Lucy Letby’s defence team claims that it has definitive proof of her innocence. Having analysed the available medical records, a group of 14 neonatologists reported that Letby had not harmed any of the babies whose deaths she was convicted of. “If you are looking for the truth you don’t need to go any further,” said Dr Shoo Lee, the leader of the panel.
The problem, however, is that Letby was not convicted solely on the basis of medical evidence. The prosecution did indeed go further to convince the jury.
“Medical or scientific evidence in a case should never be compartmentalised or examined in isolation from the wider canvas,” warned Rachel Langdale KC during the Thirlwall Inquiry, which was set up to examine the circumstances around Letby’s crimes.
Uncertainties were an acknowledged part of the case against Letby. As Nick Johnson KC, for the prosecution, noted during the trial “in this extraordinary case, context is everything”. In other words, there was no one piece of evidence that proved Letby to be a killer beyond all reasonable doubt.
Here we lay out the evidence that convinced two juries of Letby’s guilt — during the original trial and at a retrial of one of the charges.
‘He was completely stable’
Opening the case for the prosecution at the first trial, which began in October 2022, Johnson told the jury that before January 2015, the mortality rate at the Countess of Chester Hospital’s neonatal unit was two to three deaths a year, which was in line with other units around the country. However, over the next 18 months, that figure would rise significantly — to eight deaths in 2015 and seven in 2016.
It was not only the increased mortality rate that began to trouble consultants on the unit; it was also the way in which babies were collapsing. Newborns who were stable would suddenly and unexpectedly deteriorate. While some failed to respond to resuscitation efforts, others would recover. In all of these cases, medical norms and the expectations of the treating doctors were confounded.
“I couldn’t explain why this sequence of events had happened in the first place, and I couldn’t explain why the physiological responses didn’t happen as they should have done,” said Dr David Harkness, a registrar on the unit, giving evidence at the trial about the death of Baby A, the first of Letby’s victims. “Until the point that this happened he was completely stable.”
The bewilderment that Harkness described was to be repeatedly recalled by numerous doctors and nurses throughout the course of the trial as they gave evidence on the collapses and deaths of several different babies. The infants cared for at the Countess’s neonatal unit were vulnerable and premature but they were not expected to die.
Baby O and Baby P were two of triplet siblings born at 33 weeks gestation. This is roughly a month before babies are normally considered full term. A baby of 33 weeks’ gestation would normally have higher than a 99 per cent chance of survival at birth, Dr Stephen Brearey, a consultant paediatrician, told the trial. As a level two unit, the hospital normally only cared for babies older than 27 weeks’ gestation.
Dr Stephen Brearey was the lead clinician in the Chester hospital’s neonatal unit and the first to alert a hospital executive to the fact that Letby was present at unusual deaths and collapses of babies in June 2015
BBC/PA
All three of the triplets were born in good condition and were said to have been doing “very well”. But babies O and P died within days of their birth following rapid deteriorations. Both had been under the care of Letby, who was found to have killed them through a combination of blunt trauma and air injection. They would be her final murders.
Common denominator
After three babies died within the space of three weeks in June 2015, consultant paediatricians began searching for a cause. At a meeting the following month, Letby’s name was first raised because of the fact that she had been present at all three. Brearey, who attended the meeting with Eirian Powell, the ward manager, recalls eschewing the idea of any causal link, saying words to the effect of: “Not Lucy, not nice Lucy.”
But as the unexplained deaths and collapses continued, with Letby present at every one, consultants could not help but recognise that she was the “common denominator”, Johnson told the jury.
At the start of the trial, the jury was shown a chart listing on one axis 25 suspicious deaths and collapses between June 2015 and June 2016 and on the other the names of the 38 nurses who had worked on the unit. Every other nurse had a handful of crosses showing that they were on duty during incidents, but Letby had an unbroken row of crosses beside her name, putting her at the scene for every death and collapse. The nurse present at the second-highest number of collapses had been at seven.
This piece of evidence has come under criticism from statisticians, who note that the table did not include six other deaths during that period for which Letby was not charged. It has therefore been likened to the statistical illusion known as the “Texas Sharpshooter Fallacy”, an analogy of repeatedly firing a gun at a barn and drawing a ring around the densest group of holes to make it look, misleadingly, as if a target was hit.
However, the table only included cases that were deemed suspicious by the expert witness engaged by the police to look at the case, who was not aware at that point that Letby was a suspect.
Lucy Letby was removed from duties on the ward from June 30, 2016, as a precaution — and would never return to the same role
CHESHIRE CONSTABULARY/PA
These were also suspicions that were developing in real time. The staffing table was not a piece of evidence used retrospectively to identify Letby as a potential killer. As the meeting of July 2015 shows, clinicians were alert to the connection between Letby and the unusual collapses from the very beginning.
The jury also heard how collapses had at first happened at night, then began to occur during the day when Letby was moved to cover day shifts following growing concern among clinicians.
After July 2016, when Letby was removed from the unit under a cloud of suspicion and placed on clerical duties, the unexpected collapses immediately stopped and the neonatal mortality rate at the hospital returned to normal levels.
Air injection
It was the prosecution’s case that every child murdered by Letby had been killed by an injection of air, either into their bloodstream through an IV line or into their stomach through a nasogastric tube.
This was a theory developed by Dr Dewi Evans, the lead expert witness called by the prosecution, largely drawn from descriptions given by nurses and doctors of the unusual discoloration they saw on many of the babies’ skin shortly after collapse. His opinion was supported by Dr Sandie Bohin, the second expert medical witness.
