founding a conclusion on statistics. The following exchange anticipates rather than breaches paragraphs 10 – 13 of the judgment of the Court in R v Cannings:-

"Q – Am I right, Doctor, what you are saying is with a normal child, a child who isn't poorly, it would be highly unusual to get four presentations but it is not highly unusual to get four presentations in a child who might have suffered from some obstruction to their respiratory?

A – That is correct. The probability tilts very much towards it being child abuse. Without droning on too much, if I can just draw a parallel with the issue about two cot deaths which is very widely discussed in the media at the moment and the use of statistics. Now, I am not a great fan of statistics in those cases because if you have two cot deaths then the probability of that happening is small but also the probability of abuse is small. We do know that if you have two cot deaths, the relative probability between that being natural causes or abuse is roughly 50:50. It is not a huge difference in gradient.

Here we have a baby with four presentations with apparently life-threatening events, no medical diagnosis, who is well in hospital, events had occurred to one person and all the other features I have listed in my report. The probability is far, far to the right of 50 per cent; it is very typical of imposed airway obstruction"

71. Later in his oral evidence Dr. D explained his route to conclusion in the course of three answers towards the conclusion of his evidence in chief as follows:-

"As we have all said and we agreed in the experts' meeting, the diagnosis is based on the pattern of events over time and not on any one incident. We have all agreed that you can take any one snippet of information out of this and find an innocent explanation for it. The overall pattern remains, that this is unlikely to be the result of unexplained natural disease and far more likely to be the result of child abuse."

"As I said earlier, the diagnostic process is detailed history, examination and physical signs. I would say this is not simply a diagnosis of exclusion: there are clear features here to support the diagnosis of fabricated or induced illness."

"Yes, it is very difficult to construct any hypothesis, one natural disease explaining all four episodes. You would probably have to look for two or three different kinds of natural disease, all of them self-limiting to account for the different episodes, and medically that falls into the realms of being quite far-fetched."

72. Equally Dr. H demonstrated his essential premise in his written report in these terms:-

"Although potentially plausible medical explanations could be advanced for some individual episodes, there is no single explanation that unites all the features of each of the four separate presentations of this child."

73. He continued:-

"Therefore, having considered the medical evidence in this case, I have formed the opinion that, on the balance of probabilities, imposed airway obstruction is the most likely explanation for the symptoms and presenting features described. This does not preclude the possibility that an unsuspected disease may present itself at a later stage but I have formed my opinion on the basis of currently available evidence."

74. During the course of the experts' discussion there was a question:-

"How could the presentations observed been induced deliberately? What actions would have been required of the perpetrator over what period of time?"

75. The first paragraph of Dr Henderson's answer was particularly relied upon by Miss Hodgson in her submissions. It is to this effect:-

"Well the common fact is that the episodes could have been induced by obstructing the child's airway, so that she was going blue. This could have accounted for all of the presentations. This may be so even though they may differ mechanistically."

76. During the course of his oral evidence the following passages were relied upon by the respondents:-

Q –"Would it be fair – and correct me if I am wrong about this – you have looked at the whole picture of all four admissions and then gone back and considered each one as well?"

A – "I think that is fair. Part of that process is to look at the possibilities that were considered by the medical team caring for [LU] at the time and to consider whether or not there may be some other possible explanation for these events that was or was not considered in those possibilities. Therefore it is important to consider each one in its turn but also to put it into the context of the subsequent events.

So indeed you are quite right that we have to look at each one in its turn. But also, more importantly, in trying to reach an explanation for these events, we have to consider them in the context of there having been other events which were more or less similar to the first."

"So as a clinician there is no doubt that one has to take each event on its merits and try to formulate some sort of idea of what might have caused that specific presentation. However, when one is reviewing a case like this, one has to then put these into context. So, as I was saying before, the probabilities may change as time goes by. I have the advantage of being able to see all four events whereas the doctor seeing the second event did not have that advantage. So the interpretation that we put on that particular event will naturally vary by our experience of subsequent events."

"So I think we have to acknowledge that there are instances where despite intensive investigation there may be underlying pathological disorders that are not brought to light. That is why I made my concluding statement that I could not discount the possibility that something of that nature may present itself at a later stage. But again, as I stated, it is really in medical terms trying to look at a balance of probabilities between various potential explanations for a sequence of events."

Q – "So far as the primary basis of your opinion is concerned, it appears to be that, in the absence of a medical unifying explanation, you are reluctant to look at, in terms of probability, a combination of different medical explanations of these presentations?"

A – "I wouldn't use the term reluctance. I have very carefully considered all the possible explanations I can think of and have reached a hierarchy of probabilities based on my experience, my reading of the medical literature and my reading of the case."

