Facts still very much matter
Christmas has come early for fans of Court of Appeal decisions in public law proceedings, particularly for fact fans. The Court of Appeal, in the last 7 days of October, published three very notable decisions that check the ongoing push for public law proceedings to conclude quicker and quicker. Alongside these decisions, Sir Andrew McFarlane appeared before the House of Commons Justice Select Committee for about an hour on the afternoon of Wednesday 29 October, when the majority of the MPs questions were fixated on how to reduce delay, rather than any real focus on how to continue to ensure that these proceedings were fair. To add to all of that Nigel Farage introduced a private members bill on the same day continuing his quest to end the UK’s membership of the ECHR, with the enshrined protections under Articles 6 and 8 that are so often quoted in submissions and judgments.
I have written previously about the importance of facts within proceedings and how so often the push for speed within public law proceedings can see the nuance between facts and allegations be either slightly or entirely missed. I remain as surprised now as I did then about how regularly the factual basis for a set of proceedings is either missed entirely or consciously overlooked.
The three Court of Appeal decisions all have at their core the importance of a clear factual history underpinning a fair outcome to these incredibly important decisions. The decisions of which I speak are:
LJ Cobb in Re H (Final Care Orders at IRH) [2025] EWCA Civ 1342
LJ Jackson in K-H (Care Orders : Proportionality) [2025] EWCA Civ 1368
LJ Cobb in Re D (Threshold Findings and Final Orders at IRH) [2025] EWCA Civ 1362
The decisions themselves are must read decisions for all public law practitioners, but I will seek to summarise the key areas within this article, together with considering where all of this takes matters when comparing the ongoing (and often justifiable push) to avoid delay.
Re H (Final Care Orders at IRH)
On 23 October 2025 the CoA handed down this decision on appeal from a Circuit Judge in Chelmsford. The proceedings were in their 128th week when they got to an IRH which was the 12th hearing within the proceedings. The local authority & guardian pursued care orders for three boys. The parents opposed the plan, with the father seeking the children be placed in his care. The local authority in advance of the hearing were content for the case to be listed to a final hearing to determine their application, but at the hearing changed this position and sought for the judge to conclude the case at the IRH. This was unsurprisingly opposed by the parents but supported by the Guardian.
The judge went on to hear submissions from all parties and within 45 minutes (start to end) had heard those submissions, given an ex-tempore judgment and made the final care orders. The father immediately sought permission to appeal (denied) before being given permission by the Court of Appeal with a hearing taking place on October 2025.
The Court of Appeal overwhelmingly allowed the appeal on two grounds, firstly that it was wrong for the judge to have made final care orders at the IRH but also that the reasoning for concluding at the IRH was inadequate. Crucially the Court of Appeal did not say that it was never appropriate to conclude a case at the IRH, however the judgment is a robust warning that if a judge does seeks to do so then it can only happen in an appropriate case, time is given to properly air whatever issues exist and then a robust explanation given within the judgment as to why a final hearing was not listed.
Cobb LJ specifically warns the following (paragraph 38):
The judge was plainly concerned at the IRH about the undue delay, and she was right to be so … It is easy to understand her eagerness to resolve the case at that hearing, and I accept that this was well-intentioned. However, the benefits of robust resolution of public law proceedings at any stage prior to the final hearing (where evidence is tested) must never be at the expense of procedural fairness and justice (rule 1 of the FPR 2010, Article 6 ECHR and Re J at [18], above). In this case, the solution to the procedural disarray of this protracted litigation did not lie, in my judgment, in the summary termination of the proceedings at the IRH in a manner which was procedurally unfair to the respondent parents.
He went on to emphasise that in this case ‘There was a material issue of fact to be tried’ which was one of future risk, critically whether the parents had been ‘able to achieve any positive change to their lifestyles that would support the children returning to either of their care.’
There were equally issues with the lack of notice the parents were given to the plan to resolve the case at this final hearing. The court emphasised:
‘There should always be sufficient preparation and hearing time at an IRH, so that the parties are treated justly and fairly, and no-one is denied the opportunity to attempt properly to resolve the issues.’
Further to this in this specific case there were issues regarding the actual plan of the local authority for these children under the care plan. The judge gave very limited explanation with regards to contact proposals or indeed what the actual placement provisions were going to be for the children. Cobb LJ emphasises the need for the court to consider the ‘permanence provisions’ in the s31A plan and specifically the issue of contact.
