username

password

Garden Court1 Garden CourtCoram ChambersHarcourt ChambersDNA LegalHind Courtimage of 4 Paper Buildings logosite by Zehuti

Home > Judgments > 2012 archive

B (Children) [2012] EWCA Civ 1933

Appeal by local authority in care proceedings concerning two children in which the court had refused interim care orders and granted interim supervision orders. Appeal allowed.

Neutral Citation Number: [2012] EWCA Civ 1933
Case No: B4/2012/2923

IN THE COURT OF APPEAL (CIVIL DIVISION)
ON APPEAL FROM CANTERBURY COUNTY COURT
(HIS HONOUR JUDGE MURDOCH)

Royal Courts of Justice
Strand, London, WC2A 2LL
Date: Wednesday 12th December 2012

Before:
LORD JUSTICE MOORE-BICK
LORD JUSTICE ETHERTON
and
LADY JUSTICE BLACK
- - - - - - - - - - - - - - - - - - - - -

IN THE MATTER OF B (CHILDREN) 
- - - - - - - - - - - - - - - - - - - - -

(DAR Transcript of WordWave International Limited
A Merrill Communications Company
165 Fleet Street, London EC4A 2DY
Tel No:  020 7404 1400  Fax No: 020 7831 8838
Official Shorthand Writers to the Court)
- - - - - - - - - - - - - - - - - - - - -

Mr Stephen Bellamy QC (instructed by the Local Authority Governance and Law) appeared on behalf of the Appellant
Mr Jeremy Hall
(instructed by Robinson Allfree) appeared on behalf of the 1st Respondent Mother
Mr George Butler
(instructed by Daniel and Edwards) appeared on behalf of the 2nd Respondent Father
Mr Frank Feehan QC
(instructed by Ewings Solicitors) appeared on behalf of the 3rd and 4th Respondent Children by their Guardian
- - - - - - - - - - - - - - - - - - - - -

Judgment
(As Approved by the Court)
Crown Copyright ©

Lady Justice Black:
1. This is an appeal by the local authority from an order made by HHJ Murdoch on 8 November 2012 in relation to twin boys, who for the purposes of this judgment I will simply call "L" and "M".  They were born on 6 December 2011 and have therefore just had their first birthdays.  They are the subject of care proceedings brought by the local authority as a result of L having sustained injuries whilst in the care of his parents.  The final hearing is to take place in February next year.  HHJ Murdoch was considering what arrangements needed to be made for the care of the boys in the interim.  He refused the local authority's application for interim care orders in relation to both of the boys and granted instead interim supervision orders. 

2. The judge had had to determine whether the condition in section 38(2) of the Children Act 1989 for the granting of either a care or a supervision order was satisfied (that is, that there are reasonable grounds for believing that the circumstances with respect to the children are as mentioned in section 31(2) of the Children Act) and he also had to determine whether, if that was satisfied, the removal of the children into foster care under an interim care order was required (as the local authority and the guardian contended) or whether the children could safely stay at home (as the parents wished).

3. I should say that when this case commenced we did not have the benefit of a transcript of the judgment of the judge.  It was thought at that point that the judgment had not been recorded or the tapes had been lost.  However, thanks to the judge's efforts the tapes were found today and a transcript has been provided in the course of the hearing.  We have been able to look at the important parts of that transcript, although it has not been possible to move, in my case at least, entirely from the very helpful note that counsel provided of the judgment to the transcript in all respects for the purposes of giving this judgment, so I may at times refer to the notes that were provided to us earlier by counsel.

4. I need to deal with the background.  In outlining the history of the case for the purposes of the judgment I stress that none of the evidence has yet been tested.  What I am about to do is to pick out features from it which will of course come under scrutiny fully when the matter is explored by the court in February next year.  Apart from the injuries that L has sustained, there has been no cause for concern about the care given by the parents to the twins.  The local authority became involved with the family after L, then about two months old, was found to have a spiral fracture of his femur.  He had been taken to hospital by his parents because he was unsettled and they thought that his neck was painful. The triage notes from the accident and emergency department at the hospital record that he was seen there just after noon on 6 February.  He was examined by doctors and admitted to a ward, but initial investigations did not lead anyone to suspect that he had a leg injury, although his hips and legs were examined.  The leg injury came to light in the evening. His mother drew the attention of the medical staff to the fact that he had screamed when she had moved his leg to change his nappy.  An X ray was therefore carried out at about 10.30 in the evening and it revealed the femur fracture. 

