Prosecution v. Counter-evidence
Problems with the evidence
Each panel below pairs the prosecution’s claim at trial with the counter-evidence now on the record — primarily from the 14-expert International Panel convened by Dr Shoo Lee in February 2025 and from the statistical and endocrinological commentary that has accumulated since the 2023 conviction.
Air embolism — the Shoo Lee 1989 paper misapplied
Prosecution claim
The Crown argued that skin discolouration described on several infants — patches of pink surrounded by pale, almost marbled, skin — was diagnostic of air embolism, meaning air deliberately injected into the bloodstream via an IV line. The prosecution's expert drew the diagnostic criteria directly from Lee & Tanswell's 1989 Archives of Disease in Childhood paper.
Counter-evidence
Dr Shoo Lee, the lead author of that 1989 paper, has publicly stated that the skin signs described at the Letby trial do not match those in his research. The skin pattern in his paper describes a specific, large-vessel obstruction picture — not the patchy mottling described at trial. The 14-member Panel he subsequently convened reviewed every alleged air-embolism count and found none met the diagnostic criteria. Independent neonatologists add that the mottling described is a non-specific sign of any major circulatory compromise, including sepsis and natural collapse in a premature infant.
"The skin discolouration described in the Letby trial does not match the findings in our 1989 paper. There is no medical evidence of air embolism in any of these cases." — Dr Shoo Lee, 3 February 2025
Source: Lee & Tanswell (1989) Arch Dis Child; Shoo Lee International Expert Panel Report 2025; science4justice.nl scientific critique
Insulin poisoning — a screening assay used as forensic proof
Prosecution claim
Blood samples from two infants — Children F and L — returned results suggesting raised insulin with low C-peptide. Normally insulin and C-peptide are released together by the pancreas. A high-insulin-low-C-peptide pattern, the prosecution argued, is only explicable by insulin administered from outside the body. The jury was told this was proof of deliberate poisoning.
Counter-evidence
The Roche Cobas immunoassay used is a screening test. Its own manufacturer's guidance requires confirmation by mass spectrometry before a result can be treated as diagnostic of exogenous insulin. That confirmation was never done. Independent endocrinologists (including Adel Ismail and contributors to science4justice.nl) have shown the assay is prone to false positives in neonates because of interfering antibodies and cross-reactivity. The Panel concluded the insulin evidence cannot bear the weight placed on it in a criminal trial.
A screening immunoassay was never designed for forensic use. In every other British criminal case involving insulin, confirmatory mass spectrometry is performed. That did not happen here.
Source: Shoo Lee Panel Report 2025; Adel Ismail clinical biochemistry commentary; Prof. Geoff Chase (Canterbury NZ); science4justice.nl
The shift chart — selection bias presented as proof
Prosecution claim
A chart shown to the jury plotted 25 suspicious events against the nurses on duty for each. Letby was the only nurse present at all 25. The prosecution argued the improbability of this pattern, if she were innocent, was proof of her guilt.
Counter-evidence
The 25 events were selected in part because Letby was there. Collapses where she was not on shift were excluded from the chart. Statisticians including Prof. Richard Gill (Leiden, instrumental in the Lucia de Berk exoneration) and the Royal Statistical Society have characterised this as a textbook 'Texas sharpshooter' fallacy: painting the target around the bullet hole. triedbystats.com models the chart in detail — base rates of shift attendance alone mean that whoever works the most unsociable shifts will end up plotted against an unusual cluster of deteriorations, without any wrongdoing. When other nurses' attendance is plotted against the full set of collapses (including those excluded from the trial chart), the pattern dissolves.
"The chart shown to the jury is statistically meaningless. You cannot select the events because Letby was there and then use her presence as evidence of guilt." — Prof. Richard Gill
Source: Royal Statistical Society; Prof. Richard Gill open letters (2024); triedbystats.com visual analysis
The Post-it notes — stress diary, not confession
Prosecution claim
Notes found at Letby's home — including the phrases 'I am evil I did this' and 'I killed them on purpose because I'm not good enough to care for them' — were presented as self-incriminating confession.
Counter-evidence
The same scraps of paper include statements such as 'I haven't done anything wrong' and 'WHY ME?'. Psychologists who have reviewed the full set describe them as stress-diary entries typical of a nurse under accusation — oscillating between self-blame and protest of innocence. A true forensic confession would identify a method, a victim and a motive; none of the notes does. Clinical handover sheets kept at home are standard practice for many British neonatal nurses who use them for CPD and reflective logs.
Notes of the kind shown at trial are consistent with acute occupational stress and intrusive self-blame — not with a forensic confession.
Source: Mark McDonald KC submissions; independent psychology commentary; science4justice.nl
Unit conditions — a neonatal unit beyond its safe envelope
Prosecution claim
The rise in mortality on the unit was presented as attributable, in substance, to deliberate acts by one individual.
Counter-evidence
The Countess of Chester neonatal unit in 2015–2016 was a Level 2 unit caring for infants whose acuity often warranted Level 3 tertiary care. The Thirlwall Inquiry has heard evidence of chronic understaffing, a sewage-back-up incident, pharmacy errors, and patterns of late transfers of extremely preterm babies who arguably should never have been on the unit. The Panel concluded that in every case it reviewed, deterioration was fully explicable by natural causes or sub-optimal clinical care — not deliberate harm. Independent epidemiological analysis shows the mortality rise is consistent with a unit caring for babies sicker than its design.