Finally, the court emphasised the need that IF cases are to conclude at an IRH there is a need for the court to give clear reasons as to why. Cobb LJ suggests that these reasons must explain:
i. Why the IRH has been used as a final hearing, and / or why the proceedings are not being case managed to a further / final hearing, particularly if (as here) there is a dispute as to whether the proceedings should be concluded at the IRH; and
ii. The substantive final orders which are to be made at the IRH. In this regard, I suggest that the judgment should include specific reference to the threshold criteria (section 31(2) CA 1989) and a review (however short) of the evidence which supports the same, a discussion of the balancing exercise in which each future option for the child is evaluated “to the degree of detail necessary” (Re B-S (Children) [2013] EWCA Civ 1146 at [44]), the permanence arrangements in the care plan (section 31A CA 1989), the constituent elements of the welfare checklist (section 1(3) CA 1989), and the contact provisions (section 34(11) CA 1989).
Returning to my theme of facts, the court was particularly concerned with the issue of the threshold criteria. The judge suggested that the threshold was crossed in part because of the failure of the parents to respond historically to threshold. Cobb LJ described this ‘summary treatment of the threshold criteria was barely adequate (if it was adequate at all).
As all appeals the case turned slightly on the specific facts of this case (most notably the lack of coherent care plan and an actual argument to pursue at final hearing), but the judgment provides a clear cautionary warning against all decisions being taken at the IRH stage. The regularity with which cases are now listed to IRH / ‘Early final hearing’ always concerns me. Why the need for the ‘slash’? Why the need for saying it might be an ‘early final hearing’? Why is there a need for a suggestion that because it has been called this it will be anything different than the IRH that is defined within the PLO? An IRH is able to be an early final hearing, but, as this case specifically warns, it cannot always be, even when a case has been ongoing for almost 5 times the 26-week deadline.
Facts remain just as important after 128 weeks as they do at 26 weeks.
K-H (Care Orders : Proportionality)
On 29 October 2025 the CoA handed down its decision in this case. This was an appeal from a Recorder at a properly constituted final hearing, albeit again a case that had been ongoing for approximately 128 weeks. During the entirety of the proceedings the three subject children had lived with their paternal grandparents. There was no suggestion of any previous harm being caused to the children, but the local authority pursued its case solely on the risk of future sexual harm in the grandparents care due to historic convictions of the grandfather (some 40 years ago) and two sexual incidents involving an uncle approximately 10 and 18 years ago. There were also some other disputed allegations against the uncle which were not accepted and were never sought to be proven.
The appeal focussed on the risk assessment that had been undertaken by the Recorder and how, even if the unproven matters were considered, how the evidence as a whole could justify the permanent removal of the children. Indeed, the Court of Appeal were so robust in their view that they substituted the care orders to final supervision orders, ending the public law proceedings rather than remitting them. The only outstanding issue that was remitted was the making of Special Guardianship Orders to them.
The grandfather had historic convictions from 1975, 1978 and 1984. The uncle had no such convictions, but there were a number of allegations that had been made between 2007 and 2015. Expert risk assessment had been undertaken that concluded that the level of risk posed by both the grandfather and uncle was ‘very low’ if it existed at all.
Jackson LJ notes surprise at the proposed length of the final hearing being five days and even greater surprise that the case ‘lasted no less than seven days, a surprising length of time for such a clearly defined issue.’ A lengthy 50-page judgment was given by the Recorder.
The key focus of the appeal was about the risk analysis that was undertaken but also on the factual basis which underlies that analysis. Jackson LJ sets out some ‘general principles’ from paragraph 59 which can be summarised as follows:
Risk is part of life, and everyone is entitled to choose the level of risk that is acceptable to them. Courts, however, have to adhere to the legal principles governing risk assessment in the context in which they are working.
A child is likely to suffer harm if there is a real possibility, a possibility that cannot sensibly be ignored having regard to the nature and gravity of the feared harm in the particular case:
The court’s conclusions must be based on facts. Facts are evidence of which a court can be judicially satisfied to the requisite standard of proof … In my view these unresolved judicial doubts and suspicions can no more form the basis of a conclusion that the second threshold condition in section 31(2)(a) has been established than they can form the basis of a conclusion that the first has been established …. A decision by a court on the likelihood of a future happening must be founded on a basis of present facts and the inferences fairly to be drawn therefrom.