5. After this, the parents went to live with the twins with the paternal grandparents in order that the grandparents could offer supervision of the parents' care of the children. It was then determined at a multi-agency meeting on 11 April that the family should be allowed to return to their own home with the children.  They did that, and in due course the case was closed.  The social work chronology which is with the papers suggests that a child protection conference in May 2012 accepted that the explanation for the femur injury was "...likely to have been an accident whilst preventing L falling on to the floor".

6. This was a reference to an incident which had occurred on 31 January when the father was holding L.  The 31 January incident is described by the father in his statement in this way.  He said that he was in the kitchen with L in his arms.  He was making L's milk.  As he reached for the kettle L jerked backwards with his head and body and the father panicked and pulled him tighter to him.  However, L had already started to slip out of his arms, his head was towards the floor but his legs were caught between his father's arm and body.  L slipped out of that position and the father put his left hand underneath him to catch him. One hand was under the baby's bottom and the other was on his stomach.  From there he pulled him into his arms again.  He used a tight grip to stop him from falling.  That evening, he says, the mother found small bruises on L's bottom when she undressed him for a bath.  The parents were due to take L for a check with the health visitor and for vaccinations the next day, so when they arrived at the surgery they explained about the bruises and the doctor checked L and said that he was fine.  Today our attention has been invited to a note of the GP about that. 

7. I return to the chronology of this year.  The twins lived at home with their parents from April.  A further fracture was then discovered in August, this time of L's arm.  On Saturday 11 August the mother went to work on the night shift. The father put the babies to bed and heard L screaming, he says, soon after the mother went out.  It is said that he found L with his arms through the cot bars. 

8. On Sunday morning, by which time the mother was back from work, she did not think that L's arm looked right and she thought it seemed limp. The paternal grandparents seem to have reassured her that everything was OK and that she was being over-anxious.  However, on the Monday morning she was still a little anxious and so she took L to see the GP.  The GP referred him to hospital where he was X-rayed and the fracture of the elbow was found.

9. The children were then accommodated in foster care by agreement with the parents until the hearing in front of HHJ Murdoch, when as a result of his order they went home.  They therefore have been at home for approximately five weeks now.  Social services have been attending daily in order to ensure that the children are safe.

10. There is a fair amount of medical evidence assembled so far, but not all of the medical reports that are expected are yet available. The radiological expert evidence is, I think, complete.  It comes from Dr Chapman and Dr Sprigg and I will turn to it in a moment.  Dr Chapman had the X rays referred to him by the hospital where L was treated and Dr Sprigg has been instructed by all the parties in the proceedings jointly.

11. There is a report from an expert in genetics, Dr Saggar, which is anticipated by the end of 21 December.  The purpose of that is to identify whether there is an underlying genetic condition or cause for the fractures that L suffered.  So far on the genetic front, there is a letter from a Dr Irving, a consultant in clinical genetics, to L's GP in which she describes the family history that she has taken and recommends on the basis of that that there be an analysis for osteogenesis imperfecta and an additional opinion on the X rays.  She also offers to assess L in the genetics clinic.  It seems to me, however, that the matters that she said should be addressed have been overtaken by the dual instruction of both Dr Saggar and Dr Sprigg.

12. Also awaited is a report from a consultant paediatrician, Dr Essex.  That is due by 18 January next year.  The purpose of that is to provide a paediatric overview and an opinion on whether the fractures were accidental or not.

13. To return to the radiological evidence: Dr Sprigg reports that the X Rays suggested that the February fracture was no more than 24 hours old at the time of the first X-ray but that the symptoms noted suggest that it may have been present for longer than that. He would not, however, date it back as far as 31 January when L slipped whilst his father was making his feed.  He classes the fracture as a separate and much more recent event than that.  As a possible cause of it, he gives "forceful rapid rotational injury".  As for the August fracture, given the age of the baby, he does not accept that it could have happened as a self-inflicted injury in the cot bars, so he concludes that there is no clear explanation for it.   As for the possible cause for that, he says "forceful rapid over-extension of the elbow beyond its natural locking point."  He comments that each injury would have been painful at the time, although the symptoms would have been reduced by pain relief such as Calpol.  He says that as far as he "can determine radiologically there is no radiological evidence of predisposing bone disorder".  He notes, however, the family history of bone problems, which he says is "worthy of further medical investigation of the affected members".   And he notes also the suggestion that L has some abnormality of his sclera.  He advises that a paediatrician with experience in the assessment of bruising and bone disorders should provide an overview, assuming that L had not yet been referred to the genetics team at Guys Hospital.