77. The passages emphasised by the appellant, together with the passages emphasised by the respondents, give a fair flavour of the totality of the approach of the experts and the route by which they arrived at their conclusion. In our judgment the charges that they were dogmatic, or that they adopted an unprincipled approach, or that their work requires critical re-appraisal in the light of the judgment of the Court in Cannings are not made good. The judgment in R v Cannings establishes that a case of murder is not to be proved, in the case of a sequence of three unexplained infant deaths in one family, by the submission that lightning does not strike thrice in the same spot. But this is simply not such a case. Here the medical experts, and ultimately the Court, had to determine the causation of four worrying episodes affecting the same child within a relatively brief space of time. The doctor's method of assessing causation is in our judgment not open to serious criticism. Indeed, any other method would probably invite error. Whilst each episode had to be judged in isolation, no less important was a judgment on all four episodes in their sequence and in their pattern.
78. Equally we do not accept that the experts arrived at their conclusion by a process of elimination of possible innocent explanations that justify the ultimate adoption of a malign explanation. Mr Cobb helpfully supplied us with the definition of differential diagnosis extracted from a guide published by the American Medical Association. It is:-

"Differential diagnosis is the method by which the clinician considers the possible causes of a patient's clinical findings before making a final diagnosis."

79. It would be hard to think of a less contentious proposition. It amounts to no more than the earliest stage of what must be an elaborate process of evaluation in a complex case such as this.
80. At its conclusion the experts did not suggest that their explanation was certain, but only that it was established on the civil balance of probabilities elevated to recognise the gravity of its consequences.

The Additional Evidence

81. The parents have very large extended families, most of whom live in the Indian sub-continent. We were asked to admit additional evidence which was provided at a very late stage of the appeal process, comprising a history of the maternal and paternal families. The family tree is incomplete, sparse in detail and unverified. It appears from the statements of members of the families that up to 30 children have died in infancy over an unspecified period of time. The accounts are largely taken from the grandparents. All but one of these children died abroad. The medical causes of the deaths are not set out, nor even dates and places of where the children lived or the surrounding circumstances. There are no medical records or other documentary evidence to support these accounts. We were informed that a tiny percentage of births of children are registered in the other jurisdiction.

The Expert appraisal of the additional evidence

82. The three medical experts were asked to comment on the additional evidence and both Dr H and Dr D commented on it at length. They both noted that the evidence of infant death within the extended family is derived from oral reports from LB's grandparents. The circumstances and causes of the reported deaths were not substantiated. Both observed that the rate of infant mortality in the country where all but one of these infant deaths were reported to have occurred was approximately twelve times the rate of infant death mortality in the United Kingdom. Dr D made the further points that:

A proportion of the reported deaths appear to have occurred several decades ago when infant mortality rates in the country in question were as high as one in seventeen. Therefore the number of reported infant deaths does not appear significantly greater than the norm for that country.

Infant deaths are reported to have occurred in unrelated arms of the extended family, reducing the likelihood of a unifying hypothesis of a genetic condition.

83. Both Dr H and Dr B concluded that the additional evidence did not call for a revision of their opinion that on the balance of probabilities LU was the victim of imposed airway obstruction. Dr R, in a brief report, also states that his opinion was unchanged by the additional evidence. However he favoured further investigation and suggested that the experts should remain open to the possibility that further investigation might require reassessment of the fundamental conclusion.
84. Although the deaths of so many children is extremely sad for the families, the circumstances of those deaths are uncertain; the availability or otherwise of medical facilities is unclear and there may well have been varying social conditions for these families. The picture is incomplete, lacking in the most basic detail and it would not be possible for its relevance to LU to be properly assessed. We do not consider that this evidence with its obvious inadequacies can tip the balance to a degree sufficient to render necessary a further investigation of the unanimous medical evidence. The views of the medical experts is strong support for our conclusions that the additional evidence provided by the mother has not been demonstrated to be likely to have an important influence on the result of the case if it had been raised before the judge.
85. We therefore concluded that a rehearing was not warranted on the basis of the additional evidence.
86. We have not taken into account the further factor that this evidence could have been made available for the hearing before the judge.
87. There was also the issue submitted by the mother that LU had developed asthma since the hearing before Bracewell J, information said not to be known to the experts at the time. We were however informed by Miss Rowe that this development was referred to in the mother's written evidence and was considered by the experts at their joint meeting.
88. For the above reasons we are satisfied that no criticism can be made of the medical evidence in the case of LU. The judge was entitled to rely upon it, as she did, in order to come to her conclusions.

