[Home] [Databases] [World Law] [Multidatabase Search] [Help] [Feedback]
	England and Wales Court of Appeal (Civil Division) Decisions
You are here: BAILII >> Databases >> England and Wales Court of Appeal (Civil Division) Decisions >> Burne v A [2006] EWCA Civ 24 (25 January 2006) URL: http://www.bailii.org/ew/cases/EWCA/Civ/2006/24.html Cite as: [2006] EWCA Civ 24

[New search] [Printable RTF version] [Help]

IN THE SUPREME COURT OF JUDICATURE
COURT OF APPEAL (CIVIL DIVISION)
ON APPEAL FROM THE HIGH COURT
CIVIL DIVISION
SITTING IN OXFORD
HH JUDGE CHARLES HARRIS QC
Insert Lower Court NC Number Here

		Royal Courts of Justice Strand, London, WC2A 2LL
		25/01/2006

Lord Justice Sedley :

1. A was born on 7 July 1991 and is now 14. He was born prematurely, with a hydrocephalic condition that required him to be fitted at the age of 9 weeks with a ventriculo-peritoneal shunt. This is a simple but vital appliance which continuously drains from the brain cavity excess fluid which would otherwise accumulate there. A blockage in a shunt is therefore always potentially critical.
2. A was registered with Dr Burne, a family practitioner in Cowley. Among his attendances at the clinic with childhood ailments, one - in November 1992 – was because the shunt had become blocked. His condition was noted on the clinic's computerised record.
3. On 27 January 1998, when he was 6 ½ years old, A began vomiting at school and complaining of a headache. The headteacher phoned his mother, who came to collect him. By 5:00 p.m. he was no better and so she rang Dr Burne. It will be necessary to look in some detail at what passed between them then and later, but the upshot was that a diagnosis of upper respiratory infection was made by telephone. The following day A was worse. When his mother, Miss Pember, got through to the clinic she was given an appointment at 5:00 p.m., but during the afternoon his condition deteriorated to the point where an ambulance was called. In hospital it was found that the shunt had become blocked, causing a heart attack and brain damage.
4. At a trial limited to the issue of liability, in the High Court sitting at Oxford, Judge Charles Harris QC found for the claimant on the ground that the general practitioner ought by the exercise of a proper standard of skill and care to have ascertained when he first spoke to the mother on the telephone that A's symptoms might be indicative of a blockage in his shunt. Had he done so, prompt specialist attention would have prevented the occurrence of damage.
5. The judge's findings, which are central to this appeal, and his reasoning are set out with conciseness and clarity. Rather than try to summarise or paraphrase it, I will annexe his judgment to mine and refer to it as necessary.
6. The reason why the present appeal comes before this court (by permission granted by Rix LJ) is in essence this. The judge had the written and oral evidence of two expert witnesses on professional standards in general medical practice, one called by each side. Both were prepared to accept that the use of "open" questions was a proper method of diagnosis. If so, Dr Burne's failure to obtain from the mother specific answers which would have alerted him to the possibility of a blocked shunt was not professionally negligent. But the judge did not accept that this was a reasonable course for an ordinarily prudent doctor to take in relation to a child with A's known condition and history. Having considered the principal modern authority on the subject, Bolitho v City and Hackney Health Authority [1998] AC 232, he held that there was no good reason for Dr Burne not to have asked two or three specific questions which would have indicated whether symptoms of shunt blockage were present.

The problem

7. Rix LJ granted permission to appeal in these terms:

" The judge's judgment and reasoning read persuasively, but, in a case where he has set aside the joint view of both parties' experts, I am unable to say that there is not a real prospect of success."

8. It became apparent on the hearing of the appeal, however, that neither party regarded the Bolitho question as more than peripheral. Both counsel had addressed the judge on the footing that the expert evidence precluded any case for specific or 'closed' questions – what lawyers would classify as leading questions. Counsel for Dr Burne had submitted that it had been consistent with acceptable practice, as described by both expert witnesses, to work from whatever history the mother gave. Counsel for A had submitted that, even within the given model of acceptable practice, it had been incumbent on Dr Burne to ask some 'open' questions which would have elicited the danger signs. But he had not argued that the judge was entitled to reject the experts' evidence of acceptable practice on the ground that it did not make sense. That was an issue raised and resolved by the judge himself in the light of his findings.
9. Consequently both counsel addressed this court almost entirely on the facts. For my part I doubt whether Rix LJ intended to grant permission to appeal on the facts, notwithstanding that the grounds and skeleton argument annexed to the appellant's notice dealt with very little else. But by common consent, for the reason I have indicated, we have been asked on the one hand to re-evaluate and reject the judge's fact findings, and on the other to uphold them on the grounds advanced for the claimant at trial rather than in the judgment.