Because it is so rare, research into the effects on neonates of air embolism, as it is known, is scant. In reaching his conclusion, Evans referred to a paper written in 1989 by Dr Shoo Lee, a Canadian neonatologist, that detailed 53 cases of air embolism in newborn babies, some of whom had also shown signs of skin discoloration.
Lee, who only became aware that his research had been used in this way after the trial, has now said that it was misinterpreted. He and the other 14 neonatologists say that none of the babies died of air embolism, instead dying of conditions such as sepsis and thrombosis.
Dr Shoo Lee
JOSHUA BRATT FOR THE SUNDAY TIMES
The radiologist who examined X-rays of the babies agreed with the diagnosis of air embolism after finding large amounts of gas in the great vessels that could not be explained by other conditions such as sepsis or trauma.
The inquiry heard that bosses at the Countess of Chester Hospital were “truly sorry” for not contacting police earlier
ALAMY
Dr Andreas Marnerides, an expert in neonatal pathology who was asked by Cheshire police to review the case in 2017, also thought it likely that a number of babies had died as a result of air being injected into their bloodstreams. He told the court he had found “globules” in the lungs and brain tissue that were most likely to have been air, as well as air around the tip of one of the baby’s UVC catheters.
Evans’s theory was challenged at the Court of Appeal hearing this year when Lee himself appeared before the court and said that none of the descriptions of the babies’ skin discolorations given by witnesses matched the sort he had recorded in his paper. He said the only sign visible on a baby’s skin from which one could draw a conclusion of an air embolism was bright-pink blood vessels over blue skin.
But the judges did not admit his testimony, deeming it irrelevant. In their 58-page judgment, they stated that there was no prosecution expert evidence diagnosing air embolism solely on the basis of skin discolouration. “In short, the circumstantial evidence and medical evidence has to be considered in its totality, not reduced to a single issue as to skin discoloration as a basis for diagnosis,” they said.
Circumstantial evidence
Letby was never caught red-handed. No one ever saw her harm a baby. But there were occasions, it was said, when she very nearly was.
In August 2015, Letby was assigned to care for Baby E, a boy, and his twin brother, Baby F, who had been born ten weeks premature. The prosecution said the twins’ mother interrupted Letby attacking the baby when she decided to visit her sons in the unit shortly after 9pm.
Former bosses at the Countess of Chester Hospital have apologised to the families of victims at the inquiry
CHESTER HOSPITAL
She came in to find the baby distressed and bleeding from his mouth, an unusual and unexpected complication. Letby, who was the only nurse in the room, turned to her and said the words: “Trust me, I’m a nurse.”
Baby E went on to suffer significant blood loss later in the evening and died in the early hours of the following day. The fatal bleed was believed to be the result of Letby interfering with his nasogastric tube. Having so nearly been caught, Letby then attempted to cover her tracks by erasing the mother’s visit from the medical records and afterwards falsely claimed to have been in another room when Baby E collapsed, the court was told. This, the prosecution said, was the nurse trying to establish an “alibi in someone else’s medical records”.
A key part of the circumstantial evidence against Letby was discovered when police searched her home in Chester following her arrest in July 2018. Here they found a handwritten note on which Letby had written the words, “I am evil, I did this,” along with protestations of innocence, including: “I haven’t done anything wrong.”
They also found more than 250 nursing handover sheets, a blood test result and resuscitation notes relating to some of her victims. Johnson called these her “trophies”.
Insulin
Letby was found guilty of attempting to murder two babies by injecting synthetic insulin into their feed bags. Both babies survived.
The test results showing that insulin had been given to them was the only piece of concrete evidence of criminality throughout the whole of the prosecution’s case. It was the closest thing they had to a smoking gun and became the keystone for their case as a whole.
Johnson’s argument ran that if the jury could agree that Letby had deliberately poisoned two babies, they could also reasonably conclude that she had harmed others using different methods, even if the evidence for those were less concrete. In the event, the insulin cases were the first on which the jury reached a verdict.
The 14 neonatologists have challenged this, saying that the method used to test the presence of insulin is not reliable. Letby did not contest the fact that someone had poisoned those babies with insulin, only that it was her who did it.
Lady Justice Thirlwall has warned that speculation over Letby’s convictions is causing the victims’ families enormous stress
PETER BYRNE/PA
“Positive immunoassay results are not sufficient as binding toxicological evidence of foul play in a criminal prosecution for murder,” Professor Alan Wayne Jones, an expert in forensic chemistry, told The Times last month.
• What could Thirlwall inquiry mean for Lucy Letby’s convictions?
Keith Frayn, an emeritus professor of human metabolism at the University of Oxford who has been using immunoassay for insulin since the 1970s, rejected the notion that the tests were unreliable. “I don’t think many people who know about insulin assays would say you can disregard those tests,” he said. “They are very clear.”
He acknowledged there was a small margin for error but said the Letby case results were far outside that, given the two babies had insulin between ten and 40 times the normal level. Crucially, C-peptide was undetectable in one baby and very low in the other. Given that natural insulin produces with it higher levels of C-peptide, the only explanation was insulin introduced from outside. “There are most unlikely to be analytical errors,” Frayn said.
He agreed with other experts that follow-up insulin assays to confirm results would have been desirable but insisted the insulin levels in both cases were so far outside the margin of error that it was unlikely to have made a difference.