The Non-Medical Evidence

89. Mr McFarlane did not seek to challenge the findings of fact made by the judge in respect of the non-medical evidence but criticised her reliance upon those findings to support the medical evidence and in forming her conclusions. He submitted that there was more than one interpretation of the mother's emotional vulnerability, and not only a sinister one. It was submitted that there was no other evidence of concern than the incidents requiring admission to hospital.
90. The local authority and the guardian submitted that the non-medical evidence was important and the unchallenged non-medical findings of the judge significantly contributed to her ultimate conclusion that the child would be at significant risk in her mother's care.
91. In our judgment, the judge was entitled to take into account the history of the mother, her two attempts at suicide, her estrangement from her own family, her lack of support and her vulnerable situation in the household of the shocked and disapproving paternal family. The judge found that she was not frank with the court and there were a number of important discrepancies, carefully listed by the judge, between the evidence of the professionals and that of the mother. Clearly the judge was entitled to take into account her lack of credibility in particular in relation to the factual situation about the baby's health. This was highly relevant since on each occasion she was alone with the baby when the incident occurred and her version of events could not be relied upon. It was significant that on each of those occasions the father was away and she was under strain.
92. These are all important supporting elements in the jigsaw put together by the judge. As the judge said, the unanimous medical evidence and the surrounding circumstances, the stresses to which the mother was subject and her lack of credibility pointed strongly to her conclusions. Bracewell J directed herself correctly on the standard of proof and the guidance given by Lord Nicholls in re H and delivered an impeccable judgment. For those reasons we refused permission to appeal and extension of time.
93. We now turn to the case of LB.

The background to the case of LB.

94. The mother was born in 1971. Her previous history before meeting the father was set out by the trial judge who found that she had had recurrent throat and nose bleeds in 1987 and 1988, considered by the psychiatrist to whom she was referred to be self-inflicted. The mother conceded at the hearing that they had been self-inflicted. In October and November 1988 she was treated for bruising and swelling of the left wrist and hand which appeared genuine to the treating doctors but were later proved to be self-inflicted.
95. The father was born in 1966. They met in 1989 and lived together until they parted in 1992. The mother commenced training as a nurse in 1991 and subsequently qualified as a paediatric nurse. She gave birth to a child, MEB, by another man in 1995 from whom she parted soon afterwards. She re-established a relationship with the father in 1997 and MEB was adopted by them on the 7th September 2001, shortly before the events the subject of these proceedings. In 1997 the mother was employed as a paediatric nurse in the local hospital. In July 1998 their first child and the mother's second child, MAB, was born. The parents married in May 1999.
96. LB was born on the 8^th October 1999 with certain disabilities. Prior to the birth the mother was suffering from depression and was prescribed anti-depressants. The mother discharged herself and LB on the 31^st October but they were readmitted because of feeding difficulties for LB. On the 7^th December LB was admitted to hospital since she was failing to thrive, in particular, failing to put on weight and concerns were raised about the mental health of the mother. LB put on weight in hospital and was discharged after two weeks to her mother's care. She again failed to thrive and was readmitted to hospital on the 9^thJanuary 2000. At this time the mother was recorded as depressed, with thoughts of suicide and of self-harm. On the 20^th January the mother and LB were admitted to a psychiatric unit. The mother was suicidal. She viewed LB as damaged and evil. The baby's weight on admission to hospital was below the fourth percentile and after two and a half months at the unit she had improved to just below the 75^th percentile. The mother discharged herself against medical advice on the 31^st March because she did not wish to be admitted to the psychiatric unit. The mother also said that she did not want to be separated from the baby since, if separated she would not want the baby back. After their return home the baby's weight fell again.
97. On the 11^th April 2000 the health visitor and the social worker visited the home and the mother said that she felt unable to cope. She said that she had not bonded with LB and it felt as if she was looking after someone else's baby. On the 15^th April LB was admitted to hospital after the mother said she had become dizzy and had fallen backwards on to LB who appeared to have suffered no injury. The mother said that she was ill and felt differently about LB than about her other two children. Towards the end of April the social worker visited the mother who referred to LB as evil and that part of her wished that LB had been hurt so that she could have time out from caring for her. She said that she did not feel bonded with LB and resented her for causing the mother ill health. The mother had made a drawing of LB as a devil.
98. LB was readmitted to hospital on the 28^th April, her weight having fallen below the 25^th percentile. Two nurses gave evidence that they could see into the cubicle where the mother was feeding LB while she was in hospital and that they saw her watering down the milk in the bottle she gave to the child. LB should have been fed with full strength milk.
99. On the 5^th May the mother failed to attend a child protection conference and LB's name was placed on the child protection register. The parents appealed against the registration. At a reconvened case conference later in May 2000 the mother's mental health had steadily improved and the consultant paediatrician considered that no child protection issues arose. LB's name was removed from the register. The mother returned to work on the paediatric ward.
100. On the 5^th October 2000 the mother took an overdose of paracetamol and was admitted to hospital where she refused medication, threatened to run away and was sectioned under the mental health legislation and transferred to a psychiatric unit. She was discharged on the 23^rd October following an appeal and an assessment. In November she told her general practitioner that she felt panicky towards the baby; felt physically ill when she fed her; that LB's face was not right and looked like a Picasso picture. On the 5^th January 2001 she told the psychiatrist that she had thoughts of self-harm and she was readmitted to the psychiatric unit for 10 days. She declined the offer of help from a family care worker. By the end of April the mother's health had greatly improved and she returned to work. The social services closed the case file in May 2001.
101. In the third week of September 2001 LB was not well and the general practitioner referred her to the local hospital. She was admitted as an in-patient and remained there until she was transferred to Great Ormond Street Hospital for Sick Children on October 1^st 2001 for investigation. During the 6 days at the local hospital LB suffered 11 episodes of rigor with potentially life-threatening consequences. The treating consultant paediatrician became concerned and suspicious about these incidents since the hospital was unable to identify a cause for any of them. He directed that a close watch should be kept on the mother and on LB prior to the transfer to Great Ormond Street. The mother was not working on the ward during the period of LB's admission (though LB was in fact, during her admission, a patient on the ward on which the mother routinely worked).
102. LB had a cannula attached to her while she was in the local hospital which was disconnected by the mother before the child went to Great Ormond Street. No further cannula was inserted. LB suffered no further rigors at Great Ormond Street although she remained unwell. She recovered and returned to the local hospital on the 5^th October. At a further child protection conference LB's name was placed on the child protection register under the category of physical injury. LB has not returned to live with her parents and sisters. The parents agreed to LB's voluntary accommodation and she lived from the 18^th October 2001 until the 14^th July 2002 with her great grandmother. On the 25^th May 2002 the judge found that she probably suffered another rigor which began in the absence of her mother. LB now lives with her paternal grandparents under a care order made by Holman J on the 12^th September 2003.
103. The judge divided the evidence into three sections, the medical evidence, the credibility of the mother and the opportunity and knowledge of the mother to act as alleged by the local authority. Having referred to the Department of Health guidance on 'Safeguarding Children in Whom Illness is Fabricated or Induced', she said