The law

10. Judges are required to try questions of medical negligence on the evidence, not on what they themselves know or think they know. It would, for example, have been improper for Judge Harris simply to decide that Dr Burne had been negligent whatever the experts said. But judges are there to exercise judgment, and their judgment cannot be entirely dictated by expert evidence, even where the evidence is unopposed. The question always remains (provided a party raises it) whether the expert evidence makes sense.
11. Legal principles also set the question which the judge has to answer in a medical negligence case. Because medicine is a highly specialised field, what is and is not negligent has to be determined by reference to accepted or tolerated practice. It is a principal function of expert evidence to enable the judge to decide what lies within and what lies outside such practice.
12. Thus in Bolitho the House of Lords endorsed McNair J's charge to a civil jury in Bolam v Friern Hospital Management Committee [1957] 1 WLR 583, 587: a doctor

"is not guilty of negligence if he has acted in accordance with a practice accepted as proper by a responsible body of medical men skilled in that particular art…."

This test, as Lord Scarman said in Maynard v West Midlands RHA [1984] 1 WLR 634, 639, does not permit the judge to make his or her own choice between two or more respectable schools of professional practice. But, as Lord Browne-Wilkinson said in Bolitho:

"[T]he court is not bound to hold that a defendant doctor escapes liability for negligent treatment or diagnosis just because he leads evidence from a number of medical experts who are genuinely of opinion that the defendant's treatment or diagnosis accorded with sound medical practice……[T]he court has to be satisfied that the exponents of the body of opinion relied upon can show that such opinion has a logical basis. In particular in cases involving, as they so often do, the weighing of risks against benefits, the judge, before accepting a body of opinion as being responsible, reasonable or respectable, will need to be satisfied that, in forming their views, the experts directed their minds to the question of comparative risks and benefits and have reached a defensible conclusion on the matter."

A little later in the same speech he said:

"The assessment of medical risks and benefits is a matter of clinical judgment which a judge would not normally be able to make without expert evidence."

13. The last citation illustrates a further relevant proposition: that this limb of the law is about the exercise of clinical judgment. It needs to be related, therefore, to the kind of judgment which was in issue here, which was not Dr Burne's judgment as to what was wrong with the child, but his judgment as to how to go about getting the relevant information from the mother. This skill, while an important aspect of clinical practice, sits at the threshold rather than at the centre of it.

The evidence

14. The record made by Dr Burne of the mother's call was: "Cough and fever, sent home from school. Advised fluids, disprol, tca prn."
15. The judge found, however, that Miss Pember told Dr Burne on the telephone that A was throwing up a lot of phlegm, meaning (had she been asked about it) vomiting it. He expressed "real doubt" (§23) as to whether she had said the child was running a temperature (which would have been indicative of a viral infection, whereas no elevation of temperature would have pointed towards a shunt blockage), notwithstanding its mention in the GP's note. He found that the mother had not mentioned the child's irritability, drowsiness and restlessness which, since he found them to have been present, she would have mentioned had she been specifically asked about them. But he found that the mother had not learnt from the school that A had also been complaining of a headache, and so would not have mentioned it to the doctor even if he had asked about it.
16. The judge went on to find, for reasons which he gave and which are not contested on this appeal, that had Dr Burne asked whether these signs and symptoms were present he would have been sufficiently alerted by the mother's answers to see the child and, having seen him, would have had him admitted to hospital. The question was therefore, as the judge said (J§27), whether Dr Burne had been negligent in not eliciting or seeking specifically to elicit the history of vomiting, drowsiness and restlessness in making his telephone diagnosis.
17. As to this, the claimant's expert Dr Isaac and the defendant's expert Dr Lewis had agreed, after consultation, on a memorandum which set out the following hypothetical situations, the first not calling for more than was in the event done, the second calling for a visit followed in all probability by admission to hospital:

"Clinical condition 1: namely coughing and vomiting phlegm with intermittent temperature (even with mild headache) would fit with an upper respiratory tract infection to such a degree that a face-to-face evaluation would not be required. In these circumstances advice to call again as required would be reasonable.