14. He was asked to say whether he considered that the cause of the injuries was accidental or non-accidental.  He said that it could be either.  I need to quote from his report specifically.  This is D110.  He said:

"It is the absence of any clear history of a forceful accidental event that raises the issue of NAI.  It is rare for a premobile infant to fracture any bones without an evident explanation as to  how and when the injury occurred.
If there was a predisposing bone disorder the fractures would be equally painful, so the carer would know how and when the fracture occurred even if it happened with lesser force than in an infant with normal bones.  It is particularly unfortunate that he seems to have had three significant injuries by the age of eight months
."

15. Dr Chapman similarly dated the February fracture to within 24 hours preceding the first X-ray.  He said it suggested "a significant twisting injury".  Of the subsequent X-ray, timed at nearly 12 noon on 7 February, he said that the appearance suggested that the fracture was still less than 24 hours old.  The X-ray at nearly 16.30 that afternoon suggested to him that the fracture was by then approximately 24 hours old or older.  Of the elbow fracture he said that the appearances of the X-ray were consistent with a two-day old injury, which is what it would have been if it had occurred just before the father found L with his arm through the cot bars.

16. I need to quote also from his report, which is a short one and is at page D20 in the bundle.  In a section entitled "Comments" he says:

"1. Neither of these fractures is a definite non accidental injury, but both are very unusual accidental injuries at this age and the fact that [L] has sustained two injuries and at least one is unexplained (you don't tell me if there was an explanation for the femoral fracture) lead to the conclusion that non-accidental injury is the most likely diagnosis.  It is also of some concern that the recent fracture was associated with a two-day history.

2. There is no radiological evidence to support a diagnosis of osteogenesis imperfect.  Even in this condition children typically present with a fracture after a known event.

3. There is no radiological evidence of rickets.

4. The femoral fracture would have been the result of a twisting force applied on the length of the thigh.

5. The humeral fracture would have been the result of forced hyperextension of the elbow.  In an older child this is typically secondary to a fall.  This could still be an explanation in a child of this age, but the direct application of a force to the forearm, hyper-extending the elbow is more likely."

There therefore appears, in summary, to be a considerable measure of agreement between the radiologists.

17. I need, whilst dealing with the radiologists, to deal with a particular argument that is advanced by the parents with regard to the timing of the first fracture. They argue that the evidence points to that first fracture having occurred in hospital and they say this:

"That is so because the doctors who examined L on his admission to hospital did not find anything to suggest that his leg was injured despite having manipulated it when they examined him".

Indeed the suggestion was made in argument that it may have been that process which caused or contributed to the fracture.

18. Secondly, they make a point because of dating which they derived from the radiological evidence about the X-rays.  The argument on this goes in this way.  First, Dr Chapman's dating on the first X-ray puts the injury as not earlier than 10.30 pm on 5 February: that is, the day before the baby was brought to hospital. Second, his dating on the X-ray timed at nearly noon on 7 February is that the fracture is still less than approximately 24 hours old. That means it was caused not before noon on 6 February.  The 4.30 pm X-ray on 7 February caused Dr Chapman to say that the fracture was now approximately 24 hours old or older.  That means, the argument goes, that it was caused before 4.30pm on 6 February.  Putting all those three stages together, therefore, that means, the argument goes, that the window during which the fracture must have been caused was between noon on 6 February and 4.30 on that day.  We know from the triage note that L arrived at hospital to be seen at 12.09 for the first time and therefore the argument is that he was at hospital at the time of the fracture of the femur.