If the case is based on the likelihood of future harm, the court must be satisfied on the balance of probabilities that the facts upon which that prediction was based did actually happen. It is not enough that they may have done so or that there was a real possibility that they did.
Jackson LJ specifically says the following, which cannot be repeated enough:
It is therefore unarguably clear that when the court is assessing risk it must act on the basis of proven facts and inferences that can properly be drawn from them. The submission that the assessment of future risk can be based even in part on unproven facts has been repeatedly rejected … The argument that it takes time to prove contested matters is no answer. If evidence is said to justify an order, it needs to be properly assessed.
This was the point that I was seeking to make in my earlier article on facts (albeit clearly not so eloquently). Practitioners must always be alert to what actually is a fact in a case. Far too often vast swathes of unproven allegations either obviously are taken into account by professionals undertaking risk assessments, or whilst denying they have taken them into account they blatantly have.
That appears to have been the issue in this appeal as Jackson LJ goes on to consider at paragraph 65. The repeated references by the Recorder in the judgment to ‘patterns of repeat behaviour’, things ‘not being able to be ignored’ and the ‘behaviour of the uncle’ showed a lack of any appreciation of the distinction between facts and allegations. Jackson LJ specifically suggests ‘it is clear to me that she was materially influenced in her assessment by matters that had not been proved or even investigated.’
The court also raises concerns with the fact that the uncle (about which so much was said) was never even called as a witness. Jackson LJ noting that ‘it was not procedurally fair to the family to rely on the disputed matters after the court had stood him down and his statement had not been challenged.’ Only re-iterating the point that if allegations are going to be used in a risk analysis they must be proven having heard the appropriate evidence.
Importantly Jackson LJ concludes that even if all of the allegations against the uncle had been proven, the evidence still was not sufficient to justify the making of the orders that were made. This is an important re-emphasis on the fact that children can (and should) remain with parents when risks exist. The fundamental basis for a welfare evaluation is a balancing of harms. Indeed, Jackson LJ makes this specific point:
Sexual abuse is a great evil, but the court’s function is not to protect children from all risk at any price … Overall, I am left with the sense that the recorder considered that any significant risk of these children being sexually abused in the home, arising from a combination of historic events and lack of protective insight, must lead to removal. That unduly privileged one factor in the welfare checklist over all others.
Again, this point cannot be repeated enough. The parents do not need to show an absence of risk to be able to care. They do not need to be able to ‘ensure’ that the child will be kept safe in their care as they, like the local authority, cannot ensure anything. Indeed, Jackson LJ also includes this important paragraph within his conclusion:
Placement in foster carer care would protect the children from the risk of sexual abuse by family members, but it would not “ensure that the children are kept safe from the risk of sexual harm now and throughout their minority”
This is a judgment again re-iterating the point that facts are the cornerstone of all public law proceedings, but most critically those that are predicated on future risks. However, the judgment is also a key reminder for the importance of a balanced risk analysis that accepts that risks will exist from almost all realistic option that is put before the court.
Re D (Threshold Findings and Final Orders at IRH)
Finally on 30 October 2025, the CoA handed down this third decision in 7 days. Again, this is an appeal from a Circuit Judge considering a case at an IRH. However unlike Re H, on this occasion the parents were not in attendance contesting the matter, nor seeking a final hearing. However, they latterly sought permission to appeal against the relatively limited judgment that had been given in their absence. A particular focus fell on how the threshold was crossed, but also the level of reasoning given in another brief judgment at an IRH.
The CoA specifically considered the practice (actively encouraged by the Standard Form Orders template) to say that threshold will be deemed to have been accepted if they failed to file a response to the threshold. In this appeal the CoA were concerned that the judge ‘may have treated ‘deemed’ acceptance of threshold facts as ‘deemed’ proof of the same.’