Clinical condition 2: headache and 'unprovoked' vomiting (specifically no phlegm, no upper respiratory tract infection or gastroenteritis), with or without a temperature, would require a further assessment."

18. Reliance is placed by counsel for Dr Burne on the fact that in the witness box Dr Isaac had agreed that, given a history simply of coughing or vomiting up phlegm with a temperature, a doctor would not think of shunt blockage "as a first option", and that he had accepted that this case (assuming that the word phlegm was used appropriately) fell into what he and Dr Lewis had identified as clinical condition 1. Both of these elements of Dr Isaac's evidence, however, assumed something which the judge was not prepared to accept, namely that the child had been running a temperature and that Dr Burne had been told that he was. If there was no markedly elevated temperature, a differential diagnosis of a viral infection became less obvious and a diagnosis of shunt blockage correspondingly more prominent. The evidence also involved an assumption that 'phlegm', which is not a medical term of art, bore its common meaning of mucus.
19. It was in §32 and §37 that the judge dealt with the question on which his eventual decision turned. In the first of these paragraphs he accepts Dr Isaac's evidence that as a matter of good practice an "active history" should have been taken, but finds less than convincing his acceptance that, given that there could be a range of opinions about it, it was not unreasonable not to do so. In the second he expresses his own conclusion that, since it was only by asking specific questions that the doctor could satisfy himself that this child's condition was not shunt-related, further inquiry to this end had been called for.
20. He then turned to the question whether, notwithstanding that he had arrived at this view, the law allowed him to give effect to it. Having reminded himself of the decision in Bolitho he concluded that it did, because (§41) there was in his judgment "no reasonable or logical basis" for the view that it was acceptable practice to limit inquiry to "open" questions in relation to a child whose particular vulnerability was known to the doctor and who could have been readily diagnosed on the basis of a few direct questions about the child's vomiting and drowsiness.

The arguments

21. For reasons which I have touched on, the assault on the judge's decision in this court has been almost entirely concentrated on his factual premises. At the forefront of the argument of John Grace QC for Dr Burne has been the judge's alleged failure to pick up the crucial distinction between provoked and unprovoked vomiting. If A's vomiting was provoked by the ingestion of phlegm, as on the mother's account it seemed to be, it was self-explanatory. If on the other hand it was apparently unprovoked, a cause – which might then be a shunt blockage - had to be looked for.
22. The argument for the appellant goes on to urge that there was in truth no distinction between the facts as they emerged and what the experts in their joint memorandum had called clinical condition 1. The judge's belief that there was such a distinction, says Mr Grace, led him to consider that he could depart from the joint view of the experts. In particular, vomiting phlegm (one of the experts' postulates) is not unprovoked vomiting and fits clinical condition 1.
23. The approach of Gerard McDermott QC on behalf of the respondent has been that the issue is not about the generality of medical practice but about something which all sound practice must accommodate, namely the exceptional case. He accepts that there are good reasons for not as a rule asking patients what a lawyer would call leading questions: it may induce them to say what they think the doctor wants to hear. But to apply such a practice rigidly may occasionally be to neglect the very object to which the practice is directed, the diagnosis of complaints. The respondent says that this is just such a case, because A faced a serious but readily diagnosable risk which the doctor knew about and could have identified by asking a few simple but open questions. To fail to do so by inappropriately following an accepted practice was, it is submitted, negligent.
24. The questions Mr McDermott instances are those he put to the judge: Is the child getting better or worse? What do you mean by phlegm? What does it look like? (He did not and does not go so far as to say the mother should have been asked 'Does the child have a temperature or a headache?', because his own expert had accepted that it was sound practice to eschew such direct questions.) The answers, he submits, would have indicated that the child was increasingly drowsy and irritable and that the vomiting was unprovoked because what was coming up was clear liquid, not mucus; and all this would have rung an alarm bell. The basis for the submission about what A was bringing up is the mother's evidence in cross-examination that the fluid had been "clear" – and then, a little later, "clear water based … with a bit of thickness".

