The Royal Statistical Society (RSS) is calling for the Lucy Letby inquiry to cover statistical evidence used in the trial of the serial baby killer, saying it can be difficult “to draw conclusions from suspicious clusters of deaths” in hospitals.
In August, Letby was found guilty of murdering seven babies and attempting to kill six more at the Countess of Chester hospital, making her the worst child serial killer in modern British history.
After the trial, a public inquiry was announced in a bid to unpick the circumstances around the crimes and provide answers for the victims’ families. It was later announced the inquiry would become statutory, meaning it has the legal powers to compel witnesses to give evidence.
Now experts have argued the inquiry should not only cover medical evidence, but statistical evidence too.
In a letter to Lady Justice Thirwall, who is leading the public inquiry, Dr Andrew Garrett, the president of the RSS, and Sarah Cumbers, its chief executive, stress the importance of evidence based on statistics and data, and suggest a better understanding of the area could help NHS trusts act more quickly in similar cases. But the letter also notes there are potential pitfalls to avoid.
“It is far from straightforward to draw conclusions from suspicious clusters of deaths in a hospital setting – it is a statistical challenge to distinguish event clusters that arise from criminal acts from those that arise coincidentally from other factors, even if the data in question was collected with rigour,” the authors say.
The RSS sent its 2022 report Healthcare Serial Killer Or Coincidence? to both the defence and prosecution at the start of Letby’s trial. It sets out statistical issues in the investigation of suspected medical misconduct.
The letter by Garrett and Cumbers, however, suggests the judicial system could benefit from more guidance on the matter, calling for the inquiry to consider including a point in the terms of reference on the appropriate use of statistical evidence in this type of case.
“Statistical evidence is one type of evidence that NHS trusts might use to identify criminal activity and it is important that the right lessons are learned and that it is used appropriately,” they write.
Prof David Spiegelhalter, of the University of Cambridge, welcomed the letter from the RSS. “Judging whether something is too surprising to be ‘just a coincidence’ should not be a matter for human intuition – expert statistical analysis is required,” he said.
Richard Gill, an emeritus professor of statistics at Leiden University and a co-author of the RSS report in 2022, said the investigation made mistakes in handling statistical evidence, and that neither the prosecution nor the defence made use of the report’s recommendations.
Gill said that a meaningful statistical analysis was not carried out to explore whether or not there was a higher rate of incidents during Letby’s shifts than outside them, while controlling for relevant factors.
Gill also raised concerns about a lack of adherence to the rules put down by the lord chief justice for presentation of scientific evidence, adding that he supported a retrial.
Letby is seeking to challenge her convictions at the court of appeal.