19. It is, of course, not necessary for us to take any concluded view about timing of that fracture.  All that it is necessary to say at this stage is that the parents' analysis does not add up to incontrovertible evidence that the first fracture was caused in hospital and nor did the judge think that it did.  Radiological timing is not exact.  Dr Chapman speaks of the appearances "suggesting an injury was during a particular time frame" and he adds the qualification in his report of "approximately" to some of his timings.  Furthermore Dr Sprigg comments that the dating of fractures is not an exact science and he comments that, although the X-ray suggests one thing with regard to L's femur fracture, the symptoms that are reported may suggest that the fracture was earlier.  Furthermore the idea that the first injury was caused in hospital is undermined potentially by the fact that there was another fracture later which cannot be explained in that way. 

20. Drawing the threads together, therefore, with regard to the evidence as it is at the moment, there was before the judge a significant body of evidence pointing to the distinct possibility (I deliberately use a relatively neutral description) that L had sustained non-accidental injuries.  

21. There has been no appeal by the parents against the judge's determination that the threshold in section 38 has been satisfied.  The debate, therefore, in this appeal has been in relation to what type of order the judge should have imposed with the local authority, supported by the guardian, contending that an interim care order was required. 

22. It is as well to remind ourselves of the approach that should be taken to applications for interim care orders.  The principles are to be found in a number of authorities and I do not in the course of this judgment, given late in the day, intend in any way to restate principles which have been established over the course of the years in this court.

23. So, with that caveat that this is not intended to be in any way a reformulation of the test with regard to interim care orders, one might say that it is the welfare of the child that dictates the result, that dictates the order that the judge should impose at the welfare stage of an interim hearing.  The welfare is, as HHJ Murdoch says, the court's paramount consideration and what the court is looking for is whether the child's welfare demands that he or she should be removed immediately from his or her parents' care for his or her safety or whether, putting it another way, removal from their care is a proportionate response to the circumstances as they appear to be to the court.  In carrying out that evaluation the court must, as HHJ Murdoch said, bear in mind the welfare checklist set out in section 1(3) of the Children Act.

24. The local authority argue that in this case the judge failed to carry out a proper balancing exercise in relation to the various factors that he had to take into account in deciding what order to make.  They argue that he failed to attribute proper weight to certain features of the evidence, not just the mere fact of the fractures but also the absence of injury when L was not in the sole care of his parents, the absence of any identified non-sinister cause for the fractures, the opinions of Dr Chapman and Dr Sprigg, and the recommendation of the guardian that, pending full completion of the inquiries into what happened, there should be an interim care order.

25. In order to evaluate the local authority's arguments, we need to look at the factors that the judge took into account and how he weighted them and how he explained his decision as to the order that he should make.  He expressly recognised that he must have full regard to, as he put it, such risk of injury as it seemed to him there may be if the children were returned to their parents' care, as he put it "in other words a further unexplained and possible non-accidental injury.  He said that he had to evaluate that against a background of a case where there were no other welfare concerns.  He put into the balance the effect on the children of being separated from their parents.  He said that the effect of separation on young children is potentially damaging, no matter how good their foster care may be.  He commented that it is often difficult to evaluate the extent of the damage that would be caused by the resulting disruption of the bond between the child and their primary carer, particularly in the case of children who were under one year old, but he concluded that there was no doubt that there would be some damage.  He evaluated the alternatives to a return to the care of the parents.  He seems to have recognised potential difficulties with both of the proposals that the parents made, that is supervision by the paternal grandparents or the father leaving the home so that the mother could care for the children alone there, but he did not in fact go on to deal with those possible solutions because he did not consider that it was justified to limit the parents' care of the children in either of those ways.

26. He said that he had (this is a quotation from the note of judgment):

"...considered with some care the anxieties of [the local authority] and the [guardian] about the possible risks to L and M of a reuniting of them with their parents.  I have nevertheless come to the conclusion that a continued separation of the parents and their sons is not required at this stage of these proceedings.  In my judgment the risks to the twins, although far from non-existent, are not so serious as to justify a continued separation of parents and children.  I bear in mind the need for the court to be satisfied that separation is necessary and proportionate, and in my judgment in this case it is neither. I understand the concerns of the [guardian] and the [social worker] but in my judgment the risks are not so serious here as to justify a continued separation."

27. He was satisfied that the parents would cooperate with the local authority visiting as often as the local authority wished, but he did nonetheless make an interim supervision order, rather than making no order at all, as an indication of the need for the continued surveillance of the local authority of the risks that may exist in this case.

The local authority's arguments
28. The local authority criticise the judge for giving too much weight to the harm that the children would suffer by being separated from their parents. 