Discussion

25. The judge was aware of the distinction between provoked and unprovoked vomiting, as §25 of his judgment indicates. His failure to make anything of the distinction, however, is hardly surprising. It features not at all in the grounds of appeal and surfaces only in §68 of a skeleton argument which is considerably longer than the judgment.
26. It seems to me that the materiality of unprovoked vomiting depends in the first instance on how reliable the mother's initial information appeared to be. If the doctor was justified in interpreting 'phlegm' as meaning mucus, then vomiting it would indicate no more than that there was an upper respiratory tract infection, producing mucus which the child was ingesting and regurgitating: in other words, this much of clinical condition 1 would be met – though there would still remain the question of the child's temperature, as to which the judge's finding was against its having been part of what the mother reported. The judge will have had in mind in this regard that, although the oral evidence was being given several years after the events, Miss Pember had made a written statement in September 1998 which formed part of her testimony and which, strikingly, said:

"I took A's temperature and it was normal. I was surprised about this because usually if he has a virus he has a temperature of more than 100? and it is very difficult to bring it down."

27. For my part, therefore, I am not willing to fault the judge's view (§35) that neither of the two postulated clinical conditions accorded precisely with the situation. The child was coughing, but no fever had been reported. The answer the judge goes on to quote (given jointly to one of the questions posed by the solicitors), to the effect that if no vomiting was reported some form of further questioning was called for, carries as a necessary implication that the mother's report of vomiting was capable of being significant.
28. Was the information that the child was throwing up phlegm then enough to allow the doctor to diagnose an infection and not to consider a shunt blockage? The judge (§37) thought not: he considered that steps should have been taken to ascertain "whether there had been true vomiting" – by which he plainly meant unprovoked vomiting as opposed to the regurgitation of mucus. The argument to the contrary was that phlegm, once reported, calls for no further inquiry and permits diagnosis. Mr Grace contends that Dr Isaac assented to this, but Dr Isaac answered that to do this you first had to clarify what it was that was being reported. It was then put to him that throwing up phlegm meant a viral illness: he agreed but with this proviso:

"We [the two experts] both agree that if you take a careful history and that is the response he got, then his actions were reasonable" [my emphasis].

In other words, vomiting phlegm was diagnostic of a viral infection if it really was phlegm.

