The clinical picture the Crown presented
Between June 2015 and June 2016 the Countess of Chester Hospital neonatal unit experienced a cluster of unexpected collapses and deaths in infants. The pattern was not consistent with the unit’s previous years’ experience. The consultants who worked on the unit — Dr Brearey, Dr Jayaram, Dr Gibbs, and others — were the first to notice that the cluster did not match the clinical picture they expected.
When they reviewed the cluster internally in February 2016, they identified a common factor: one specific nurse had been present at every one of the events. A second thematic review in July 2016 reached the same conclusion. The consultants wrote to the executive team in September 2016 asking for police referral. That request was declined and the Trust commissioned the RCPCH review instead. The police were eventually contacted in May 2017.
The Crown’s causation expert, Dr Dewi Evans, reviewed the medical records of 17 infants across 22 counts and concluded that the pattern of collapses was not explicable on standard neonatal-clinical grounds. He identified specific alleged mechanisms — air embolism, insulin administration, nasogastric over-feeding, endotracheal tube dislodgement — for individual counts. A second expert, Dr Sandie Bohin, substantially endorsed his conclusions.
The insulin counts
Two blood samples taken from infants on the unit (Children F and L) returned insulin/C-peptide ratios that the Crown’s experts (Prof. Peter Hindmarsh and Royal Liverpool biochemists) described as consistent only with exogenous insulin administration rather than endogenous pancreatic secretion. The Baby F reading was 4,657 pmol/L; Baby L’s reading also showed the high-insulin / low-C-peptide pattern. The Crown argued these results could not be explained by any natural mechanism.
The Child K retrial
At the June–July 2024 retrial on Child K, Dr Ravi Jayaram gave eyewitness testimony that he walked into the nursery and found Ms Letby standing over a 25-week infant whose endotracheal tube had become dislodged and whose saturation alarm had been silenced. The jury convicted on that single count. The Child K count is the only prosecution count with a direct-eyewitness element.
The circumstantial overlay
Beyond the medical evidence the Crown presented a body of circumstantial evidence. The shift-rota chart showed Ms Letby had been on duty for 25 selected suspicious events. Handwritten notes recovered from her home included the phrases “I am evil” and “I did this.” She had searched Facebook for the surnames of parents of infants who had died or collapsed. She had retained 21 handover sheets relating to the indicted infants. In total the pattern — presence, written self-reference, searches, retention — was presented to the jury as consistent with the psychology of a nurse who had deliberately harmed infants in her care.
Why the jury returned guilty verdicts
The jury heard 10 months of evidence at the first trial. It heard Dr Evans and Dr Bohin give causation opinions on each count. It heard the consultants who had raised the alarm give their accounts of what they had observed and why they had escalated. It heard the biochemistry evidence on the insulin counts. It heard from Ms Letby, who denied harming any infant. It was given detailed directions by the trial judge Mr Justice Goss.
On 18 August 2023 it returned guilty verdicts on 14 counts (seven murders and seven attempted murders), not-guilty verdicts on two counts, and failed to agree on six. The July 2024 Child K retrial added one further attempted-murder conviction. The Court of Appeal refused leave to appeal in May 2024 on the evidence then available to it.
If you read that summary and find it compelling, you have understood the case the Crown made. The post-conviction review does not claim the Crown’s case was unintelligible or that the jury was unreasonable. It claims the case was constructed on evidence that has since been substantially undermined — by expert witnesses who were not available to the 2023 and 2024 juries, by analytical frameworks that were not deployed at trial, and by institutional developments (the Panel report, the Joint Insulin Report, the CPS January 2026 decision, the Thirlwall evidence) that have changed what a reasonable reviewer would now conclude.
What we think the prosecution case actually proves
Stated as carefully as we can: the Crown established that Ms Letby was on duty during each of the indicted events; that the unit experienced a cluster it did not expect; that consultants identified her as a common factor; that the clinical records show deterioration patterns that the Crown’s expert considered not routinely explicable; that certain biochemistry results (on the Crown’s reading) were consistent with exogenous insulin; that she wrote notes and made searches that on the Crown’s reading were consistent with guilt; and that Dr Jayaram gave direct-eyewitness testimony on one count at retrial.
That is a non-trivial body of evidence. It is the evidence on which the jury convicted. The post-conviction question is not whether the prosecution was frivolous. The question is whether the specific pieces of evidence — taken individually and collectively, and evaluated against the independent expert evidence that has since emerged — still justify conviction to the criminal standard (beyond reasonable doubt).
What happens from here
The CCRC applies the “real possibility” test under section 13 of the Criminal Appeal Act 1995. It is asking whether, if the Court of Appeal saw the current evidential picture including the Panel report, the Joint Insulin Report, the Thirlwall evidence, and the January 2026 CPS decision not to extend the pattern, there is a real possibility it would overturn the conviction. The prosecution case summarised above is the baseline against which that question is asked.