"It seems to me that it is entirely inappropriate to try to shoehorn the mother into a particular syndrome instead of ascertaining what actually happened. The question to be addressed is not: "Does mother fit any definition?" but instead: "Has the local authority proved that the mother deliberately injured her child?" I do not find labels helpful at this stage of the inquiry in determining the crucial issues. Thus the relevance of mother's previous history of self-harming and depression is not in respect of any issue of labelling but is to credibility and the history of her commitment to her child."

The medical findings

104. We set out the medical findings of the judge later in this judgment.

The non-medical findings

105. The judge held that the mother had been present or nearby on each of the 11 occasions when the child suffered rigors.
106. The judge found that there were many positives in respect of the mother. After her depression was cured all the witnesses spoke of a good relationship between the mother and LB and no one observed any behaviour which might have given cause for suspicion. Many of the nurses questioned the opportunity for the mother to injure the child. One matron thought it unlikely but possible. The other matron thought it was impossible. The social workers found the mother to be co-operative and spoke of high quality contact between her and the child. They trusted the mother to comply with local authority plans. The wider family spoke highly of the mother's quality of care for LB and other witnesses spoke of a warm and loving relationship. Neither of the other two children had ever been the subject of care proceedings or of any concerns of the local authority.
107. The evidence was that the mother suffered from an uncommonly severe depression. Her general practitioner, Dr W, who had been so since the birth of LB was clear that the mother did not always have insight into her condition; was reluctant to accept help when she needed it; insisted on breastfeeding which was unsuccessful and was unable to give LB sufficient attention. The judge found that the improvement of the mother's health which enabled her to return to work in June 2001 was short lived and illness again overtook the mother.
108. The judge found that it was plain from the evidence that LB was not an easy baby to care for by reason of her illnesses and her various disabilities. The judge said

"I find that these recurrent and genuine illness of [LB], combined with mother's recurring depression, were not a happy combination."

109. The judge considered it to be significant that on the 21^st November 2000 when Dr W saw the mother, the mother was in floods of tears, feeling frightened, panicky towards LB and feeling physically ill when she fed her. This was the occasion when she said that LB looked like a Picasso picture. The judge said

"At this time I am satisfied that there was a real rejection of [LB] by mother, which was unobserved by Dr B, who thought that the relationship was good at this time."

110. Dr W was concerned as to how the mother might behave towards LB and she wrote in her notes

"It seems she is still able to be angry and take advantage. She continues to baffle me."

111. The judge found that the mother had had a conversation with Dr W in which she admitted causing recurrent bleeding to her gums, a conversation which the mother denied.

The credibility of the mother

112. The judge assessed the credibility of the mother. She set out in detail the circumstances in which the mother lied or exaggerated. The judge said

"Mother is undoubtedly a skilled and persistent liar, who over the years has sought to and succeeded in conning doctors, teachers and family in respect of illnesses which she claimed to be genuine but which were self –induced. She has wasted hospital resources and submitted to procedures she knew were unnecessary…

The lies led to significant admissions to hospital with extensive tests…

In effect, at this time [1987-1988] her whole life revolved round a deceit. It is significant that the professionals have found her plausible and sensible until the truth was discovered and she was unmasked.

The truth was that she was attracting attention by manufacturing symptoms and complaining when she was not believed and when she protested her genuineness."