29. The way the claimant's case was put to the judge was constrained by the unanimity of the expert witnesses about the undesirability of asking closed or direct or leading questions in arriving at a diagnosis. Mr McDermott, as the transcript of his final speech shows, was suggesting some 'open' questions to explore the symptoms further before reaching a diagnosis. The judge, however, did not deal with these. Instead he took the bull by the horns and asked (§37) whether in a case like this there was any alternative to direct questions about the detail of the vomiting, and whether there was drowsiness or headache, if shunt blockage was to be excluded. The answer he had been given by the experts (§38) was that such questions were not necessarily called for.
30. It offends one's sense of justice to be obliged to accept the unacceptable. I can entirely understand why Judge Harris thought it unacceptable that the culture of general medical practice should be so suspicious of self-serving reportage that it encouraged doctors to ask nothing specific even where the caller was the mother of a child whom the doctor knew to have a shunt in place and the child might have symptoms caused by a blockage. But if this was his initial response, as it evidently was, it was incumbent on him to do one or both of two things: to ask the claimant's counsel whether, should his findings reach this point, he was invited to consider whether the expert evidence in support of the doctor made sense; and, if counsel said yes, to ensure that the doctor's side had a proper opportunity to respond. From what we are told by both leading counsel, this is not what occurred. In fairness to the judge it should be said that he did, over four pages of transcript, canvass with counsel for Dr Byrne the difference between what seemed to him the commonsense approach to questioning Miss Pember and what the experts had considered acceptable. But the possibility in law of his discounting their view was not raised.
31. We have been asked by Mr Grace, contingently on our holding the judge to have been entitled to proceed as he did, to hold that he erred in law in so doing. The written submission on which he relies contends that Judge Harris's decision on this part of the case was in any event an unwarranted extension of the Bolitho principle. If this were all the appeal turned on, I would not agree. For the reasons set out in the section of this judgment dealing with the law, I consider that the judge's approach to the issue was legally correct. The real problem is that he ought not to have embarked upon it, and that if he was to do so it could only be after the defendant's expert witness (and quite possibly the claimant's too) had had a proper opportunity to explain why medical practice, or at least one respectable school within it, took this position.
32. This alone requires the judgment to be set aside. But it is only if the judge could have come to no conclusion other than one in favour of the defendant that we would be required to allow the appeal in full; and this I do not consider to be the case. At the very lowest Mr McDermott was entitled to a decision on his submission that, within the four corners of the diagnostic practice accepted by both experts, Dr Burne had fallen short of a proper standard of care and skill by failing to ask further 'open' questions indicated by the initial information. And if it was in the judge's mind that, failing this, he was not bound to follow the experts' joint view if it did not make sense, he was required to put it to the parties and to act only on their considered responses.
33. I am unable to accept Mr Grace's submission that he is in any event entitled to succeed because it has not been shown that to have asked the questions postulated by Mr McDermott would have produced a correct diagnosis. The judge expressly found (§25-26) that if unprovoked vomiting, drowsiness and restlessness had been elicited, a competent GP would have visited the child and had him admitted to hospital. Everything therefore turns on whether what the judge found the doctor had been told by the mother called for further inquiry. The judge's view that it called for specific further inquiry was not open to him unless it was also open to him – as in the circumstances it was not - to override the experts' view to the contrary. The claimant's submission that further non-specific inquiry was called for was not adjudicated upon. I do not consider that this court, albeit equipped with the judge's careful fact findings, is in a position to give the answers.
34. I would therefore allow this appeal, set aside the judgment in the claimant's favour and direct a retrial. It is necessary, however, to add three riders.
35. First, in the situation which has now been reached this case calls out for alternative dispute resolution. Both parties need to take stock of their position and to enter into mediation in the light of it. No further step should be taken in the remitted action until this has been done.
36. Secondly, if for some reason mediation fails, the remitted trial should be on causation as well as liability. The district judge's procedural direction was that "breach of duty" be tried as a preliminary issue. But in negligence there is no liability unless a breach of duty has caused damage. There are negligence cases in which causation is so legally complex or so evidentially discrete that it makes sense to split the issues for trial; but it will be apparent from the earlier part of this judgment that the present case is not one of them. Indeed, much of the evidence and argument before Judge Harris was concerned with aspects of causation. Subject to any submissions of counsel, I would direct that the retrial be on the issue of liability including causation.
37. Thirdly, although it will not be helpful to anticipate in any detail the issues which will arise within this remit, this much seems clear. If the point is taken on the claimant's behalf that, notwithstanding the expert evidence, the practice of asking only 'open' questions was not acceptable by Bolitho standards in the particular circumstances known to the defendant, and if the point succeeds, the judge will have to decide whether 'closed' or leading questions would have elicited enough information to have prevented the eventual outcome. If the point is not taken, or if it is taken but fails, the judge will still have to decide whether there were other 'open' questions which the doctor ought to have asked and which, if asked, would have elicited enough information to have prevented the eventual outcome.
38. I reiterate, however, that it is to be hoped that ADR will see an end of this anxious and distressing case.

Lord Justice Wilson:

39. The central reasoning behind the judgment of this highly respected judge proceeded in four stages.
40. First, in paragraph 35, the judge described it as "important" that in the joint memorandum of the two experts their answer to one question was:

"We agree further enquiry would be necessary to arrive at [the defendant's] course of action. This may or may not include specific enquiry regarding vomiting."

41. Second, in paragraph 37, the judge held that, in order to conduct the "further enquiry" to which the experts had referred, the defendant should have asked the mother a series of specific or "closed" questions.
42. Third, in paragraph 41, the judge, applying dicta in the Bolitho case which had not been cited to him and upon which he had not invited submissions, held that there was no logical basis for the joint view of the experts that it was proper for the defendant not to have asked the mother closed questions.
43. Fourth, in the same paragraph, the judge held that, had the defendant asked closed questions, he would have elicited answers which would or should have alerted him to the possibility of a blockage in the claimant's shunt.
44. The inference to be drawn from the second, third and fourth stages is that the judge rejected the claimant's case that open questions to the mother would have elicited such answers.
45. In my view the judge's reasoning went awry even at the first stage. For he misunderstood the question to which the experts had given the "important" answer about the need for further enquiry. In purporting in paragraph 35 to recite that question he omitted part of it which, had he taken account of it, would have saved him from error. In quoting it as follows I italicise the part which he omitted:

"If [the mother] is wrong and no vomiting was reported to [the defendant], should he have specifically enquired as to whether the claimant had been vomiting …?"