29. As part of that criticism, they argue that the judge failed to take into account that the children were settled and thriving in foster care.  I do not see that particular criticism of the judge as entirely valid, as I indicated in the course of argument this afternoon.  In my view, this experienced family judge was entitled to take into account that any prolonged separation of young children from their parents is likely to be harmful, although it was obviously necessary for him to look closely at how that harm measured up against the risk of physical injury.  And, in so doing, the local authority are no doubt correct to say that he had to consider how things were actually working out for the twins in foster care, which may have limited the harm that they were suffering from the separation from their parents.

30. For me, the key issue in this case is whether the judge gave sufficient weight to the risk of injury that the children faced and whether he sufficiently explained his approach to that and to why it was he felt that he was able to take the risk of putting them back in the care of their parents or whether the local authority are right in arguing that he did not.  I remind myself firmly that, in determining an appeal, we are not making the decision that the trial judge had to make ourselves, particularly so in a case like this one where the judge heard oral evidence and argument over the course of a day.  We are considering whether the judge failed to apply the proper approach to the issues that he had to determine and/or whether his decision was plainly wrong and/or whether he has failed to explain why he took the decision that he did sufficiently or at all.

The parents' arguments
31. The parents have responded valiantly to the urgent listing of this case and have advanced their arguments in opposition to the local authority's appeal.  They raise a number of points.  I have dealt with one already, which concerns the strength, or, as they would say, deficiencies in the evidence as to whether the injuries were occasioned in their care at all, that is to say the timing of the leg fracture. They argue that the judge considered the whole matter entirely properly and justified why it was that he took the decision that he did. He clearly recognised that there was a risk of non-accidental injury if the children were returned to the parents' care.  He said so in terms in his judgment, but he then went on and said that there were other features which had to be borne in mind including that there had been no prior welfare concerns with regard to the parents.

32. Another feature that potentially went against there being a risk and that that he was entitled to consider, the argument goes, was the open question with regard to genetics.  A further relevant matter was the fact that it was the mother who had drawn the attention of the doctors to both of the injuries.  That should be seen as a protective factor in the home which would assist in maintaining the safety of the children. 

33. The parents argue that this was an exercise of discretion, that the judge took into account the proper factors and that his decision should not be interfered with.

The guardian's arguments
34. The guardian puts the matter very simply, drawing attention to the seriousness of the injuries that L has suffered and the evidence of how they are likely to have occurred.  I have gone through that evidence already insofar as it is presently available.  The guardian submits that the judge had to carry out a proportionality analysis.  The harm was serious and the guardian herself could not, as she told the judge, support a return of the children to the parents' care until enquiries were over because of the risk that they would again suffer such serious harm.  She submits that the judge did not address that proportionality analysis properly in his judgment. 

Discussion
35. The judge heard, as I have said, oral evidence from the mother and the father and from the paternal grandfather as well as from the social worker who had only recently handed the case on.  It is perhaps not surprising, given that it was an interim hearing and he was not there to make findings of fact, that the judge says little about that oral evidence in the course of his judgment.  He rightly reminded himself that he should not make findings of fact at this stage, but he did go on, again rightly, to consider the evidence about the injuries in some detail.  He noted the incident of 31 January.  He went through the timing of the February fracture from Dr Chapman's evidence and he took into account the parents' belief that the clinical examinations on 6 February may have exacerbated an injury that was already present because of what happened on 31 January.  He referred also to their thought that the radiological timing may indicate an injury in hospital.  He said, however, that he thought that the parents may be attaching too much significance to the timings in Dr Chapman's report and he said, quite rightly, that Dr Sprigg's report was necessary. 

36. He noted the fraught situation in the grandparents' house when the parents had lived there and the fact that the father on 18 March had got very upset and punched a kitchen cupboard, injuring his hand, but he said there was nothing else in the papers suggesting that either parent suffered from a short temper or was given to displays of aggression or violence. 

37. He reviewed the evidence in relation to the August fracture and he quoted verbatim the concluding passage from Dr Chapman's report, which I have already gone through in this judgment.

