Lucy Letby's Alleged Methods: What Was Proved, Explained

Lucy Letby faced 22 charges relating to 17 babies, built around five alleged methods of harm: injecting air into the bloodstream, introducing air into the stomach via a feeding tube, poisoning with insulin, deliberate overfeeding with milk, and physical trauma. A jury convicted her on 14 of those counts in August 2023, and on a fifteenth at a retrial in July 2024.

It's worth being precise about what "proved" means here. A criminal conviction means a jury was satisfied beyond reasonable doubt, based on the evidence put before it — not that a scientific fact has been independently and permanently established. That distinction matters more in this case than most, because since 2025 the specific medical reasoning behind several of the convictions has been publicly challenged by an international panel of neonatal specialists, including the very researcher whose work the prosecution relied on for its central theory.

Five Alleged Methods

According to the official case summary later set out in Court of Appeal judgments, the prosecution's alleged methods included: air embolus, affecting Child A, Child B, Child D and others; air introduced via nasogastric tube, affecting Child C; insulin poisoning via intravenous fluid bags, affecting Child F and Child L; deliberate overfeeding with milk, affecting Child G; and physical trauma, affecting Child E and one of a set of triplets, Child O. Prosecutor Nick Johnson KC told the jury the seventeen collapses and deaths under consideration were not "naturally occurring tragedies" but "the work… of the woman in the dock."

Air Embolism: The Central Allegation

The most frequently alleged method, and the one underpinning most of the murder convictions, was air embolism — the introduction of air directly into a baby's bloodstream, argued to be capable of fatally obstructing circulation. Prosecutors pointed to Child A, Child B and Child D among others, and the trial heard evidence that Letby had researched the effects of air embolism online before some of the collapses occurred.

Air embolism as a genuine clinical phenomenon in neonatal medicine is well established. What has always been harder to establish is how confidently a specific collapse can be attributed to deliberately introduced air after the fact, rather than to one of the many other causes of sudden deterioration common among extremely premature infants.

The Skin Discolouration Evidence

Because air embolism leaves little conventional forensic trace, much of the supporting evidence at trial was clinical observation: several treating doctors and nurses described unusual patchy skin discolouration or mottling in babies shortly before they collapsed, consistently and — by most accounts — honestly reported by staff who found it genuinely puzzling at the time. That gap, between the genuine puzzlement of treating clinicians in the moment and the retrospective medical interpretation offered years later at trial, is central to the dispute that followed.

Dr Shoo Lee's 1989 Paper — and His 2025 Reversal

The prosecution's key expert witness, retired consultant paediatrician Dr Dewi Evans, supported the air embolism diagnoses partly by reference to a 1989 academic paper co-authored by Dr Shoo Lee, a Canadian neonatologist, describing specific skin signs associated with air embolism in infants — a pink-red pattern against a purplish-blue background, since termed the "Lee sign," and pale patches on the tongue, the "Liebermeister sign."

In February 2025, Dr Lee held a press conference in London alongside a panel of fourteen international neonatologists and paediatricians, including Professor Neena Modi, a former president of the Royal College of Paediatrics and Child Health. Dr Lee said the trial evidence had misrepresented his research, that neither of the specific signs described in his own paper had actually been present in the cases reviewed, and that a follow-up paper he published in 2024 found no link between the general skin discolouration described at Letby's trial and venous air embolism at all. "The notion that these babies can be diagnosed with air embolism because they collapsed and had these skin discolourations has no evidence in fact," he told the assembled press.

Letby's legal team had already tried to introduce Dr Lee's evidence at her first appeal. Three senior Court of Appeal judges declined to consider it, ruling that the prosecution had not, at trial, relied solely on skin discolouration to support the embolism diagnoses — a narrow procedural finding about what the original trial record showed, rather than a full assessment of whether Dr Lee's underlying scientific criticism is correct. That distinction is why the argument remains alive in the material now before the Criminal Cases Review Commission.

Insulin Poisoning: Child F and Child L

Two attempted murder charges concerned insulin: Child F in August 2015 and Child L in April 2016, both of whom survived. The Crown's case was that intravenous fluid bags administered to each baby had been deliberately adulterated with insulin. This was supported by blood tests showing the ratio between insulin and C-peptide — a substance the body produces alongside its own naturally occurring insulin, but not alongside insulin from an external source — indicating the insulin found had been administered rather than produced naturally.

This test result has become one of the more technically contested elements of the case since conviction. Critics, including some of the experts convened by Dr Lee, have questioned whether the immunoassay method used was validated for proving deliberate administration in a forensic rather than purely clinical context. Immunoassay testing of this kind is well established for clinical decision-making at the bedside; using the same test result years later to support a criminal allegation beyond reasonable doubt is a different evidential bar, and one defence experts have argued the original trial didn't sufficiently interrogate.

Overfeeding and Trauma

A separate method alleged deliberate overfeeding, causing Child G to vomit, collapse and suffer permanent brain injury during a resulting cardiac arrest in September 2015 — a charge that relied on nursing records and witness accounts of milk volumes rather than a specific forensic marker. The most severe alleged method involved physical trauma: Child E, who died in August 2015, was alleged to have suffered acute internal bleeding, and one of a set of triplet brothers, Child O, was found at post-mortem to have a ruptured liver.

What the Jury Actually Convicted Her Of

Of the 22 original charges, the jury convicted Letby on 14 counts in August 2023 — seven murders and seven attempted murders — acquitted her on two, and couldn't reach a verdict on six. One of those, relating to Child K, was retried in 2024 and resulted in a fifteenth conviction. The other five were not pursued further. This mixed verdict is itself informative: the same jury, hearing the same general evidence about Letby's conduct and character, was not persuaded beyond reasonable doubt on every count put to it.

Where the Methodological Challenge Stands

As of August 2026, none of this technical dispute has resulted in a referral back to the Court of Appeal. The Criminal Cases Review Commission's review remains active, having received multiple rounds of expert reports and legal submissions on exactly these questions well into 2026. Whether that material meets the legal threshold the CCRC must apply — a "real possibility" that the Court of Appeal would now find the convictions unsafe — is a decision that remains pending. A CCRC referral, if one comes, sends a case back to the Court of Appeal for reconsideration; it doesn't itself overturn a conviction, and the Court of Appeal could, in principle, hear the fresh medical evidence and still uphold the original verdicts. Both outcomes remain genuinely possible.

Read the full timeline → · Read about the trial and sentencing → · Browse DOUBT: The Case of Lucy Letby →