The premise behind that second question was the converse of the premise behind the first question posed to the experts, namely "if the court accepts the claimant's case that [the mother] told [the defendant] that the claimant had been vomiting …". The claimant's case, as set out in the particulars of claim, was that the mother told the defendant that he had been "vomiting phlegm". In the event the judge's finding was that she told the defendant that he had been "throwing up a lot of phlegm". The crucial points are however that the references to "vomiting" in both questions to the experts are references to vomiting (or throwing up) phlegm; and that, whereas in this respect the judge accepted the claimant's case (with which at trial the defendant in effect agreed), the premise behind the second question is that, contrary to the claimant's case, the mother never told the doctor that he had vomited (or thrown up) phlegm. Thus in the event the second question never arose; and the answer to it, far from being "important" and indeed a valid foundation for the decision, proved to be irrelevant.

46. Both experts agreed that the non-medical word "phlegm" indicated an excess production of mucus in the upper respiratory tract and that the mother's use of the word suggested that the problem was a viral infection there. In answer to the judge Dr Isaac, the expert instructed by the claimant, said:

"The … problem is that, if she used the word throwing up or vomiting "phlegm", then that, in my mind, would then throw you – you go down a path."

Earlier he had explained that, whereas an adult will usually cough phlegm up out of the respiratory tract, a child will often swallow it and then indeed vomit it up out of the stomach.

47. But, in addition to the mother's use of the word "phlegm", there were two further features or alleged features of her conversation with the defendant upon which he strongly relied in denying that his diagnosis of a viral infection, reached without further enquiry, had been reached negligently.
48. The first was that, in the mother's own words,

"I told [the defendant] that he had the same thing a couple of weeks before but had seemed to get over it then."

Speaking for myself, I find it hard to conceive a comment more likely to lull the defendant into concluding that the problem did not relate to the claimant's shunt.

49. The second was that, according at least to the defendant, the mother told him that the claimant had, or during that day had had, a temperature. His computerised note of the conversation had, after all, expressly referred to "fever" as one of the two symptoms, the other being "cough". The experts agreed that an elevated temperature, intermittent or otherwise, was also symptomatic of a viral infection rather than of a blocked shunt.
50. The judge, however, refused to find that the mother had told the defendant that the claimant had, or during that day had had, a temperature. He said that he had real doubt about it. The judge's refusal to make this finding was arguably surprising. True it is that, in her written statement made only eight months after the conversation, the mother said that she had told the defendant that the claimant did not have a temperature. In oral evidence, however, she qualified that as follows:

"I am sure from my statement that I said that he was coughing up phlegm and no temperature, but it may well have been elevated during the day. But at that time no temperature."

Indeed in the Particulars of Claim the claimant had expressly averred that his mother had told the defendant that "the claimant's temperature was normal at some times and elevated at others"; and in the Defence the defendant had expressly admitted that she had said that the claimant was feverish. The defendant's oral evidence was indeed that the mother had told him that the claimant had had a temperature during the day. And then, no doubt more significantly, there was the note of "fever" made by him at the time, which in cross-examination Mr McDermott never even suggested to him represented an inaccurate record of the conversation.

51. In this appeal the defendant asks this court to replace the judge's refusal to make a finding on this point with a finding that the mother told him that the claimant had on that day been feverish. Since in the end, however, I have been persuaded by the judgments of my Lords that the action should be reheard de novo by a different judge, it becomes out of the question for this court to impose a finding in this respect even if it might otherwise be open to it to do so. The new judge will consider this matter afresh and will decide for himself whether to make the finding.

Lord Justice Ward:

52. I gratefully adopt the recitation of the material facts set out by Sedley L.J. in the draft judgment which I have read. The thrust of the claimant's case was that the defendant failed to take a proper history from the claimant's mother so as to satisfy himself whether the child might have a blocked or blocking shunt which needed attention: see paragraph 15 of the judgment. The claimant's case was that it was negligent to proceed to make a diagnosis without that full history having been elicited. The defendant on the other hand was asserting the information was adequate enough to make a diagnosis and that the diagnosis he made that the child was suffering from a minor viral infection was not a negligent one. It seems to me that in order to resolve those disputes the judge really did have to ask and answer a number of questions.