The judgment references the final threshold statement which was a short four paragraph document. Cobb LJ notes that it was defective in material ways. Firstly, it included reports of ‘alleged facts’ which were relied upon to predict future harm, very similar to the point made the day before in K-H. Secondly, it suggested that professionals asserted that if the child returned, she would be at risk. Cobb LJ makes the point that this should never be in a threshold document. Thirdly, it failed to link the facts relied upon with the statutory threshold grounds. Notably Cobb LJ reiterates the point from Re A:
A case based on a lack of honesty with professionals must feed through into a conclusion that the child is suffering or likely to suffer a particular type of significant harm. In this case, as I regret in many others of its type, “the conclusion does not follow naturally from the premise”
Simple dishonesty does not automatically mean that there is a risk. The threshold document needs to link how that fact means that the child actually faces a risk of harm. Whilst clearly incorrect dishonesty is often very difficult to link to a threshold risk, it is far better to be argued as creating an inability to protect against other proven future risks.
The judgment however isn’t a restatement of Re A and the inevitable ongoing regular deficits seen in Threshold documents, some 10 years after it was handed down by Sir James Munby. Its principal focus is upon the way that the final basis of threshold is determined when the parents do not engage either with a response or the process more generally.
Cobb LJ emphasises the dicta of Sir Andrew McFarlane in Re B (Adequacy of Reasons) regarding the importance of the threshold findings when considering the basis for the welfare analysis when he said:
‘The task of evaluating threshold goes to the core of the judicial exercise in every case. It is, in essence, what the case is about. Unless the court has a clear and detailed understanding of the basis upon which it finds, if it does, that a particular child ‘is suffering or is likely to suffer significant harm’, substantial difficulties will be encountered when the court then moves on, as it must, to evaluate future risk of harm at the welfare stage. Public law proceedings under CA 1989, s 31 are engaged in the business of ‘child protection’. Unless a court has made detailed findings as to what it is that a particular child is to be protected from, in terms of significant harm, it is unlikely that the court will be able to undertake a focussed and bespoke evaluation of any plan to protect the child from that harm.’
This makes the point that unless the court knows what the risks are that underpins its decision, how can it analyse those within the welfare element of the judgment.
The parents in this case had almost completely failed to engage. They’d refused to be assessed; they’d refused to attend many contacts, and they had dismissed their lawyers. The parents then chose not to attend the final hearing and there was a warning that the case could be concluded at the IRH. Despite this Cobb LJ reiterates:
The ultimate resolution of the proceedings had to be done in a fair and just way; if final orders were to be made, they needed to be clearly – even if briefly reasoned, and those reasons laid out in a judgment. Sadly, the judgment under review falls far short of what is expected.
Cobb LJ appears to conclude that the reasons for the failures in the judgment are caused by an insufficient consideration of the threshold document and thus the basis for the entire case. Cobb LJ emphasises that:
- Determining whether the threshold is crossed is ‘squarely on the court’. Whether it is crossed should not be ‘determinately resolved by agreement between the parties, nor by default; the court must scrutinise the documents and satisfy itself of proof of the same and say why it is so satisfied’;
- The judge must exercise discipline in scrutinising a proposed threshold. In this case the judge had failed to spot the deficits highlighted above, and should have rejected the threshold as presented to him;
- The failure of the parents to engage with the process may have led the judge to ‘lower his judicial guard’ and distract him from the duty he owed under the CA 1989.
The most notable section is towards the end of the judgment where Cobb LJ queries the current Standard Form Orders regarding the ‘deemed acceptance of threshold.’ He notes that the current wording is as follows:
If [name] fails to comply with paragraph [para number of parent’s response direction] of this order they shall be deemed to accept the threshold allegations made by the local authority and to not be putting forward any alternative carers unless this paragraph is varied upon application.”
Cobb LJ says that he is troubled by this provision as:
Such an order may well have the effect (as shown by this case) of reducing or discouraging judicial engagement in conducting analysis by reference to the burden of proof of evidence necessary to establish the threshold facts. The effect is all too easy to see – that the determination of threshold becomes more of an administrative than a judicial act. The standard form formula has some of the characteristics of a default judgment under Part 12 of the CPR (esp. rule 12.3), which would be wholly inapt in proceedings concerning children.
He suggests that this wording appears to have the unintended consequence of reversing the burden of proof. Cobb LJ goes on to suggest an alternate wording which is as follows:
‘If the parents fail to respond [to the schedule of findings in support of the threshold criteria], the court may proceed to consider [at the next hearing / at the IRH / at the final hearing] whether the section 31(2) Children Act 1989 threshold criteria are established by reference to the written evidence filed by the local authority.’