113. The mother contended that the incidents were as a result of her unhappiness as a child and that she had been frank and that the history was irrelevant to the present proceedings. The judge found that

" However, I am satisfied that the extent to which mother admits anything depends upon her own perception of the best presentation of her case. She cannot any longer deny the self-harm, because it is well documented, but I find she did lie to the social services team manager, and to Mr S, the social worker, when she denied using a knife to cut her forearms.

I find mother does manipulate events in order to get what she wants."

114. The judge then set out examples of manipulation, such as an occasion when the mother wanted a hospital bed and when it was refused she took an overdose; the incident on the 15^th April when she exaggerated an injury to LB in order to get the child admitted to hospital and to give her some rest from the child. On an occasion in November 2001 she went to the wrong department for a gynaecological consultation and in order to get attention repeatedly said that she had a child who was dying.
115. The judge was impressed by and accepted the evidence of a friend of the family, Mrs N, who cared for LB on occasions and who described the mother as appearing to seek attention and to exaggerate. Mrs N observed that the mother was very negative towards LB until the May case conference when the mother was spurred into determination to prove that she had recovered. The judge considered that the evidence of Mrs N that the mother was increasingly obsessed to prove that LB's failure to thrive was not due to the mother's illness but by reason of the illness of LB was very telling. Mrs N also expressed concern about the mother's inadequacy in preparing suitable meals for LB and in feeding her, whereas Mrs N did not find LB difficult to feed. Eventually Mrs N complained to the authorities about the care of LB by her mother. The judge held that Mrs N had a better view of what was happening to LB than the professionals since she saw it at first hand.
116. At the incident of the 25^th May 2002 when the child was in the care of her great grandmother and suffered a rigor, the judge held that the reaction of the mother when she realised that LB was suffering a rigor was significant. She did not call an ambulance and her concern was her own self-protection and not concern for the welfare of the child. When the child was taken to a doctor, she did not give any previous history of the child's medical problems and stated: "It was not for me to diagnose." This answer the judge found to be extraordinary and wholly inconsistent with any real concern for LB. The judge's view was that a concerned parent would have told the doctor what had happened in the past and that the child had previously suffered from rigors.
117. The judge's assessment of the mother was

"I find mother an unimpressive witness who did not tell the truth about many aspects."

Opportunity and knowledge

118. The judge pointed out that the rigor attacks in September 2001 always commenced when the mother was on or about the ward and some around shift handovers. Although the mother and child were supposed to be observed, it was not close or strict and some nurses did not know why they were supposed to monitor her. The mother was a colleague. The judge held that the nurses did not monitor closely and surveillance was minimal. The mother had access to all parts of the ward and the judge was satisfied on the totality of the evidence that the mother had the opportunity and the knowledge to insert some substance into the cannula.
119. The judge considered that the incident of the dilution of the milk feed on the 28^th April 2001, which the mother strenuously denied, was important in the context of the situation. The mother knew there was the threat of an emergency protection order. She had earlier said she wished the child had been hurt on the 15^th April so she could have the child admitted to hospital. She was according to the judge in a desperate emotional state, on anti-depressants and to use a colloquialism 'at the end of her tether'. The judge said that

"I find that it was very much part of her agenda to show that [LB's] problem was organic as opposed to neglect."

120. The judge did not accept that the failure to thrive was the result of a milk intolerance and that any intolerance to quantities of milk was exaggerated by the mother in order to demonstrate illness. In relation to the issue of the removal of the cannula prior to the transfer to Great Ormond Street there was a stark conflict of evidence between the mother and nurse in charge, Nurse W. The judge had no hesitation in preferring the evidence of Nurse W. The judge found that the mother removed the cannula without permission; that she knew she should not have done so, and would not have been given permission if she had asked. The judge found that the mother had lied to the court about the removal of the cannula.
121. The judge also said that the mother had an agenda of wishing LB to be ill and gave the instance when, on the 17^th September 2002, the father told Mrs T that his wife got bees in her bonnet and was angered by the fact that LB was well.
122. The judge carefully set out her conclusions as follows

"To summarise the factors which cause me to reach my conclusions that to the appropriate standard and burden of proof the local authority have demonstrated that mother deliberately interfered with the cannula of [LB], I find the following factors.

1. On several occasions [LB]failed to thrive in the care of mother for non-organic reasons, and was neglected by insufficient food intake and attention when, in contrast, she thrived in hospital. A telling description of [LB] at one of her stays in hospital is that of an anorexic waif.

2. The relationship between mother and [LB] before May 2001 showed not only a lack of bonding and resentment of the child, but indeed rejection of her.

3. Mother has a history of lying and deceiving others in respect of herself over a substantial period of time.

4. Mother's agenda to demonstrate [LB]was ill, to divert blame from herself and her tendency to exaggerate.

5. The absence of any rigors prior to the insertion of the cannula.

6. The presence of mother on the ward around the time of the rigors with opportunity and necessary knowledge how to interfere with and contaminate the cannula.

