B & G (Fact-Finding)

Factual and Procedural Background

This is an appeal by the First Appellant (mother) and Second Appellant (father) against fact-finding conclusions made by Judge Kushner in care proceedings concerning two children referred to in the judgment as B (a boy, born 2020) and G (a girl, born 2024) (hereafter "Child B" and "Child G"). Child B has been diagnosed with autism and a form of diabetes. In September 2024 Child B was brought to primary care with multiple bruises; Child G was later found to have bruising and chronic subdural haematomas on imaging.

Following hospital investigations and child protection medicals, the local authority commenced care proceedings under the Children Act 1989 (section 47 assessment and Part IV application). Interim care orders were made in April 2025 and the children were placed in foster care. Expert reports (paediatric, haematology and neurosurgical) were ordered. A contested fact-finding hearing on causation ran in January 2026. The family court judge delivered an oral judgment in February 2026 making multiple findings, including that the listed injuries had been inflicted by the mother and/or father and that the perpetrator of the shaking injury to Child G failed to seek medical attention. The parents obtained permission to appeal limited to findings about the cause of the injuries. The Court of Appeal heard the consolidated appeals and reserved judgment, handing down the decision on 9 July 2026.

Legal Issues Presented

1. Whether the family court judge conducted a sufficiently rigorous analysis of the complex expert medical and non-medical evidence when reaching findings that the children's injuries were inflicted.
2. Whether the judge adequately assessed and explained her evaluation of the parents' oral and documentary evidence (credibility and reliability) and how that assessment informed her fact-findings.
3. Whether the judge's reasoning process and reasons were sufficient in law (including whether conflicts in expert evidence were resolved with adequate reasoning).
4. Whether the judge's findings as to disproportionality of bruising (including bruising while under 24/7 supervision and in foster care) were irrational or unsupported by the evidence.
5. Whether the judge impermissibly treated absence of an explanation as proof (i.e. reversed or impermissibly shifted the burden of proof) and failed to take into account forensic limits of the expert evidence.

Arguments of the Parties

First and Second Appellants' Arguments (consolidated)

• The judge failed to identify key features and conflicts in the medical evidence and mischaracterised expert opinion (notably neurosurgical opinion).
• The judge did not properly assess the parents' credibility or explain which parts of their evidence were accepted or rejected.
• The judge relied on flawed or incomplete bruise-logging (foster carer logs) and gave insufficient weight to the treating haematologist's clinical view (that the children bruise easily).
• The judge made substantive amendments to the transcript after delivery in a way that improperly affected reasoning, and used counsel's notes in corrections.
• The judge's overall reasoning was inadequate and could not be salvaged by requests for clarification; the fact-finding should be set aside or remitted.

Local Authority's Arguments

• The judge carried out an extensive survey of the evidence and was "steeped in the evidence"; her corrections to the transcript were permissible and reflected her thinking.
• The judge was entitled to prefer the paediatric expert's (Dr Robinson's) assessment of bruising and to treat the treating haematologist's view as a clinical "snapshot".
• The haematology evidence (joint expert) indicated that, on balance, the genetic findings were unlikely to explain the subdural bleeding; the judge's conclusions were open to her.

Children's Guardian's Position

• The guardian supported allowing the appeal on grounds relating to identification of the perpetrator (pool finding), noting the judge gave little analysis of the parents' oral evidence and impressions formed from them.

Table of Precedents Cited

Court's Reasoning and Analysis

The Court of Appeal (Judge Baker giving the lead judgment, with Judge Andrews and Judge Holgate concurring) examined (i) the expert evidence (paediatric, haematology, neurosurgery), (ii) the non-medical evidence (parents' statements and oral evidence, foster carer logs, supervised-care records), and (iii) the manner in which the family court judge prepared and corrected her oral judgment.

Key analytical points (as recorded in the opinion):

• The medical evidence was complex and uncertain: both children had two genetic variants of uncertain significance; experts disagreed or expressed degrees of uncertainty; the neurosurgical expert (Mr Jayamohan) identified four possible aetiologies for Child G's subdural haematomas and could not say which was most likely.
• The family court judge had delivered an ex tempore judgment from notes and later produced an approved transcript containing substantive amendments. The Court expressed disquiet that corrections may have been informed by looking at grounds of appeal and counsel's notes, and emphasised that any corrections must genuinely reflect the judge's reasons at delivery rather than being an ex post facto rationalisation.
• The judge had mischaracterised aspects of expert evidence (notably overstating Mr Jayamohan's position), which skewed her assessment of the medical issue.
• The judge failed to provide an adequate assessment of the parents' credibility and reliability despite the parents' oral evidence occupying a substantial part of the hearing; there was insufficient indication of what impression the judge formed of them and why parts of their evidence were accepted or rejected.
• The judge gave insufficient explanation for discounting the treating haematologist's clinical observations that the children bruise easily (Dr Dickens) in favour of Dr Robinson's analysis based largely on foster-carer logs; the foster logs were shown to record only "significant" or large bruises and therefore were an incomplete dataset for pattern analysis.
• The judge did not resolve whether there was a persistent pattern of unexplained/disproportionate bruising in supervised-care or foster-care periods; that unresolved factual issue undermined the haematological analysis and the judge's conclusions about disproportionality.
• Applying established authorities, the Court held that while medical uncertainty does not preclude findings on the totality of the evidence, the judge's totality analysis here was incomplete and flawed: misinterpretation of key expert opinion, inadequate evaluation of non-medical evidence, and insufficiently reasoned conclusions.

Conclusion of analysis: the Court found the judge's treatment of the evidence regarding the cause of the injuries was flawed in ways that rendered her conclusions on those issues unsafe.

Holding and Implications

DISPOSITION: The Court of Appeal allowed the appeal in part. It set aside the family court findings recorded as (2) and (3) in the judgment schedule (findings that each of the listed injuries was inflicted by the mother and/or father; and the finding that the perpetrator of the shaking injury, knowing Child G had been shaken, failed to seek medical attention).

Implications and direct effects:

• The remaining findings (relating to emotional harm, risk from the parents' relationship, father's drug history and mother's mental health) remain in place.
• The case is remitted to the Family Presiding Judge for the South Eastern Circuit (Arbuthnot J) to consider whether a rehearing of fact-finding on the cause of the injuries is necessary and, if so, its scope.
• The judgments of the Court of Appeal contain cautionary observations about judges delivering complex, medically‑intensive decisions ex tempore and the need for clear, reasoned written judgments that demonstrate how the court assessed medical and non-medical evidence together.

No wider legal precedent was created beyond the Court's application of existing principles concerning adequacy of reasons, evaluation of expert evidence in context, and the need for proper assessment of parties' credibility in care proceedings.