The relevant questions in this case

53. In my judgment these questions arose:

(i) Precisely what was said during the telephone conversation on 27 January? This was a pure question of fact which the judge had to resolve, difficult though it was given the passage of years and the dimming of memory but in the questions that follow the judge had to apply the Bolam test and ask himself whether the defendant was acting in accordance with a practice accepted as proper by a responsible body of general medical practitioners. Thus:

(ii) Was the information received by the defendant adequate to act upon it or should he have supplemented the information he had received by making further enquiry?

(iii) If the information was adequate, was the diagnosis of minor viral infection negligent?

(iv) If the history received was inadequate, (a) would a reasonably competent medical practitioner have made further enquiries asking by open questions only? If so, (b) precisely what further questions should he have asked?

(v) If not, was the judge entitled to apply the Bolitho exception and find that the opinion of the experts was without logical basis, thus allowing him to use his common sense to decide what questions should have been asked?

(vi) What history would have been obtained in the light of proper questioning, the answer to which depends on whether the questioning should have followed the experts' view of its permissible scope or his own?

(vii) What diagnosis would then have been made and in particular would the claimant have been admitted to hospital that evening?

The judge's approach

54. As to the actual conversation (question (i) above), he found in paragraphs 17 and 18 that nothing was said by the mother to the defendant which might have led him to know that the claimant had had a headache. She did tell him that he had been throwing up a lot of phlegm at school: see the first sentence of paragraph 20. He did not expressly make any finding as to what, if anything, she said about his being sick at home. In paragraph 19 the judge recites her evidence and appears to accept that she said either that he had been "throwing up a lot of phlegm" or possibly "coughing up phlegm". She did not use the expression "vomiting" to the doctor. There was no mention about irritability, drowsiness or restlessness.
55. It is clear the judge would have answered my question (ii) by finding that the information was inadequate. He said in paragraph 31:-

"I can find no good reason, after having heard his [the defendant's] evidence, why he did not ask more specific questions."

Since he did not advert to Bolitho and to the experts' evidence it seems to me this finding standing alone is incomplete.

56. In paragraph 35 the judge seized upon and thought important the answer given by the two experts to question 2 as put to them by the claimant's solicitor where they agreed that if no vomiting was reported to the defendant then further enquiries would be necessary and that might or might not have included specific enquiry regarding vomiting. For my part I do not understand why that answer was material still less why it was important. Question 2 was formulated, I have no doubt, to clarify what the position would be if the note in the computer record which referred simply to "cough and fever" was an accurate account of the conversation between the mother and the doctor. The judge's finding was that it was not accurate. The question had become hypothetical. Consequently the answer did not necessarily govern the actual situation.
57. As it seems to me, the judge was driven in paragraph 27 to say:-

"The case, therefore, comes down to this. Was the defendant negligent in his enquiries during the telephone conversation in not eliciting or seeking specifically to elicit the history of vomiting, drowsiness and restlessness."

So, whether or not this was the view of the experts, the judge clearly felt that further enquiry was necessary.

58. The implication of the judgment must be that if, contrary to his conclusion, the defendant was entitled to rely only on that which he was told in the telephone conversation then he would not have been negligent. On that assumption my question (iii) would have been answered in the defendant's favour.
59. As to the kind of questions that could be asked to clarify the position (my question (iv)), the judge noted in paragraph 32 the opinion of Doctor Isaac that:-

"… it was not unreasonable not to go down the path of asking specific questions about the major symptoms of potential shunt blockages."

The judge appears to be referring to the evidence given at page 17 of the transcript (page 203 in the bundle before us). Whilst the judge may not have been impressed by that evidence, it was, I think, common ground between the experts and it was certainly the defendant's own view. I suspect the judge would have been bound to answer my question (iv)(a) by saying that application of the Bolam test required him to accept that closed questions did not have to be asked. As for (iv)(b), he does not precisely formulate the kind of questions that a reasonably competent general practitioner would have asked. In paragraph 41 he poses specific questions he would think it sensible to ask like "Has he vomited? Is he drowsy? Has he got a headache?" These are, however, the kind of closed questions he was being told need not be asked.