7. The frequency and severity of the rigors with no medical explanation to account for them naturally.

8. The absence of rigors at Great Ormond Street when no cannula was inserted.

9. The incidence of feeding diluted milk in April 2000, the attempt to get [LB]'s admission to hospital on April 15^th 2000, the exaggeration of milk intolerance, the removal without permission of the cannula on October 1^st 2001, and the reaction to events on May 25^th 2002.

10. The evidence of [Mrs N].

11. The conclusion of Dr S and Dr. M that the most likely explanation for the rigors is deliberate interference with the cannula in the absence of any medical cause.

12. My findings of lack of credibility of mother and my rejections of her explanations.

In those circumstances, I find the threshold criteria are established to the appropriate standard of proof in the following respects.

1. [LB] suffered non-organic failure to thrive from birth until May 2001

2. Mother deliberately administered some unidentified infected substance to [LB], thereby causing the rigors which were potentially life threatening while [LB] was in hospital between September 25^th and 1^st October 2001. In consequence thereof, [LB] has suffered significant harm within the meaning of s.31 of the Children Act, and I find the threshold criteria are made out by the local authority."

The Appeal

123. On appeal, Mr Cobb challenged the conclusions of the judge on the medical and on the non-medical evidence. He did not seek to challenge the judge's findings as to the credibility of the mother.

The Medical Evidence

124. The judge described the medical problems of LB in the following terms.

"All the doctors have found this a difficult, complex and puzzling case. Medically, [LB] is an unusual child, as agreed by all the experts. She has a chromosome abnormality. She is of dysmorphic appearance, with some deafness and delay in speech. She has a general development delay, a tongue-tie, a lesion on her palate and some webbed toes. She has had numerous episodes of chest infections, blepharitis, recurrent styes, otitis externa, otitis media, with oozing and pussy ears. She has suffered from skin rashes and infected spots which are not typical of eczema. She has had oral thrush, a suspected cow's milk intolerance, asthma, spiking temperatures and overnight sweats. Questions arise for determination whether at all, and if so to what extent, these pre-existing conditions can be or are relevant to the rigors suffered by [LB]."

125. The medical evidence in this case was as complex and as controversial as could be conceived. Rigors are rare, yet LB experienced no less than eleven during six days of admission to the local hospital; she suffered from the genetic abnormalities set out above; she suffered a further rigor six months after the critical six-day admission which could not possibly have been caused by her mother. In these circumstances the available expert evidence was divided, only two of the six experts called concluding, on the balance of probabilities, that the rigors had been induced by LB's mother.
126. How then did the judge deal with this most difficult of issues? This was her stated approach: -

"The evidence falls into three sections. 1, the differential diagnosis of [LB's] symptoms, which depends on the medical evidence; 2, the opportunity and knowledge of the mother to act as alleged by the local authority; 3, the credibility of mother."

127. She set out her views of the medical experts. They were all distinguished in their field, some with a narrow expertise. They were agreed that the most likely cause of rigors was infection. There was agreement that other suggested causes were extremely unlikely. She set out in detail the differences in the opinions of the doctors to which we refer in more detail below. Dr S was the jointly appointed expert instructed to give an overview. He was a consultant in child health and the expert chosen by the mother. Dr M was one of the treating consultant paediatricians at the local hospital. Dr B was a community paediatrician and the other treating consultant. The judge found that her evidence was sympathetic to the mother, rather trusting of her and rather defensive, nor was she provided with the full information. The judge found that Dr B's objectivity was to some extent compromised.
128. Dr K was a consultant in infectious diseases. The judge found that he stressed that he was commenting purely on the medical facts and ignored all the social factors.
129. The judge found that Dr Sp, a consultant in clinical genetics, did not express any clear opinion on the disputed issue nor did Dr F, who was a professor of forensic toxicology.
130. The judge reviewed the medical evidence in some detail without stating any clear conclusion. However she had impliedly signalled the way she was going when early in her judgment she had said: -

"However, concentration on a very narrow area of expertise can sometimes render it difficult for the expert to see the whole picture. It is for that reason that I find Dr S is best placed to view the overall picture."

131. Dr S was one of the two doctors to hold the mother responsible for the rigors, the other being Dr M.
132. Another indication of the judge's leaning is to be gleaned from her review of the subsequent rigor of 25^th May 2002. Dr M had conceded in cross-examination that, if LB indeed suffered a rigor on 25^th May,

"I think it would knock my view in respect of imposed illness."

133. However the judge stated that she preferred the view of Dr S that the rigor on May 25^th

"did not inform as to the cause of the earlier rigors."

134. At the conclusion of her review of the medical evidence the judge offered the following summary: -

"The conclusions of the medical evidence considered in isolation from other areas of evidence are as follows:

1. All experts agree that infection is the most likely cause of the rigors.

2. All known natural conditions have been eliminated or are extremely unlikely in identifying the cause, despite very extensive investigations.