38. His conclusion was that the threshold for an interim care order was crossed.  It seems that he considered that this was so on the basis of both of the fractures.  He stressed that he was not making a finding on the balance of probabilities about that and he said that there were plainly, on the evidence, matters that went in the opposite direction.  However, he said at paragraph 31 of the now draft transcript :

"I make it plain that there are plainly on the evidence matters which might be going in the opposite direction.  But it appears to me that both of these fractures and the circumstances surrounding them suggest that there are grounds for believing that one or the other of the parents may have caused those injuries."

That came towards the conclusion of his consideration of the threshold criteria. 

39. When the judge went on to consider the welfare issue, he said this at paragraph 33:

"When, however, I come to look at the second stage of the decision making process at this hearing, I must look at the matter in the round.  I must look at the existence of arguments which go in the other direction in respect of the femoral fracture and the possibility that there is that the findings at the fact finding hearing in February may not be to the effect that non accidental injury has been caused."

40. One might have expected that that passage in the judgment would then have been followed by an enumeration by the judge of the various features which gave the judge reassurance in placing the children with the parents in the interim period or at least a closer examination of the risk that there was to the children in the parents' care, including the features that gave rise to concern, not just in the shape of the medical evidence available so far but also the other matters such as the existence of the 31 January incident and the absence of injury whilst under the supervision of the grandparent or, subject to a hand swelling which is noted in the clinical records, in the care of the foster parents.

41. In short one would have expected the judge, faced with the seriousness of the injuries which L had suffered so far and which he had found there were grounds for believing had been caused by one or the other of the parents, to go on at that point to explain why nevertheless he felt the risk was one that he could take.  One would have expected him at that stage, I think, to have explained what he thought the risk was and what, if any, he thought was the chance of such harm as the children risked actually happening, whether it was predictable as to whether it would happen and what protective features there were in the case that would guard against it.  The judge does not go on to deal with matters in that way. He sees the matter in terms of a balance between the risk of physical harm and the risk of harm to the children's bond with their parents.  He clearly arrived at the view that the risk of the harm to the bond was greater than the risk of the physical harm, but he does not explain in his judgment how it was that he arrived at that evaluation.  Given the gravity of the circumstances here I see that as a fundamental flaw in his evaluation of the matter or at least in his articulation of how he saw the respective risks.

42. We were asked to say that no judge could have arrived at the decision that was arrived at in this case.  I am reluctant ever to say never in a family case, because each case depends upon a sophisticated mixture of the particular facts in the particular case.  I may have taken a lot of persuasion to have countenanced a return of children in circumstances such as these, but I would not translate that into saying that no judge could take that course.  But what a judge would need to do in those circumstances is to spell out very clearly why it was that he felt that the risk could be taken.  That is missing from this judgment and I would therefore overturn the decision made by the judge and would hear further submissions, insofar as those are necessary, with regard to what needs to happen next.

Lord Justice Etherton:
43. I agree.  The judge followed the two-stage process appropriate in this case: that is, (1) determining whether section 38 (2) of the Children Act 1989 is satisfied in that "there are reasonable grounds for believing that the circumstances with respect to the child are as mentioned in section 31(2)" namely the significant harm tests, and (2) if the section 38 requirement is met, examining the paramount welfare of the child, applying the welfare checklist in section 1(3), and having regard to delay (section 1(2)) and the no order principle (section 1(5)).

44. At the second stage, there were three important issues to bring into the balance in the present case.  One was the strength of the local authority's case, the stage (1) threshold not being one which is particularly high; secondly, the frequency or likely frequency of any harm occurring and its predictability; and, thirdly, the seriousness of harm should the risk materialise.  The Judge, in carrying out his evaluation exercise, did not address those issues separately but broadly stated his conclusion that the risk of harm was not such as to outweigh the risk involved to the children in a continued separation of them from their parents.   He said the following in paragraph [46] of his judgment:

"I have considered all of these possible regimes and I have considered with some care the anxieties of the local authority and the guardian about the possible risks to L and M of a reuniting of them with their parents. I have nevertheless come to the conclusion that a continued separation of the parents and their sons is not required at this stage of these proceedings.  In my judgment the risks to the twins, although far from non existent are not so serious as to justify a continued separation of parent and children."

45. There is, therefore, absent from the otherwise careful judgment of this experienced judge an analysis justifying that baldly stated conclusion.  In those circumstances I too agree that the order should be set aside.

Lord Justice Moore-Bick:
46. I agree that the appeal should be allowed for the reasons given by Black LJ.

Order: Appeal allowed