It remains to be seen whether this wording is adopted by the ‘Standard Orders Group’ but the case is again a crucial re-iteration of the importance of the factual basis in public law cases that should flow from the threshold document. Whilst inevitably this is another example of a limited judgment being given at an IRH (in the absence of the parents) the more relevant takeaways must be on the reminder to judges that they should themselves determine that threshold is not only crossed, but more critically how. The way in which it is crossed is the factual history of the case, which is then the basis for what if any future risks may exist in the care of a parent.
A failure by local authority social workers (or Guardian’s) to understand the basis on which threshold is pleaded or accepted is the most common reason for defective analysis to exist.
Thoughts on these cases – Delay
As I have said at the top of this article it appeared to me very ironic that these three appeals were all published around the same time that Sir Andrew McFarlane was being questioned by MPs. The MPs on the committee appeared fixated on the delays in public law proceedings, wanting to know how the perceived delays were being reduced and what more could be done to tackle any backlogs. I didn’t hear any questions in the hour-long session about whether the focus on 26 weeks was leading to incorrect decisions being made or whether proceedings remained fair for those involved.
Whilst K-H doesn’t relate to the same issue, both Re D and Re H are clear examples of judges seeking to robustly manage cases to conclude them without further delay. On both occasions the pressures of the system led highly experienced Circuit Judges to produce analysis thought to lack sufficient reasoning to justify the conclusions that they had reached. The point about the Standard Orders underlines another attempt to short circuit the crucial issue of threshold, again for the purpose of simplifying proceedings and reducing delay, whilst at the same time appearing to cut across the crucial determination of s31(2) by the court.
It would be interesting to see a challenge to the CoA on the increasingly standard order that connected persons assessments will not be completed if not identified at the outset of the proceedings by a parent. Indeed, I recently saw one court centre’s standard order being that the local authority were prevented (by court order) from doing further assessments unless specifically authorised by the court. It may be that other elements of the Standard Orders come under scrutiny following the decision in Re D.
Re D and Re H are important reminders that regardless of the pressures from the legislature to reduce delays the process must be properly followed in all cases and cases must be fairly determined. Analysis even when parents do not engage need to be sufficiently detailed to justify the serious (often lifechanging decisions) that are being made. Article 6 rights to a fair trial, and ultimately an actual trial on evidence, remains regardless of the attempts to either simplify or speed up justice.
As Pauffley J said in 2014, ‘Justice must never be sacrificed upon the altar of speed.’ These cases reiterate that this is clearly still the case. It seems unlikely that following both Re D and Re H many judges will be willing to conclude significant disputes at IRHs, even if they are also called an ‘early final hearing.’
Thoughts on these cases – Facts, facts, facts
What underlies all these three cases however is a re-emphasis on one of my oldest and most repeated refrains, facts. It continues to surprise me that more is often not made about the factual basis of a case, when it has clearly led to an unfair risk analysis. Social workers must be regularly reminded of what the actual factual history is within a case, rather than what they have presumed it is.
All these cases specifically highlight the importance of the factual basis set out within the threshold document, and this being used both in the judicial analysis of risk but also the analysis undertaken by the professionals.
In Re H, the facts were important to be analysed because the fact in dispute was what improvements (if any) the parents had made during the proceedings. These facts were crucial because they were the crux of the welfare decision and had to be analysed with reference to hearing oral evidence. In K-H the facts were critical to the welfare analysis, and it was entirely unclear that either the professionals or the Recorder had properly separated proven facts from unproven allegations. In Re D, the lack of focus by the court on what the actual facts were (because of a reliance on the standard order mechanism) led to a paucity of analysis in the final judgment.
The words of Sir Andrew McFarlane in Re B (Adequacy of Reasons) are particularly important to restate, noting that the threshold is not just something to checkoff before making public law orders. The basis on which it is crossed should be the basis upon which future risk is analysed at the welfare stage.
Inadequate understanding, or analysis upon an unproven factual basis, still remains the easiest way for parents to expose a ‘tottering edifice built on inadequate foundations.’ If these cases emphasise nothing else, it is that there continue to be difficulties in understanding this despite the many previous warnings of the Court of Appeal.