3. Dr S and Dr M opine that deliberate introduction of infection via the cannula is the most likely cause of the rigors but they cannot identify the substance.

4. The other experts cannot suggest any cause as more likely than any other and do not know what caused the rigors."

135. There is nothing in that fair summary to support a judicial conclusion that the mother had caused the 11 rigors. It is only when the judge had completed her review of the second and third sections of the evidence that the judge summarised

"the factors which cause me to reach my conclusions that to the appropriate standard and burden of proof the local authority have demonstrated that mother deliberately interfered with the cannula of [LB]."

136. We have set out the twelve factors above. Numbers 7 and 11 relate to the first of the judge's three sections of the evidence, that is to say the medical evidence.
137. In his skeleton argument supporting his grounds of appeal Mr Cobb QC mounted a convincing critique of the judge's elevation of a controversial medical opinion into a judicial finding. The fallacy of the judge's finding was evident from her initial adoption of the doctor's differential diagnostic approach. This submission was considerably expanded in his supplementary skeleton argument, which reviewed the transcript of the oral evidence of the experts analytically and in detail.
138. The evidence of Dr S and Dr M revealed that both relied substantially on "extraneous and circumstantial" evidence to arrive at their conclusions. Yet even that reliance did not bring them to more than a "marginal" opinion.
139. Two quotations from Dr S suffice: -

(a) "All this is postulation, that is all that you can say about it. There is no clear objective evidence that faeces were injected, I was just postulating for the court on what mechanisms could be involved."

(b) "I think it marginally comes down on the situation that it is more likely than not that this was fabricated illness…there was never any evidence to show that the mother was seen to do anything, and that is important because in most of the cases I have dealt with there has been that evidence."

140. Both Dr S and Dr M accepted that there were considerable difficulties with their postulations. Dr S had originally suggested that the mother had injected LB with water contaminated with faeces but in his oral evidence he withdrew that hypothesis. Accordingly Dr S conceded

"I do accept that there was that inherent problem with my proposition that this was faecal".

141. Likewise Dr M accepted that his own

"theory causes problems…all the theories leave cause for doubt."

142. In consequence it is hardly surprising that Dr M conceded that

"no cause for this very serious illness was determined."

143. The experts who were of the contrary opinion were Dr's K, B, Sp and F. The judge's conclusion that Dr B's views were unduly favourable to the mother were fully reasoned and must stand. Dr Sp and Dr F were highly specialist in their respective fields which were not predominantly engaged. By contrast Dr K's field of speciality was particularly engaged. Dr K alone had had previous experience of a case of faecal administration, but it did not follow the pattern of LB's rigors. That led Dr K to the following important conclusions encapsulated in the following quotations from the transcript: -

"So it comes back to my major point, that again I think I have said right the way along that it doesn't fit neatly with a bacterial infection, it doesn't fit neatly with a viral infection, it doesn't fit neatly with a drug, it doesn't fit neatly with injecting material."

"None of it quite fits together with any one story".

"It doesn't quite fit neatly with anything".

"I don't think it fits well with anything".

144. Of great importance is the fact that Dr K's expertise in this vital area was unmatched by that of any other medical witness. Both Dr S and Dr M deferred to Dr K on issues relating to the physical manifestation of a child who had been injected. Dr S emphasised that he was a general paediatrician and that he was not a toxicologist. He said:

"I wouldn't argue with Dr K …as I have said in my evidence in chief, I don't have personal experience in dealing with those cases."

145. To like effect was the evidence of Dr M who said repeatedly that he was not an expert on the topic and that difficult questions would have to be addressed to Dr K to whom he would defer. Indeed he accepted that the hypothesis of faeces as the likely agent for bacterial infection was speculation on his part.
146. In offering an overview Dr S looked not just at the critical six days of life threatening illness but also at the mother's medical history at large, including the episodes of self-harming and the later episodes of psychiatric illness. It was from this wider picture that Dr S drew the circumstantial factors that supported his conclusions. However Dr S conducted only a paper review. He did not examine either LB or the mother. Accordingly he posed a problem for the judge when, in the course of his oral evidence, he suggested for the first time that his assessment was incomplete without a psychiatric component. He said that in his experience of other difficult cases he had been accustomed to working in conjunction with a child psychiatrist "such as Dr David Jones". This development led to an investigation of the possibility of obtaining a forensic psychiatric opinion but when that proved impossible without an unacceptable delay, the judge, inevitably and without dissent, directed that these aspects were to be confined to assessment of the mother's credibility and commitment to LB.
147. In dealing with this development the court may have been influenced by the well known decision of Wall J in the case of Re: CB and JB (care proceedings: guidelines) [1998] 2FLR 211 in which he formulated guidelines for the conduct of preliminary hearings. Wall J dealt with psychiatric assessments at 217H in the following terms: -

"(iv) Evidence of propensity or psychiatric or psychological assessment of one of the parties is unlikely to be of any assistance in resolving a purely factual issue. There will in any event be before the court evidence from the Local Authority and the parents relating to the history of the case and the backgrounds of each of the parents. A psychologist or psychiatrist instructed to undertake an assessment of a parent for the first stage of a split hearing is unlikely to have a complete knowledge of the facts.