60. In paragraph 37 of his judgment he expressed what in paragraph 38 he described as his "conclusion", namely:-

"In those circumstances, it does seem to me that an ordinarily careful doctor should have taken steps to clarify or elicit whether there had been true vomiting and whether there were other significant symptoms, such as drowsiness or headache. Only by asking those questions, which are not complicated, could the doctor have satisfied himself that this was probably not shunt related. This, it seems to me, was or ought to have been the "further enquiry" to arrive at Doctor Burne's course of action referred to in the G.P. experts' joint report. I do not see how this could reasonably exclude specific enquiry, for example, about vomiting. If it was appropriate to discover, which it clearly was, whether there had been vomiting or indeed headaches or drowsiness, the only reliable way to find out was by asking explicitly and clarifying any unclear answer."

61. The judge's view on my questions (vi) and (vii) is expressed in paragraph 41, namely:-

"As I have said, answers to specific questions like "Has he vomited? Is he drowsy? Has he a headache?" would swiftly have equipped the doctor with material to know whether this vulnerable child was or might have been showing the symptoms of shunt blockage and not merely a cold. It would have been common sense to ask them. Had Doctor Burne asked, he would, as I have found, been likely to have discovered a history of vomiting and drowsiness although he would not have been told of the headache. Had he discovered the vomiting and drowsiness he would or should have seen the claimant and had he done that he would most probably have referred him to a hospital forthwith. "

62. What the judge does not deal with is whether or not any line of further enquiry would have elicited the whole history the mother was capable of giving which included her belief that he had not shaken off the virus he had had earlier in the month when he had been admitted to hospital. In her evidence she agreed at page 29 that she felt that "the bug had come back" and at page 36 "When [I] got home [from school] and he was coughing up phlegm, [I] thought it was part of the same problem [as his having been sick a few weeks previously]". She said in paragraph 20 of her statement, "I told Dr Burne … that he had had the same thing a couple of weeks before but had seemed to get over it then." The judge then also has to deal with what effect, if any, that history would have on the proper diagnosis.

The Bolitho question (v)

63. Reading the judgment as a whole it seems to me to be apparent that if the judge had followed the views given by the experts, then he would have acquitted the defendant of professional negligence. He baulked at this, as it seems to me from reading the transcript as a whole, because he was much influenced by the fact that the boy had twice been referred to the hospital because of complications in connection with the shunt. He may have been in error in saying that one of those occasions was May 1994 but that does not much matter. If one concentrated on the fact that this was a boy with a particular problem, or as the judge said in paragraph 36 of his judgment, "this case did not involve an ordinary patient", and at paragraph 25, "there should have been a low threshold for seeing the claimant", then one is easily drawn into expressing surprise that obvious questions were not asked to eliminate problems with the shunt as a possible diagnosis of the boy's distress. Those problems were identified in the evidence of Dr Burne at page 78 of the transcript as unprovoked vomiting, headaches, irritability and drowsiness. One then sees at page 69 on the second day that the judge explains to Mr Hockton, counsel for the defendant, what has been troubling him. He said:-

"I frankly do not understand why if there are four, as it were, leading indicative symptoms, if that is not a tautology, of potential blockage, you do not then, if you have potential blockage in your mind, try and exclude that by asking specific questions, viz: has he vomited, is he drowsy, this sort of thing."

Now I confess I have a great deal of sympathy with the judge in that view. Like Rix L.J. I at first felt that "the judge's judgment and reasoning reads persuasively." I am not, however, convinced that we are entitled to rely on what seems common sense to judges and consequently dismiss the views of the experts as illogical. At least we should not do so unless the Bolitho point has been properly taken in the court below and the experts given an opportunity to explain and justify their practice. There is, I am afraid, a force in Mr Grace's submissions that it would be unfair to allow the case to be disposed of in that way. For the reasons given in the preceding paragraph I do not know what answers further questioning might have elicited and how that would have affected a proper diagnosis.

Conclusion

64. I see no alternative but to send this matter back for a full re-hearing. I do so with considerable regret because it is nearly eight years since this tragedy occurred. If the matter is contested every inch of the way a final judgment could be years away assuming that the preliminary issue is resolved in the claimant's favour and there are then further battles on issues of causation and damages. For that reason I agree that it would be sensible to extend the rehearing to the determination of causation as well as breach of duty.
65. On the issue before us there are powerful arguments either way and I express no view whatsoever as to the eventual outcome. I do, however, feel very strongly that this is a case which must be referred to alternative dispute resolution before it is restored for the re-trial. Both parties should take stock of the strengths but also the weaknesses of their respective cases which are now plain for all to see and I hope mediation will bring a swift conclusion to a tragic event.