In every case we reviewed, deterioration was explicable by natural causes or identifiable clinical error — not deliberate harm.
Source: Shoo Lee Panel Report; Thirlwall Inquiry evidence on unit capacity; science4justice.nl
Dr Dewi Evans — the prosecution's lead expert
Prosecution claim
Dr Dewi Evans provided causation opinions on most counts. The Crown presented him as the neonatal expert whose reading of skin signs, radiological findings and clinical patterns established the mechanism of harm.
Counter-evidence
Dr Evans had not worked in routine neonatal intensive care for over a decade at the time of trial. He reportedly approached Cheshire Police offering his services before being instructed. In 2023 a separate family-court judgment described an unrelated Evans expert report as 'worthless' for its methodology. The Shoo Lee Panel and multiple other practising neonatologists have concluded that his methods — particularly for air embolism — fall well below the standard expected in modern neonatology.
It is our view that the methods used to infer cause of death in this case fall well below the standard expected in modern neonatology.
Source: Shoo Lee Panel Report 2025; Private Eye investigations; Court of Appeal (2024); family-court judgment cited in Private Eye
Child K — a dislodged breathing tube in a 25-week infant
Prosecution claim
At the Child K retrial (July 2024), Dr Ravi Jayaram testified that he walked into the nursery to find Letby standing over the infant, whose endotracheal tube had become dislodged and whose oxygen-saturation alarm had been silent. The jury convicted on this single count.
Counter-evidence
The Panel reviewed Child K's medical notes and Jayaram's contemporaneous 2016 records. In extremely preterm infants (25 weeks), spontaneous ET-tube dislodgement is a frequent and expected event; UK neonatal guidance specifically warns clinicians to assume the tube has moved whenever such a baby deteriorates. The Panel found no objective evidence of interference. Analysts comparing Jayaram's 2016 notes with his 2024 testimony have highlighted material differences in his account of where he was, what he saw and when the alarm was sounding.
A dislodged endotracheal tube in a 25-week infant is a routine event, not evidence of wrongdoing.
Source: Shoo Lee Panel Report; Thirlwall Inquiry transcripts of Jayaram evidence; lucyletbyinnocence.com Child K archive
No CCTV, no fingerprints — conviction by inference only
Prosecution claim
Letby was the 'constant presence' at each collapse; the jury was invited to infer the physical act of harm from this pattern of presence.
Counter-evidence
There was no CCTV on the neonatal unit. No fingerprint or DNA evidence was recovered from any syringe, feeding bag or item of equipment implicated in the alleged attacks. No colleague, parent, or visiting clinician witnessed a single physical act of harm in any of the indicted cases. Every conviction rests on inference from patterns of medical deterioration and shift attendance. By contrast, the Panel's medical review concluded the deteriorations themselves were explicable without deliberate harm.
In every case, the alleged physical act is inferred from outcome — never directly witnessed, never forensically documented.
Source: Trial transcripts (lucyletbyinnocence.com archive); Shoo Lee Panel Report; defence closing submissions
The 'Facebook searches' — routine nurse behaviour, reframed
Prosecution claim
The prosecution showed the jury that Letby had searched parents' names on Facebook after some deaths, presenting the pattern as evidence of a morbid or predatory interest.
Counter-evidence
Searching family social media after a serious ward event is common among nurses — it is how many trainees and senior nurses contextualise grief, check for safeguarding concerns, or verify names. Defence analysis showed the searches were spread across many more patients than those charged, and were not unusually time-clustered around deaths. Internal NHS guidance does not prohibit such searches. Without that base-rate context, the presentation at trial was selection bias.
The Facebook pattern cited at trial is consistent with ordinary grieving-nurse behaviour, not with predation.
Source: Defence closing submissions; Mark McDonald KC post-trial commentary; science4justice.nl
Datix records — the system that saw a struggling unit
Prosecution claim
The prosecution argued it did not rely on the Datix incident record because, it said, Letby covered her tracks.
Counter-evidence
Datix is the mandatory NHS incident-reporting system — every crash call, equipment failure, deterioration and medication error is supposed to be logged. The 2015–2016 Datix record for the unit — partly examined at the Thirlwall Inquiry — shows a unit under severe clinical strain: staffing gaps, sewage and plumbing failures, late transfers of extremely preterm infants, pharmacy mix-ups. The Panel argues that this picture, which the jury never saw in full, is itself exculpatory: it explains the cluster of deteriorations without a deliberate-harm hypothesis.
The Datix record is evidence not of wrongdoing by one nurse, but of a unit operating outside its safe envelope.
Source: Thirlwall Inquiry evidence bundles; NHS-reviewer submissions; science4justice.nl
Child G — deterioration expected at 23 weeks
Prosecution claim
Letby was convicted on two counts of attempted murder of Child G, whom the Crown alleged she had over-fed with excessive milk via nasogastric tube, causing aspiration and collapse.
Counter-evidence
Child G was born at approximately 23 weeks — at the absolute edge of viability. The Panel and independent paediatricians note that infants of this gestation commonly suffer severe deteriorations including aspiration, intraventricular haemorrhage, and necrotising enterocolitis without any deliberate act. The volumes of feed the jury was told were abnormal fall within commonly observed ranges for the clinical context. The severe long-term disability that Child G suffered is itself the expected outcome of a 23-week gestation complicated by these events.
Serious deterioration in a 23-week infant is entirely expected. No independent evidence established a deliberate act.
Source: Shoo Lee Panel Report; Panel-member commentary