(vi) Furthermore, such a witness may, as here, express opinions as to propensity or as to responsibility for a child's injuries which are both prejudicial and wrong. The assessment of adult credibility as to the responsibility for a child's injuries (often the critical factual issue) remains the function of the judge. In my judgment, therefore, a psychiatric or psychological assessment of the parties should not be permitted at the first stage of a split trial unless the particular facts of the case demonstrate that such evidence is or is likely to be directly relevant to the factual issue to be tried."

148. The preparation of the case that Wall J considered fully merited the judge's stricture and no doubt at that time warning against the unnecessary introduction of psychiatric evidence at the first stage of split trial was very necessary. However, as Wall J himself recognised, there will from time to time be complex and seemingly insoluble factual questions for the resolution of which the trial judge would be assisted by a psychiatric or psychological assessment of a possible adult perpetrator or of the wide family dynamics.
149. Given the direction that she had given as to the relevance of such psychiatric evidence as there was, in our judgment, the judge fell into error in bolstering her ultimate conclusion with the 7^th factor. The frequency and severity of the rigors in themselves proved nothing. The absence of medical explanation to account for them naturally was equally probative of nothing. To leave out extraneous factors and to conduct a purely scientific analysis provided only negative conclusions. As Dr K said repeatedly "nothing fits". If nothing fitted there could be no positive conclusion.
150. The 11^th factor revealed precisely the same fallacy. Dr S and Dr M could not ultimately rationalise their conclusion of deliberate interference by pointing to "the absence of any medical cause", since even as experienced general paediatricians their knowledge and experience of medical cause is inevitably incomplete. Had Dr S and Dr M restricted themselves rigorously to "the medical evidence considered in isolation from other areas of evidence" they could not have convincingly advanced the opinion which they did. Of course the judge had a wealth of other evidence to assess but in, at that stage, restricting herself to the medical evidence in isolation, the only tenable conclusion was that the rigors were inexplicable and unexplained.
151. In our judgment it is clear that the eleven rigors which L suffered in those six September days defied the knowledge and experience of all six experts summoned to explain them. Four, led by Dr K, could find no scientific explanation to fit all the established circumstances. Two, led by Dr S, postulated an explanation of deliberate interference. But the postulation rested partly on the foundation that science had no answer and partly on extraneous circumstances culled from the mother's medical history.

The Non-Medical Evidence

152. In our judgment, for the reasons set out above, the medical evidence taken alone was not sufficiently cogent to be relied upon to pass the test of balance of probabilities in accordance with the principles set out in re H. Although the judge fell into error in relying upon the evidence of Dr S and Dr M as part of her findings, she was entitled to take into account that there was no natural or medical explanation for the rigors and to assess the weight and credibility of the non-medical evidence. The judge paid particular attention to the non-medical evidence and relied upon the other 10 factors, set out above, in order to come to her conclusion that the local authority had made out its case that the threshold required by section 31 had been crossed.
153. It is clear that the mother could not challenge the findings of fact made by the judge and her assessment of the credibility of the mother nor did Mr Cobb seek to do so. The judge found the mother to be a liar, that she exaggerated events and manipulated them in order to get her own way. The mother suffered from a very serious pre and postnatal depression during which she rejected her baby. She described LB as damaged and evil; she felt ill when she fed her and resented the baby causing her ill health; she felt she was looking after someone else's baby. In April 2001 she diluted the milk in the bottle when the child had been admitted for failing to thrive and weight loss. Although the psychiatric evidence was that the mother had recovered from depression by May 2001 and the local authority closed the file, the judge made findings that the mother continued to lie, to manipulate others and to divert blame from herself. She deliberately removed the cannula from the child before the transfer to Great Ormond Street. The mother's reaction to the rigor suffered on the 25^th May 2002 and her failure to take the normal steps of a concerned mother undoubtedly raise question marks. As late as September 2002 the father said that she wanted the child to be ill and was angered that she was well. The judge found that the mother had the opportunity and also the knowledge to introduce a harmful substance into the cannula attached to her child. The totality of the evidence, all of which is admissible in family proceedings, is relevant and cogent. The judge was clearly entitled to rely upon it in coming to her conclusions.
154. Having found the facts set out above, the judge made a careful self-direction which cannot be faulted. In our judgment this Court cannot say that the non-medical evidence was not sufficient for the judge to conclude that the mother had been responsible for the rigors suffered by the child between 25^th September and 1^st October 2001.
155. We therefore grant permission to appeal and extension of time. We dismiss the appeal.