Why it matters in the Letby case
Gill — central to the de Berk acquittal — has publicly engaged with the Letby case, arguing the statistical reasoning shows structural similarities.
Statistical explainer
The Lucia de Berk statistics lesson
Lucia de Berk was a Dutch nurse wrongly convicted of multiple homicides on the basis of shift-pattern statistical evidence. She was acquitted in 2010 after sustained methodological critique led by Prof. Richard Gill.
Legal explanation2Expert/professional
Last updated
Prosecution position
As in de Berk, the case connected a nurse to a cluster of adverse events through her presence on shift.
Expert challenge / post-conviction reading
Lucia de Berk was convicted in the Netherlands on shift-pattern statistical evidence and acquitted in 2010 after sustained methodological critique led by Prof. Richard Gill. Gill has since engaged publicly with the Letby case and argues the statistical reasoning shows structural similarities.
What remains uncertain
De Berk's acquittal followed a full judicial re-examination in a different legal system. That a critique succeeded there says nothing about how the CCRC or the Court of Appeal will treat this one.
Common questions
Who is Lucia de Berk?
A Dutch paediatric nurse convicted of multiple murders and attempted murders, largely on statistical evidence about her presence at adverse events. She was acquitted in 2010.
What overturned it?
Sustained methodological critique of the statistical reasoning, led by Prof. Richard Gill, which showed the original calculation was unsound.
What is the structural similarity Gill identifies?
A cluster of events, a nurse present at them, and a statistical argument built on that presence without an independently defined event set or a comparator.
Is a Dutch acquittal relevant to an English conviction?
Not as precedent. It is relevant as a documented instance of the same class of statistical reasoning being found unsound on re-examination.
Related reading
Found an error or missing source? Send a correction with the page URL and supporting source.