https://www.ft.com/content/6797b800-3fe8-4144-b63d-435ae52ba157

September 7 2024

Doctors and scientists raise concerns over evidence used to convict neonatal nurse of infant murders ahead of inquiry next week

Laura Hughes and Michael Peel in London

The public inquiry into infant deaths at a hospital in England begins next week amid a growing debate around the scientific evidence used to convict former neonatal nurse Lucy Letby of their murder.

Letby, found to be one of Britain’s deadliest child serial killers, was sentenced in August last year to life imprisonment for murdering seven babies and attempting to end the lives of six more between 2015 and 2016. She was convicted of a further attempted murder in a recent retrial. 
Last year, the UK government announced an independent statutory inquiry into the circumstances surrounding her crimes at the public Countess of Chester hospital.

Letby has a new legal team and is planning to apply for another appeal at the Criminal Cases Review Commission, the body responsible for investigating potential miscarriages of justice, according to her lawyer.
The inquiry begins on Tuesday against the backdrop of concerns from expert commentators, statisticians and medical practitioners over Letby’s conviction, highlighting long-standing questions over the use and presentation of scientific data in her trial.

Many specialists have stressed they are not arguing that Letby is innocent, but rather that the evidence against her may not have reached the criminal standard of “beyond reasonable doubt”. They also pointed out that they had not seen all the evidence and proceedings viewed by the jury. 
The inquiry, led by Lady Justice Thirlwall, aims to determine whether suspicions over the deaths should have been raised earlier by hospital staff, and how NHS managers responded to clinicians’ concerns months before Letby’s arrest. 

The proceedings will also look at whether the police or other external bodies should have been informed of events in the ward sooner. 
In July, a group of neonatal doctors, including nurses and statisticians wrote to ministers urging them to postpone the inquiry or change its terms of reference, expressing concerns over how medical and other data had been used in court.

Stressing they did not want to relitigate the case, the signatories said they were focused on “patient safety, healthcare management, and the potential for miscarriages of justice in complex medical cases”.

“We believe that legal systems are particularly vulnerable to errors when dealing with intricate scientific evidence, especially in cases involving statistical anomalies in healthcare settings, and the complex physiology of neonates,” they wrote.

A spokesperson for the inquiry said it would “follow the terms of reference set by the secretary of state”.

Experts are often particularly important in cases of alleged medical misconduct because of the range of evidence involved in these hearings — and the risk it can be misunderstood.

“Statistical evidence, however misleading, can often seem convincing,” said Philip Dawid, emeritus professor of statistics at Cambridge university. “It’s quite hard to bring out the subtleties and show that apparently obvious interpretations are not to be trusted.”
Peter Skelton KC, a barrister representing some of the bereaved families at the inquiry, has accused conspiracy theorists of spreading “grossly offensive” claims which are deeply “distressing for the families of her victims”.

The crown court judge who handed Letby her life sentence described her as the architect of a “cruel, calculating and cynical campaign of child murder”. Two attempts to appeal her conviction have so far been dismissed by three appeal court judges.

A fundamental problem with the Letby case is that arguments now being raised outside the courtroom were never heard inside, as the defence declined to call any expert witnesses.

Statisticians have zoned in on the validity of a roster used by the prosecution to show Letby was on duty when the incidents in question took place.

Some experts said it looked like a case of the “Texas Sharpshooter Fallacy” — named after the analogy of repeatedly firing a gun at a barn and drawing a ring around the densest group of holes to make it look, misleadingly, like a target was hit.

The roster’s central flaw is seen as being its selective use of data and exclusion of other serious incidents, including deaths, which happened when Letby was not present.

Potential alternative explanations for her presence at the incidents were ignored, they argued, including other plausible reasons for rising death rates on the unit and the possibility that she was trusted to handle the hardest cases.

The impression created by the roster that all the suspicious incidents could be attributed to Letby was crucial to how the proceedings unfolded, said Peter Green, an emeritus professor of statistics at Bristol university
It had “opened the door to confirmation bias creeping in,” said Green who, as former president of the Royal Statistical Society, was involved in overturning the wrongful conviction of Sally Clark’s in 2003 for murdering her children.

“It may have meant the other evidence wasn’t treated with the scepticism it should have been,” he added.

Jane Hutton, a statistics professor at Warwick university, suggested that unusual clusters of deaths in medical contexts could be referred to the Health Security Agency, the public health body responsible for protection against infectious diseases, for initial analysis by its epidemiologists.
This would give a measured view on the statistical significance of events before the police are involved, she added.

“If you get a healthcare cluster of deaths, you should not assume murder,” Hutton said. “Because when you go to the police you have the presumption there was a crime.”

Another contentious point is that Letby was found guilty of poisoning babies with insulin and causing embolisms after injecting air into their bloodstreams. 

Dr Michael Hall, a visiting professor in neonatal medicine, was an expert witness appointed by her defence but was not called to give evidence during the trial.

Speaking in a personal capacity, Hall said he did not believe Letby had been given a “fair trial” because the jury did not hear from medical expert witnesses who would challenge the prosecution’s arguments.
“How can a jury determine how credible the prosecution case is if they have nothing to compare it with?” he added.

Hall said he would have raised concerns over the prosecution’s claim that skin discolouration found in some of the babies was evidence they had been injected with air.

He added that the argument was primarily based on academic research on the consequences of oxygen entering babies’ bodies via the lung, when it was alleged that some victims had air injected into a peripheral vein.
Air, largely composed of nitrogen, is processed differently by the body from pure oxygen.

“I do think the relationship between the courts and medicine is an uncomfortable one because, as far as I know, judges are not trained in evidence-based medicine,” Hall noted, adding that the interface between medicine and the law “bears further scrutiny”.

Letby’s solicitors in her initial trial declined to comment, citing legal privilege.

Notes written by Letby, used as evidence of her guilt, have also been queried. The Guardian newspaper reported this week that these were made on the advice of hospital mental health counsellors as a way of coping with stress. 

Letby had scribbled: “I am evil I did this” and “I killed them on purpose because I am not good enough to care for them and I am a horrible evil person.” She also wrote: “I haven’t done anything wrong” and “I feel very alone + scared”. 

An explanation as to why these notes were written was never cited by her defence team during the original trial.

A government spokesperson said: “This was a horrendous case and there were clear failings across the NHS and with regulators”. The inquiry would “review all aspects of the case”, they added.